-
https://clinton.presidentiallibraries.us/files/original/721f89a11a777497464366cbb95141d7.pdf
36230e944f7a6841c318dc564502b0f0
PDF Text
Text
FOIA Number: 2006-0810-F
FOIA
MARKER
This is not a textual record. This is used as an
administrative marker by the William J. Clinton
Presidential Library Staff.
Collection/Record Group:
Clinton Presidential Records
Subgroup/Office of Origin:
Health Care Task Force
Series/Staff Member:
Subseries:
OA/ID Number:
1235
FolderlD:
Folder Title:
Population Based Prevention and Public Health [8]
Stack:
Row:
Section:
Shelf:
Position:
S
51
6
7
1
�Clinton Presidential Records
Digital Records Marker
This is not a presidential record. This is used as an administrative
marker by the William J. Clinton Presidential Library Staff.
This marker identifies the place of a tabbed divider. Given our
digitization capabilities, we are sometimes unable to adequately
scan such dividers. The title from the original document is
indicated below.
Divider Title:
8
�FOR OFFICIAL USE ONLY
H.
OCCUPATIONAL SAFETY AND HEALTH
Control o f Asbestos Exposure During Brake Drum Service. John
W Sheehy, e t a l .
.
Case Studv: Control o f Methylene Chloride Exposures During
Commercial F u r n i t u r e S t r i p p i n g . Cheryl F a i r f i e l d and Amy
Beasley
�OCCUPATIONAL SAFETY AND HEALTH
�CONTROL OF ASBESTOS EXPOSURE DURING
BRAKE DRUM SERVICE
John W Sheehy, e t a l .
.
Data from the National I n s t i t u t e f o r Occupational Safety and
Health (NIOSH) estimates t h a t a t l e a s t 150,000 brake mechanics
and garage workers i n the U.S. are p o t e n t i a l l y exposed t o
asbestos, a known h e a l t h hazard. This study examined the
asbestos exposure l e v e l s experienced by auto mechanics using
various techniques t o c o n t r o l asbestos brake dust.
Five methods o f reducing exposure t o asbestos were t e s t e d ,
i n c l u d i n g v e n t i l a t i o n and vacuum devises. Each of the methods
c o n t r o l l e d the mechanics exposure d u r i n g brake s e r v i c i n g t o less
than 20 percent of the (NIOSH) standard.
�rev
j V^G.
Control of Asbestos Exposure
During Brake Drum Service
John W. Sheehy, Thomas C. Cooper, and Dennis M. O'Brien
National Institute for Occupational Safety and Heatth, Division of Physical Sciences and Engineering, Engineering Control
Technology Branch, 4676 Columbia Parkway, Cincinnati, Ohio 45226
An estimated 150,000 brake mechanics and garage workers in
the U.S. are potentially exposed co asbestos, a known health
hazard. Earlier studies of airborne asbestos exposures to mechanics during brake maintenance operations showed overexposure to asbestos fibers during brake servicing, especially brake
assembly cleaning. Most brake service operations are performed
by small businesses that lack resources to evaluate control devices.
The resulu from the evaluation of five control methods for
containing asbestos brake dust during brake maintenance are
presented. These controls included rwo commercial enclosure
devices with ventilation provided by high-efficiency particulate
air (HEPA)filter-equippedvacuums, a HEPAfilter-equippedvacuum cleaner without an enclosure, a wet brush/recycle system
which recirculated the cleaning solution, and an aerosol spray
for wetting the brake assembly. Detailed surveys of the controls
were conducted at five separate facilities during actual brake
servicing operations to automobiles, pickup trucks, vans, and
vehicles with a four-wheel rear axle. Personal and area air samples
collected during brake repair to each vehicle were analyzed by
phase contrast microscopy (PCM) and transmission electron microscopy (TEM).
Personal sampling results for the brake mechanics show that
concentrations determined by PCM analysis ranged from less chan
0.004 f/cc to 0.016 (/cc. AU exposures were below the National
Institute for Occupational Safer) and Health (NIOSH) recommended limit of 0.1 f/cc. Analysis by TEM indicated not only the
presence of asbestos fibers not detected by PCM but also at levels
well below 0.1 f/cc. The highest exposures measured were found
for workers servicing heavy duty trucks.
The results from this study indicate that all the devices tested,
in combination with the work practices used, controUed the
mechanics' asbestos exposure during brake servicing to less than
the NIOSH recommended exposure limit and the Occupational
Safety and HeaJth Administration permissible exposure limit.
Sheehy, J.W., Cooper, T.C.; O'Brien, D.M.: Contro* ot Asbestos Exposure During
Brake Drum Service. AppMnd.Hyg. 4:313-319; 1989.
11
the United States are poienually exposed to asbestos.'
Repair facilities from small service sutions to fleet garages
follow similar brake servicing procedures The vehicle is elevated, wheels are removed, and the brakes are inspected Tradiuonally, brake dust was often cleaned from the drums and
brake assemblies by blowing with compressed air or dry wiping'
brushing. ' At the time of this study, however, most brake servicingfacilitieswe observed used wet brushing, wet wiping, squirtbottle wash-off, or high-efficiency paniculate air (HEPA) filtered
vacuum cleaning systems. After cleaning, pans are replaced or
repaired as needed, and the brake sysiem is reassembled and
adjusted.* ^
The brake repairman and other service personnel in the garage
area are potentially exposed to asbestos dust during and following the brake drum removal if the dust buildup inside the drum
and brake assembly is removed and disposed of using appropriate control equipment and work practices, the risk can be
minimized However, small businesses and the many home mechanic that perform brake repair lack resources to perform an
evaluation of these control measures.
The objeaive of this study was to determine asbestos exposure
levels experienced by automobile mechanics using various control techniques during maintenance and replacement of drum
brakes. The study focused on vehicles with brake drum sizes of
•12 inches or less. Repairs of disk brakes were not evaluated in
the study.
12
Description of Control Methods
introduction
Any procedure used to remove and collea the dust during brake
servicing should limit asbestos brake dust emissions and effectively remove the brake dust that adheres to the brake parts, as
well as the loose dust that collects in the drum. In this study, five
methods for controlling exposure to asbestos during brake repair
were evaluated. These included two commercial enclosure devices with ventilauon provided by a HEPAfilter-equippedvacuum, a HEPAfilter-equippedvacuum cleaner with no enclosure,
a wet brush/recycle sysiem with recirculating cleaning solution,
and an aerosol spray to wet the brake assembly
Vehicle brake maintenance procedures provide an arena for exposure to airborne asbestos fibers. From data contained in the
National Occupational Hazard Survey, the National Insutute for
Occupational Safer,' and Health (NIOSH) estimates that a work
force of a least 150.000 brake mechanics and garage workers in
Each control method was evaluated at a differentfacility.This
introduced many uncontrolled variables such as building layout,
traffic panem, and ventilation sysiem. In addition, the types of
vehicles and wheel sizes were noi identical among the control
methods tested.
APPL IND. HYG. VOL 4, NO. 12 • DECEMBER 1389
313
�Vacuum Enclosure Units
TAB Li I. Air Sampling Summary
Two vacuum enclosures were evaluated Vacuum enclosure A
consisted of an adjustable height, dear plastic, two-glove box
with an overlapping neoprene seal; a double motor HEPA filterequipped vacuum unit; and connections inside the enclosure for
an air hose and a vacuum hose. The glove box fit over the brake
drum and backing plate on all vehicles except the large salt truck.
Vacuum enclosure B consisted of a steel cylinder, with a single
glove at one end and an iris-t\pe seal at the other end, conneaed
to a single motor HEPAfilter-equippedvacuum cleaner. A plastic
viewing window was located on one side, and inside the enclosure was an air hose connection. With enclosure B, the drum
must be removed before the enclosure can be placed over the
backing plate.
Vacuum Cleaning
The HEPAfilter-equippedvacuum cleaner is used to vacuum dust
from inside the brake drum and from around the brake assembly
as well as dust that falls io the floor during brake drum removal.
Vacuuming is done after the hubcap, wheel, and drum are removed. Air is drawn into the vacuum at about 50 cubic feet per
minute. No enclosure and no compressed air or wet methods
are used.
Wet brush/recycle
This system, shown in Figure 1, consists of a fluid reservoir, a
pump, a delivers' ss-stem (a low-velocity nozzle or brush), and a
catch basin. An aqueous solution containing an organic solvent
is pumped through a nylon filter, direaed through a flexible
tube, and forced out berween the bristles of a brush. It provides
for a gende flooding of the brake assembly area to wash down
the dust, the brush is used to assist in removal of the dust. The
solution captured in a catch pan is returned and recirculated
from a reservoir.
Aerosol Spray
This control method consists of a solvent (methyl chloroform)
spray to wet and control potential asbestos exposures during
brake maintenance. Typicalh', the operator dispenses the solvent
from a refillable, hand-held sprayer. The sprayer is filled with
approximately one quan of solvent. Shop air is used initially to
pressurize the sprayer. In the brake cleaning procedure, a catch
pan is placed under the brake assembly, and the exposed surfaces
are wened thoroughly. The sprayer is held about 18 inches from
BRAKE DRUM
RINSE K B E
a
Samples
Flow
Rate
(Umin)
Duration
(min)
2*
2*
2*
2*
2.5-3.0
7.0
7.0
30
120
120
140-240
260-480
Number
Sample
Type
Petsonal
Near-source
Background
Ambient
Location
Worter
Fender/axle
Inside garage
Outdoors
Minimum
Volume
(U
300
840
1000
800
'perwhicie
•per day
the brake drum and other components so that brake dust is not
blown off to become airborne before it is wened Washing is
then performed by moving the sprayer to about 12 inches from
the parts. Some mechanics wipe the wened pans and drum with
rags.
Airborne asbestos concentrations were also measured during
brake servicing of a van by a "home mechanic" using a spray can
solvent and garden hose.
Methodology
Air Sampling and Analysis
Two personal air samples for asbestos were colleaed side-byside in the breathing zone of each worker for the duration of a
single brake job, or for 2 hours, whichever was longer (Table 1).
Samples were colleaed on 25-mm diameter, 0.8 (xm pore size,
cellulose ester membrane filters at flow rates of 2.5-3 0 IVmin
using a personal sampling pump. The minimum volume colleaed
(300 L) allowed a limit of deteaion (LOD) of approximately 0.004
fee by phase contrast microscopy (PCM) analysis.
To determine if fibers escaped into the work environment,
two area samples per vehicle (near-source samples) were colleaed; one near the fender and the other under the axle. Each
utilized the above-described media at a flow rate of approximately 7 lymin using rotary vane vacuum pumps for the duration
of a brake job, or 2 hours, whichever was longer.
Two background samples were colleaed in the garage at least
10 feet from brake repair activities, at flow rates of 7-10iymin
for a 4-hour sample period. These background samples were
used to determine the effects of general shop cleanliness and
overall effectiveness of the dust control procedures. Samples
were also colleaed out-of-doors (ambient samples) to determine
environmental concentrauons of fibers. These ambient samples
were colleaed at a flow rate of approximately 3 L/min using
personal sampling pumps for up to 8 hours.
All filtered air samples were analyzed by PCM using NIOSH
Method 7400.'*' In addition to PCM analysis, approximately 80
percent of these samples were analyzed by transmission electron
microscopy (TEM) using NIOSH Method 7402. To facilitate
analysis by PCM and TEM on the same samples, the direa transfer
method of sample preparation ^ was used, modified by the omission of the filter etching. (Six samples, etched and recounted by
TEM, showed no difference in fiber count from the unetched
samples.) TEM analysis was performed to identify' asbestos fibers
and measure concentrations for fibers too small to be deteaed
by optical microscopy analysis.
For PCM analysis, all fibers with a 5:1 (or greater) length to
width aspea ratio were counted using NIOSH Method 7400-B.
(A small number of samples were analyzed using Method
7400-A counting rules because the routine laboratory procedure
for Method 7400 was changed to " A" counting rules before these
m
1
CATCH BASIN
PUW>
BARRIER FILTER
FG R 1.
JUE
314
Wet brush/recycle system.
1
APPL IND. HYG. VOL 4, NO. 12 • DECEMBER 1963
�samples were analyzed) While the Occupational Safety and Health
Administration (OSHA) Permissible Exposure Limit (PEL) and the
NIOSH Recommended Exposure Limit (REL) are expressed in
terms offibershaving a 3:1 (or greater) aspect ratio, the difference in counting rules has little practical significance in the case
of brake dust, since fewfiberswere idenufied in this study with
aspea ratios between 31 and 5:1 (see "Results" section)
For TEM analysis,fiberswere identified by morphology, seleaed area electron diffraction (SAED), and energy-dispersive
X-ray (EDX) analysis. Fibers were classified in one of four categories: chrysotile, amphibole, ambiguous, and no identification.
SAED panems were observed for all fibers.' ' Fibers were also
identified by asbestos structure (fibers, bundles, clusters, and
matrices) and sized by length and width. Allfiberswith a 31 (or
greater) aspea ratio were counted using TEM The analysis was
performed using a magnification of 17,600 x (grid area of
0.000081 cm ) and counung either a minimum of 10 grid openings or 100fibers,whichever camefirst.The limit of resoluuon
was approximately 0.06 fim, thus the minimum fiber length that
could be measured was three times the limit of resolution, or
about 0.2 >im. One or two field blanks were prepared for each
vehicle sampled and submined for PCM and TEM analysis.
A bulk brake dust sample from each vehicle and a senled dust
sample from each repair facility were colleaed and analyzed for
asbestos by TEM. The percentage of asbestos in the bulk samples
were qualitatively established by estimating the ratio of the number of asbestosfibersto dust panicles, and the percentage of
asbestos of the totalfiberspresent was quantified.
6
2
Results
Asbestos Concentrations as Determined by PCM
Table II presents the average and range of personal sample concentrations for airborne asbesiosfibersdetermined by PCM for
the five different control methods used during brake service.
These results include exposures encountered while workers serviced brakes of small- and medium-sized vehicles and two large
vehicles. Area (near-source) samples (PCM) colleaed near the
fender and over the axle of all vehicles were less than 0.002
fee.
Of 83 personal samples colleaed on brake mechanics in our
study, the highest concentration determined by PCM was 0.016
fee. This is about four times the LOD for the personal samples
(0 004 fee) Personal sample concentrations represenl only those
exposures which occurred while servicing brakes (usually 2-3
hours per shift) and not the time-weighted average (TWA) exposure for the entire work shift. Usually, only one brake repair
was performed per day, thus the TWA exposure of the mechanic
would be lower. The arithmetic mean fiber concentration, while
using either of the vacuum enclosures or the wet brush/recycle
TABLE II. Personal Sample Fiber Concentrations by Control
Method (PCM)
Arithmetic
Control Method
Vacuum enclosure A
Vacuum enclosure B
Vacuum only
Wet brush/recycle
Aerosol spray
Water hose and solvent
Number of
Samples
Mean
18
22
13
20
8
2
< 0.004
< 0.004
0.007
< 0 004
0.007
0 007
Range
(t/cc)
fl/cc)
APPL IND. HYG. VOL 4. NO. 1 • D C M E 1989
2
EE BR
<
<
<
<
<
< 0.004
0.004-0.006
0.004-0.016
0.004-0.006
0.003-0.016
0.006-0.008
£111. Personal Sample A
Method (TEM)
Control Method
Vacuum enclosure A
Vacuum enclosure B
Vacuum only
Wet brush/recycle
Aerosol spray
Waler hose and sotvenl
s Concentrations by Control
Number
of
Samples
ArtthmetK
Mean
(t/cc)
16
12
13
10
6
2
0.021
0.044
0.022
< 0.013
0.052
0.039
Range*
(t/cc)
Geom.
Mean
(t/cc)
0.017
0.010-0.065
< 0.013-0.139
0.028
0.019
< 0.011-0.045
< 0.013-< 0.014 < 0.013
0.013-0.079
0.045
0.039
0.039
"Limits ot detection vary win the sample vWune
with recirculating solution, was 0.004 fee. Arithmetic mean exposures while using the vacuum only and the aerosol spray methods were 0.007 fee. The exposure for a "home" mechanic using
a spray can solvent and garden hose during a brake replacement
averaged 0.007 fee using Method 7400 "A" rules.
The OSHA PEL of 0.2 fee* ' (8-hour TWA) and the NIOSH REL
of 0.1 fee' ' for asbesios are based on PCM analysis of asbestos
using "A" counting rules (3:1 aspea ratio). "B" counting rules
(5:1 aspearatio)were utilized in this study, except where noted,
and the results cannot be compared direaly to the OSHA PEL or
the NIOSH REL However, TEM analysis of the filter air samples
showed that 82 to 95 percent of all fibers counted using a 3:1
aspea ratio would also have been counted if a 5.1 aspea rauo
was used. This analysis indicates that there would be litde difference infiberconcentration using eitherfiberaspearatiocriteria
Personal samples anah'zed by PCM indicated that the mechanics' TWA exposures all would be below the NIOSH REL for asbestos of 0.1 fee and the OSHA PEL of 0.2 fee even if they performed brake servicing for the entire work shift. These levels
were achieved with the work practices normally used by the
mechanic.
7
8
Asbestos Concentrations as Determined by TEM
Asbestos coneentrations obtained in the breathing zones of the
mechanics and analyzed using TEM are summarized in Table III
for each control method evaluated. (These results exclude exposures encountered while workers serviced brakes to the two
large vehicles.) All fibers identified as chrysotile or amphibole
asbestos with an aspea ratio of 3:1 or greater were counted
(fibers s 0.2 u.m in length). Amphibole asbestos was found on
only 7 of 219filteredair samples analyzed (one or two amphibole
fibers per filter).
Arithmetic mean asbestos exposures ranged from less than 0.013
fee while using the wet brush/recycle with recirculating solution
to 0.052 fee for the aerosol spray method. The arithmetic mean
exposures for the aerosol spray and vacuum enclosure B were significandy higher than that for the wet brush/recycle (p < 0.05).
Geometric mean asbestos exposures ranged from less than 0.013
fee to 0.O45 fee for thefivecontrol methods evaluated
Asbesios concentrations near the vehicle fender and axle (excluding the two large trucks) are presented in Table IV. Arithmetic
mean asbestos concentrauons near the fender rangedfrom0.006
to 0.115 fee; arithmetic mean asbestos concentrations near the
axlerangedfromless than 0.006 to 0.027 fee. The aerosol spray
method average fender concentration (0.115 fee) wasfivetimes
the average fender concentration for vacuum enclosure B and
an order of magnitude higher than for the other three control
methods. Dust was observed escaping from the seal of vacuum
enclosure B during brake cleaning with compressed air.
315
�TABLE IV. Near-Source Sample Asbestos Concentrations (TEM)
Fender
Control Method
Vacuum enclosure A
Vacuum enclosure 6
Vacuum only
Wet brush/recycle
Aerosol spray*
Water hose and solvent
Number of
Samples
8
11
5
10
2
1
0010
0.024
0.008
0.010
0115
0.006
Axle
Arithmetic
Mean
(t/cc)
Range
(t/cc)
<
<
<
<
0.004-0.031
0.005-0.092
0.004-0.015
0.004-0.040
0.063-0166
—
Number of
Samples
Arithmetic
Mean
(t/cc)
8
11
5
10
2
1
0.009
0.027
0.008
< 0.006
0023
0.017
Range
(t/cc)
<
<
<
<
0 003-0 028
0.004-0.164
0.004-0.020
0.004-0.016
0.022-0.023
—
Three tnte jobs evaluated.
TEM results are not direcdy comparable to the PCM data because 1) TEM counts include allfibers,regardlessof length, whereas
PCM include only fibers greater than 5 nm in length; 2) TEM
counts includesfiberstoo thin to be seen using PCM; and 3) TEM
data include only fibers identified as asbestos, whereas PCM data
include any fiber type. TEM analyses indicated that only 8 of 57
personal samples contained asbestos fibers 5 fim or longer.
Large Vehicles
Two vehicles with rear wheel brake drums 16-17 inches in diameter were evaluated in this study. A salt truck was sampled
while using vacuum enclosure A, and a boom truck was sampled
using the aerosol spray method, fiber concentrations determined
ASBESTOS EXPOSURE (FTBESS/OC)
1.00
E3 DRUM SIZE >16"
m M U M SIZE; < i r
0.75
0.88
by PCM for both large vehicles were below the 0 004 f/cc LOD;
however, asbestos exposures determined by TEM analysis were
0.15 fee for the salt truck and 0.88 fee for the boom truck. These
results are based on two simultaneous personal samples taken
during brake service to the rear wheels of the respective vehicles.
In Figure 2, the results for these two vehicles are compared to
the maximum asbestos concentrations (TEM) measured during
brake service to vehicles with 8- to 12-inch drum sizes using the
same controls.
Indoor and Outdoor Ambient Concentrations
The arithmetic mean asbestos concentrations inside the garages
were 0.006 fee or less. As determined by TEM, indoor ambient
arithmetic mean asbestos concentrations are compared to arithmetic mean asbestos exposures in Figure 3 for the five control
methods evaluated. These data indicate that neariv- all of the
mechanics' asbestos exposure was due to job tasks and not indoor
ambient asbestos concentrations. Outdoor ambient arithmetic
mean concentrations were 0.006 fee or less.
Bulk and Settled Dust Samples
CL50
0-Z5
0.15
0.055
0.00
ENCLOSURE A
AEROSOL SPRAY
CONTROL METHOD
FG R 2. Asbestos exposures: trucks vs. smaller vehicles (analysis by
IUE
TEM).
ASBESTOS (FIBERS/CC)
0.105
i PERSONAL
J INDOOR AMBIENT
0.084-
A bulk sample of brake dust was colleaed from each vehicle
serviced to determine if the friction materials contained asbestos.
Each brake dust sample consisted of a few grams of dust from
the drum of each brake serviced Bulk samples were analyzed
for asbestos by TEM. Generallv, less than 1 percent of the panicles
present in the bulk brake dust samples was asbestos although
several samples contained as much as 1 percent asbestos. These
results are summarized in Table V. Senled dust samples were
colleaed from each facility to indicate potential building contamination. Senled dust samples were analyzed for asbestos by
TEM.
Most of the fibers present in both the brake and senled dust
samples were chrysotile asbestos. Two of seven senled dust samples contained amphibole asbestos; one senled dust sample contained 20 percent (of total fibers) amphibole asbestos. Other
fibers in both the brake dust and settled dust samples were
determined to be nonasbestos.
For most vehicles, the percentage of fibers longer than 5 M-m
was less than about 3 percent. TEM analysis of the bulk brake
dust samples showed that the aspea ratio of 90 to 97 percent of
fibers was greater than or equal to 5:1 for each of the five major
controls evaluated.
161
0.063 -I
Discussion
ENCL A
FG R 3.
IUE
316
ENCL B
\MZ WET BRUSH SPRAY
CONTROL METHOD
Personal vs. background asbestos levels (analysis by TEM).
All the control techniques srudied prevented exposures in excess
of the OSHA PEL or NIOSH REL, as determined using the PCM
analytical method
APPL IND. HYG. VOL 4. NO. 12 • DECEMBER 1989
�EV. TEM Analysts of Bulk Brake
Settled Dust Samples
Brake dust
Settled dust
Brake dust
Settled dust
9
1
11
1
Vacuum only
Brake Dust
Settled Dust
Wet brush/recycle
Brake dusl
Settled dust
Brake dusl
Settled dust
Brake dust
6
1
9
1
Control
Vacuum enclosure A
Vacuum enclosure B
Aerosol spray
Water hose and solvent
'Indudes liber bundles
Comparison to Historical Data
TWA exposures of about 0.2 f/cc and peak exposures of about
15 f/cc while using dry brushing, wet brushing, or compressed
air during brake repair have been repo^ed. Analyses were
performed using NIOSH Method P&CAM 239 (PCM). The reponed TWA asbestos concentrations during compressed air cleaning
of brakes ranged from 0.03 to 0.19free;concentrations during
wet brush brake cleaning ranged from 0.23 to 0.28free.A reported
asbesios exposure using a squin bottle to wash the brake drums
was 0.21 free. Fiber concentrations determined by PCM in our
study were one to two orders of magnitude lower. A statistical
comparison berween our study results (for the servicing of vehicles such as pickup trucks, vans, automobiles, and several large
trucks) and historical data (for brake servicing using compressed
air, dry brushing, and squin bottles) showed that if variables,
such as workers, vehicle types, and facilities, had been controlled
berween our study and the htsiorical study, then exposures in
our study would be significantiy lower (p < 0.005) for PCM exposure data
(,,
19101
In two 1979 surveys, the brakes of automobiles were cleaned
using compressed air. The asbestos exposure of the mechanics
ranged from 0.14 to 0 95 free as determined by TEM using SAED
and EDX. The sampling times ranged from 2 to 4.5 hours. The
TEM exposure data from our study were an order of magnitude
lower than those of the 1979 surveys. This, again, demonstrates
a major difference between exposures when using the controls
evaluated in our study and the methods used in the earlier studies.
Personal sample concentrations during brake repair were significantly higher than indoor background levels as determined
by PCM (p < 0.03) and TEM (p < 0.005) when using the aerosol
spray method. Personal sample concentrations during brake repair for the other control methods were not statistically different
than indoor background levels based on both PCM and TEM
results.
A minimal control (water hose and solvent) brake repair operation was evaluated in our study. The measured asbestos exposures as determined by PCM and TEM were comparable to
the five major conirol methods. The results for the minimal control method is based on a single brake job.
Vehicle and Drum Size
Most of the vehicles evaluated in the study were automobiles and
light trucks with 8- to 12-inch drum sizes; two large vehicles with
16- to 17-inch drums were evaluated. Persona) sample concenAPPL IND. HTG. VOL 4. NO. 12 • DECEMBER 1989
6
3
1
Percent
Asbestos
in Total
Dust
Percent
Asbestos
of Total
Rbers
< 1
NA
Sample
Type
Number
of
Samples
54-100
85
&-100
60
< 0.1-1
< 1
< 0.1-1
< 0.1
< 1
NA
< 1
< 1
NA
N = Not
A
24-100
68
83-100
99
74-100
5&-84
84
available
Percent
of Fibers*
^ 5 jim
1-17
5
0-6
0
0-9
0
0-3
7
1-16
0-7
0
trations during brake service to the latter, as determined by PCM,
were at or below the deteaion limit (0.004free),as were most
of the personal sample concentrations (PCM) during brake service io the smaller vehicles. There was no difference based on
PCM measurements between large vehicles and small- and medium-sized vehicles. However, asbestos exposures as determined
by TEM during brake service to the large vehicles were an order
of magnitude greater than during brake service to vehicles with
smaller drum sizes. Larger brake shoe surfaces and bulkier drums
probably contain more residual dust.
Sutistical analysis of TEM data showed that when using the
aerosol spray method, ihe average personal sample concentration
for brake repair of the large boom truck was significantly higher
than the highest exposure during brake repair of the smaller
vehicles (p<0.01); but for vacuum enclosure A the exposure
during brake repair of the large vehicle was not shown to be
statistically different than that for the smaller vehicles. Except for
the effea of drum size, differences in personal and near-source
sample concentrations due to vehicle model, miles traveled, number of drums per vehicle, etc., were small.
Control Method Design Strengths and Weaknesses
Vacuum enclosures permit the use of compressed air, which
appeared to effeaively remove brake dust adhering to the brake
components and the shoes. However, when pointed at the back
of the enclosure, the compressed air blast may be strong enough
to deflea some seals, resulting in the escape of brake dust during
brake cleaning. Vacuum enclosure manufacturers could incorporate a means to regulate the air pressure in compressed air
cleaning hoses. Brake dust is colleaed dry, and care must be
taken to avoid exposure during enclosure maintenance and vacuum filter replacement Available vacuum enclosure sizes limit
their use io certain brake drum sizes. The two-glove vacuum
enclosures are superior because both hands can manipulate pans
and tools within the enclosure.
The principal advantages of aerosol spray systems are low cost
and the capability to be used on all sizes of brake drums Care
must be taken to hold the sprayer at a proper distance from the
brake to wet the brake, yet not so close that the force of the spray
res us pen ds the dust. The effeaiveness of the solvent spray systems as an exposure control method appears to be more dependent on work praaices than the other techniques In some
cases, the solvent may contain potentially hazardous components).
The wet brush/recvele svstem can be used on all sizes of brake
317
�drums. Limited wetting of the brakes can be accomplished with
the drum in place. Dust thai otherwise would have fallen to the
floor is wened and colleaed in the catch basin beneath the wheel.
This may provide bener control when difficult-to-remove drums
are encountered. The low velocity delivery of the wa brush/
recycle effectively cleans the brake components. The contaminated fluid provides a dustfree,though bulkier method of disposal.
HEPAfilter-equippedvacuum cleaners can be used on brakes
of any size. These systems do not use compressed air, nor do
they generate dust that must be contained as do the vacuum
enclosure systems. However, the drums must be removed before
the vacuum cleaner can be used, thus there is a potential for
asbesios release during drum removal. They do not clean the
brake components as effeaively as some other systems and require small vacuum nozzles to reach smaller pans of the brake
assembly. As with the vacuum enclosures, maintenance and replacement of HEPA and primaryfilterspresent an occasion for
asbestos exposure.
>0.2 - 1
»1 - 4
>4 - 8
>8 - 64
<0.1
30%
26
2
p 'i Kt
>0.1 - 0.25
e
11
«1
Bjj <i Hfi
»0.25 - 0.5
i
12
<1
R
>0-5 - 1.5
o
3
2
>1.5 - 2.5
0
0
0
•
>64
o WB
0
<i
1
I'
LOW TUMOR RATE IN ANIMALS
@
MODERATE TUMOR RATE IN ANIMALS
H
HIGH TUMOR RATE IN ANIMALS
11
In this study, afibersize analysis (includingfibersnot identified
as asbesios) was performed on a composite sample of all personal
TEM samplesfromthefivemajor controls (Figure 4). The majority
of thesefibersare less than 0.1 jim in diameter and less than
4 jtm in length.
The potential health impaa of thefiberspresent in brake dust
is not completely understood. However, Stanton' > anempted to
estimate the tumorigenic potential in humans according to a
fibers diameter and length matrix. Stanton noted that long, thin
fibers produced the greatest tumorigenic incidence in experimental animals. Figure 4 plots the fiber sizes measured in this
study with an overlay of Stanton s classification based on his
animal studies. About 1 percent of thefibersin the samples from
our study fit the classification of medium and high tumorigenic
potential. Pott* has summarized other studies concerning the
carcinogenic potency of asbestosfibers.These studies indicate
tumor induction extends tofibersshoner in length than those
given by Stanton Pon also notes that, although the carcinogenic
potential of shonfibersmay be low, many shonfibersmay induce
a tumor as easily as a few long fibers.
11
121
Conclusions
All of the five methods tested, in combination with the work
practices used, controlled the mechanics' asbestos exposure during brake servicing to less than 20 percent of the NIOSH REL
and 10 percent of the OSHA PEL In our study, the brake mechanics' exposures for 2-hour sampling periods ranged from less
than 0.004 fee to 0.016 f/cc (counting rules 7400B). Personal exposures, as determined by PCM, were at least an order of magnitude lower than personal exposures reponed in the literature
for brake service operations involving compressed air, dry brush,
or wet brush cleaning.
TEM results include asbestos fibers of all sizes. Brake mechanics' average asbestos exposures, as determined by TEM, for
five control methods ranged from less than 0.013 fee to 0.052
fee. These low exposures funher suppon the PCM results which
showed that these methods controlled asbestos emissions during
brake repair of small- and medium-sized vehicles such as automobiles and light trucks.
Brake service to two heavy dun- trucks (16- to 17-inch drum
size) showed higher asbesios concentrations, as determined by
TEM, than for smaller size vehicles The large trucks had greater
318
DIAMETER (fim)
FG R 4. Fiber size distnbution for composite of all personal sarr
IUE
(TEM). (* Stanton et al. ").
Fiber Size and Potential Health Effects
13)
FIBER LENGTH (ym)
brake shoe surface area and bulkier drums. Although mechanics'
exposure in our study were well below the NIOSH REL for asbestos as determined by PCM, the TEM results point to a greater
potential for asbestos exposure while servicing heavy duty trucks.
Except for the effea of drum size, differences in personal and
near-source sample concentrations (as determined by TEM) among
the vehicles evaluated were ven' small, and no trends with respea
to mileage, model, model year, etc., were observed.
Not all control devices of a given type will be equivalent
Differing controls and work praaices, techniques, and drum sizes
contributed to the observed test differences. Other faaors to
consider are equipment cost, reliability, ease of use, production
rate, and method of disposal of brake residue.
Fibers in dust samples obtained from the brake drums of 40
of the 43 vehicles tested in this study were mostly chrysotile. The
other three vehicles appeared to have nonasbestos-type brake
shoes. Fiber size distribution for allfibers,including those not
identified as asbestos, showed that only 4 percent of the fibers
measured during brake service would have been counted using
PCM. Furthermore, 90 percent of the fibers idenufied by TEM
had an aspea rauo of 5:1 or greater.
Recommendations
Mechanics should use asbestos control methods during brake
servicing irrespective of the lining material (The health effects
of many asbestos substitutes have not been fully investigated.)
Workers involved in brake repair must be trained and super.',
in the correa work practices and proper use of the control system
Compressed air cleaning or vacuum cleaners without HEPA
filters should never be used to clean brakes Any spills of brake
dust or contaminated soluuons containing brake dust should be
cleaned up immediately by either vacuuming (HEPA) or thorough
wet mopping and remopping. The difficulty of total cleanup even
on well-sealed concrete suggests the use of a pan or a disposable,
impervious floor cover sheet under each wheel area Disposal
of asbestos-contaminated material should be done in accordance
with federal and state regulations
Acknowledgment
This study was supponed under Interagency Agreement No.
APPL IND. HTG. VOL 4, NO. 12 0ECI
11989
�DW75931956-01 wrth the Chemical Engineering Branch, Office
of Toxic Substances, U. S. Environmental Proteaion Agency.
References
1. National Instirute for Occupational Safety and Health: Survey Analysis
and Supplemental Tables. In: National Occupauonal Hazard Survev',
Vol. III. DHEVC' (NIOSH) Pub. No. 78-114 Cincinnati. OH (1977).
2. Reid. R.. How to Protea Maintenance Workers from Asbestos Occupational Hazards. Occupational Hazards, pp. 39-42 (November 1987).
3. Roberts, D R., Zumwalde, R.D.: Asbestos Exposure Assessment for
Brake Mechanics. Industrywide Study Repon 32 4 DHHS, PHS, CDC,
NIOSH, Cincinnati, OH (1982).
4. National Institute for Occupational Safetv- and Health: Methods 7400
and 7402. In: NIOSH Manual of Analyucal Methods, 3rd ed, Vol.2.
DHHS (NIOSH) Pub. No 84-100 Cincinnau, OH (August 15 .1987).
5. Burden. GJ.; Rood, AP.: Membrane-Filter, Direct-Transfer Technique
for the Analysis of Asbesios Fibers or Other Inorganic Panicles by
Transmission Electron Microscopv. Environ. Sci. Technol. 17(11)^43
(1983).
6. Shtrom. E_: Personal communicauon. PEI, Chester Road, Cincinnati,
OH, (513) 782-4700 (January 28, 1988).
Am.
HYG. VOL 4. NO. 11 • DEC
7. U.S. OccupauoruJ Safety and Health Administration: Occupational
Exposure to Asbestos. Tremolite, Anthophyllite, and Aainolite. Final
Rules, 29 CFR 1910.1001 and 29 CFR 1926 58 Fed Reg. 51:22612 (June
20. 1986).
8. National Insutute for Occupational Safety and Health: Revised Recommended Asbestos Standard, p. 93 DHEW (NIOSH) No. 77-169.
Gncinnau. OH (December 1976).
9. Roberts, D R.: industrial Hygiene Report Asbestos—Reading Brake
and Alignment Service. Industrywide Studv Report 32.56. DHHS, PHS,
CDC, NIOSH. Cincinnati. OH (1980).
10 Roberts, D R.: Industrial Hygiene Repon Asbestos—Allied Brake Shop.
Industrywide Study Repon 32 58 DHHS. PHS, CDC, NIOSH. Cincinnau, OH (1980).
11. Stanton, M.F.; Layard, M.; Tegeris, A; et al.: Relation of Panicle Dimension of Carcinogenicity in Amphibole Asbestos and Other Fibrous
Minerals J. Natl. Cancer Inst. 67:965 (November 1981).
12. Pon. F.: Some Aspects on the Dosimetry of the Carcinogenic Potency
of Asbestos and Other Fibrous Dusts. Staub-ReinhalL Luft 38:486 (1978).
Recetved 4/4/88: review decision 10/17/88: revision 7/V89: accepted 7/1V89
319
�Case Study: Control of Methylene
Chloride Exposures During Commercial
Furniture Stripping
Cheryl L. Fairfield
Amy A. Beasley
National Institute for Occupational Safety and Heaitn
Cincinnati, Onto
T
he furniture stripping Industry Includes
an estimated 20.000 workers who are
employed by approximately 6.200 small
businesses, averaging three employees each. Furniture strippers. In gener sd. do not have an occupational safety and health program as an
Integral part of their business. Therefore, these
small businesses are unlikely to develop innovative controls to protect their workers.
At the National Institute for Occupational
Safety and Health (NIOSH). researchers have
documented time-weighted average exposures to
methylene chloride In the furniture stripping Industry ranging from 12 parts per million (ppm) to
over 2.000 ppm. The Institute recommends that
methylene chloride be regarded as a potential
occupational carcinogen and that methylene
chloride be controlled to the lowest feasible limit.
Researchers from the Engineering Control Technology Branch of NIOSH are therefore conducting
field research to develop, document, and evaluate
effective controls for methylene chloride In furniture stripping facilities.
• Removal of a panel of charcoal filters that
caused a pressure drop, thus hindering
the effectiveness of the ventilation, and
• Improved work practices.
Figure 1.—Naw vantllatlon hood Including a slot hood and
a downdratt hood.
NIOSH researchers implemented a control at
one facility on a retrofit basis. Workers who
stripped furnituretathis facility using the existing
ventilation system had exposures to methylene
chloride of 600 to 1.150 ppm. NIOSH researchers
designed. Installed, and evaluated a new ventilation system (Fig.l). which incorporated these
primary improvements:
The new ventilation system was tested over
three days with charcoal sorbent sampling tubes
for methylene chloride. The sampling results indicated that the new ventilation system lowered
exposures from 600 to 1.150 ppm to a geometric
mean of 25 ppm. with lower and upper confidence
intervals of 11 and 58 ppm. respectively (Fig. 2).
Conclusions and recommendations at the end of
the case study include the following:
• A new local ventilation hood.
• An Increased amount of makeup air to the
stripping area.
199
�CL FAIRFIELD 4 A.A. BEASLEY
JVEW
OLD
1*00
• This facility should use the installed local
ventilation as a combined hood,
M«CI C o n c e n t n t i o n (ppm)
• Elimination of crossdrafts Is an Important
factor in Improved control, and
• Further reductions can be achieved by
Installing ventilation controls in the rinse
area.
Oot e s
Nov 8 9
Jan SO MarMISO
Oslo e t Toot
M«r(2)S0
Several other surveys have been conducted of
ventilation controls for methylene chloride in furniture stripping, and reports from these surveys
are being prepared. In one study, a ventilation
system was designed at the same time that a new
stripping tank was installed. The additional design
flexibility allowed for control to similar levels as
this other case study but with considerably less
exhaust air. A final report will be completed In late
1991.
MirOISO
Flgur* 2—Mathylcn* chloride exposures for old and new
control designs.
200
�FOR OFFICIAL USE ONLY
I.
MENTAL HEALTH AND SUBSTANCE ABUSE PREVENTION
A MulticommunitY t r i a l f o r Primary Prevention of Adolescent
Drug Abuse. Mary Ann Pentz, e t a l .
Risk and P r o t e c t i v e Factors f o r Alcohol and Other Drug
Problems i n Adolescent and Early Adulthood: I m p l i c a t i o n s f o r
Substance Abuse Prevention, J. David Hawkins
The Search f o r E f f e c t i v e Prevention Programs: What We Learned
Along the Way, Richard H. Price, e t a l .
14 Ounces o f Prevention: A Casebook f o r P r a c t i t i o n e r s , eds.
Richard H. Price, e t a l .
�MENTAL HEALTH
AND
SUBSTANCE ABUSE PREVENTION
�A Multiconmtunitv T r a i l f o r Primary Prevention of Adolescent Drug
Abuse, Mary Ann Pentz, et a l .
F i f t e e n communities i n the Kansas C i t y area p a r t i c i p a t e d i n a
community-based program f o r drug abuse prevention. The program
included c h i l d and parent education through the school system and
a media campaign. A f t e r the f i r s t year of the program, the use
of drugs was s i g n i f i c a n t l y lower i n the i n t e r v e n t i o n schools than
the n o n - i n t e r v e n t i o n schools.
�A Multicommunity Trial for Primary
Prevention of Adolescent Drug Abuse
Effects on Drug Use Prevalence
Ma7 Ann Pentz, PhD; James H. Dwyer, PhD; David P. MacKinnon, PhD; Brian R. Flay, DPhil;
William B. Hansen, PhD; Eric Yu I. Wang, MS; C. Anderson Johnson. ohD
The entire early adolescent population of the 15 communities that constitute the
Kansas City (Kansas and Missouri) metropolitan area has participated in a
community-based program for prevention of drug abuse since September 1984.
The Kansas City area is the first of two major metropolitan sites being evaluated
in the Midwestern Prevention Project, a longitudinal trial for primary prevention
of cigarette, alcohol, and marijuana use in adolescents. The project Includes
mass media programming, a school-based educational program for youths,
parent education and organization, community organization, and health policy
components that are introduced sequentially inio communities during a 6-year
period. Effects of the program are determined through annual assessments of
adolescent drug use in schools that are assigned to immediate intervention or
delayed intervention control conditions. In the first 2 years of the project, 22 500
sixth- and seventh-grade adolescents received the school-based educational
program component, with parental involvement in homework and mass media
coverage. Analyses of 42 schools indicate that the prevalence rates of use for all
ree drugs are significantly lower at 1-year follow-up in the intervention condirelative to the delayed intervention condition, with or without controlling for
•ace, grade, socioeconomic status, and urbanicity (17% vs 24% for cigarette
smoking, 11 % vs 16% for alcohol use. and 7% vs 10% for marijuana use in the
last month), and the net increase in drug use prevalence among intervention
schools is half that of delayed intervention schools.
k
DRUG abuse in the United States, including the abuse of tobacco, alcohol,
marijuana, and illicit substances such as
cocaine, is implicated in one third to half
of lung cancers and coronary heart disease in adults and in the majority of
violent deaths (homicides, suicides, and
ar cidents) in youths. Recognizing the
role that drug abuse plays in chronic
diseases and premature mortality, the
1979 report by the US Surgeon General
entitled Healthy People? and subsequent national reports' targeted the primary prevention of tobacco, alcohol,
and other drug abuse as a high priority
for health promotion efforts in the United States. The objective is a significant
increase in the health status of the enu
From me institule (or He»hh Piomotion end Diseue
AMmion Research. Depanment o Pteventive Medi<
^^Wniveitity of Southern California. Pasadena (Ota
^ ^ • P Dwyer. MacKinnon. Hansen, and Johnson and
wWanj); and the Prevention Research Center, School
C Public Health. University ot Illinois at Chicago (Dr
ftty).
Reprint requests to Institute lor Heatth Promotion and
Disease Pieventioft Research. Department ol Pieveiv
t<vt Medicine. University ol Southern Califomia. 35 N
Lake Ave. Suite 200. Pasadena, CA 91101 (Or Pentz}
JA.MA. June 9 1989-Vbl 261 No »
"gateway drugs" (tobacco, alcohol, and
marijuanaX" ' The early adolescent
years have been identified as the first
risk period for use of these drugs." The
t jrm "gateway" refers to the predictive
relationship of tobacco, alcohol, and
maryuana to the use of more illicit substances later in life, including cocaine:
Unfortunately, the effectiveness of
large-scale, school-based programs in
producing significant and sustained
changes in drug use behavior might be
questioned. Reported effects have been
minimal to moderate and, in some cases,
short-lived or delayed.***" Several explanations have been hypothesized, including the brevity of most school programs, the lack of integration of schoolbased programs with community programs, and mass media and other
environmental influences outside of
school that conflict with drug prevention messages learned in school.If
not mobilized to promote non-drug use
(MAfA. 1989;261:2259-3266)
norms, these collective environmental
influences—including parents, community leaders, mass media programmers,
tire US population by 1990.* Accom- and local school and government adminplishing this goal is expected to require istrators—might have diluting or abthe dissemination of prevention tech- breviating effects on an otherwise ponologies on the scale of community or tentially powerful prevention program
city to populations at prin^ryriskfor that is initiated in the school.***'
onset of drug use.*
A preventive intervention model that
Recent research shows promise for uses multiple environmental influences
school-based programs that teach youth might be required to effect long-term
peer pressureresistanceand social com- changes in adolescent drug use. These
petence skills for avoiding drug use.** influences could be used to support and
More than 20 of these studies have extend prevention skills learned initialshown significant reductions, between ly in a school program and to promote a
29% and 67%, in experimental smoking consistent community norm for not usrates among adolescents, with more ing drugs. In studies of communitymoderate reductions in alcohol and mar- based programs for prevention of unyuanause." A meta-analysis of 143 drug " wanted pregnancy and cardiovascular
prevention studies that employed vari- disease, use of at least some of these
ous approaches and populations has influences in a small number of commushown that effect size is highly related nities has led to significant reductions in
to the inclusion of peers in the teaching estimated pregnancy rates and dgaprocess.* The most promising programs rette smoking. However, most studseem to be those that are initiated in the ies of programs for drug prevention
early adolescent years, particularly the have been limited in the size and numyears marking transition to middle or ber of sites required to produce suffijunior high school (sixth or seventh cient power to estimate effects on whole
grade), and that focus on delaying the communities. Statistical control of deonset or abuse of one or more of the mographic variables in communityctt a
1
4
1
�*4l«wm«nto<
Community Drug UM
ProWem and Pra^nton
Nwds md Bi»ouroo«
I
Introdjclory Traimng d
Community Laadori in PioOtem
AMI i« nou and Program Nood
ConvnunHy
Entry ind
Ptvpvtton
T
<
EtlaMitrimont o(
Community Coordmaung
Strucura
T
Training of
II "isir II *~ 11^. I
I
I
I
Program
I
torrautna
I I
4
1
I j
Qrganjaiion
j
Oganj.tKy.
Admrnatratora
^
Environmantai Support
Uy Changing Social
Norm el Drug Uaa
mdiroctStuilito
Support Ra»*lanc« Practice
Target Batwviora
H^mir
Laidora
Roductcnef
Drug Use
Pravalenoe
I
FMuclionol
Drug Uaa
I
Proniofconof
Non-OrugUaa
Social Norma
R g 1.—Community program intervention model. Training and program components progress
from left to right over a 6-year period. Bold lines
indicate phases implemented at the time of this
study: and the dashed line, indirect involvement
of parents through homework.
�-
- _
—
dependent of the program. Baseline
data were collected approximately 1.6
months prior to program implementation.
group, with school as the unit of analy- cioeconomic status, but this difference
was reversed for marijuana. Results
sis, are presented in Table 1. The percentages presented are means of school were similar for rates of use in the last
week or month, but the grade and race
level percentages. Nine of 24 program
schools and 5 of 18 control schools were differences were not significant.
Statistical Analysis
middle schools; baseline measurement
Intercorrelations among demographProgram effects were estimated with for students in these schools was in sixth ic variables are presented by study
grade. The remainder were junior high group in Table 3 at baseline and followseveral alternative statistical models.
These included conditional (covariance) schools; students in these schools were up, with school as the unit of analysis.
first measured in seventh grade. Ihere Except for alcohol use in the program
and unconditional (change score) models with both the proportion of drug us- were no significant differences between group, autocorrelations of rates of drug
the program and control schools in co- use between baseline Mid follow-up
ers in each school and the logit of that
variates or grade-adjus.ed drug use.
were positive in both study groups, as
proportion as the metric-" ConcluSince the proportion of girls was similar were intercorrelations among use rates
sions concerning program effect were
in the two groups, this variable was not
of different drugs. Socioeconomic status
uniform across these alternative apincluded as a covariate in subsequent
was related negatively to the preproaches. Results from a multiplevalence of cigarette and maryuana use
group unconditional model applied to analyses.
in both groups at both baseline and
proportions are reported herein since it
Self-reports of lifetime drug use at
follow-up, consistent with other recent
avoids the substitution of arbitrary valbaseline are summarized for individrr search." Urbanicity, socioeconomic
ues for the logit of zero and does not
uals, rather than schools, by covariate
status, and race were strongly interassume consistent covariance patterns levels in Table 2. The lifetime measure
across groups. Change in the proportion reflects experience with a drug at any correlated, but differed somewhat in
of usersfrombaseline to follow-up was level of use. More seventh- than sixth- theirrelationshipsto drug use. Fbr exregressed on covariates within each
grade students reported druguse. More ample, socioeconomic status showed a
group. A common zero-point was used whites reported use of alcohol than non- negative association with cigarette use
at follow-up in both study groups
for covariates in both groups. The inter- whites but fewerreportedever using
cept in each group was then the mean or marijuana. Students from families with (r = - .66 and - .29), while race showed
a negative relationship in the program
adjusted mean of change in the propor- a higher socioeconomic status were
tion of users in each group. The differmore likely to report use of alcohol than group (r = - .22) and a positive relationence between the intercepts is then the students from families with a lower so- ship in the control group (r = .28). Howchange in the program condition minus
^e change in the control condition (net
T«ble 2.-Individual Mean Percentages ot Uletime Drug Use by Demographic Predictors at Baseline
ngesX Covariates included grade at
(sixth grade, - 1 ; seventh
Lifetime Prevalence Rata, %'
grade, +11), socioeconomic status, perPredictor
Cigarette Uae
Marfluana Uee
Alcohol Uae
cent white, and urbanicity. Except for Grade
grade, the value of each covariate was
S
«et
»t
72T
computed as its deviation from the
7
56
TB
13
grand mean. Estimates were obtained
Race
53
White
sot
lit
by ordinary least-squares analysis.
1
66
RESULTS
The total sample of individuals observed at baseline in 42 schools was 79%
white, 17% black, 2% Hispanic, 1
%
Asian, and 2% other; 51% female; and
39% reported fatherfc occupation as
professional/managerial, 10% white collar support/sales, and 51% blue collar
labor. Baseline characteristics by study
66
17
»t
Urtwractty
Urban
Urban/suburban
Suburban
Outlying suburban
«8t
76
81
ist
17
11
ao
6
61
52
45
Socioeconomic statue
Professional occupation
42t
su
se
Nonprofessional occupation
77
St
14
'Based on the percenuge of individuals who ever used a aubstance. with individual as the urtt of analysie
(N-S065).
tP< .01 for the test of equal proportions across the levels of each predictor variable lor each subctanea.
Table 3 —Correlations Between School Demographic Characteristics and Lasi Month Drug Use in Study Group**
IMS
Cigarette
Uae
Alcohol
Uee
Marijuana
Uee
Cigarette
Alcohol
Marijuana
Uee
Uee
Uae
Program School! (n>24)
Grade
Urbanicity
Socioeconomic
Statue
Race
1
Cigarene U M
Alcohol uae
Marijuana uee
1985
ntuaneuee
Urbanicity
Sodoeconomic
.67
SI
M
m
i
-
-
.43
XL
XI
Si
-.14
-.12
.19
Jl
46
-ia
sa
-.46
.16
JSI
• 25
.18
J7
-XA
Ot
-J9
-J04
sa
.40
.IC
.16
58
xi
M
-.29
-04
-at
• • -.
JOT
Ot
-m
A\
-28
Control Schools .79• IS)
(n
JK
at
.14
£2
.25
-ts
-ao
.18
JOS
at
at
.54
-St
Ay
-JB
-at
-.49
-/«1
.52
JS3
XI
-x\
34
-.14
-.68
— .19
-JST
J03
-.72
-JM
-.79
-.41
-21
-36
-22
25
-.44
.14
-JH
M
V>.47 (oortrol) or r>.40 (program) are signifcart al P<.05 level (two tailed).
.IAMA
l , ™ 9 1989-Val261.No.22
Prevention of Druo Ahuse—PpniTorn!
32B3
�- 5%, -2%, and (Mb after 6 months to
-8%, -4%, and -3% after 1 year for
cigarette, alcohol, and marijuana use,
respectively, in the last week; results
were similar for prevalence rates of use
in the last month. Preliminary results of
2-year follow-up in the tracking study
(1986 through 1987) suggest that program effects have been maintained
(-9%, -2%, and -3%). The maintenance of group differences over time
suggests that the program effect is not a
novelty effect or an initial interim effect
that regresses back to control group levels over time." lhe increase in the net
difference between groups over time
suggests that the resistance skills and
social environment intervention model
1964-1965
followed in this program, through the
addition of multiple successive components, might be shifting the social normative climate in the youthful population toward a non-drug use norm.
Recently, a defined population study on
predictors of adolescent substance use
concluded that the substantial drug use
inlluence represented by perceived social environment might be most modifiable through community-based prevention efforts that promote drug use
resistance skills and non-drug use social
norms."
The public health implications of
these results can be substantial. Based
on our 1985 through 1986 results on cigarette smoking and 1986 dollar values,
1985-1986
1984-1985
Marijuana Used in Last Week
198&-1986
Marijuana Used in Last Month
Fig 4.—Prevalence rales tor marijuana use in the last week and the last month, adjusted tor grade, race,
urbanicity, and socioeconomic status. Solid line indicates program group; and dashed line, control group.
and assuming maintenance of a net reduction in smoking prevalence rates
across the original 1984 cohort and all
subsequent program cohorts into middle adulthood, the estimated total savings in the Kansas City area from preventing an adolescent from becoming
even a moderate smoker is $40 679 for
each boy and $13232 for each girl."
These projections should be observed
with caution, since the MPP is still in its
early stages and the formula used to
calculate savings was based on the entire adolescent population as it moves
into adulthood. Nevertheless, the projections illustrate the sizable potential
of primary prevention efforts for impacting on long-term morbidity and
mortality rates in the United States and
the feasibility of primary prevention as
a means for accomplishing this nation's
goals toward improved health status of
the population by 1990.
Two primary threats to the vali^t v '
thefindingsshould be considered. The
first is the possibility of nonequivalence
of study groups, since the majority of
schools were assigned to program or
control conditions based on administrator flexibility in changing school-year
schedules that already were in progress
when the MPP was initiated. The equivalence of study groups on the dependent
and potential confounding variables at
baseline, the direction of the crossover
effect on use rates between study
groups over time, and the lack of a differential effect of assignment or attrition on study groups over time suggest
. underreporting or other types of
confounds probably are not operating in
this study. The threat of nonequivalence will be addressed further by the
randomized replication in Indianapolis.
The second threat derives from the
study's reliance on self-report measures
to estimate program effects. Self-reported drug use might not represent
actual drug use prevalence rates in a
community. Furthermore, it is possible, particularly during the course of an
intervention that is aimed at changing
social norms for drug use behavior, that
Table 4.—Change in Proportion of Drug Users After Intervention
Proportion of Drug Users, % (95% Confidanca Interval)
Unadjusted«udy
Group
Program
H4IWI
Control
Clgaienae
Alcohol
3 4 (02-6.6)
12.9 (7.6-17.9B
4 3 (1.M.9)
104 (5.9-14.6)*
Adjustedf
Clgarattae
Alcohol
Marijuana
4 0 (1*6.7)
8.7 (5.7-11 a)J
Marl|uana
Users tn Last Month
3.2 (1.0-5.4)
7.3 (4 5-102)*
3.4 (0 2-6.6)
13.1 (7.5-18.8)*
4 2 (1.6*6)
9.4 (6.5-12.4)*
3.4 (1.6-5.3)
7.1 (4.0-103)5
2 0 (0 2-3.9)
4.5 (2 7-62)J
Users In Laat Week
2.2 (0.7-3.7)
4.7 (2.7.6.7)$
4 2 (1.74.9)
10.5 (5.6-15.5»
2.1 (02-4.0)
4.9 (3 « . 4 ) |
2 4 (1.2-3.6)
4.7 (2.5-6.8){
•Change in proportion is adjusted for grade.
tChange in proportion is adjusted lor grade, race, urbanidty, and socioeconomic siatus (% of fathers in professionalAnanageria! occupation*).
*P< .01,t
on one-tailed
iP<J05. based on one-tailed
�Risk and P r o t e c t i v e Factors f o r Alcohol and Other Drug Problems
i n Adolescent and Early Adulthood: I m p l i c a t i o n s f o r Substance
Abuse Prevention. J. David Hawkins
Although studies have shown many curriculum programs t o be
e f f e c t i v e , t h i s a r t i c l e suggests t h a t those programs would be
less e f f e c t i v e w i t h high r i s k populations. An a l t e r n a t i v e type
of prevention program would be t o address the f a c t o r s t h a t are
known t o lead t o probable substance abuse. Such a prevention
program may include e a r l y childhood education, parenting s k i l l s
t r a i n i n g , promotion of academic and s o c i a l competence, and
r e o r g a n i z i n g schools w i t h a high number of h i g h - r i s k students.
A l l of these approaches have been e f f e c t i v e t o o l s of substance
abuse prevention.
�PREPRINT
Risk and Protective Factors for Alcohol and Other Drug Problems in
Adolescence and Early Adulthood: Implications for Substance Abuse
Prevention
J David Hawkins. Richard F Caialano. and Janet Y. Miiier
Social Developmem Research Group. School of Social Work
UniverMiy of Washingion
The authors suggesi [hat ihe most promising route to effective strategies for ihe preventun of
adolescent alcohol and other drug problems is through a risk-focused approach This approach
requires the identification of risk faaors for drug abuse, identification of methods by which risk
factors have been effectively addressed, and application of these methods to appropnate high-risk
and general population samples in control led studies. The authors review nsk and protective factors
for drug abuse, assess a number of approaches for drug abuse prevention potential with high-risk
groups, and make recommendations for research and practice.
In spiteof general decreases in the prevalenceof the nonmedical use of most legal and illegal drugs in recent years, the abuse
of alcohol and other drugs during adolescence and early
adulthood remains a senous public health problem (Adams.
Blanken. Ferguson. & Kopstein. 1990). The consequences of
drug abuse are acute on both a personal and a societal level. For
the developing young adult, drug and alcohol abuse undermines motivation, interferes wuh cognitive processes, contributes to debilitating mood disorders, and increases risk of accidental injury or dealh. For the society at large, adolescent substance abuse extracts a high cost in health care, educational
failure, mental health services, drug and alcohol treatment, and
juvenile cnme.
\dded to the immediate personal and social costs of adolescent drug abuse are the longer range implications for youngsters who continue to abuse alcohol and drugs into adult life.
Drug abuse is involved in one third to one half of lung cancer
and coronary heart disease cases in adults (R. Blum. 1987).
Alcohol and other drugs are major faciors in acquired immunodeficiency syndrome (AIDS), violent crimes, child abuse and
neglect, and unemployment. The problems associated wuh ai
cohol and other drug abuse carry costs in lost productivity, lost
life, destruction of families, and a weakening of the bonds that
hold the society together.
Given the serious consequences of drug and aicohol abuse,
considerable effon has been directed toward identifying effective treatment. Until recently, applied research in the substance
abuse held has consisted primarily of experimental trials of
various forms of treatment for alcohol and other drug abuse.
The goal has been to identify ways to increase the effectiveness
of treatment and to prevent relapse following treatment. Strategies ranging from self-help to aversive counterconditioning
have been advocated and assessed.
Many of these studies have demonstrated how abstinence
can be achieved
it long-term maintenance of abstinence has
been more difficult. The reinforcing properties of alcohol and
other drugs are themselves often reinforced by norms and behaviors of family members and others in the communities in
which recovering people live. These combined reinforcemenis
often overcome shon-term treatment gains. According to the
surgeon general: "For many drug-dependent persons, achieving
at least brief penods of drug abstinence is a readily achievable
goal. Maintaining abstinence, or avoiding relapse, however,
poses a much greater overall challenge" (Surgeon general. 1988.
Preparation of this article was supported in pan by Grant DA03721
p. 311).
from the National Institute on Drug Abuse and by Grant 87-JS-CXAdded to disappointment with the staying power of drug
K084 from the Office of Juvenile Justice^«s«<«*/and Delinquency /
treatment is a growing recognition of the high cost of treatment
Prevention. Points of view or opinions expressed are those of the auand of the inability of existing treatment programs to keep up
thors.
with increasing demand. In recent years, these considerations
Our thanks to George S Bridges. Elise Lake. Randy Gainey. Tim
Murphy, John Campos, and Cheryl Yates for their assistance ir the
have stimulated interest in pnmary prevention of alcohol and
preparation of this anicle and to Patricia Huling for her management
olher drug abuse.
of the document's creation. Thanks also to Barry Brown. David
This anicle focuses on the prevention of alcohol and other
Farnngton. Michael Goodstadt. Rolf Loeber. Roger W'eissberg, riedrug abuse among adolescents. A number of views have been
lene White, and two anonymous reviewers for their comments on
advanced about what constitutes substance abuse when condrafts.
sidering adolescents (Hawkins. Lishner. & Caialano. 1985). In
Correspondence concerning this anicle should be addressed to J
this anicle. adolescent drug abuse is defined as the frequent use
David Hawkins. Social Development Research Group. School of Soof alcohol or other drugs during the teenage years or the use of
cial Work. University of Washington. 146 N. Canal Street. Suite - I I .
alcohol or other drugs in a manner that is associated w ith prot>
Seattle. Washington QSlO?.
001
�oo:
D H-\W KIN'S, R. C A T A L A N O . A N D J. M I L L E R
lems and d\s)'unciions. This conception oTthe problem is not
meant to condone the infrequent use of alcohol or other drugs
by teenagers, which is a violation of the law. The present definition simply reflects a recognition that a relatively large proportion of teenagers try alcohol or other drugs without becoming
involved in the frequent use of these substances or developing
drug-related problems (Newcomb <k Bentler. 1988. Shedler &
Block. 1990).
Clonmgerand his colleagues (Clontnger. Bohman. Sigvardsson. & von Knorring. 1985: Cloninger. Sigvardsson. &. Bohman. 1988) have identified two types of alcoholism. One type is
associated with frequent impulsive-aggressive behavior and follows an early onset of alcohol use and alcohol problems m adolescence. This type of drug abuse is considered in this anicle. It
is distinct from alcoholism thai develops after age 25. which is
not a focus of the current article.
Precursors of drug and alcohol problems have been described as risk factors for drug abuse. Risk factors occur before
drug abuse and are associated statistically with an increased
probability of drug abuse. A risk-focused approach seeks to
prevent drug abuse by eliminating, reducing, or mitigating us
precursors. This anicle suggesis that a p.omismg line for prevention research lies in testing interventions targeting multiple
early risk factors for drug abuse.
A risk-focused approach in drug abuse prevention research
and policy is warranted given the apparent success of this approach in reducing risk factors for problems as divergent as
hean and lung disease (Bush et al.. 1989: Vaniainen. Pallonen.
McAlister. & Puska. 1990) and school failure (Berrueta-Clement. Schwemhan. Barnett. Epstein. & Weikhan. 1984). The
apparent failure of early prevention interventions, such as drug
informalion programs that did not address known risk faciors
for drug abuse (Stuan. 1974: Weaver & Tennant. 1973). also
argues for this approach.
Many of the risk factors for adolescent drug abuse also predict other adolescent problem behaviors (Hawkins. Jenson. Catalano, & Lishner. 1988). There is evidence thai adolescent drug
abuse is correlated with delinquency, teenage pregnancy, and
school misbehavior and drop oui (Elliott. Huizinga.& Menard.
1989: Jessor & Jessor. 1977; Zabin. Hardy. Smith. & Hirsch.
1986). Comprehensive risk-focused efforts probably can prevent other adolescent problem behaviors besides drug abuse.
If prevention of drug abuse (as defined above) is the goal, then
risk factors salient for drug abuse rather than for the occasional
use o f alcohol or other drugs should be targeted. A relatively
small proportion o f adolescent drinkers or users are frequent or
problem users (Johnston, O'Malley, & Bachman. 1988: Shedler
&. Block. 1990). The following review focuses on factors that
have been shown to precede drug abuse.
Risk Factors for Adolescent D r u g Abuse
Most studies to date have focused on small subsets of identifiable risk factors for drug abuse. There is little evidence available regarding the relative imponance and interactions o f
various risk factors in the etiology o f drug abuse, although
current studies are seeking to measure a broader range of identified risk factors. At this time, it is difficult to ascertain, for
instance, which risk factors or combination of risk factors are
most virulent, which are modinable. and which are specific to
drug abure rather than generic contributors to aooiescent problem behaviors. Current know-ledge about the risk factors for
drug abuse does not provide a formula for prevention, but it
does point to potential targets for preventive imervennon. Implications for intervention are considered in this article afier a
review of know n risk factors for drug abuse in adolescence and
early adulthood.
These risk factors can be roughly divided into two categories.
First are broad societal and cultural (i.e.. contextual) factors,
which provide the legal and normative expectations for behavior. The second group includes factors that lie within individuals and their inierpersonal environments. The principal interpersonal environments in children's lives are families, school
classrooms, and peer groups. The risk factors have been described elsewhere (Hawkins. Lishner. Caialano. &. Howard.
1986: Kandel. Simcha-Fagan. & Davies. 1986: Newcomb. Maddahian. & Bentler. 1986: Simcha-Fagan. Gersten. &. Langner.
1986) and are summarized here and in the left half of Table 1
This is not intended as a critical review of the methodologies of
::.. studies but rather as an overview of the evicei.c ' " - r - n t l y
available on risk factors for adolescent drug abuse.
Contextual Factors
Individuals and groups exist within a social context: the values and structure of their sociely. For example, shifts in cultural
norms, in the legal definitions of certain behaviors, and in economic factors have been shown to be associated with changes in
drug-using behaviors and in the prevalence of drug abuse. The
following risk factors (1 through 4 below) exist in the broad
social context:
/. Laws and norms favorable lonard behavior. Recent research on the effects of laws on alcohol consumpuon has. focused on three interventions by law: (a) laxauon. (b) laws stating
to whom alcohoi ...ay be sold, and (c) laws regarding how alcohol is to be sold.
Alcohol consumption is affected by price, specifically the
amount of tax placed on alcohol at purchase (Levy & Shellin.
1985). Cook and Tauchen (1982) found that increases in taxes
on alcohol led to immediate and sharp decreases in liquor consumption and cirrhosis m o n a l n y
Studies examining the relationship of minimum d r i n k i n g
age and adolescent drinking and driving have generally shown
that lowering the drinking age increases teen drinking and
driving and teen traffic fatalities and raising it decreases teen
driving while intoxicated citations (DWIs) and deaths (Cook &.
Tauchen. 1984: Joksch. 1988: Kneg, 1982: Saffer &. Grossman.
1987) .
Studies of restriction on how alcohol is sold have show n that
allowing patrons to purchase distilled spirits by the drink increased the consumption of distilled spirits and the frequer.c;.
of alcohol-related car accidents (Holder & Blose. 1987). However, there was i o increase in accidents involving males under
the legal drinking age of 2 I (Blose & Holder. 1987).
Two general explanations of how laws affect the use of substances have been advanced. The first posits that laws reflect
social norms and that use is lareelv a function of group norms
(Watts Sc Rabow. 1983). Alcohol consumption rates varv among
�l.ihlc
Risk
Fm li'is
for
idolrstrnl
Siihsnnuv
\hii\r
II / / / / ( KIH SIUHKI HI; I'rnriilion
l
iiulmm
Imervennon
Risk
n i e c ls o n risk f a c l o r or
Risk Factor
I
(.))
Etiological sludy
I aws a n d n o r m s
Taxiilion
K v i d c m e (findings)
l e v y A S h e d m . 141(5
I'i!. i n c r c u s i i n tan o n a l c o h o l l e d
l o 1/2% decrease i n
liuplicaiions
K u i s e laxes o n
jlcohol
Study
u v / a l m s c (liinlmgs)
I evy A Sheflin,
19X5; C o o k
A
T a u c h e n . 1982,
consumpuon.
H i g h e r a l c o h o l taxes was i c l a i e d
l o clrcrcj.ses i n c o n s u m p l i o n
a m i accompanying cllcclY
Saffer &
G r o s s m a n . 19X7
Cook A Tauchen.
1982
SalTer & G r o s s m a n .
1987
•J
(b)
Laws regulating
lo w h o m liquor
is s o l d
(c) l a w s r c g u l a i i n g
h o w l i q u o r is s o l d
(d) C r i m i n a .
'ws
SafTer & G r o s s m a n .
1987; K r i e g . 1982;
C o o k SL T a u c h e n .
1982; Joksch.
1988
H o l d e r & Blose,
1987; Blose &
H o l d e r . 1987
P o l i c h c i a l . , 1984
making drugs
illegal
(e) C u l t u r a l n o r m s
Increase i n a l c o h o l l a x l e d I n s l u r i i
decrease i n c o n s u m p l i o n a n d
cirrhosis m o r i a l i l y .
H i g h e r laxes associated w i t h l o w e r
leen d r i n k i n g a n d falalnies
H i g h e r t a * m o r e salient than
drinking j
H i g h e r d r i n k i n g age a s s o c i a t e d
w i l h fewr - leenage traffic
ralaliiics, d r i v i n g while
i m o t i c a l e d citations.
>
Increase a n d
e n f o r c e age
reslriclions o n
purchase of
alcohol
I i q u o r - b y - t h e - d r i n k sales
increased c o n s u m p l i o n o f
distilled spirits but n o l
p r o p o n i o n o f drinkers in lhe
population.
Discourage
Neither doubling of inlerdiclion
nor increased arresls o f d r u g
dealers w o u l d affcci retail prices
or a v a i l a b i l i t y o f illegal drugs.
D e c k e r et a l . . 1 9 8 8 .
K n c g . 1 9 8 2 ; Saffer
& Grossman,
1987, C o o k A
T a u c h e n . 1982;
J o k s c h . 1988
I n c r e a s i n g age r e s l r i c l i o n s o n
a l c o h o l p u r c l u s e s t a n reduce
alcohol-related Iraliic
falaliues.
n
—i
o
Deemphasi/e
Walls & Rabow,
1983; Rasher &
M a i s l o . 1984;
Robins, I9H4.
V a i l l a m , 1983
Alcohol c o n s u m p l i o n and olher
alcohol-related effecis arc
associated w i l h
sociodemographic faciors.
einnic and olher group norms
A l k m e l a l . , 1984
O
-o
•po
O
H
m
liiiii'u
by I h e - d r i n k
LO
sales
-n
O
TO
O
c
o
interdiction and
criminal
sanctions
Foslcr social
-o
M o r e e x p o s u r e t o ads p r n m o i i n g
alcohol a m o n g Iccns r e p o r t i n g
h i g h e r d r i n k i n g levels
norms
opposing
d r u g use
J o h n s o n A Solis,
1 9 8 3 ; P e r r y c l al
1988
CoiniiiuDily healih pronioiion
associated w i l h c c s s a n o n or
redm lion of smoking
B l a c k . 19X9
O
Saluralion ailverliMng
a c c o m p a n i e d incieascs in
negalive alliludes l o w a u l
dnifss. i l m g users; t a l k i n g
a h o u l drugs w i l h parenis.
le.uliers. siblings a m o n g
college s l u d c n l s . c h i l d i e n b u l
less for 1.1- I 7-year o l d s
C o m p r e h e n s i v e school policies
e m p l u s m n g pievenlion
reslriclions on opporliinilies
for ust- m a y i e d u c e s m o k i n g
P e n i / . H r a n n o n , el
al . 1 9 X 9 .
Moskowti/ A
Jones. I9XX
03
I—
m
O
�>
ro
Table I ( c o n l i n u e d )
Inlcivenlion
Risk
Risk Faclor
Etiological study
Evidence (findings)
Implicanons
Study
Hansen el al.. 1988;
Perry, 1986; Penlz,
Owyer, et al ,
1989; Botvin,
1986; K l e p p c l al..
1986
Robins, 1984;
Johnston, 1991
2. Availability
•o
3. Extreme
economic
deprivation
Gorsuch & Builcr,
1976
Maddahian et al .
1988; E>embo el
al.. I 9 7 9 ; G D
Gotlfredson. 1988
Bursik & Webb.
1982; Farnngton
el al , 1985
Bachman el al..
1981; Zucker &
Harford. 1983. I )
M . Murray et al..
1987
Robins & Ralclilf.
1979
Increased alcohol availability led
to increases in d r i n k i n g
prevalence, amount o f alcohol
consu ned, heavy use of alcohol.
Enforce drug and
alcohol laws;
resistance skills
Iraining
EflTects on nsk faclor or
use/abuse (findings)
Some social induencc resistance
programs include n o r m a l i w
change componems, (e.g.,
depicting drug use as socially
unacceplable, use of peer
leaders lo leach c u r r i c u l u m )
For programs sec Risk Faclor
14
Norms anlitheucal lo use are
assiK'iated w n h rcduclions m
prevalence of frcqiienl
niari|iiana, wthei illicil drug
use
(Risk faciors l b , 2, 6, and I arc
rclalcil, all involving social
influences to use For studies
and resulls sec Risk Faclor
14 )
Availabilily aflected use of alcohol
and illegal drug..
U
•z.
>
>
r-
Poverty associated with childhood
conduct problems, delinquency,
chronic offenses.
I'arenial education and
occup. lion positively conelaled
wilh t en alcohol and marijuana
use.
Exlrei.ie poverty one o f Ihree
faciors increasing risk of adult
alcohol and drug abuse in adulls
who weie aniisocial as children.
Target
intervenili
io
economically
disadvantaged
children
Lazar el al., 1982;
Ramey cl al.,
1988; Seilz et al..
1985
Swift, 1988, Olds et
al , 1986; Bronson
ct al . 1984;
Pierson el al .
1983, S c i u ei al .
1985
Berrucla Clement et
al , 19X4
Inlerventions w i l h low-mcoine
families, including day caie.
preschool, parcnling, home
visitors, health tare, showpromising effects on
anlistRial behavior in
adolescence, aggression,
special educaiion placcmcms.
c n m i n a l involvement
Early family support
inlcrvenlions have shown
positive cllects on child
abuse, eaily school
performance and allendance.
family si/e, maternal
eniploymenl
l o l l o w - u p of low-irH onie 5- (>•
ycai -olds from Perry Preschool Program al age 19
reveals less menial
lelaldallon. st hoot drop ouls.
crime, and wellare reliance
anti greater hleracy.
c m | i l o y m e n l . college/
^ o c i l i o n i i l st fiool lor
p:iincip;tnls
>
o
>
o
�lablc
I
(nuiliiHifJ)
Intervention
Risk
03
C
Risk Faclor
Eliological sludy
l
Brook el al.. I9 )0
4. Neighborhood
disorganization
p
V Physiological
faciors
(a) Biochemical
C.A. M u r r a y . 1983;
Merling A Guesl,
1985; W i l s o n *
Herrnsiein. 1985.
Fagan, 1988;
Simcha-Fagan A
Schwartz. 1986;
Sampson el a l ,
1981; Sampson.
1986
Cloninger el al..
1988
Zuckerman. 1987;
von Knorring ci
al., 1987;
TabakofT A
HolTman, 1988
Suwaki A Ohara.
1985; Schuckil.
•a
o
CO
1987
(b) Genetic factors
Blum el al.. 1990
Pollock ei al . 1983
Schuckil et al . 1983.
Schuckil. 1980;
Schuckil A Rayes.
1979. Schuckil.
1987
K a i j . I960; llrutx-c
A Omenn. I9HI
I vidcncc (lindings)
Implicalioi
Low socioeconomic siatus in
childhood related .o greater
drug use in adolescence.
Characteristics of neighborhoods
such as populaiion density,
mobilily. physical deterioration,
low auachment, high crime are
related lo juvenile crime and
drug tralficking
l a r g e , families in
high-risk
neighborhoods
Study
Fllccis on risk factor or
use/aliiisc (findings)
(Sec siudics under
Kisk Factor 3.
Economic
deprivation)
v.
o
Sensalion ' - e k i n g and low harm
avoidamc predicl early-onset
alcoholism.
Sensalioi seeking, early-onsel
alcoholism Iinked to plalelel
monoamine oxidase
Aldehyde dehydrogenase
diflerences found in Asians wilh
lower rates of alcoholism than
controls
Genetic susceptibility to al leasl
one form of alcoholism
suggeste t>y polymorphic
pattern of dopamine D2
receptor gene
More slow-wave
electroencephalogram activity
in children of alcoholics lhan
non-alcoholics.
DilTerences between children of
alcoholics and nonalcoholics in
serum prolactin response,
muscle response, and levels of
acclaldehyde after
adminislralion of alcohol
Monozygotic Iwins were more
lhan iwice as likely as dizygotic
Iwins ti) be concordant for
alcoholism (all males)
Targcl youngsters
wi-.h certain
central nervous
system disorders
or biochemical
levcis and w i l h
low
socioeconomic
stalus and
central nervous
syslem disorders
Target
inlcrvenlions lo
children of
alcoholics,
especially ln'> s
73
o
Tl
-t
o
73
o
73
o
73
r
Ci
T3
73
o
O
o
�luhlc I (aHtitmted)
Inlcivenlion
Risk
Kisk Taclor
Eliological sludy
C u r l i n g el al., 19X1
G o o d w i n el al.,
1974; G o o d w i n el
al., 1977;
Bohman, 1978;
Cadorel el al..
1980; Cadorel &
G a l h . 1978
Goodwin, 1985
M a r l e y e i al.. 1986
R. M . Murray &
Slabenau, 1982
6. f a m i l y drug
behavior
C o t t o n . 1979;
G o o d w i n , 1985;
Cloninger el al ,
1985; Johnson cl
al., 1984; Kandel
el al., 1978;
M c D e m i o l i . 1984
Evidence (findings)
Sludy
Effecis on risk lacloi or
use/abuse (lindings)
Concordance rales for alcoholism
of 2 1 % for monozygotic and
25% for dizygotic twins, when
Ixiih males and females were
included.
Rales o f alcoholism ranging from
18% io 27% found for adopted
sons o f alcoholics compared
w i t h 5% to 6% for adopted
males w i i h o u l biological
alcoholic pareni.
About half of hospitalized
alcoholics do not have a family
history o f alcoholism.
Evidence from animal siudics o f
heniability in predisposition to
barbiturate and morphine abuse.
No consistent evidence for genetic
transmission of alcoholism in
females reponed.
Parental and sibling alcoholism,
use o f illicil drugs increase risk
of alcoholism, drug use
i n i t i a t i o n , d i n g abuse in
childien.
Ahmed ei al.. 1984
Drug salience in Ihe household
besl predicioi of children's
cjipcclalions lo use and actual
use o f alcohol, tobacco, and
marijuana.
Hansen cl al.. 1987
I'arcnul modeling direclly rclalcd
lo Inends' use of drugs, which
in turn was related to adolescent
subjects' dri ; use.
Oldest hrothe
nd parenis each
had indcpend.nt eflect on
younger brother's use Both
drug modeling and drug
advocacy hy older brothers had
independent elfecls and
inleiacted with parental drug
use |o provide a risk/proiecnve
elleel
Brook el al.. 1988
Implicalions
D
$
V.
o
Target
inlcrvenlions to
children whose
parents or
siblings are
users/abusers
DeMarsh &
Kumpfer, 1986
Narcotic- and polydrug-abusing
parenis given parcnling skills
training dcvelo|ied more
effective discipline melhods.
Iheir children had fewer
behavior problems alter
treatment and reported
decreased mlenlion lo smoke,
use alcohol
(Social inlluence resistance
intervenlions can also I;MKCI
family drug use See studies
and results under Kisk l-'aiior
14, associaiion wilh diugusmg iH-eis.)
V
p
o
�l.ihlc
I
(nniiinnnl)
Imerveniion
Risk
Risk F a c l o r
Ltiologicul siudy
H r o o k c l a l . 1990
M c D e r m o l t . 1984;
Hansen el al..
1987. Barnes A
Welle. 1986;
B r o o k et a l . . 1986;
7
I amily
managemeni
praciices
Jessor e i a l . , 1 9 8 0
Kandel A Andrews.
1987, B a u m n n d .
1983; Penning A
B a r n e s . 1982
Ziegler-Driscoll.
1979; K a u f m a n A
K a u f m a n . 1979
B a u m n n u . 1983
F l l c c i s o n iisk f a c l o r or
I.vidence (lindings)
Implicalions
Study
use/abuse (lindings)
I a / a r el al . 1982;
K a m c y c l al .
1 9 8 8 : S o i l / ct a l .
1985. Berrucla( l e m e n l c l al
1 9 8 4 ; S w i f t , 1988
t ' a r l y c h i l d h o o d i n l e i ^ enl ions
i n c l u d i n g a p a r c n l i n g skills
c o m p o n e n l p i o d u c c d positive
o u l c o m c s l o i lii|:li-ri.sk . I . m
i m ome childien
P a t t e r s o n ct a l .
1982. B a u m A
F o r e h . i m l , 1981
Pareni skills n a m i n g i m p r o v e d
f a m i l y i n l e r a c l i o n . reduced
c h i l d p r o b l e m lieh.iviors.
Alexander A
Functional family
F a t h e r s ' n o n d r u g use, e m o l i o n a l
s l a h i l i l y e n h a n c e d c l l e c l s o f peer
nonuse o f drugs.
P e r c e i v e d par. r t p e r m i s s i v e n e s s
t o w a r d d r u g . . I c o h o l use m o i e
i m p o r t a n t lhan actual pareni
d r u g use i n d e l e r m i n i n g
a d o l c s c e m d r u g , a l c o h o l use
l ack o f or inconsisienl parenial
liscipline. l o w paiental
educalional aspirations for
children predict iniliation m i o
d r u g use
O v e r i n v o l v e m c n l by o n e pareni
a c c o m p a n i e d by distance or
permissiveness by the other
associated w i l h n s k .
Parent a u l h o r i l a t i v e n e s s related to
I each f a m i l y
managemeni
skills l o p a r e n i s
Q
Ihcrapy
c h i l d r e n ' s prosocial. assenive
reduced dclini)ucncy for
behaviors; parent
Reilly. 1979
Parsons. 1973;
K l e i n et a l . , 1977
juvenile ollenders; prevented
nondirectiveness,
f)ermissiveness associated w i l h
h i g h e r d r u g use.
C o m m o n characteristics o f
families o f adolescent d r u g
abusers; n e a l i v e
c o m m u n i c a t i o n panerns.
inconsisienl. unclear behavior
limits, unrealistic parental
expccialions.
D r u g users s a w f a t h e r s as m o r e
h o s l i l e . a d v e r s a r i a l ; p a r e n t s as
b
Norem-Hebeiscn cl
a l . , 1984
d c l i i u ) u c n c y l o i t h e n siblings.
Pallcrson A
Fleischman, 1979;
ITcischman. 1981;
Pallcrson A Reid,
1973; P c e d c i al..
1977
P a r e n t a l d r u g use i n u n r e w a i d i n g
family structure more linked l o
m a r i j u a n a use t h a n i n a
rewarding family coniexl
P a r c n l i n g skills I r a i n i n g laiighl
parenis lo monitor cluldien's
b e h a v i o r , l o use c o n l i n g e n l
discipline for undesired
behavior, and lo
ICWMHI
prosocial iK'havior
n
o
73
o
TO
o
73
r
Ci
Tt
73
O
CD
T r e m b l a y el al.. 1990
P a r c n l i n g s k i l l s , s t x i a l skills l o i
k i n d e i g j i l c n boys reduced
school adjuslmcnl problems,
delayed dclin<|ucn( tx-haviois.
PcnW. D w y e r . cl a l . .
1989
Parenis participated in
Midwestern prcvenlion
P r o j e c l : 8()'''. o f c x p e n m e n l a l
families involved in
less c a r i n g , m o r e r e j e c t i n g
Tec, 1974
/.
o
IIOIIH WOIk
.issignmcnls
Paienling componenl nol
ass.-sst-d s c p a u l c l y . h " i
| > i o ) ; r a n i p.u kaj'.c a s s o c i a l c . l
w i t h I n w o c i l l,Kes o f l o h . i c c o .
a l c o h o l , a n d i n . u i i u a n . i ust'
�>
Table
I
s
(lontimicd)
Inicrvcnlion
Risk
I ' d e c l s o n l i s k f a c t o r or
Risk f actor
Ftiological study
Evidence (findings)
B r o o k ct a l . . I W O
Parent-adolescent
Shedler A
Implications
use/abuse ( f i n d i n g s )
Sludy
r e l a t e d l o less m a r i j u a n a use.
Quality of mothers' inleractions
w i l h 5-year o l d s r e l a t e d t o
Block.
1990
attachment
m a r i j u a n a use al 18.
Psychological slahilily o f molhers
olTsel effecis o f peer d r u g use
H i g l a n , G l a s g o w , el
al , 1987
N o cllecl on c h i l d s m o k i n g ol
l o u r p a r e n i messages m a i l e d
h o m e for s l u d c n l s i n v o l v e d i n
social i n d u e n c c resistance
program Pareni compliance
level n o l m e a s u r e d
Karoly A Rosenthal,
1977; M a r t i n ,
1977; Patterson cl
al.. 1982; W a l l e r s
A G i l m o r e . 1973
B r o o k et a l . , 1990
P a r c n l i n g skills I r a i n i n g reduced
prcadolcsccnls' problem
behaviors, suggcsling lhal
parenting skills I r a i n i n g can
b u l f e r nsk f a c l o r o f c h i l d h o o d
behavior problems.
D i s h i o n cl ,
•o
8
Family conllict
B a u m r i n d . 1983:
Penning A
Barnes. 1982:
R o b i n s , 1980
Wilson A
H e r r n s i e i n . 1983
M c C o r d . 1979;
Ruller A Ciller.
1983; W . M c C o r d
A M c C o r d . 1959.
P o r t e r A O l eary.
1980;
l
)
L o w h o n d i n g to
family
H e t h e r i n g l o n el
al.. 1979;
Wallerslein A
K e l l y , 1973;
S i m c h a - F a g a n el
a l . . 1986
K a n d e l e l al . 1978:
B r o o k et a l . .
1 9 8 0 ; B r a u c h t el
a f . 1978; P e n n i n g
A B a r n e s . 198 2
C h i l d r e n f r o m h o m e s b r o k e n by
m a r i t a l d i s c o r d are at h i g h e r
risk o f d e l i n q u e n c y , d r u g use.
No independent contribution of
parenis' m a m a l dissolution lo
delint|uenl behavior
Reduce family
c o n d i c l . services
lo children
whose families
are i n c o n l l i c t
1989
Parent t r a i n i n g groups
improved paient-child
i n t e r a c t i o n , level o f loba i o
use, r e d u c e d d e p r e s s i o n lor
al risk y o u l h s
n
(See s i i . d i e s o n p a r e n i I f a i m n g .
f a m i l y i n l e t v e n h o n s ,itM»\c.
Risk F a c l o r 7).
v.
O
F a m i l y conflict stronger predictor
of delinquency lhan family
structure (intact parental
marriage)
Lack o f p a r c n l - c h i l d closeness,
lack o f m a t e r n a l i n v o l v e m e n t
related l o d r u g i n i t i a t i o n
P
>
>
v.
o
m
73
Strengthen
bonding
family
( I . l l e i l i v e c.uly i h i l d h o o d and
family s u p p o n programs and
pareni Iraining progiams
have been f o u n d lo incicase
p a r e n i - c h i l d I v n n l i n g See
SlndlCS a n d l i n d i n g s n i n t c f
l< isk I a c l o i s 1 ( e c o n o m i c
d e p l i v:il i n n I a n d 7 f l.iul l U
managemeni I
�I.ililo
I (i
ini/Diiin/)
Imervennon
Risk
Risk Faclor
Eliological sludy
l l i r s c h i , 1969
Ellioit el al.. 1985;
Brook el ul.. 1990
Jessor & Jessor.
1977; K i m . 1979.
Norcm-Mebeiscn el
al.. 1984; Gorsuch
& Buller. 1976.
Selnow. 1987.
Brook. Brook, el al .
1990; Brook,
( i o r d o n . e l al.. 1986.
I l u n d l e b y & Mercer,
1987
10
Karly and
jK-isislfnl
problem
behaviors
Robins. 1978
l erner Si Vicary,
1984
(•vidcncc (findings)
Implicalions
Sludy
Elfects on i isk lactor or
use/ah use (findings)
Bonding inhibits delinquency
l amily bonding inlcracls w i l h
peer variables (o inlluence
delinquency and drug use
Family involvcincnl and
atlachmcnl discourage youlh drug
initiation and level of use.
7-.
D
More variety, frequency o f child
aniisocial behavior portends
adull antisocial behavior
liKcrvcnc early
wilh children
with problem
behaviors.
Strategics
include pareni
Iraining, social
competence
training lor
children
I'altcrson cl al.,
1982. Baum A
f o r e h a n d , 1981;
Fleischman. 1981.
Caucrson A Reid,
1973; Pecdcl al .
1977
O
o
O
" D i l l k u l t " temperament in 5ycar-olds contnluiles to drug
problems in adulthood
73
o
73
r
Shedler Sc Block.
1990
Kellam A Brown,
1982; Lewis,
Robins, & Rice,
1985; Nylander,
1979; Loeber,
1988; Spivack,
1983
Loeber & D i s h i o n ,
1983
18 year-old frequcn marijuana
users had emolio j l distress in
childhood.
Aggressiveness in boys age 5-7
predicts frequent drug use,
delinquency in adolescence;
drug problems in adulthood
O n l y 30-40% of boys w i l h
problem aggressive behavior
maintained it 4 io 9 years later.
O
-u
73
llallistich cl al., n d ;
Bieiman A
F u r m a n . 1984;
Gcsten ei al.. 1982,
Ladd. 1981, Ladd
A Asher. 1985;
Rothcram 1982b,
Shure A Spivack,
1982; llierman et
al , 1987,
Weissbcrg A
Caplan. 1989,
WcisslK-rg ct al .
1981; Weissbcrg
A Allen. I 985
o
00
�ro
1 ahle I ( c o n u m w d )
Intervention
Risk
c
R . s k f actor
/ '•
Eliological sludy
Loeber. 1988;
McCord. 1981;
Barnes & Welle.
1986; Kandel.
1982
Loneyel al., 1979
G i l t e l m a n cl aL.
1985
o
•a
o
Brook ct al.. 1990
i-
Evidence (lindings)
Implications
Early aggressive or antisocial
behavior persisling into early
adolescence predicts later
adolescenl aggressiveness, drug
abuse and/or alcohol problems.
Hyperactivity, attenlion-deficil
disorder raises delinquency risk
if combined with conduct
problems
Hyperaclivity m children,
especially accompanied by
conduct problems, increases
substance abuse risk in late
adolescence
Sludy
Allen et al , 1976
N o eflecls on adjustment of
social competence skills
inicrvcnlion.
Lochman, 1988
Aggressive boys given anger
managemeni program had
lower alcohol, marijuana use.
fewer negalive consequences
of alcohol al age 14 ( .1 years
(Xislinlei ven I ion. compared
w i l h matched group of
nnirealed boys) No posuivc
Vyear follow up eilects on
aggression or general
deviance
Children who are irritable,
dislraciible, have lemper
lantrums. fight with siblings,
engage in predelinquenl acls
more likely lo use drugs in
adolescence.
G. D. Goltfredson.
1981
Inlelleciual ability and . .
delinquency have inverse
relalionship. after controlling
for socioeconomic status
Fleming el al.. 1982
High reading and IQ scores in
Grade I predicted earlier, more
frequent adolescenl alcohol use
in African-American inner-cily
sample.
Higher lest performance
associalcd w n h higher lifetime
levels ol cocaine use in young
adulls
Drug use in boys related to IQ
decline irom age I i lo age i 8.
Kandel A Davies.
1991
Block el al . 1988
>
r-
>
p
>
s
Kelchel A Bicger,
1989; K i m el al
1989
I I. Academic failure
(a) Intelligence - -
ElEecls on risk faclor or
use/abuse (findings)
Target
inlcrvenlions
toward some
higher ability
youlhs
Social coni|H-leiH e. skills
Iraining in 4III grade
produced lower use ol
alcohol, cigarettes, marijuana,
especially in Grades 5-7
�Table I (l onrimietl)
Inldrvcntion
Risk
Risk Faclor
Ftiological sludy
(h) School failure
Jessor, 1 9 7 6 ; S m i t h
& Fogg. 1978;
R o b i n s . 1980
F e l d h u s c n el a l . .
1973
Lvidciv e (findings)
Failure in school
predicted
Promote academic
B e r r u c l a - C l c m e n l el
al , 1 9 8 4 ; Seilz el
al , 1 9 8 5 , H o r a c e k
ct a l . . l 9 8 7 ; G o t l s .
1 9 8 9 ; L a z a r el a l .
1 9 8 2 . R a m e y ct
a l . . 1988
Larly c h i l d h o o d educaiion and
B l c c h m a n et a l . ,
1981. Bien & Bry,
1987
P a r e n t i n v o l v c m e n l led t o
i m p r o v e d academic elforl.
grades, a l l e n d a n c e i n s l u d c n l s
w i t h low school c o m m i l m e m .
Brophy A
1986
Studenl achievemeni fains
l i n k e d lo teacher's active
i n s t i u c l i o n . d i r e c t sii|x.'rvision
o f learning
a c h i e v e m e n i in
f r e q u e n c y a n d levels o f use o f
a v a r i e t y o f ways
illicil drugs
may
p r e d i c l later a c a d e m i c f a i l u r e
H u n d l e b y A Mercer.
1987
Sludy
adolescent d r u g abuse.
( arly amisoci;.! behavior
F r f c c l s o n risk l.iclor o r
use/abuse ( l i n d i n g s )
Implications
G o o d school performance reduced
l i k e l i h o o d o f f r e q u e n t d r u g use
i n m m h graders.
Good.
family suppon inlerventions
resulted in higher
achievemeni
school
(See R i s k
I actois 3 and 7 above )
I lawkins A l am,
1987, H a w k i n s .
Ooueck, A
I i s h n e r , 1988
Interactive leaching, proactive
classroom managemeni, and
c o o p e r a t i v e l e a r n i n g led l o
grealci m a l h ai.liicvcmcnl
gains in s c v c n l h grade
studenls.
DeVries A Slavin.
1978; M a d d e n A
S l a v i n . 1983,
Ziegler. 1981;
D o l a n el a l . 1989
S c h a p s el a l . . 1986
C o n l r o l l e d siudics showed
posilive cllecl ol i ooiH'i.ili v e
learning on achievemeni and
alliludes l o w a i d school and
|H-ers
"Jigsaw'' c o o p e i a l t v i - I r a i n i n g
method did nol pievenl d i n g
use.
Coie A
19X4
Comer.
Krehbicl.
19X8
F r e i l n - r g c l a l . . 19X9
T u t o r i n g o f socially rejected,
l o w - a c h i e v i n g f o u r l h graders
p r i y l u c e d i m p r o v e m e n i s in
reading, malh achievement
a n d r e d u c e d peer r c j e c l i n n .
tlisruplive Ixhaviot
( i a i n s in sliiilcnl a c u l c m i e
achievemeni ilemonsltaled
o v e r 12-year p e r i o d a f t e r
c r c a l i o n o f sc h o o l g o v e r n a n c e
a n d m a n a g e m e n t teams in
urban school district.
( ooperalive learning, classroom
managemeni. siudcmleacher m o t i v a l i o n . p a r e n i
c o n i . K is i n i c i . i c l i v c l e u l u n g
discipline pievenlion icsultcd
in i m p i o v c d academic
.K I n e v e m c n t
>
O
-a
73
o
H
m
n
>
n
H
o
73
tn
-n
O
73
D
73
r
o
-X3
73
O
CD
�>
-a
I able
I
[conlinued)
Inicrvcnlion
Risk
EfTecIs o n risk f a c t o r o r
n
Risk Faclor
Etiological sludy
Evidence (findings)
Implicalions
Sludy
use/abuse ( l i n d i n g s )
03
C
Felner A A d a n ,
1988, Felner,
A d a n , A Evans.
1987, Felner,
Weissbcrg, A
A d a n , 1987
12
T3
l ow c o m m i t m e n t
io school
J o h n s t o n el a l . . 1985
IV C. G o n f r e d s o n .
1988
K e l l y & B a l c h . 1971
F n e d m a n . 198.)
Use o f a v a r i e t y o f d r u g s is
signiticantly lower a m o n g
studenls enpeciing to attend
college.
4-(i7i' o f variance i n t r u a n c y was
associan J w i l h d r u g
i n v o l v e m e n . c o n t r o l l i n g for
c l h n i c i l y . parent education,
delinque i y
I'romoie
commilmem
school
lo
I n i c r v c n l i o n l o ease i r a n s i i i o n s
lo middle, junior high, and
high schools, w i t h advocacy
a m i schools-within-schixils
t o decrease f r a g m e n t a t i o n
I'arlicipaling sludcnls hail
better a c a d e m i c p e r f o r m a n c e
lhan lumparticipaling smdenl
i n the s a m e s c h o o l s
B l c c h m a n c l al .
1981. Bien A Bry,
1987
Parent i n v o l v c m e n l i m p r o v e d
effon and ai. ulancc of
studenls w n h low school
commilmem
Teacher i n i e r a c t i v e leaihin,-..
proactive classroom
management, conpcralivc
l e a r n i n g resulted in higher
c o m m i l m e n l lo school and
fewei susix-nsions. expulsions
i n sevenlh grade
enperimemal compared wnh
c o n i r o l classronis
I lawkins A l am,
1987
H o w m u c h students like school
r e l a t e d t o levels o f d r u g use.
l i m e spent o n h o m e w o r k ,
p e r c e p i n m o f relevance o f
c o u r s e - w i r k r e l a t e d l o levels o f
d r u g use
>
o
p
DeVries A Slavin,
1978; M a d d e n A
Slavin. 1981;
Z i e g l e r , 1981
I lawkins. Doueck. A
I i s h n e r . 1988
Goojx'ralive learning pnHliued
jxisilive cllecls o n alliludes
t o w a r d s c h o o l a n d |X-eis.
Coopcralive learning produced
r c d u c l i o n s in susix-nsions.
e x p u l s i o n a m o n g lowachievers
I)
(' Goltfredson.
I9K6
M u l l i c o m p o n e n l school
p r o g r a m p r o m o l i n g shared
decision making sludcnl
services a c a d e m i c
innovations produced lowci
i ales o l d i ug a l n i s e .
drlinciucncy. alien.mon.
higher rales o f a l l a c h i n e n l lo
si h o o l . e d u c a l i o n a l
c x p c i l a l i o n s . I K IICI in school
i ulcs i n c \ | x - i i i n c n l . i l l l i : i n
c o M i p : i n s i m s, ht >t tls
�.ihlc
I
(ronliniit'd)
Inicrvcnlion
Risk
E l l e c l s o n risk l a c l o i or
Risk K a c l o r
Eliological sludy
Evidence (lindings)
Implicalions
Study
E). C
Goltfredson
use/abnsc I
findings)
&
Curriculum resnuclunng
i n c r e a s e d p o s i l i v e sellc o n c e p i a i l a c h m e n l to
s c h o o l , belief in rules, m a l h
scores, decreased s c h o o l rate
o f d e l i n < | ! i c n c y . d r u g use.
suspensions
Eel ncr A A d a n , 1988
S c h o o l t i a n s i l i o n s l u d y (see
R i s k 1 a c t o r I I b. a b o v e )
resulls show lower
a b s e n t e e i s m , d r o p o u l for
participaiing sludcnls
compaicd wnh
n o n p a r t i c i p a t i n g students in
the same schools
Cook.
1986a.
1986b
11.
Peer r e j e c t i o n i n
e l e m e n t a r y grades
Parker & Asher.
1 9 8 7 , C o i e , 1990
K e l l a m et a l . , 1900
f
B r o o k ct a l . . 1986
C a i r n s cl al., 1988;
H a r l u p . 1983;
T r e m b l a y , 1988
E o w a c c c p t a n c b y peers seems l o
elevate risk 1. r s c h o o l p r o b l e m s
and criminalily.
C h i l d r e n w h o h a d b e e n aggressive
as first graders o r aggressive a n d
shy h a d h i g h e r levels o f d r u g
use l h a n t h o s e w h o w e r e j u s i shy
C h i l d h o o d iraii o f social
i n h i b i t i o n , is. j i i o n . a n d
aggression not associalcd w i t h
a d o l e s c e n t d r u g use stage, b u l
aggression, lower i n l n b i l i o n . a m i
lower isolation in adolescence
a s s o c i a l c d w i l h h i g h e r d r u g use
stage.
A g g r e s s i v e n e s s m a y t>e a s s o c i a l c d
w i l h a c c e p t a n c e by o t h e r
aggressive peers w h o c o u l d
f o s i e r d r u g use; s i m i l a r l y ,
s o c i a l l y r e j c c i c d c h i l d r e n mayf o r m friendships w i l h other
rejccicd c h i l d r e n in adolescence,
leading lo lalcr d e l i n i | u c n l
txhavinr
Provide
o p i x i r t u n i l i e s for
socialization,
social
com|K'lencc
skills
B a l l i s t i c h el al , n d . ;
Bierman A
E u r m a n . 1984.
Geslen cl al., 1982;
I add. 1981; Ladd
A Asher. 1985;
R o l h e r a m , 1982a.
1982b; Shure A
Spivack, 1982;
Weisslxrrg A
C a p l a n . 1989
I o c h m a n . 1988
Social c o m p e t e n c y siudics have
p r o d u c e d p o s i t i v e eflet Is o n
c h i l d r e n ' s inter|>ersonal
b e h a v i o r . (See R i s k I a c l o i
10 )
a
-v
TO
O
—1
m
O
H
<
>
n
o
73
o
TO
In spile o f positive alcohol and
m a r i j u a n a elfecls (see Risk
l a c l o i 10). I h e i e w e r e n o
p o s i l i v e e l f e c l s for aggressive
i K h a v i o i or g e n e r a l
I x - h a v i o r a l d e v i a n c e al 3-ycar
follow-up
o
73
<a
73
o
00
O
O
�>
-o
>
Table I
(conlinued)
CD
c
Risk Faclor
14
- J
O
•a
Association wilh
d r u g - u s i n g peers
Eliological study
Barnes & Weltc,
1986; K a n d e l ,
1978, 1986;
Kandel
&
A n d r e w s , 1987;
E l l i o l l c l a l , 1985:
Jessor e t a l . , 1980:
Brook et al.,
Newcomb
&
O
•a
Intervention
Risk
rr
Bentler.
Byram &
1984
Evidence (findings)
I'eer use o f s u b s t a n c e s is a m o n g
Ihe strongest p r e d i c t o r s o f
substance use a m o n g y o u t h .
Study
Social influence
resistance s k i l l s ;
social
competence
s k i l l s (sec also
Risk Factors I,
B o t v i n . 1986;
B u k o s k i , 1986.
Elay, 1985,
M o s k o w u z , 1989,
T o b l e r , 1986
Reviews o f social influence
resistance strategics have
found modesl bul significam
r c d u c l i o n s i n l h e onset a n d
prevalence o f c i g a r e l i c
s m o k i n g lor groups receiving
t r a i n i n g In c o m p a r i s o n w i l h
untrained conlrols
M c A l i s t e r el a l . ,
1980; B o t v i n .
1986; Hansen cl
al . 1 9 8 8 , P e n i z .
D w y e r , e l al ,
1989. Perry. 1986.
J o h n s o n et al .
1984
S o c i a l i n f l u e n c e strategies
s h o w e d beneficial eilects in
p r e v e n l i n g or d e l a y i n g lhe
ousel ol alcohol and
m a r i j u a n a use
2. and 6)
1990
1986;
Fly,
H a r f o r d , 1985
E > e m b o e t a l . . 1979
B r o o k c l a l . . 1990
I n l l u e n c e o f peers o n d r u g use
s t r o n ^ l h a n l h a l o f parenis for
Whiles.
African-Americans,
Asians. Hispanics
N o n d r i n k i n g African-A merican
y o u l h s reported fewer d r i n k i n g
friends lhan A f r i i a n A m e r i c a n
youlhs who drank.
F r i e n d s ' use o f a l c o h o l a n d
m a r i j u a n a was r e l a t e d to use by
African A m e r i c a n and Puerto
Rican youths.
The mosl powerful linkage in
causal p a t h w a y to m a r i j u a n a
n o n u s e was a s s o c i a t i o n w i t h
n o n - d r u g - u s i n g peers
Effects o n risk f a c t o r o r
use/ahuse ( f i n d i n g s )
Implications
B o l v i n , 1987; K l c p p
ct a l . , 1 9 8 6 ;
McAlister, I98J;
D M M u r r a y el
.:l , 1984
B o t v i n c i al , 1989
Flay ct lil . 1989
O
X
$
7".
73
n
Social i n f l u e n c e resistance
I r a i n i n g g r o u p s led by peers
a c h i e v e d greater r c d u c l i o n s
i n d r u g use l h a n n o n - p c c r - l c d
groups
l ife skills I r a i n i n g to prevenl
s m o k i n g for A f r i c a n
A m e r i c a n urban junior high
s t u d e n l s i c s u l t c d i n fewer
postlest s m o k e r s i n t r e a l m e n i
l h a n c o n t r o l g r o u p o n lhe
basis o f a d j u s t e d m e a n s f o i
s m o k i n g s t a l u s i n pasl m o n i h
Siv year f o l l o w - u p o f s o c i a l
i n l l u e n c e resistance s m o k i n g
prcvenlion piogiam found
no overall diflerences
lx-lwecn j w o g i a m and conirol
g i o u p s I a i l v p i o g i a n i cllecls
(( i i . i d e 8) l i . u l d i s a p i K - . u e d by
( ir.nle I 2
>
o
>
m
73
�Tahle
I
(ainiinitrd)
Inicrvcnlion
Risk
F l l c i Is o n risk lac l o r or
Risk f a c l o r
Ftiological sludy
F v i d i n c c (findings)
Implications
usc/ahusc (lindings)
Sludy
Flhckson
&
Dell.
I 'MO
S c h i n k e . B o t v i n el
a l . . I9XX
Social i n l l u e n c e resistance
c u r i i c u l u n i lor s c v c n l h
g i a d e i s Willi e i g h t h grade
booster p i o g i a m
Kesulls
showed modesl rcduclions in
d r i n k i n g l o i s l u d c n l s al three
l e v e l s o l b a s e l i n e use
i m n i e d i a l e l y after i n i t i a l
p i o g i a m ( l a v m i n g letwi leader
o v e r a d u l l le:ider f o r m a l )
c o m p a i c d wilh conlrols. A l
f o l l o w - u p in eighth grade.
diHcrcnces had disapiK'aied.
f o r s m o k i n g , sigmficant
r c d u c l i o n s were observed
across a l l s u b s e ( | u e n l s m o k i n g
l e v e l s for b a s e l i n e
experunenlers bul nol
b a s e l i n e n o n u s e r s or s m o k e r s
f o r the laller. s m o k i n g
i n c r e a s e d m o r e for t r e a i m e n t
lhan conirol sludcnls. f o r
maiijtiana. boih inilialion
a n d c u r r e n i use w e r e MI'J l o
fiO'S l o w e r lor (he I r e a l n i e n l
group
Kicullunil compclcnce drug
abuse p i e v e n l i o n p r o g r a m for
Native A m e n c a n adolcscenls
f o u n d greater s u h s l a n c c use
knowledge, alliludes.
i n t e r a c t i v e s k i l l s , l o w e r sellr e p o r t lates o f d r u g s , t o b a c c o ,
a l c o h o l I'm | i : i r l i c i p . i n l s l h a n
conlrols
7^
V.
o
po
O
n
>
o
H
O
73
ist
~n
O
po
O
73
—
(
d
73
O
"i
�-a
TMc
I
(amiiniied)
Inicrvcnlion
Risk
F l l c c i s o n nsk f a c l o r or
Risk Factor
Ftiological sludy
Evidence (findings)
Implications
Sludy
I). M
use/abuse ( f i n d i n g s )
M u r r a y el al
1987,
1988; D
M
M u r r a y , P i r i e , el
af.
1989
o
•o
o
Schinke, Bcbcl.
Orlandi & Hoivin,
1988
V a n i a i n e n et a l . .
198.1. 1986. 1990
•c
I V
Alienation and
rebelliousness
Jessor & Jessor.
1977; K a n d e l .
1982; Penning &
Barnes, 1982.
Jessor el a l . . 1980.
R o b i n s . 1980
B a c h m a n c l al .
1981; Kandel.
1982; Smiih A
Fogg 1978
A l i e n a t i o n f r o m d o m i n a n l socieial
values, l o w r e l i g i o s i t y p o s i l i v e l y
r e l a t e d l o d r u g use. d e l i m i u e n l
Ix'havioi
Rebelliousness, rcsislance lo
l i a d i l i o u a l .lulhoiily posilively
r e l a t e d to d r u g use. d e l i m i u e n l
Kh.ivioi
P r o m o t e prosocia
iinoKeinenl
Schaps cl al . 1986
I i.bier. r>Xf>
F o l l o w - u p of study c o m p a r i n g
four smoking prcvenlion
strategies (three v a r i a t i o n s ol
social i n n u e n c c s m o d e l a n d
a healih consequences m o d e l )
ami no m l c r v e n l r o n I wo io
5 years p o s l i n t c r v e n l i o n .
m i x e d i c s u l i s h i v o r c d (X'cr led
social inlluenccs p i o g r a m
( w i i h o u l videos) in i c s t i . l i n i n g
s m o k i n g l o i baseline
n o n s m o k e r s aiul
C M K ' r i m c n l e r s . A l S a n d 6year f o l l o w - u p , d i d e i e n c e s
hail disap|>caicd. w u h no
p r o g r a m cllects.
Smoking pievenlion program
for c h i l d r e n o f blue collar
families c o m p a r e d skillsbased w i t h discussion-based
a n d c o n i r o l groups. Skills
based p - o g r a m h a d l o w e r use
rales l o r p a r l i c i p a t a n l s al 6.
12. 18. a n d 24 m o n t h s
follow up
Social inlluenccs s m o k i n g
prevention program; Iwo
levels o f i n i c r v c n l i o n . one
comparison condiiion. I w o
and 4-ycai follow-up
r e p o n e d lower m o n i h l y
s m o k i n g rales l o i
i n l e r v e n l i o n over c o m p a r i s o n
s c h o o l s Hy 8 year f o l l o w - u p .
o n l y c l l e c l s for b a s e l i n e
nonsmokers weie evidcul
0 | > e r a l i r i g a s c h o o l store
( c o m b i n e d w n h cross age
l i i l o n n g ) d i d nol prevenl drug
use t n p r c d u i n i n a n t l y W h i l e
middle-class eighlh n m i l i
graders
I h c r c is snrnc c v i J c n r c l h . i l
m l c n s i v c poig.Kiois lh.il
e m p o w e r liig,h risk y n u l l i s l o
m . i s l c l new s k i l l s .irc
associalcd w i l d i m p i o w , !
K ' l c o lor . i n d i i c h i c . c i n c n l
D
V.
n
>
>
7
P
>
D
m
73
�able I (amimucd)
Inicrvcnlion
Risk
Risk Faclor
F f l c i Is o n risk l a c l o i o
Alliludes
favorable lo drug
use
I7
Parly onsel of
drug use
Evidence (findings)
Jessor 4 Jessor.
1977; Jessor. 1976;
Palon & Kandel.
1978
Shedler 4 Block,
1990
16
Eliological siudy
High tolerance of deviance,
rcsislance to authority, strong need
lor independence, normlessness
Iinked wilh drug use.
Interpersonal alienation al age 7
predicted frequent marijuana use
at 18.
Iniliation into substance use is
preceded by values favorable tn
its use.
Kandel el al , 1978;
Krosnick & Judd,
1982; S m i l h 4
Fogg, 1978
Rachal el al., 1982;
Kandel, 1982;
Fleming el a f .
1982; Robins 4
Pryzbeck. 1985
Brunswick 4 Boyle.
l979;0 Donnell
A C I a y l o n , 1979
Kandel el af, 1976
,
Misusers of alcohol begin drinking
earlier than users; earlier onsel
of drug use predicts greater and
more persistent use of more
dangerous drugs
Earlier i n i l i a l i o n of drug use
increases probability of dcvianl
aclivilies.
I ater onsel of drug use predicts
lower drug involvcmenl anil
higher probabilily of
disconlinualion of use.
Implications
Sludy
usc/ahusc (lindings)
V.
o
-o
o
H
m
O
Foslei antidrug use
alliludes and
(See s i u d i c s a n d r e s u l l s u n d e r
R i s k F a c t o r s l e a n d 14 for
social rcsislance a n d c u l l u i a l
norms inlcrvenlions )
Dircci inlervenlion
lo younger
children and
Iheir parenis;
large! risk
factors
developmenially
before initiation
of drug use
(Sec R i s k F a c t o r s .1 a n d 7 for
early c h i l d h o o d , pareni
s u p p o n . pareni Iraining
interventions.)
o
73
O
73
D
73
<~
o
TJ
73
o
U3
s
�00 1 s
0. H A W K I N S . R. C A T A L A N O . A N D J. M I L L L R
different ethnic groups—tor example, in association with differences in the extent to which members hnd consumption
socially acceptable (Flasher & Maisto. 1984; Vaillam. 1983).
The second view of the law's effect focuses on supply and
demand. As noted above, legal restrictions that influence the
availability or price of alcohol or other drugs, such as taxation
or laws regarding sales, appear to limit consumption.
Legal restrictions on the purchase of alcohol and norms unfavorable toward alcohol use clearly are associated with a lower
prevalence of alcohol abuse. Conversely laws and norms that
express greater tolerance for the use of alcohol are associated
with a greater prevalence of alcohol abuse. Johnston (1991) has
suggested a similar relationship between norms regarding illegal drugs and the prevalence of illegal drug abuse.
2. Availabilily The availability of drugs is dependent in
part on the laws and norms of society. Nevertheless, availability
is a separable factor. Whether or not panicular substances are
legal, their availability may vary and is associated with use.
Research has shown that when alcohol is more available, the
prevalence of drinking, the amount of alcohol consumed, and
the heavy use of alcohol all increase (Gorsuch &. Butler. 1976).
With regard to illegal drugs. Maddantan. Newcomb. and
Bentler (1988) found man adolescent sample that two measures
of drug availability were significantly related to the use of cigarettes, alcohol, marijuana, and other illegal drugs, even after
controlling for the amount of money available to the subjects.
Dembo. Farrow. Schmeidler. and Burgos (1979) reponed that
the availability of drugs affected substance use indirectly
among junior high school youths. G. D. Goltfredson (1988)
found that drug availability varied in different schools and that
drug availability influenced the use o f drugs beyond the influence of individual characteristics of subjects.
J. Exireme economic deprivation. Indicators of socioeconomic disadvantage, such as poveny. overcrowding, and poor
housing, have been shown to be associated with an increased
risk of childhood conduct problems and delinquency (Bursik &
Webb. 1982: Farnngton et a l . 1990). However, research on social class and drug use has not always confirmed popular stereotypes. A slight positive correlation between parental education
and high school seniors" manjuana use has been reponed
(Bachman. Lloyd. & O'Malley. 1981). R. A. Zucker and Harford
(1983) found that parental occupational prestige and education
were positively related to teenage drinking. D. M. Murray.
Richards. Luepker. and Johnson (1987) found that mother's
occupation was positively correlated with monthly alcohol use.
heavy alcohol use. and marijuana use among seventh-grade students. The 1988 National Household Survey on Drug Abuse
revealed significantly higher lifetime prevalence rates for marijuana use among those wuh some college educaiion as compared with those who had less than a high school education
(Adams et aL 1990). In contrasi. Robins and Ratcliff (1979)
found lhat extreme poveny. though not lower-class status per
se. was one of three factors that increased the risk of adult
aniisocial behavior, including alcoholism and illegal drug use.
among children who were highly antisocial in childhood.
In summary, whereas there appears to be a negative relationship between socioeconomic siatus and delinquency, a similar
relationship has not been found for the use of drugs by adolescents. Only when poveny is extreme and occurs in conjunction
with childhood behavior problems has it been shown to increase risk •"ir later alcoholism and drug problems.
4 \eishborhooddisorganization. Neighborhoods with high
population density, lack of natural surveillance of public places
(C. A. Murray 1983). high residential mobility phvsical deterioration, low levels of attachment to neighborhood (Hening &
Guest. 1985). and high rates of adult crime also have high rater,
of juvenile crime (Wilson &. Herrnsiein. 1985) and illegal drug
trafficking (Fagan. 1988). Simcha-Fagan and Schwartz (1986)
assessed the conte.xiual effects of neighborhood on delinqueno
and found that communiiy economic level and communitv disorder-criminal subculture were significantly related to orticially recorded delinquency.
When neighborhoods undergo rapid population changes,
victimization rates increase, even after accounling for race and
age differences (Sampson. 1986: Sampson. Castellano. & Laub.
198 1). Neighborhood disorganization has been hypothesized to
contribute to a deterioration m the ability of families to transmit prosocial values to children (W McCord & McCord. 1959.
Reiss. 1986; Shaw & McKay 1969). Although few studies of
i.^. ..borhood disorganization have explicitly ex^ni.-vH its relationship wuh drug abuse, a deterioration in parental labialization and supervision associated with neighborhood disorganization could also be expected to produce high raies of drug
involvement. More research is required to determine ihe effects
of neighborhood disorganization on adolescent drug abuse.
t
Individual
and Interpersonal
Faciors
Certain characteristics o f individuals and of their personal
environments are associated with a greater risk of adolescent
drug abuse. These characteristics are summarized below as
Risk Factors 5 through 17.
5. Physiological factors. Sensation seeking and low- harm
avoidance predio ^arly-onset alcoholism (Cloninger et al..
1988). Poor impulse control in childhood predicts frequent
marijuana use at age 18 (Shedler & Block. 1990). Zuckerman
(1987) has suggested that sensation seeking is linked biochemically to platelet monoamine oxidase (MAO) activity, which has
also been found to be associated wuh early-onset alcoholism
(Tabakoff &. Hoffman. 1988: von Knorring. Oreland. & von
Knorring. 1987).
The enzyme aldehyde dehydrogenase (ALDH). important in
the decomposition of ethanol in ihe body (Li. 1977). has also
been linked to alcoholism. Asians without one A L D H enzyme
d n n k less and have lower rates of alcoholism than conlrols
(Harada. Agarwal. Goedde. & Ishikawa. 1983: Schuckil. 1987.
Suwaki & Ohara. 1985).
Researchers have also studied differences in genetically mediated biological responses to alcohol among children of alcoholics and nonalcoholics. Pollock. Volavka. and Goodwin
(1983) reported more slow-wave activity on the electroencephalogram (EEG) fcr children of alcoholics compared wuh children of nonalcoholics. Schuckil. Parker, and Rossman (1983)
found differences in children of alcoholics and children of nonalcoholics in serum prolactin response io administration of alcohol. Schuckit (1980) reported greater muscle relaxation in
response to ethanol. and Schuckit and Raves (1979) found in-
�0019
RISK AND PROTECTIVE FACTORS FOR DRUG PROBLEMS
creased levels of acetaldehyde after admimsiration of alcohol in
sons of alcoholics when compared w ith sons of nonalcoholics.
Researchers have sought to assess the incependent coninbuuon of genetic factors to the development of alcoholism
through twin and adoption studies. Kaij (I960) and Hrubec
and Omenn (1981) found that among males, monozygotic
twins were more than twice as likely as dizygotic twins to be
concordant for alcoholism, although in a study of both females
and males. Gurling. Clifford, and Murray (1981) reponed concordance rates for alcoholism of 21% in monozygotic and 25%
in dizygotic twins.
Adoption studies in Denmark. Sweden, and the United
Slates have provided more consistent evidence for genetic
transmission of alcoholism in males, reponingratesof alcoholism ranging from 18% to 27% for the adopted sons of alcoholics
compared with only 5% to 6% for adopted males without a
biological alcoholic parent (Bohman. 1978: Cadoret. Cain. &.
Grove. 1980: Cadoret & Gath. 1978: Goodwin et aL 1974;
Goodwin. Schulsinger. Moller. Medmck. & Guze. 1977). No
consistent evidence for genetic transmiss;~i ' ilcoholism in
females has been reponed (R. M. Mun-., i Stabenau. 1982).
Note that the adoption studies suggesting a genetic factor in
male alcoholism also reveal that less than 30% of the sons of
alcoholics themselves become alcoholic. Furthermore, about
half of hospitalized alcoholics do not have a family history of
alcoholism (Goodwin. 1985). suggesting that factors other than
genetic predisposition also contribute to alcoholism.
It is beyond the scope of this article to review thoroughly the
recenl developments in this area of alcoholism studies. Research continues to point toward differences in physiological
responses to ethanol among sons of alcoholics (Schuckit. 1987)
and to other possible genetic and biochemical "markers" of risk
for alcoholism (K. Blum ei aL 1990: Tabakoff & Hoffman.
1988). Early-onset alcoholism lhat is associated with impulsi. .. -nd aggression apparently has a partial foundar • in individual physiological characteristics.
Little research has been conducted on genetic predisposition
and the abuse of drugs other than alcohol in humans, although
there is evidence from animal studies of a heritability in predisposition to barbiturate and morphine abuse (Marley. Miner.
Wehner. & Collins. 1986).
6 Family alcohol and drug behavior and attitudes. Families
affect children's drug use behaviors in a number of ways.
Beyond the genetic transmission of a propensity to alcoholism
in males, family modeling of drug using behavior and parental
attitudes toward children's drug use arc family influences related specifically to the risk of alcohol and other drug abuse.
Poor parenting praaices. high levels of conflict in the family,
and a low degree of bonding between children and parents
appear to increase risk for adolescent problem behaviors generally, including the abuse of alcohol and other drugs (Brook.
Brook. Gordon. Whiteman. & Cohen. 1990). In this section,
lhe family risk factors specific to alcohol and other drug abuse
are reviewed. Family faaors more generally predictiveofadolescent problem behaviors are reviewed in subsequent seaions.
Parental and sibling alcoholism (Cloninger. Bohman. Sigvardsson. & von Knorring. 1985: Cotton. 1979; Goodwin.
1985) and illegal drug use (G. M. Johnson. Schoutz. & Locke.
1984) increase the risk of alcoholism and drug abuse in chilr
r
dren. Parental drug use is associated with initiation of use by
adolescents (G. M. Johnson et al.. 1984: Kandel. Kessler. &
Margulies. 978: McDermott. 1984) and wuh frequencv of
marijuana :se (Brook et al.. 1990). Similar findings have been
reponed for adolescent drmking habits (Rachal et al.. 1982.
R. A. Zucker. 1979). G. M. Johnson et al. (1984) found lhat
parental use of marijuana was associated wuh adolescenis' use
of other illegal drugs, including cocaine and barbiturates.
Ahmed. Bush. Davidson, and lannotti (1984) examined the
effects of parental modeling of drug use on children's expectations to use drugs and on their drug use. In a study of 420
children in grades kindergarten (K) to 6. they found "salience."
a measure of the number of household users of a drug and the
degree of children's involvement in parental drug-taking behavior, to be the best predictor of both expectations to use and
actual use of alcohol. Salience was also a predictor of children's
cigarette and marijuana use. The imponance of number of
household users varied across substance. As the number of family members who used alcohol or manjuana increased, so did
the p'-obability that the child used or expected to use these
sutitances. For cigarette smoking, having or. household
member who smoked cigarettes almosi doubled the probability
that a child smoked or expecied to smoke, but additional
smokers in the home did not increase this probability further.
Note that Hansen et al.'s (1987) structural equation modeling
analyses using cross-sectional data indicated indirect, but not
direct, effects of parental modeling of drug use on children's
drug use in early adolescence. Parenial modeling was directly
related to friends' use of drugs, which, in turn, was related to
subjects' drug use. This finding is consistent with the findings
of Brook et al. (1990). whose combined longitudinal and crosssectional studies revealed that nondrug use and emotional stability in fathers enhanced the effect of peer nonuse of drugs and
that psychological stability in mothers offset the effects of peer
drug use.
Brook. Whiteman. Gordon, and Brook (1988) examined the
role of older brothers in younger brothers' drug use and found
that older brothers' advocacy o drugs and modeling of drug use
were both associated with younger brothers' use. They also observed an interaction pattern in which some of the negative
effeas of parental drug use were offset by the older brothers'
nonuse. Older brothers' and peers' drug modeling both were
more strongly associated with younger brothers' use than was
parental modeling of drug use.
McDermott's (1984)researchindicated that although parental drug use and adolescent drug use are related, suggesting the
modeling effect discussed above, permissive parental attitudes
toward drug use as perceived by youths may be of equal or
greater importance than actual parental drug use in determining the adolescent's use of drugs. This finding is consistenl with
Hansen et al.'s (1987) results. Similarly, Barnes and Welte (1986)
found that par :ntal approval of drinking was a significant predictor of the amount of alcohol consumed by teenage drinkers,
and Brook. Gordon. Whiteman. and Cohen (1986) found that
parental tolerance of drug use prediaed adolescent drug use.
This relationship has been shown for Whites. Hispanics. African Amencans. Native Americans and Asian Americans (Jessor. Donovan. & Windmer. 1980).
" Poor and inconsisienl familv managemeni practices
r
�00 2 u
D H A W K I N S . R. C A T A L A N O . A N D J M I L L E R
kandel and Andrews (1 987) found that lack of maternal involvement in activities wuh children: lack of. or inconsistent, parental discipline (see also Baumrind. 1983. Penning &. Barnes.
1982): and low parental educational aspira .ons for their children predict initiation of drug use. Stanton (1979). Kaufman
and Kaufman (19791. and Ziegler-Driscoll (1979) suggested
that familial risk factors include a pattern of overinvolvement
by one parent and distance or permissiveness by the other.
Differences between the effects of mothers' and fathers' disciplinary techniques were observed by Brook et al. (1990). Maternal control techniques were more important than paternal
techniques in explaining adolescent marijuana use. Specifically, mothers" control patterns that included setting clear requirements for responsible behavior led to less marijuana use.
and mothers' use of guilt to control was correlated with greater
drug use.
Baumnnd (1983) classified parenting styles as authoritative,
authoritarian, or permissive and found that children who were
highly prosocial and assenive generally came from authoritative families. She found that parental nondirectiveness or permissiveness contributed to higher levels of drug use. Reilly
(1979) found that common charactensncs of families with adolescent drug abusers included negative communication patterns (criticism, blaming, lack of praise), inconsistent and unclear behavioral limits, and unrealistic parental expectations of
children.
Shedler and Block (1990) found that the quality of mothers'
interaction with their children at age 5 distinguished children
who were frequent users of manjuana at age 18 from those who
had only experimented with manjuana use. Mothers o f children who became frequent users were relatively cold, underresponsive. and underprotective with their children, giving their
children liltle encouragement but pressuring them to perform
in tasks.
Norem-Hebeisen. Johnson. Anderson, and Johnson (1984)
Jso found that the quality of adolescents' relatior^hips wjth
their parents was related to patterns of drug use. Generally,
drug users perceived their fathers as more hostile, rejecting, and
adversarial than did nonusers.
The evidence suggests an independent contribution o f family
interactions to adolescent drug use. separate from the effects of
parental drug use. Tec (1974) found that parental drug use in a
rewarding family structure only slightly promoted frequent
marijuana use but that in a unrewarding context, there was a
clear association between levels of drug use by parents and their
children.
In summary, the risk o f drug abuse appears to be increased
by family management practices charaaerized by unclear expectations for behavior, poor monitoring of behavior, few and
inconsistent rewards for positive behavior, and excessively severe and inconsistent punishment for unwanted behavior.
8. Family conflict. Although children frorr homes broken
by marital discord are at higher risk of delmq icncy and drug
use (Baumrind. 1983; Penning & Barnes. 1982. Robins. 1980).
there does not appear to be a direct independent contribution
of "broken homes" to delinquent behavior (Wilson & Herrnsiein. 1985). Conflict among family members appears more
imponant in the prediction of delinquency than does family
structure per se (Farrington. Gallagher. Moriey, Ledger. &
West. 1985; McCord. 1979: Rutter &: Ciller. 19S3V Rutter and
Giller have noted that parental conllict is associated wuh antisocial behavior in children even when the home is unbroken
(see also W McCord & McCord. 1959: Poner & O'Leary. 1980)
and that even in samples in which all homes are broken, the
extent of family conllict is associated wuh the likelihood of
antisocial behavior in the children (see also Hetheringlon. Cov.
& Cox. 1979: Wallerslein &. Kelly 1980). Similarly. Simcha-Fagan. Gersten. and Langner (1986) found that the use of heroin
and other illegal drugs was strongly associated wuh parental
marital discord. In summary, children raised in families high
in conflict appear at risk for both delinquency and illegal
drug use.
9. L o * bonding to family Parent-child interactions characterized by lack of closeness (Brook. Lukoff. & Whiteman. 1980:
Kandel et al.. 1978) and lack of maternal involvement inactivities with children (Braucht. Kirby. & Berry. 1978: Penning &
Barnes. 1982) appear to be related to initiation of drug use.
Conversely, positive family relationships—involvement and attachment—appear to discourage youths' initiation into drug
use -ooketal.. 1986; Gorsuch & Butler.! 976: Jes;jr . .'essor.
1977; K i m , 1979: Norem-Hebeisen et al.. 1984; Selnow. 1987).
Hundleby and Mercer (1987) found that adolescents' repons of
parental trust, warmth, and involvement explained small portionsofthe variance in the extent of tobacco, alcohol, and marijuana use.
r
Bonding to family may inhibit drug involvement during adolescence in a manner similar to the way in which family bonding inhibits delinquency (Hirschi. 1969). Brook et al. (1990)
pointed to the salience of parent-child attachment in describing the pathways to marijuana use frequency in their combined
longitudinal and cross-sectional studies. They reponed a
causal pathway in which parental internalization of traditional
values led to the development of strong parent-child attachment: this m u f a l attachment led to the child's internalization
of traditional nonr.; ?"d behavior, which in turn led t ' youi..ster to associate with non-drug-using peers, which led to
nonuse.
L
10 Early and persistent problem behaviors. The greater the
variety, frequency, and seriousness of childhood antisocial behavior, the more likely antisocial behavior is to persist into
adulthood (Robins. 1978).
A longitudinal study of 5-year-olds followed into adulthood
(Lerner & Vicary, 1984) found lhat a difficult temperament,
including frequent negative mood states and withdrawal, contributes to drug problems. Children charactenzed by withdrawal responses to new stimuli, biological irregularity, slow
adaptability to change, frequent negative mood expressions,
and high intensity of positive and negative expressions o f affect
more often became regular users of alcohol, tobacco, and marijuana in adulthood than "easy" children, who evidenced
greater adaptability and positive affect early in life. Similarly.
Shedler and Block (1990) found that frequent manjuana users
at age 18 were characterized in childhood by emotional distress. Lerner and Vicary (1984) suggested that the negative
mood and withdrawal responses o f the difficult child may be
analogous to the depression and social alienation frequently
reponed for drug abusers (Knight. Sheposh. & Bryson. 1974;
�002!
RISK AND PROTECTIVE FACTORS FOR DRLG PROBLEMS
Paton& Kancel. 1978. Paton. Kessler. & Kandel. 1977-. Smith &
Fogg. 1978).
BrooK et al. [1990) found that children who were irritable,
easily distractible. had temper tantrums, fought often with siblings, and engaged in predelinquent behav ior were more likely
to use drugs in adolescence.
Aggressive behavior in boys appears to signal another path
toward later antisocial behavior. Aggressiveness in boys as early
as ages 5-7 (Grades K-2) has been found to predict later antisocial behavior including frequent drug use in adolescence (Kellam & Brown. 1982). drug problems in adulthood (Lewis.
Robins. & Rice. 1985; Nylander. 1979). and delinquency in adolescence (Loeber. 1988; Spivack. 1983). However, early aggressiveness is not invariably followed by senous antisocial behavior. Approximately 30% to 40% of the boys engaged in maladaptive, aggressive behaviors continue that behavior 4 to 9 years
later (Loeber & Dishion. 1983).
Few youths develop highly physically aggressive behaviors in
late childhood or adolescence if not engaged in such behaviors
m e-rlier childhood, and most boys grow out of early aggressive
behaviors. How .cr, if aggressive behavioi cc itinues into early
adolescence (age 13). it is a relatively strong predictor of continued aggressive behavior in late adolescence as well as of later
alcoholism (Loeber. 1988; McCord. 1981). Funhermore. ifant.social behavior persists and becomes more varied in early adolescence to include fighting and school misbehavior, drug abuse
is more I ikely (Barnes & Welte. 1986; Kandel. 1982). Bames and
Welte found that school misconduct was one of the three most
imponant predictors of alcohol-related problems in a study of
subjects from six ethnic groups in Grades 7-12.
Hyperactivity and attention-deficit disorders have been
shown to increase risk for delinquency when combined with
conduct problems including aggression (Loney. Kramer. & Milich. 1979). Giltelman. Mannuzza. and Bonagura (1985) found
a higher prevalence of substance abuse disorders in late adolescence among subjects diagnosed as hyperactive in ..'iild ?od.
As with delinquency, those at highest risk were those with both
hyperactivity and conduct disorders. The Giltelman et al. finding that hyperactivity, without accompanying conduct problems, predicts an increased risk of substance abuse has not been
replicated. However, it suggesis funher investigation into the
relationship beiween attention-deficit disorders, conduct problems, and substance abuse.
/ / Academic fadure. Although there is an inverse relationship between intellectual ability and delinquency after controlling for socioeconomic status and race (G. D. Goltfredson.
1981). a similar relationship has not been reported for drug use.
in spite of the covariation in delinquent and drug-using behaviors. In fact, in an African-Amencan inner-city sample, higher
scores on reading readiness and IQ tests in Grade I predicted
earlier and more frequent use of alcohol in adolescence (Fleming. Kellam. & Brown. 1982). Similarly, in a national probabilily sample, high intelligence, as assessed by the Armed Forces
Qualifying Test, was associated with higher lifetime levels of
cocaine use among young adulls age 19-26 (Kandel & Davies.
1991).
Nevertheless, failure in school has been identified as a predictor of adolescent drug abuse (Jessor. 1976; Robins. 1980).
Poor school performance has been found to predict frequency
u
and levels of use of illegal drugs (Smith < Fogg, 1978). Holm&
berg (19;?5). m a longitudinal studv of 15-year-olds, reponed
that truancy placement m a special class, and early drop out
from school were prognostic faciors for drug abuse. In contra' •;.
outstanding performance in school reduced the likelihood of
frequent drug use among a ninth-grade sample studied b>
Hundleby and Mercer (1987).
W hat is not clear from the existing research is when, developmentally, poor school achievemeni becomes a stable predictor
of drug abuse. The available evidence suggesis that social adjustment is more imponant than academic peribrmance in the
early elementary grades in predicting later drug abuse. Eariy
aniisocial behavior in school may predict both academic failure
in later grades (Feldhusen. Thurston. & Benmng. 1973) and
later drug abuse. Academic failure in late elementary grades
may exacerbate the effects of early antisocial behavior or contribute independently to drug abuse.
12. Low degree of commitmeni io school. A low degree of
commitment to education also appears to be related io adolescent drug use. Annual surveys of high school seniors by Johnsion. O'Malley, and Bachman (1985) show that ihe useo!> 'lucinogens. cocaine, heroin, stimulants, sedatives, or nonmedically prescribed tranquilizers is significantly lower among
students who expect to attend college than among those who do
not plan to goon to college. G. D. Goltfredson (1988) found that
truancy for both boys and girls was associated with drug i nvolvement. after accounting for effects of ethnicity, parental education, and delinquency. Factors such as how much students like
school (Kelly & Balch. 1971). time spent on homework, and
perception of the relevance of course work are also related to
levels of drug use (Friedman. 1983), indicating a negative relationship between commitment to educaiion and frequent drug
use among junior and senior high school students.
13. Peer rejection in elememary grades. Although it would
be premature to posit a direct link between peer rejection and
substance abun IOW acceptance by peers seems to put an adolescent at risk for school problems and criminality (Coie. 1990;
Kupersmidt. Coie. & Dodge. 1990; Parker & Asher. 1987).
which are also risk factors for substance abuse (Hawkins.
Lishner. Jenson. & Catalano. 1987).
Little research has been done on the direct link between peer
rejection and substance use. but trails of ihe child that have
been associated with peer rejection—aggressiveness, shyness,
and withdrawal—have been examined for their relationship to
drug use. For example. Kellam. Ensminger. and Simon (1980)
found that children who had been shy in first grade reported
low levels of involvement in drug use. whereas those who had
been aggressive or had shown a combination of aggressiveness
with shyness in first grade had the highest levels of use. Brook ei
al. (1986) found lhat childhood traits relevant to peer rejection
—social inhibition, isolation from peers, and aggression
against peers—were not significantly associated with adolescent drug use stage. However, aggression against peers during
adolescence was associated with stage of use. and teenagers
who were less socially inhibited and less isolated from peers
were likely to be at a more advanced stage of use.
These studies suggest that the link between peer rejection
and subsequent drug use may not e a simple one. Shyness, bv
isolating a child from his or her peers, may proiect the child
w
�oo
-<\,
D HAWKINS. R CATALANO. A N D J. M I L L E R
agams; drug use bv eliminating one source of influence to use:
drug-using peers. Aggressiveness, on the other hand, ihough
resulting forsomechildren in exclusion from groupsofconventional peers, mav be associated wuh acceptance by other aggressive and perhaps delinquent peers who could foster drug
use (Cairns. Cairns. Neckerman. Gest. & Gairepy. I98S).
Hartup (1983) suggested that rejected children form friendships with other rejected children during the preadolescent
years and that these friendship groups become delinquent during adolescence. However, this process is as yet unconfirmed
(Tremblay. 1988).
14. Association with drug-using peers. Peer use of substances has consistently been found to be among the strongest
predictors of substance use among youth (Barnes & Welte.
1986; Brook et al.. 1990; Elliott. Huizinga. & Ageton. 1985:
Jessor et al.. 1980: Kandel. 1978. 1986: Kandel & Andrews.
1987). Studies among specific ethnic groups confirm this relationship. Newcomb and Bentler (1986) reported that the influence of peers on adolescent drug use was stronger than that of
parents for^Caueasian^'African Amencans. Asian Americans
and Hispanic Amencans. Similar findin- were reponed by
Byram and Fly (1984). Harford (1985) tound that AfricanAmerican youths who did nol dnnk alcohol reponed fewer
school friends who drank than did those who drank, and
Dembo et al. (1979) found that friends' use of alcohol and marijuana was related to a youth's own use for both African-American and Pueno Rican-Amencan youths.
15 Alienation and rebelliousness. Alienation from the dominant values of society (Jessor & Jessor. 1977; Kandel. 1982;
Penning & Barnes. 1982). low religiosity (Jessor et aL 1980:
Kandel. 1982: Robins. 1980). and rebelliousness (Bachman et
al.. 1981: Kandel. 1982) have been shown to be positively related
to drug use and delinquent behavior. Shedler and Block (1990)
found that interpersonal alienation measured at age 7 predicted
frequent marijuana use at age 18. Similarly, high tolerance of
de^ance (Jessor & Jessor. 1977). a strong need for independence (Jessor. 1976). and normlessness (Paton & Kandel. 1978)
have all been Iinked with drug use. All these qualities would
appear to characterize youths who are not bonded to society.
16. Attitudes favorable to drug use. Research also has shown
a relationship between drug use initiation and specific attitudes
and beliefs regarding drugs. Initiation into use of any substance
is preceded by values favorable to its use (Kandel et aL 1978:
Krosnick & Judd. 1982; Smith & Fogg. 1978).
17. Early onset of drug use. Eariy onset of drug use predicts
subsequent misuse of drugs. Rachal et al. (1982) reponed that
misusers of alcohol appear to begin drinking at an earlier age
than do users. The earlier the onset of any drug use. the greater
the involvement in other drug use (Kandel. 1982) and the
greater the frequency of use (Fleming. Kellam. & Brown. 1982).
Earlier initiation into drug use also increases the probability of
extensive and persistent involvement in the use of more dangerous drugs (Kandel. 1982) and the probability of involvement in
deviant activities such as crime and selling drugs (Brunswick &
Boyle. 1979; O'Donnell & Clayton. 1979). Robins and Przybeck
(1985) found that the onset of drug use before the age of 15 was a
consistent predictor of drug abuse in the samples they studied.
Conversely, a later age of onset of drug use has been shown to
predict lower drug involvement and a greater probabiiitv oi
discontinuation of use ( Kandel. Sinele. & Kessler. I 976).
Implications of Research on Risk
A risk-focused prevention approach requires identification
of those risk factors io be addressed. L-nfonunatelv. all ihe information needed to select the most promising risk factors for
intervention is not vet at hand. Experimental research is
needed to discover which risk factors are causal and which are
spurious in the etiologv of drug abuse. Onlv by addressing risk
factors in experimental trials and observing the effects on drug
abuse can one determine whether a precursor of drug abuse is
causally related to drug abuse. Experimental prevention research is therefore necessary both to understand the etiology of
drug abuse ana to determine which risk factors should be targeted in preven' on policy and programs.
Several general conclusions regarding risks for drug abuse
can be drawn, which have implications for prevention. First,
the risk factors reviewed above have been shown to be stable
over ' ne in spite of changi";: norms. For example, despite
general changes in norms regarding the use of drugs such as
marijuana over the past 20 years (Johnston. O'Malley, &. Bachman. 1989). studies conducted in different times and places
have shown these factors to predict adolescent drug abuse relatively consistently. This suggests the risk factors" stability as
predictors and their viability as targets for preventive work.
Second, risk factors from several domains predict drug
abuse. Some factors are characteristics of the individual; oihers
are characteristics of families and their interactions, schools
and classroom experiences, peer groups, and broader community, legal, economic, and cultural factors.
Third, different risk factors are salient at different periods of
development. For example, poor academic achievement in
Grades 1 and 2 does not appear to be a stable predictor of
teenage dru^. abuse (Kellam & Brown y$2). though poor
achievement in the later grades predicts drug abuse. Aggressiveness ai ages 5-7 predicts later drug abuse and. if u continues,
becomes more strongly predicnve of drug abuse with increasing age.
Founh. there is evidence that the more risk factors present,
the greater the risk of drug abuse (Bry. McKeon. & Pandina.
1982: Newcomb et aL 1986). Rutter (1980) found a multiplicative effect of added risk factors on the likelihood of childhood
psychopathology, and Newcomb. Maddahian. Skager. and
Bentler (1987) reponed a similar contribution of combinations
of different risk factors to overall risk for adolescent drug use. It
is plausible that a greater length of exposure to environmental
risk factors exacerbates risk as well . Current research focuses on
how risk factors interact in the etiology of drug abuse (Loeber &.
Stouthamer-Loeber. 1986). Greater precision in estimating how
much vanous nsk factors contribute to drug abuse will help to
focus prevention effons on those risk factors that are most virulent.
A risk-focused prevention approach does not require that risk
factors be manipulated directly It may be impossible to reduce
or chanee cenain risk factors directly through preventive intervention. In these instances, the goal of prevention efforts will
�RISK \ N D PROTECTIVE FACTORS FOR DRLG PROBLEMS
be to mediate or moderate the etiects of the identified but nonmampulable risk factors. A familv historv of alcoholism, for
example, may be difficult or impossible to change. Nevenheless. it may be possible to moderate the effects ofa famils historv of alcoholism bv intervening wnh children who are at risk
because of their exposure to this environment. One task of
risk-focused prevention research is to determine which risk factors can be manipulated, which risk factors cannot be changed
but can be mediated or moderated, and which risk factors cannot be affected at all.
Protective Factors Against Drug Abuse
Because some risk factors for drug abuse may be resistant or
impossible to change, the results of research on protective factorsare important for prevention policy Protective factors mediate or moderate the effects of exposure to risk (Cowen & Work.
1988:Garmezy. 1985: Rutter. 1985-. Werner. 1989). To the extent
that protective factors are identified that inhibit drug abuse
among those at risk, strategies can seek to address risk by enhancing these protective factors. Research with populations exposed to multiple risKS has identified substantial subgroups of
individuals who are able to negotiate risk exposure successfully,
escaping relatively unscathed (Werner. 1989). These observations have led to interest in the etiological importance of factors
that may protect against health problems including drug abuse.
Concepts of vulnerability and resiliency have been advanced
to identify the extent of individual susceptibility to risk (Rutter.
1985). Vulnerability denotes intensified susceptibility to risk;
resiliency is the ability to withstand or surmount risk. From
this perspective, protection involves enhancing resilient responses to risk exposure. The hypothesis is that certain characteristics or conditions mediate or moderate the effects of exposure to risk, thereby reducing the vulnerability and enhancing
the resiliency of those at risk and protecting them from undesirable outcomes. To illust.uie. Werner and Smith (1982) found
that in rural Kauai. Hawaii, being raised in a small family with
low conflict, having high intelligence, and being a firstborn
child buffered the effects of extreme poveny and other risk
factors for poor educational, economic, a'-d health outcomes.
For the concept of protective factors—as distinct from risk
factors—to be useful, it must apply to differences in outcomes
among individuals exposed to the same risks. Though some
have viewed protective factors simply as the opposite of those
variables identified as nsk faaors (Labouvie & McGee. 1986).
this conception does not appear pamculariy useful. Designating two distinct constructs (eg., risk and protective factors) tc
distinguish extreme levels of a single variable bearing a linear
relationship to drug abuse adds little. It is not necessary to
postulate protective factors if belter outcomes are observed in
those not exposed to risk. On the other hand, if protective factors are viewed as sources of differences in response to a given
amount of exposure to risk, the construct stimulates attention
to nonlinear and interactive relationships among risk and protective factors.
In urging a focus on protective mechanisms. Rutter (1985)
described interacuve processes to identify multiplicative interactions or synergistic effects, in which one variable potentiates
00:3
the effect of another. The idea of identifying protective processes or specifying panicular interactions among variables
that produce an enduring shield or resilience in the face 01 risk
for negative outcomes has direct relevance for risk-focused drug
abuse prevention. It suggests that the goals of risk-focused prevention may be accomplished both through direct effons at
risk reduction and through the enhancement of protective factors that moderate or mediate the effects of exposure to risk.
Preventive work that seeks to address risk factors for drug abuse
must clearly hypothesize how a panicular intervention is expected to address risk: by directly eliminating or reducing a risk
factor or by mediating or moderating its effects through the
enhancement of proteaive factors or processes.
Little research has focused specifically on protection against
adolescent drug abuse defined in this way However, recently
Brook et al. (1990) identified two mechanisms by which protective factors reduce risk for adolescent drug use. The first is a
"risk/protective" mechanism through which exposure to risk
factors is moderated by the presence of protective factors. They
reponed that the risk posed by drug-using peers was moderated
by a strong attachment or bond between parent and adolescent
and by parent conventionality. The second is a * protec i.ve< protective" mechanism through which one protective factor potentiates another protective factor, strengthening its effect. They
reponed that a strong bond of attachment between adolescent
and father enhanced the effects of other protective factors such
as adolescent conventionality: positive maternal charactensncs.
and marital harmony m preventing drug use.
In related research areas. Garmezy (1985) has identified protective factors among children exposed to extreme stress because of highly disturbed family circumstances. These include
a child's own positive temperament or disposition, a supponive
family milieu, and an external suppon system that encourages
and reinforces the child's coping effons and strengthens them
by inculcating positive values. Rutter (1985) has suggested that
resilient children display a repenoire of social problem solving
skills and belief in their own self-efficacy.
In designing interventions to reduce the negative effects of
identified risk factors, it is imponant to focus attenlion on the
potential positive effects of such protective factors. The available evidence suggests that to be viable, a prevention strategyrequires attention 10 risk and protective factors related to individual vulnerability, poor child rearing, school achievement,
social influences, social skills, and broad social norms, all of
which are implicated in the development adolescent drug
abuse. Because risks are present m several social domains and
cumulate in predicting drug abuse, multicomponent prevention strategies focused on reducing multiple nsks and enhancing multiple protective factors hold promise. Such strategies
would be designed to build up proteaion while reducing risk.
Each risk factor targeted should be addressed dunng the developmental period at which it begins to stabilize as a predictor
of subsequent drug abuse. Interventions must also target populations at greatest risk—groups and individuals who are exposed to a large number of risk factors—if the prevalence of
drug abuse, as defined here, is to be reduced through prevention efforts. Although intervention with people who are not
exposed to multiple risk factors may delay or prevent the onset
�1)0 24
D. H A U K INS. R. C A T A L A N O . A N D J. M I L L E R
of drug use in the general population, a desirable goal in its o* n
right, it may fail to reduce signiricanily the prevalence of drug
abuse.
The e\ idence suggests a developmentally adiusted. multiplecomponent risk-reduction strategy that cuts across traduionai
health, educaiion. and human service delivery systems. The
strategv must reach those at highest risk bv virtue of exposure
to multiple risk factors. It must address the mosl significant risk
factors faced by those groups. Finally, the strategy may explicitly seek to increase protective faciors as mediators or moderators against risks that cannot be changed by imervennon.
Using Theory to Guide Prevention
Research and Practice
To design a multicomponent intervention strategy that seeks
to reduce multiple risk factors and simultaneously enhance protective factors among those exposed to risk, it is useful to be
guided by a theory of causation and prevention. Theory supplies the explanatory framework for the observed evidence regarding risk and protective factors for drug abuse by hypothesizing causal relationships among these variables that lead toward or away from drug abuse. Theory is also useful in guiding
the design of complementary prevention interventions in different social units when multiple interventions are desired. To
guide prevention intervenlions. theory should (a) identify the
factors that predict drug abuse, (b) explain the mechanisms
through which they operate, (c) identify the factors that inlluence these mechanisms, (d) predict points to interrupt the
course leading to drug abuse, and (e) specify the interventions to
prevent onset of drug abuse (Kazdin. 1990).
It is not our goal to review theories of drug abuse (see Lettieri.
Savers. & Pearson. 1980. for a review). Nevertheless, an example
illustrates how theory can provide clear direction for preventive
interventions of the type described here.
As noted by Kazdin (1990). our delinquency and drug abuse
prevention efforts have been grounded in the social development model (Farrington & Hawkins. 1991: Hawkins & Lam.
1987; Hawkins & Weis. 1985). An integration of control theory
(Hirschi. 1969) and social learning theory (Bandura. 1977). the
social development model emphasizes the role of bonding to
prosocial family, school, and peers as a protection against the
development of conduct problems, school misbehavior,
truancy, and drug abuse. This concepi of bonding is closely
related to the concept of attachment as defined by Bowlby
(1969. 1973) and as observed by Brook et al. (1990) to inhib't
adolescenl drug abuse. It is also consistent with Garmezy's
(1985) identification of familial and external support and value
systems as protective faaors against exposure to stress in childhood.
Four elements of social bonding have been shown to be inversely related to drug use. These are strong attachment to parems (Brook. Brook, et aL 1990; Brook. Gordon, et al. 1986:
Hundleby & Mercer. 1987; Jessor & Jessor. 1977; Norem-Hebeisen et al. 1984); commitment to schooling (Friedman. 1983.
Johnston. Bachman. & O'Malley. 1981: Kim. 1979: Krohn &
Massey. 1980): regular involvement in church activities (Schlegel & Sanborn. 1979; Wechsler & McFadden. 1979): and belief
in the generalized expectations, norms, and values of society
(Akers. Krohn. Lanza-Kaduce. & Radosevich. 1979: Krohn i
Massey. 1980) Research has not yet determined whether these
elements of social bonding are best viewed simply as the opposite extremes of variables already identified as risk factors for
drug abuse (e.g.. low commitment to schooling and alienation
and rebelliousness) or whether sociai bonding represents a distinct protective lactor capable of buffering the effects of other
risk factors such as a familv history of alcoholism or extreme
poveny. Further research on this question is needed.
The social development model specifies hypotheses regarding the processes that produce bonding to a social unit. Interactions among (a) opportunities for involvement offered in each
social unit, (b) the skills used by individuals in these social
units, and (c) the reinforcements offered in these units are hypothesized to produce social bonds of attachment, commitment, and belief in the values of the social units in which young
people develop (see Catalano & Hawkins. 1986)
Guided by this social development perspective, our own riskfocused prevention work has two purposes: (a) to understand
better the processes by which risk and protective factors contribute to the etiology of drug abuse in adolescence and (b) to test
promising approaches to prevent adolescent drug abuse. W'e
hy pothesize that children who develop strong bonds to social
units holding norms antithetical to drug abuse will be less likely
to abuse drugs.
To enhance social bonding, we manipulate social settings
and individual capacities using the principles of social learning
theory in developmentally appropriate ways. For example, in
the school setting, we train teachers in methods of proactive
classroom management, interactive teaching, and cooperative
learning, including students in peer teaching. The explicit,
theory-driven objectives of these intervention elements are (a)
to make available opportunities for children to be involved in
prosocial activities, (b) to provide skills needed to undertake
these activities successfully, and (c) to provide positive reinforcement for successful involvement. All of these objectives serve
the broader goal of strengthening bonding to the social unit, in
this case the school. From a social development perspective, the
same three objectives guide interventions with parents, daycare providers, youth ministers, recreation workers, and oihers
participating in the socialization of children. The framework of
the social development model thus fosters a multicomponent
prevention approach, grounded in knowledge of risk and protective factors and consistent in goals, across a variety of social
settings.
Current Risk-Focused Drug Prevention
During the 1960s and 1970s there was little explicit attention
to risk or protective factors for drug abuse in the design and
development of preventive interventions. More recent research
on drug abuse prevention has focused on risk reduaion. but has
not included attention to multiple nsk or protective factors and.
for the most part, has not addressed risk faaors that appear
developmentally before the age of likely drug use initiation.
Most recent prevention research has targeted only two risk factors for drug abuse, both of which are most salient just at the
point of drug use initiation: (a) laws and norms favorable to drug
use and (b) social influences to use drugs. Approaches targeting
�RISK AND PROTECTIVE FACTORS FOR DRLG PROBLEMS
these risk factors art designed tor a relatively quick "return.'' in
that if they are effective, they should reduce or curtail drug use
immediately Prevention approaches that target these risk factors are summarized below: the right half of Table 1 summarizes the elfects of these approaches on the risk factors addressed.
Supply .Vlanipulmion.
Strategies
Inwrdiction.
and
Enforcement
Attention to the laws and norms of society related to the use
of alcohol and other drugs is clearly warranted, given the link
between these factors and rates of alcoholism and drug abuse. If
reduction of the prevalence of abuse of drugs is the goal, the
evidence does not suppon those who advocate the legalization
of currently illegal drugs such as marijuana and cocaine (Clayton, in press). Rather, the evidence suppons effons to limit
behavior that is inconsistent with existing legal sanctions. This
has been attempted through effons to control the supplies of
both legal and illegal drugs.
Since the repeal of prohibition, the supply of alcohol has
been manipulated in several ways, including taxation, age restrictions on consumption, restrictions on hours of purchase,
and restnctions on liquor-by-the-drink sales. As noted earlier,
restricting availability and increasing the price of alcohol by
increasing taxes on the purchase price can reduce rates of alcohol abuse as indicated in rates of cirrhosis of the I iver and alcohol-related traffic fatalities. Although increasing age restrictions on alcohol purchases and restrictions on liquor-by-thednnk sales appears less effective than taxation in limiting
alcohol abuse, these strategies also have shown desirable effects
in reducing alcohol-related traffic fatalities (Blose & Holder.
1987; Decker. Graitcer. & Schaffner. 1988: Kneg. 1982).
This evidence might appear to imply that supply manipulation strategies such as drug interdiction and arresls of drug
u^jiers would have a similar desirable effect on the abuse of
illegal drugs by raising the price of these drugs to the user.
However, existing evidence does not suppon this contention.
Analysis by the Rand Corporation resulted in the conclusion
that neither a doubling of interdiction nor increased arrests of
drug dealers would affect retail prices dr the availability of illegal drugs (Polich. Ellickson. Reuter. & Kahan. 1984). Dais
from the Drug Enforcement Administration confirm this conclusion. In spite of an increase in federal spending on interdiction and law enforcement from $1.807 billion in 1986 to $3,770
billion in 1989. the average street price ofcocaine fell from $100
to $75 dollars per gram during the same period. Although at
some level well beyond current spending, interdiction and enforcement might reduce drug supplies and drive up pnees. the
fiscal costs, effects on U.S. international trade, and constraints
on individual rights required would be excessive. Moreover, in
this scenario, if demand for illegal drugs were not reduced, it is
plausible that domestic producers of synthetic drugs would step
in to fill demand as interdiction began to reduce drug supplies,
ihus continuing to hold down prices to users.
In our view, the most powerful effect of interdiction and enforcement activities is to communicate general social norms of
disapproval for the distribution and use of illegal drugs. Social
norms antiihetical to use appear associated wuh reductions in
0025
the prevalence of the frequent use of marijuana ( Robins. . 9s41
and other illegal drugslJohnston. 1 99 11. Supply-reduction stra:egies communicate an imponant message to citizens but
should not be expected, by themselves, to eliminate illegal c:ui
supplies, to significantly raise the price of illegal drugs, or ;o
eliminate drug abuse. Those who are at greatest risk of drug
abuse by v mue of low social bonding to society may view tne
relative benefits of drug dealing and drug use as wonh the risks
of apprehension. The prevention of alcohol and other drug
abuse among those at greatest risk requires attention to the
ractors that distinguish these people. The risk factors encountered by these persons at highest risk must be addressed to
reduce the demand for illegal drugs.
Changing Social
Norms
A second approach currently emphasized is changing social
norms about drug- and alcohol-influenced behaviors. The approach includes "Just Say 'NO!' " activities, community coalitions against drugs, media campaigns, and cenain policy
changes.
C. A. Johnson and Solis (1983) and Pern.; Klepp. and Shultz
(1988) reviewed a number of com munity health promotion programs aimed at reducing cardiovascular disease by changing
smoking and other risk-related behaviors. These programs included involvement of the mass media, risk-factor screening
programs, and education programs for adults and youths. The;,
have been associated with lower smoking onset rates among
youlhs (Perry, Klepp. & Shultz. 1988) and cessation or reduction of smoking (C. A. Johnson & Solis. 1983).
Of panicular interest in this area is the influence of advemsing on drug use. There is some indication lhat higher exposure
to "life-style" ads promoting alcohol consumption is found
among adolescents who repon higher levels of dnnking(Atkin.
Hocking, & Block. 1984).
The media ar.. /.dvenising industries have cooperated in a
national project to encourage negative attitudes toward the use
of illegal drugs through the use of antidrug advertising. Results
of mall intercept surveys indicate that saturation advertising in
10 markets was accompanied by significant normative changes
over a 1-year period (Black. 1989). College students and children were more negative in their attitudes toward drugs, viewed
drug users less positively, and perceived less drug use among
their friends in 1988 compared with 1987. Moreover, in areas
that received saturation advertising, 9% to 15% more children
reported conversations about drugs with parenis. teachers, and
siblings in 1988 than did in 1987. In the balance of the United
States, there were no increases in such communications. Teenagers age 13 through 17 showed the fewest changes in altitudes
in association with saturation advertising, though ihey became
more positive in their views toward nonusers and perceived
greater risks from marijuana and cocaine use (Black. 1989). Of
course, these differences could have been produced by other
factors operating in communities sufficiently concerned about
drug abuse that their broadcast media would run a saturationadvertising campaign against drugs.
Social norms regarding the use of specific drugs and their
attendant risks and benefits can change over a relatively short
period of time. From 1978 to 1983. the proponion of the na-
�00 2 ^
1 H A W K I N S . R. C A T A L A N O . A N D J. M I L L E R
3
uon's high school seniors who perceived ihere io be a great
health risk associated with the regular use ot marijuana rose
from 58 "': to 63". Over the same period, the proportion of the
nation's seniors who used marijuana dailv dropped from 10.7%
to 5 5 (Johnston. 1985).
%
An important question for studv concerns the role of broadly
focused norm-change effons. such as media campaigns, in producing such changes in norms and the frequent use of drugs.
Studies are needed that examine how these efforts affect children at greatest risk for drug abuse. It is not known how children who come from poorly managed families, who have failed
in school, who are aggressive, or who have lost commitment to
school respond to "Just Say 'NOr " or other antidrug messages
in the media or in their personal social environments.
Changes in social norms have also been codified in school
policies regarding drug-using behavior (Moskowitz & Jones.
1988). Recent studies of school policies regulating smoking
have shown that more comprehensive policies, which emphasize prevention of use and restrictions on opportunities for use
in or near school groundsappear to reduce the amount of smoking by students (Pentz. Brannon. et aL. 1989). although effects
on smoking prevalence are less consr'^nt. These policies appear to affect smoking behavior primarily through the clear
specification of norms regarding smoking rather than through
the enactment of punitive consequences for policy violations,
which have not shown effects in reducing smoking (Pentz.
Brannon. et al.. 1989). Additional research is needed on the
effectiveness of school policies in preventing or reducing the
use of drugs other than tobacco and on the effects of such
policies on those at highest risk for drug abuse.
Social Influence Resistance Strategies
As noted earlier, among the strongest correlates of teenage
drug-using behavior is association with others who use drugs. If
the relat'onship between association with drug-using peers and
drug-using behavior is actually causal, the manipulation of a
factor that accounts for a great deal of variance in drug use
would hold promise forproducingsignificant reductions in adolescent drug ose.
Prevention strategies focused on social influences to use
drugs also are appealing from a cost-effectiveness perspective.
Because peer influence to use drugs is salient developmentally
at the point of onset of drug use. long delays are not required
before effects of such interventions can be observed.
The most heavily researched strategy for addressing social
influences to use drugs is classroom-based skills training for
adolescents in Grades 5 through 10, most often Grades 6 and 7.
The training leaches students through instruaion, modeling,
and role play to identify and resist influences to use drugs and.
m some cases, to prepare for associated difficulties and stresses
anticipated in ihe process of resisting such influences (Botvin.
1986). Grounded in social learning theory (Bandura. 1977),
social influence resistance strategies view drug use as a socially
acquired behavior, initiated and reinforced by drug-using
oihers (Bukoski. 1986).
Whereas all programs of this type offer skills in resisting
social influences to use drugs, many also seek to promote
norms negative toward drug use (Hansen. Johnson. Flay. Gra-
ham, Sobel. I 988: Perry 19S6 i. These normative-change components have included efforts to depict drug use as socially
unacceptable: identification of short-term negative consequences of drug use: the provision of evidence that drug use is
not as widespread among peerc as children may ihink: encouragement ror children io make public commitments to remain
drug free: and. in some instances, the use of peer leaders to
teach the curriculum i Botvin. 1986: KJepp. Halper. & Perry
1986).
Social influence resistance approaches also have been combined wuh training in problem-solving and decision-making
skills, skills to increase self-control and self-efficacy, adaptive
coping strategies for relieving stress and anxiety, interpersonal
skills, and general assenive skills (Botvin. 1986: Flay, 1985). In
this regard. Boivm's skills training program has combined elements of both social influence resistance training and social
competence skills training discussed later (Botvin &. Wills.
1985). Recenl projects have also combined classroom-based social influence resistance curricula with mass-media programming and parent involvement strategies in comprehensive interventions seeking to change norms toward drug use and increase
.cj.fiance to drug-prone influences among aaoicsc'—< ' ?ntz.
Dwyer. et al. 1989).
Most published studies of social influence resistance strategies have found modest but significant reductions, in comparison with controls, in the onset and prevalence of cigarette smoking after training (see Botvin. 1986: Bukoski. 1986: Flay. 1985:
Moskowitz. 1989: D. M. Murray. Davis-Hearn. Goldman. Pine.
& Luepker. 1988: Tobler. 1986, for reviews). A few studies have
reported beneficial effects of the strategy in preventing or delaying the onset of alcohol or marijuana use (Botvin. 1986: Ellickson & Bell. 1990: Hansen et al. 1988: McAlister. Perry, Killen. Slinkard. &. Maccoby. 1980: Pentz. Dwyer. et al. 1989).
Student- or peer-led social influence resistance training interventions have achieved greater reductions in drug use, compared with inlerventions led by teachers (Botvin. Baker. Filazzola. & Botvin. 17 JKJ: Klepp et al.. 1986: McAlister. 1983: D. M.
Murray. Johnson. Luepker. & Mittelmark. 1984). This difference may reflect greater fidelity in implementation of the curriculum by peer leaders (resulting in greater skill acquisuion by
students: Botvin et al.. 1990). or the finding may reflect peer
leaders' stimulation of classroom norms antithetical to
drug use.
A number of research issues remain to be addressed regarding the effects of social influence focused intervention. One
question is whether smoking prevention programs, without
content specific to other drugs such as alcohol or marijuana,
have effects on the use of these drugs. Some studies suggest
there may be a generalized effect of these prevention programs
on alcohol and marijuana use by subjects (G. M. Johnson et al.
1984; McAlister et aL 1980). A related question is raised by
Ellickson and Bell (1990), who sought to extend the social influence model of smoking prevention to alcohol and marijuana.
Results were mixed. Modest reductions in drinking forstudents
at three risk lev. is were observed immediately after the teen-led
version of the program but disappeared at 1-year follow-up.
Exposure to the curriculum was associated with reductions in
smoking among baseline experimenters but increases in smoking among baseline smokers. Curriculum exposure was also
p
�oo:'
RISK A N D PROTECTIVE FACTORS FOR DRLG PROBLEMS
associated wuh reductions in both initiation and current use of
marijuana. The investigators speculate that the apparent erfectiveness of social influence approaches f' r tobacco and marijuana may reflect the generalized social norms against those
two substances but for alcohol social inlluence training is less
effective because societv has not developed a consensus against
its use.
Biglan. Glasgow, et al. (1987). on the other hand, found no
generalization of effects to alcohol or marijuana use of a smoking refusal skills training program. Others have found that cognitive and interpersonal skills training interventions reduced
tobacco use but had no effects on alcohol, marijuana, or other
drug use (Gersick. Grady. &. Snow. 1988). More research is
needed to determine whether preventing the onset of an early
behavior in a sequence, such as smoking in the progression of
drug use initiation, has effects on later behaviors in the sequence.
f
Another question is whether classroom-based social influence resistance interventions have significant effects on adolescents at greates* '•isk for drug ab'^e. In most social influence
resistance studies, risk groups have !
'.r.jd b;. different
levels of baseline use. usually in smoking behavior (i.e- regular
tobacco users, occasional tobacco users, and nonusersfjEHickson. Bell. Thomas. Robyn. & Zellman. 1988). Although this
approach can reduce smoking among students with parents
and friends who smoke (Botvm & Wills. 1985). few assessments
are available of effects on those at greatest risk for drug abuse by
virtue of exposure to multiple risk factors earlier in development.
Some social influence focused studies have looked at the
effects of preventive interventions on groups with special demographic characteristics that may be related to higher risk. Botvin et al. (1989) addressed a special population of urban African-American youngsters with a smoking prevention program
'hat was based on life skills training using cognitive behavioral
technique: O f several smoking outcomes examnu . .;.c only
significant effect observed was a smaller proponion of smokers
at posttest in the treatment than in the control group on the
basis of adjusted means for smoking status in the past month.
Some intervention effects were also observed for cognitive and
attitude variables such as knowledge of smoking consequences
and normative expectations.
In another study designed to examine effects on a specific
population. Schinke. Botvin, et al. (1988) tested a social competence/skills building intervention designed with cultural relevance for Native American adolescents. They found at posttest
and 6-month follow-up that subjects who received the intervention improved more than control subjects on measures of substance use knowledge, attitudes and interaaive skills, and selfreponed rates of tobacco, alcohol and drug use.
The same group o f investigators examined still another special population, children of blue-collar families, using a schoolbased social skills smoking prevention program (Schinke. Bebel. Orlandi. & Botvin. 1988). Lower use rates were observed
and validated among pupils who received the SKills-based intervention (as compared with discussion-based groups and control
groups) at 6. 12. 18. and 24 months follow-up. Although they
have isolated panicular populations and baseline user, none of
the social influence intervention studies have examined
whether program effects varv for groups characterized bv multiple risk factors predictive of heavy drug use.
Another question regarding social influence resistance programs i n v ves the durability of effects. Evidence from a number of rece u studies points to deterioration of initially positive
program effects as eariv as 2 to 3 vears and as long as S years
postinterventiont Botvm etal.. 1990: Flay etal.. 1989: Hansenei
al.. 1988: D. M. Murray. Pine. Luepker. & Pallonen. l989i
Long-term results from the Nonh Karelia Vbuth Smoking Prevention Project (Vaniainen. Pallonen. McAlister. Koskela. i :
Puska. 1983. 1986: Vaniainen et al.. 1990) offer a somewhat
more promising picture, with 4-year results favoring the schools
given two versions of the social influence intervention over
matched companson schools. At 8 years postinterveniion. onl--.
baseline nonsmokers showed significant program effects
(Vaniainen e t a l . 1990).
That program participants mav have equal or higher rates of
substance use than program controls by 2 years after intervention raises the question of whether social influence focused
interventions will have effects on drug abuse as defined here. To
date, virtually no prevention studies targeting drug abuse have
followed subjects long enough to assess effects in late adolescence or eariy adulthood.
Even if they are successful in reducing the prevalence of teenage drug use. social influence resistance strategies may have no
significant effect on the prevalence of teenage drug abuse as we
have defined it. The lifetime prevalence of alcohol use among
U.S. high school seniors in 1987 was 92.2%. but only 4.8% used
alcohol daily (Johnston et a l . 1988). Social influence resistance
programs could show positive effects in reducing the prevalence of alcohol use in the general population without affecting
this 5% at greatest risk for alcohol-related problems.
Although social influence focused interventions have been
most widely tested, many of these same questions apply to
interventions targeting other risk factors. To address these questions, research methodologists must grapple with the complexities of the multiple risk factors and causal pathways implicated
in substance abuse etiology.
Methodological
Research
Challenges
for Risk-Based
Imervennon
Although it is not our purpose to offer a methodological
evaluation of specific prevention studies, a number of research
issues that pertain to risk-focused investigations should be considered when weighing the evidence, individually and cumulatively, from the studies available (Moskowitz. 1989). Risk-focused prevention studies require research designs that address
threats posed by mixed units of analysis, differential attrition,
and differential implementation as well as the interpretive challenge presented by heterogeneous effects across risk groups and
along the developmental life course. Careful theoretical specification and rr jltiple and varied statistical analysis techniques
can be used to meet these challenges.
Mixed units of analysis. In many published studies of social
influence resistance programs, the basic premise of experimental design—that the randomized expenmental unit is the unit
of analysis—is violated. Schools or classrooms are generally the
unit of random assignment io experimental or control condi-
�o :s
o
D HAWKINS. R CATALAN' .. AND J MILLER
non. but the unit of analysis often used is individual students.
School or classroom differences are thus confounded wnh program effects on individuals (Biglan & Ary 1985). Some studies
have addressed this problem by assigning multiple schools or
classrooms to each condition, then analyzing at the classroom
level i Biglan. Severson. et al.. 1987; Botvin. Baker. Renick. Filazzola. & Botvin. 1984; Hansen et al.. 1988; Pentz. MacKinnon, et al.. 19891.
When scarce resources impose limits on the number of units
that can be randomly assigned, some alternative solutions have
been suggested. Randomized block and factorial designs can
be used to stratify' schools by factors known to affect key outcomes (McKiniay. Stone. & Zucker. 1989). Alternatively, to account for variability attributable to the school, multiple investigators conducting similar studies with different populations in
comparable or contrasting school settings could build a collective case for the general elfectiveness of a given approach. Clear
specification of the relevant features of the school settings and
careful attention to implementation integrity would be critical
elements of this approach. Dwyer et al. (1989) ha-e proposed
the use of linear regression models fit to aggregated d^ta to
assess bias in standard errors when individual data are analyzed
but schools are assigned to conditions. D. M. Zucker (1990) has
argued tha: use of the individual as the unit of analysis when
classrooms or schools are the units of assignment will always
lead to positively biased tests, compromising the internal validity of analyses. D. M. Murray. Hannan. and Zucker (1989) and
D. M. Murray and Hannan (1990) have concluded from this
that the most prudent course remains ensuring that the unit of
analysis and unit of assignment are the same.
Classroom and school effects should be carefully examined
when analysis at the level of assignment to conditions is precluded by small samples. The relative importance of classroom
comext variables and individual-level variables, along with
their potential inleractions. can be addressed directly using con.extual analysis. Such an approach requires a clear thecrydnven specification of the nature of the predicted comexiual
effects (Bursik. in press). This requirement, however, poses a
problem for current school-based research, in which the contribunon of school and classroom variables, both to risk and to
iniervention effectiveness, are not well understood or coheremly organized in theory.
Homogeneity of effect across levels of risk. When sample size
is sufficiently large, researchers can investigate directly the differential effects of intervention on groups at different levels of
risk. When subgroups are not lar?e enough for such analysis.
Dwyer et al. (1989) have proposed statistical methods, using
conditional proportional odds models with interaction between intervention dummy variable and baseline behavioral
level. Although this solution is proposed for assessing differential effects across baseline drug use levels, it may be applicable
to other quantifiable risk factors as well.
Systematic attrition. Problems of attriticn are acute in
school-based studies that are designed to foil )w longitudinal
cohorts of individual students but use the school or classroom
as the unit of random assignment. The external validity of results from social influence resistance evaluations has been
compromised in many studies by systematic attrition of those
at highest risk for drug abuse. Many published studies of this
i> pe have not addressed attrition, reponing results only for subjects remaining m experimental and comparison classrooms
i Biglan. Severson. etal.. 1987). Where attrition has been investigated, studies have consistently shown that subjects with a
higher mean rate of tobacco smoking and marijuana use arcmost likelv to be lost at follow-up ( Biglan. Severson. et al.. 198".
Hansen et al.. 1988). raising questions as to the generahzabihiv
of reported results to those at greatest risk. Several solutions to
this problem have been proposed. McKiniay et al. 11989) recommend the "intention-to-treat" approach, in which all subjects in
the original cohon are retained for the analysis to avoid the bias
of differential attrition and preserve the integrity of the randomization. Alternatively, direct observation of the effects of
missing data because of attrition may be obtained by including
a dummy-coded variable for subjects lost to the study in the
analysis (Raymond. 1987).
Intervention implementation and intensity Studies should
also address questions of differential intervention implementation and intensity (McKiniay et aL 1989). Byrandomlyand
independently selecting samples of classrooms or schools at
•>ich oomt in time, variable doses of the intervention may be
e.xarmned (e.g. length of exposure, level of teacher iraining.
variety of media used). All intervention studies demand systematic attention to implementation integrity. We have proposed and used three steps in examining implementation: (a)
collection of data to assess degree of implementation, (b) reporting of data on implementation for each dimension of the
interventions, and (c) inclusion of implementation data in the
tests of efficacy (Hawkins. Abbott. Catalano. & Gillmore.
1991; Hawkins & Lam. 1987).
Measuring developmental change and intenennon effecis.
Designs nesting cross-sectional intervention studies within
longitudinal panel studies have special relevance and appeal
for risk-focused prevention work. Such designs are well suited
to explore questions of group differences as well as change over
time, thus ^rov ding for tests of intervention effectiveness a^d
for estimation of uc. "'opmemal sequences. Cross-^. jjuentiai
designs, conceived particularly to study developmental problems (Schaie. 1965: Tonry, Ohlin. & Farrington. 1991). allow
estimation of age. cohon. and period effects, thereby producing data on both the development of and changes in drug risk
and use patterns over time and on the effects of interventions
across cohorts of adolescents (Hawkins. Abbott, et al.. 1991).
A major challenge for sociai influence resistance studies at
this time is to overcome methodological weaknesses. Much
work is proceeding along the lines described above. Meanwhile, the failure of social influence strategies to establish durability of effects or to show consistent results with substances
olher than tobacco, as well as the questions remaining about
the strategies' effects on drug abuse as opposed to initianon or
occasional use, suggests that a second major line of prevention
research should be pursued.
Social influence resistance approaches address risk factors
salient developmentally just before or simultaneous with initiation of drug use. It is not known whether these approaches can
protect children made most vulnerable by previous exposure to
other risk factors. For example, a social influence resistance
program that demonstrated significant reduaions in tobacco
use for baseline experimenters and nonsmokers was actually
�R.'Sk A \ D PROTFCTI\ E FACTORS FOR DRUG PROBLEMS
associated with increases in tobacco use lor baseline smokers
(Ellickson & Bell. 1990). These youngsters, having already dehneo themselves as pan ofa smoking subculture with attachmenis io tobacco-using peers, may have rejected drug resistance skills as antithetical to their social group identity Learning skills to resist prodrug social influences may be a necessar>
but not sufficient element of prevention for children who have
been "set up" for drug involvement by exposure to earlier mdiwdual. family: or community risk factors (Block. Block. &
Keves. 1988: Shedler & Block. 1990). This possibility suggesis a
search for imervennon strategies that have been effective in
reducing other factors, developmentally earlier than social influences to use drugs, that predict drug abuse. Such strategies
should be investigated for their long-term effects in preventing
drug abuse.
Prevention Approaches
Targeting Early Risk Factors
The foi lowing strategies merit attention from drug abuse prevent, wfi researches, both because they address risk factors seen
to occur befor ug initiation and because iney use intervention methods that have demonstrated positive effects. The appeal of many of these interventions is strengthened by the fact
thai they target risk factors implicated in a range of disorders,
including drug abuse as well as antisocial behavior, delinquency, and later adult criminality. These interventions have
shown positive effects on targeted risk factors in controlled intervention trials using expenmental or quasi-experimental designs. As noted below, in some of the studies the interventions
appear to have increased protective factors against drug abuse
in populations at high risk.
/ Early childhood and family support programs. Several
interventions focusing on the prenatal and early infancy periods with a variety of componems ranging from health care:
mT
n; child care; social suppon for mothers: educational,
career, and fai,.ily planning services: and home .isits irom
health or social service workers have produced significant differences between high-risk-program and control- or comparison-group families. Positive intervention effects have been reported on child abuse and neglect (Olds. Henderson. Chamberlin. & Tateibaum. 1986: Swift. 1988). early academic
performance (Bronson. Pierson. & Tivnan. 1984). maternal
employment and smaller family size, child's higher rate of
school attendance and lower rate of school special services, as
well as lower mother-rated antisocial behavior and lower
teacher-rated aggression (Pierson et aL 1983: Seitz. Rosenbaum. & Apfel. 1985).
Early childhood education has produced reductions in risk
factors for drug abuse. Horacek. Ramey. Campbell. Hoffmann,
and Fletcher (1987) randomly assigned socially and economically deprived children, at infancy and again at kinderganen. to
intenennon or control groups, allowing evaluation of the effects of different amounts of intervention. They successfully
identified at binh children at high risk for school failure. Rates
of retention in grade for high-risk children in the control group
were almost four times higher than for an average-risk group of
peers. The intervention significantly reduced grade retention
and improved test scores in math and reading, showing a
greater impact on children who had pamcipated in both inter-
0029
vention phases. The high-risk children who received the mos;
intervention achieved a rate of grade advancement neariv equa!
to that of the average-risk group.
The Perry Preschool Project focused on enhancing the inte •
lectual and social development of 3- and 4-\ear-old AfricanAmerican children from backgrounds of extreme poveny. The
expenmental intervention reduced academic failure, adolescent pregnancy rates, and cnminal behavior when randomlyassigned expenmental and control subjects were followed up
and compared at age 19 (Berrueta-Clement et al.. 1984). The
experimental program consisted of daily panicipation in a preschool over a l-to-2-year penod and weekly home visits by
trained teachers to teach mothers skills in child management.
By age 19. expenmental preschool panicipants had lower arrest
rates and fewer arrests as well as lower rates of self-reponed
fighting. They had higher rates of secondary school completion, lower rates of placement in special education classes, and
higher grade point averages than their randomly assigned control counterparts. Other studies of early childhood education
programs have shown similar positive effects on children's intellectual development (Gotts. 1989; Lazar. Darlington. Mi- v.
Royce. &L Snipper. 1982: Ramey, Bryant. Campbell. Sparling. &
Wasik. 1988).
These findings suggest that early childhood and parent support programs can buffer the effects of extreme poverty and
neighborhood disorganization by reducing three risk factors for
adolescent substance abuse: childhood behavior problems, family managemeni problems, and academic failure. Research is
needed to evaluate the long-term effectiveness of early childhood and family support programs for high-risk subgroups in
preventing adolescent drug abuse.
2. Programs for parents of children and adolescents. Controlled studies have shown that family management problems
and child behavior problems can be reduced through parenting
skills training and functional family therapy. Parenting skills
training has pi.uuced shon-term improvements in family interaction and reductions in children's problem behaviors
(Baum & Forehand. 1981; Patterson. Chamberlain. & Reid.
1982). Parenting skills training combined wuh social skills
training for disruptive kindergarten boys reduced school adjustment problems and delayed the onset of delinquent behavior (Tremblay et aL 1990). Funclional family therapy has reduced delinquency among juvenile offenders (Alexander &
Parsons. 1973) and prevented it among their siblings (Klein.
Alexander. & Parsons. 1977).
Most systematic evaluations of parent training have involved
children with condua problems. Parenting skills training focused on teaching parenis to monitor their children's behavior,
to use moderate contingent discipline for undesired behavior,
and to consistently reward prosocial behavior (Patterson &
Fleischman, 1979) has resulted in increases in parent-child attachment, decreases in children's skill deficits, and decreases in
the children's targeted behavior problems (Fleischman. 1981.
Patterson &. Reid. 1973: Peed. Roberts. & Forehand, 1977).
Randomized experimental tests of parenting skills training
have shown significant reductions in preadolescents' problem
beha.-ic": when compared wuh controls (Karoly &. Rosenthal.
19". Martin. 1977: Patterson et a... 1982: Walters & Gilmore.
1973). These results suggest that parenting skills training can
�0030
D
HASKINS.
R. C A T A L A N O .
buffer the risk of childhood behavior problems lor adolescent
drug abuse by reducing family management problems and increasing family bonding.
To date, little experimental research on the elfectiveness of
parent training for drug abuse prevention has been conducted.
In one study, parenting skills training was tested with parents
who were narcotic and polydrug abusers participating in treatment programs. A preliminary evaluation reponed that parents were successfully trained to develop more etfective discipline methods, that their children had fewer behavior problems
after treatment, and that the children reponed decreased intentions to smoke and use alcohol (DeMarsh & fCumpfer. 1986).
although the effects of the intervention on the children's actual
drug use was not reponed. These preliminary results suggest
that parents whose children are at high risk by vinue of parental addiction can be successfully taught parenting skills. Another study with a sample of youths at risk for substance abuse
has reponed preliminary positive trends for parent-training
treaiment groups on parent-child interaction, levels of tobacco
use. and reduction of depression (Dishion. Kavanagh. & Reid.
1989).
Parent involvement has been shown to L . beneficial in improving academic effon. grades, and attendance of students
evidencing low commitment to school (Bien & Bry. 1980:
Blechman. Taylor. & Schrader. 1981). Bry (1982) reponed a
reduction in juvenile justice system involvement o f experimental subjects receiving an intervention package consisting o f (a) a
parenting program involving regular contacts with the family
emphasizing training and encouragement for parenis to reward
school progress, (b) teacher goal setting for students, and (c) a
schedule o f rewards for students' goal attainment. These findings suggest that a promising method for increasing parental
involvement is through training and reinforcement for parents
io promote the classroom performance of their children.
Biglan. Glasgow, et al. (1987) found no effects on children's
smrk'ng of four parent messages mailed to lhe homes of students, designed to reinforce social influence resistance skills
and commitment to nonsmoking taught in a classroom curriculum. However, no implementation assessment was conducted,
and it is not clear that parents used the mailed messages. The
chosen intervention method may nol have been sufficiently
potent to enlist parental panicipation.
Pentz. Dwyer. et al. (1989) involved parents in the expenmental drug abuse prevention package tested in the Midwestern
Prevention Projea. The classroom prevention curriculum included 10 homework assignments in which students were expected to involve their parents using aaive interviews and role
plays. Using interview data from teachers, the authors estimated that 80% o f the experimental families participated in the
homework assignments. The experimental program was associated with lowered prevalence rates o f tobacco, alcohol, and
marijuana use, although the independent contnbution of the
parenting component was not assessed.
The existing evidence suggests the promise of parenting
skills training and involvement approaches for preventing adolescent drug abuse. Training adjusted to the developmental
stage of the child should help parents develop skills to (a) set
clear expectations for behavior, (b) monitor and supervise their
children. ic> consistently reinforce prosocial behavior, (d) create
A N D J. M I L L E R
opportunities for family involvement, and (e) promote the development of their children's academic, sociai. and refusal skills.
Acquisition and use of these skills by parenis in managing their
families could be expected to reduce children's behavior problems in preschool and elementary school years, to mcreasechildren's academic pert'ormance in elementarv and middle school,
and to empower children to deal erfectiveK wnh sociai influences to use drugs encountered in late elementarv and middle
school grades.
Problemsof nonparticipation. attrition, and implementation
in parenting skills training programs have been well documented (Bry: 1983: Fraser. Hawkins. & How-ard. 1988; Grady
Kelin. & Boratynski. 1985: Hawkins. Caialano. Jones. & Fine.
1987; Perry. Crockett. & Pine. 1987; Perry. Luepker. et al..
1988). Parenting training interventions for parents of preschool, element-ry. and middle school children that seek to
overcome these difficulties should be tested in experimental
drug abuse prevention tnals.
i . Social competence skills training The evidence that aggression and other behavior problems in the early elementary
grades is associated wuh an increased risk of later drug abuse
has stiniulated suggestions thr.. educational strategics seeking
to enhance the social compeiencies of youngsters during childhood could reduce the risk of later drug abuse (Hawkins. Jenson. Catalano. & Lishner. 1988). It has been argued that children who are aggressive, disruptive, and rejected by peers in
elementary grades are deficient in basic interpersonal skills
that can be taught (Spivack & Shure. 1974).
Advocates of skills training for interpersonal competence
cite evidence that socially competent children engage in less
school misbehavior and have better cognitive skills in such
areas as basic problem solving essential for academic achievement (Asher & Renshaw. 1981). They assert that learning certain basic moral values like concern for the rights and needs of
others is essential to the development of prosocial behavior
(Battistich. Solomon. Watson. & Schaps. r ' -.nd that children
from families in which family managemeni practices arc poor
and family conflict is great do not learn basic interpersonal
competencies at home.
Social competence promotion approaches have used a variety of methods. For example, socially rejected youths have been
taught and coached in social interaaion skills to increase the
frequency of their social interactions (Ladd &. Asher. 1985). and
classroom instruction has been used to teach cognitive processes and behavioral skills to handle interpersonal problems
(Weissbcrg & Allen. 1985). Such methods have been implemented in programs of widely varying duration, with samples
ranging from all students at a particular grade level to identified students with behavior problems. They have been tested
with inner-city, low-income samples (Shure & Spivack. I982)as
well as with White middle-class samples (Rotheram. 1982a). To
date, most of these tests have focused on proximal outcomes
such as school adjustment rather than on drug use behavior.
Some studies have found positive effects immediately after
training for both suburban and inner^-ity samples (Weissberg et
a L 1981).
Social competence promotion approaches have yielded varying result. Some investigators have reponed positive effects on
interpersonal behavior (Battistich et al.. n.d: Bierman & Fur-
�RISK AND PROTECTIVE FACTORS FOR DRL'C PROBLEMS
man. I9S4: Bierman. Miller. & Slabb. 198": Gesten et al.. 1982:
Ladd. 198 I: Ladd & Asher. ^85: Rotheram. 1982b: Rotheram.
Armstrong. & Booraem. 1986: Shure & Spivack. 1982: Weissberg £ Caplan. 1989). Others have found no effects on adjustment i Allen.' Chmsky. Larcen. Lochman. & Selmger. 1976)
There is promising evidence for social competence promotion as a drug abuse prevention strategy. Social competence
promotion has shown a significant impact both on students'
willingness to try nondrug or nonalcohol options when confronted with problem situations and on the certainty of their
intention not to use drugs or alcohol, when compared wuh a
matched control group (Ketchel & Bieger. 1989). .Kim.
McLeod. and Palmgren (1989) evaluated the effects of an intervention in 4th grade on students in Grades 5 through 12. The
program consisted of one session a week for 9 weeks focusing
on social skills. In the last session, students applied the skills
learned to drug use choices. Across all grade levels, students
showed significantly lower use o f alcohol, cigarettes, and marijuana compared with students who had not participated in the
program. The largest impact was found in Grades 5 through 7;
the intervention's positive effects declined rapidly at 9th grade.
Lochman (1988) provided an anger management program
during school hours for boys identified as aggressive by their
teachers. The program included role playing, goal setting, videotaped modeling to develop self-statements, social problem
solving skills to cope with anger arousal, and group-produced
videotapes illustrating alternative ways of coping with an
anger-arousing situation. When compared with a matched
(though not randomly assigned) comparison group of untreated
aggressive boys, the treated subjects were found to have significantly lower rates of alcohol and marijuana use as well as fewer
negative consequences of alcohol use at age 14. 3 years after the
intervention.
This is an important area for additional prevention research.
I f school-based social competence promotion strategies reliably
reduce aggressive and othe- problem behaviors as well as frequent and problem drug use during adolescence, they represent
a viable prevention strategy.
A question for future research is the age of children for whom
such interventions are most effective. To illvstrate. from about
age 5. aggressiveness in boys predict^later deviance including
drug abuse (Kellam & Brown. 1982). Few boys appear to become seriously aggressive if they did not manifest aggressive
behavior in childhood. However, aggression declines in prevalence with age. Some boys desist from aggressive behavior between the ages of 5 and 14. Social competence promotion strategies that seek to reduce aggressiveness among young boys run
the risk of false-positive error, in that they may target for intervention some youngsters who w i l l not show later deviant outcomes (Loeber & Dishion. 1983). Conversely, programs that
wait to intervene until aggression has crystallized as a pattern
of behavior may minimize false-positive errors in identifying
those at high risk for later drug abuse, but these interventions
may have less chance of successfully eliminating the aggressive
behavior and other problems of adjustment and achievement
produced by aggressive behavior during the elementary school
grades. Lochman's (1988) research, which showed lower rates of
drug involvement at 3-year follow-up among aggressive boys
who received problem-solvmg-skills training at age 1 1. did not
003'
show similar positive effecis on aggressive behav ior or general
behavioral deviance at 3-year follow-up.
J Academic achievemeni promouon. Three strategies have
shown positive effects on the risk factors of academic achievement and problem behaviors in school and thus hold promise
for preventing drug abuse. The strategies include early childhood education, as previously discussed (Berrueta-Clement et
al.. 1984). alterations in classroom teachers' instructional practices in elementarv and middle schools (Hawkins. Doueck. &
Lishner. 1988; Hawkins & Lam. 1987), and academic tutoring
of low achievers (Coie & Krehbiel. 1984). The latter two are
discussed here:
a Alieraiions in classroom insiructionai praaices The use
of cenain methods o f instruction in classrooms has been
shown in experimental and quasi-experimental studies to improve achievement and social bonding to school and to reduce
student misbehavior. A number of studies have linked achievement gains to the amount of active instruction and direct supervision of learning effons that teachers provide to students
(Brophy &. Good. 1986). Classroom teachers' use o f a package
of instructional methods consisting of interactive teaching,
proactive classroom management, and cooperative learning resulted in significantly greater achievement gains in math and in
levels of commitment to school and in significantly lower rates
of suspensions and expulsions from school among urban seventh-grade students in experimental classrooms when compared with control classrooms (Hawkins & Lam. 1987).
Cooperative learning methods have been included in some
classroom interventions seeking to enhance achievement and
commitment to school. An intervention targeting children
starting preschool attempted to bring all children in the study
up to grade level by third grade (Slavin. Madden. Karweit.
Liverman. &. Dolan. 1990). The program focused on language
development, academic readiness, and improved self-concept
with preschool and kindergartners. In Grades 1 through 3. the
intervention rep'aced pull-out programs and special education
classes with in-class tutors and alternative classroom strategies,
including grouping o f students across grades by ability. The
program's effects were evaluated by companson with a
matched school and with individually matched students within
the control school. The results showed significantly higher test
scores for intervention students at all grade levels.
Freiberg. Brady, Swank, and Taylor (1989) found positive results on academic achievement from a program combining cooperative learning strategies with improved classroom management, student and teacher motivation, parent contacts, interactive instruction, and discipline prevention. Students in five
target schools were compared with the students of five matched
controls. At the end o f the I'/i-year intervention, students in the
test schools surpassed the control students on standardized test
scores in every subject.
In the most effective cooperative-learning approaches, students of differing abilities and backgrounds work together in
small groups of 4 to 6 children to master the curriculum material, and they receive recognition asa team forthe performance
of all members of the group. Cooperative-learning strategies
have been designed to encourage students to help and suppon
peers of diverse ability, ethnicity, and background toward the
achievement of academic success.
�oo?:
D
HAW. K I N S . R. C A T A L A N O . A N D J. M I L L E R
Controlled studies have shown positive erfects of cooperative
learning on achievement and attitudes tow-ard school and peers
! DeVries & Slavin. 1978: Dolan. Kellam. & Brow n. 1989: Madden & Slavin. 1983: Ziegler. 1981) and. in combination with
other instructional methods, have produced reductions in rates
of suspension and expulsion from school among low achievers
(Hawkins. Doueck. & Lishner. 1988). However, the use of the
"Jigsaw" cooperative learning method did not prevent drug use
in the Napa Project (Schaps. Moskowuz. Malvin. & Schaeffer.
1986).
Cooperative learning methods hold promise for changing
peer influence patterns in schools, for reducing academic failure, and for increasing commitment to school and attachment
to prosocial others. Given these effects on risk factors for drug
abuse, cooperative learning should be investigated funher for
drug abuse prevention effects in spue of the apparent lack of
effect of the "Jigsaw" method in the Napa Project.
b. Tutoring. I ndividual tutoring for low achievers wuh behavior problems has been used in conjunction with, as well as
separately from, social competence skills training approaches
discussed earlier (Coie & Dodge. 1988). Coie and Krehbiel
(1984) found that even without social skills training, tutoringof
socially rejected, low-achieving fourth graders reduced peer rejection and disruptive behavior in the classroom and produced
significant improvements in reading and math achievement.
The existing evidence suggests that both improvements in
methods o f instruction used by teachers in mainstream
classrooms and individualized tutoring programs hold promise
for reducing academic failure and problem behaviors of children. Both types of intervention with children in elementary
and junior high school grades should be studied for effects in
preventing adolescent drug abuse.
5 Organizational changes in schools. Schools with the highest rates o f studenl misbehavior and drug abuse are typically
"demoralized" organizations (G. D. Gotlfredson. 1988). Such
schools are likely to have great difficulty implemennng strategies such as improved classroom leaching, lutoring programs,
or parental involvement in promoting students' classroom performance. Because school organizational characteristics apypear to be related to student behavior, achievement, and bonding to school and because they influence the ability of schools
to implement changes thai might prevent drug abuse, they are
factors that should be considered potential targets for drug
abuse prevention efforts.
There is evidence lhat school organizational factors can be
changed to reduce drug abuse risk faaors (Comer. 1988: D. C.
Goltfredson. 1986, 1988). In one school (D. C. Gotlfredson &
Cook. 1986a, 1986b) a program o f curriculum restructuring,
increased opportunities for student involvement, greater
school-faculty-community integration, and changes in school
discipline procedures resulted in significant changes. Despite
implementation problems resulting from district staff cuts,
companson with a control school showed increases in positive
self-concept, attachment to school, and belief in riles as well as
higher standardized test scores in math. Significant decreases
were found in the school's rates of delinquency, drug use. and
suspensions.
In another study, D. C. Goltfredson (1986) evaluated an intervention that included (a) the establishment of an organizational
structure to facilitate shared decision making and managemeni
m schools. (b) the use of curriculum and student concerns specialists, (o academic innovations including cooperative learning, reading, and test-taking programs and career exploration,
and id) direct services to targeted high-risk students to increase
academe involvement and achievement, including individual
ireatment plans, behavioral objectives, and monthiv monitoring by specialists.
High-risk students in participating schools were randomly
.assigned to treatment and control groups. At the end of 3 years
of intervention, students in experimental schools reponed
lower rates of drug use. delinqueni behavior, and alienation and
higher rates of ailachmenl to school, educational expectations,
and belief in school rules when compared with students in comparison schools. However, the direct services for targeted highrisk students did nol produce significant effecis on risk factors
or behavior (D. C. Goltfredson. 1986).
Comer (1988) demonstrated positive gains in student academic achievement over a 12-year period after the creation and
suppon of school governance and management teams in two
New Haven schools serving predominantly low-income African-American populations. The teams consisted of principal,
parents, teachers, and a mental health worker and developed
and implemented comprehensive school plans for academics,
social activities, and special programs. Consistent schoolwide
gains were achieved in scores on standardized reading and
math tests in comparison with national norms, though control
or comparison schools were not used.
Felner and his colleagues (Felner. Adan. & Evans. 1987.
Felner. Ginter. & Pnmavera. 1982: Felner. Weissberg. & Adan.
1987) have altered the school environment for students making
normal transitions from elementary to middle or junior high
schools or from these to high schools. The intervention kept
groups of transitioning students together in homeroom and
core courses in circumscribed schools-within-a-school. in
which all homeroom and core teachers' classrooms were within
close proximity to one another. In addition, the homeroom
teacher served as an advocate-counselor for all students in his
or her homeroom, contacting each homeroom student's family
before the school year began and serving as a counseling link
for homeroom students, their parents, and the rest of the school
(Felner & Adan. 1988). The intervention produced positive effects on academic performance, absenteeism, and school drop
out when participating students were compared wuh nonparticipating students in the same schools.
These results indicate that altering the organizational characteristics of schools can reduce risk factors for drug abuse as
well asdrug use itself. Organizational change in school management and the school environment should be funher investigated for drug abuse prevention effects.
6. Youth involvement in alternative activities. Activities in the
school setting that provide opportunities for youths to participate in contributing—roles such as involvement in school government, experience-based career education programs in
which students are provided hands-on opportunities to learn
about the world of work, and tutoring programs in which students are enlisted to help other studenls—have been hypothesized to increase commitment to school and reduce alienation
and rebelliousness. When such activities have been provided
�RISK \ N D PROTECTIVE FACTORS FOR DRUG PROBLEMS
for adolescents, the emphasis has been on active involvement
(i.e.. shifting the student's role from consumer of information to
producer of some benent). From a risk-focused perspective, this
strategy might be expected to increase commitment to school
and to reduce the likelihood of violation of school standards of
behavior, including proscriptions against drug use.
Similarly, it has been suggested that physically challenging
risk-taking activities, such as those offered bv Outward Bound
programs, might provide effective drug-free alternatives to
those at risk for drug use by virtue of the personal characteristics of high novelty seeking and low harm avoidance.
Evidence for the effectiveness of such approaches in preventing drug abuse is mixed. Neither cross-age tutoring nor operating a school store prevented drug use among predominantly
White middle-class students in eighth and ninth grades in the
Napa Project (Schaps et aL 1986). On the other hand, there is
some evidence that when delivered at high intensity, alternative
programs that empower high-risk subjects to master new skills
are associated with improved behavior and achievement
(Tobler. 1986). Programs such as Outward Bound, which provide risk-taking challenges as opportunities to learn skills,
could be enhanced by continuation interventions designed to
build mastery over the environments in which youths routinely
function. These latter approaches should be investigated further for drug abuse prevention effects.
7. Comprehensire risk-incused programs.
Because drug
abuse is a phenomenon influenced by multiple risk factors, its
prevention may be most effeaively accomplished by a combination of interventions promoting consistent prevention principles across units of socialization. Comprehensive prevention
programs that combine multiple interventions focused on different sources of social influence have shown beneficial effeas
on smoking (Puska et aL 1982).
Pentz and her colleagues (Pentz. Dwyer. et a L 1989) have
tested a multicomponent communitywide program involving a
.-iculum of social influence resistance skills training for students in Grades 6 or 7 that includes (a) homework assignments
to be conducted with parents, (b) booster sessions in the year
after initial intervention, (c) organizational and training opportunities for parents in positive parent-child communication
skills and in reviewing school policies! (d) training o f community leaders to organize drug abuse prevention task forces, and
(e) news coverage. The multicomponent program produceu
lower prevalence rates o f weekly cigarette (-8%). alcohol (-4%).
and marijuana (-3%) use after the 2nd-year intervention (Pentz.
Dwyer. et aL 1989) and significantly lower prevalence of
monthly cigarette (-6%) and marijuana (-3%) use 3 years after
the initial school intervention, though the prevalence o f alcohol
use was not significantly reduced at this measurement point
(C. A. Johnson et aL 1989). The comprehensive intervention
appears to have been equally effective in lowering tobacco and
manjuana use prevalence among those at risk because o f exposure to parental drug use. drug-using peers, and early initiation
of use (C. A. Johnson et aL 1989). Although the unique contribution of each individual component in this comprehensive
program has not been determined, the results indicate that a
multiple-component strategy focused on reducing risks is more
effective in reducing drug use prevalence than is mass-media
coverage alone.
0035
Similarly, research on a comprehensive teacher-, parent-, and
peer-focused prevention program grounded in the social development model has shown that the comprehensive program produced significantly lower rates of school suspension and expulsion among seventh-grade experimental subjects (Hawkins.
Doueck. &. Lishner. 1988). signilicantlv low-er prevalence of
early aggression among second-grade subjects iHawkins. Von
Cleve. & Catalano. 1991). and significantly lower prevalence of
self-reponed delinquency (-6.7%) and alcohol use (-6.6%) bv
Grade 5 among children exposed to the comprehensive program in Grades 1-4 in companson with controls (Hawkins et
aL in press). The intervention consisted of (a) teacher training
in methods of classroom managemeni and instrucnon consistent with the principles of the social development model, ibi
training for parents in developmentally adjusted curricula focused on development of family management skills consistent
with the social development principles, and (c) involvement of
student subjects in classroom-based peer leaching and skill development. The results suggest that by promoting consistent
opportunines and expectations for prosocial behavior at home
and school, by enhancing ski 1 development, using peer involve1
ment and teaching and parent monitoring, and by stressing
positive reinforcements for prosocial involvement from family,
school, and peers, the incidence of early initiation of drug use
and delinquency can be reduced. Analysis indicates that these
outcomes are accompanied by effects on family- and schoolbonding variables that are hypothesized in the theory on which
the comprehensive program is based (Hawkins et al.. in pressi.
Summary
A risk-focused approach to drug abuse prevention holds
promise for identifying effective prevention strategies. Implementing and testing approaches that seek to reduce or buffer
the effecis of known antecedents of adolescent drug abuse will
increase our kn---' ledge of which are causally related to drug
abuse and what prevention strategies reliably address these risk
factors.
Research has identified these antecedents of adolescent drug
abuse: laws and norms favorable toward drug use: availability
of drugs: extreme economic deprivation: neighborhood disorganization: cenain physiological characteristics: early and persistent behavior problems including aggressive behavior in
boys, other conduct problems, and hyperactivity in childhood
and adolescence: a family history of alcoholism and parental
use o f illegal drugs: poor family management praaices: family
conflict: low bonding to family; academic failure: lack of commitment to school; early peer rejeaion: social influences to use
drugs: alienation and rebelliousness: attitudes favorable to drug
use: and the early initiation of drug use. There is some evidence
that cenain factors including personal attributes and a social
bond to conventional society may protect against drug abuse,
though more research is needed to determine the relationships
beiween risk and protective factors as related to adolescent drug
abuse.
Evidence from studies of the etiology of adolescent drug
abuse suggests that a viable prevention model would include
simultaneous attention to a number of risk factors in different
social domains to be addressed during the developmental pe-
�0034
o
HAWKINS, R. C A T A L A N O . A N D J. M I L L E R
nod when each begins io stabilize as J predictor oi'subsequent
drug abuse. The evidence funher suggests that prevention efforts target populations at greatest risk of drug abuse because of
their exposure to a large number of risk factors during development. A theory of adolescent drug abuse that accounts for the
existing empirical evidence regarding risk and protective factors for adolescent substance abuse should be used to organize
and integrate the complex work of developing and testing prevention interventions.
Most drug abuse prevention effons have addressed two risk
factors for adolescent drug abuse: laws and norms favorable to
drug use and social influences to use drugs. These effons include supply manipulation, interdiction and enforcement strategies, efforts to change social norms regarding drug use. and
social influence resistance skills training. Of these approaches,
social influence resistance strategies have been evaluated mos.
extensively for preventive effects in controlled studies.
Several available studies of social influence resistance skills
training for drug abuse prevention have produced shon-term
effectson ratesof drug inuiation. including reductions in smoking and. in a few cases, in alcohol and marijuana use. Although
such results are promising, the limits o 'hese programs should
be considered. Peer influence resistance skills training methods do not change the basic developmental conditions experienced by children. Although these methods have shown shonterm effects on the incidence of drug initiation in the general
population, they may have little effect on drug abuse among
higher risk groups. Children who are at highest risk for adolescenl drug abuse by vinue of poor family management, early
and persistent behavior problems, low bonding to family, academic failure, and low commitment to school may be unmotivated to refuse or avoid drug use by late childhood.
If the goal is to reduce drug abuse and its accompanying
social and health problems among children at high risk, it is
imponant to test preventive approaches that have successfully
addressed risk factors present earlier in child development.
Promising nsk-focused approaches that should be investigated
for drug abuse prevention effects are early childhood education
and early family suppon. parent training, school-based social
competence promotion, school-based academic competence
promotion, and school organizational change strategies. Recent studies indicate that coherent multiple-component or
comprehensive strategies, including but not limited to social
influence resistance, hold significant promise for preventing
drug abuse and its attendant costs. These approaches should be
implemented in varying combinations and settings, and their
effects on the initiation, use, and abuse of drugs should be studied funher in controlled field expenments.
f
References
Adams. E. H.. Blanken. A. J.. Ferguson. L. D.. & Kopstein. A. (1990).
Overview of seleaed drut; trends. Rockville. MD" National Institute
on Drug Abuse.
Ahmed. S. W. Bush. P. J.. Davidson. F. R. < lannotti. R. J. (1984.
&
November). Predicting childrens use and iniennons io use abusable
substances Paper presented at the Annual Meeting of the American
Pubiic Health Association. Anaheim. CA
Akers. R. L.. Krohn. M. D. Lanza-Kaduce. L.. i : Radosevich. M.
1 979) Social learning and dev iant behavior: A specitic ies; oi a
1
general theory American Soaoiovcui Review. •'•> 636-65:
Alexander. J. F. & Parsons. B. V (I9"j). Shon-term behavion sr.:*:.
vention with delinquent families- Impact on family process ana
cidmsm. Journal ol -tbnnrmai Puchnlngw /.? r!9-22:.
Allen. G. J.. Chmsky. J. VL. Larcen. S W. Lochman. J. E.. & Sdir.i;-. J
V(1976) Community pswholoeyandthe Khwis- A hehavmra:.'.
emed mulnlevel prevt'niivt' (sppnitich. Hillsdale. NJ. Erlbaum
Asher. S. R.. SL Renshaw-. P D. (19811. Children w-uhout friends: Social
knowledge and social-skill training. In S. R. Asher & J. M Gouman
(Eds.). The developmem ofchildrens inendships ipp. -7}-296i. New
York: Cambridge University Press.
Atkin. C. Hocking. J.. & Block. M. (1984). Teenage drinking: Does
advertising make a difference Journal of Communicauon. i-i. 157167.
Bachman. J. G.. Lloyd. D. J.. & O'Malley. P. M.(198l). Smoking, drinking, and drug use among Amencan high school students. Correlates
and trends. 1975-1979. Amencan Journal of Public Health. ' I . 5969.
Bandura. A. (1977). Self-efficacy: Toward a unifying theon- of behavioral change. Psychological Review: 84 191-215.
Barnes, G. M..& Welte. J. W (1986). Patterns and predictors of alcohol
use among 7-12th grade students in New- York State. Journal of Studies on Alcohoi. 47. 53-62.
Battistich. Y, Solomon. D.. Watson. M.. & Schaps. E. (n.d.) Erieas oj an
elememary school program to enhance prosocial behavior on childrens cognitive social problem-solving skills and strategies Unpublished manuscript. San Ramon. Developmental Studies Center.
Baum. C. G.. & Forehand. R. (198 I). Long-term follow-up assessmeni
of parent training by use of multiple outcome measures. Behavior
Therapy. 12. 643-652.
Baumrind. D. (1983. October). Why adolescenis take chances—and'whv "
they don't Paper presented at the National Institute forChildHealth
and Human Development. Bethesda. MD
Berrueta-Clement. J. R.. Schwemhan. L. J.. Barnett. W S.. Epstein.
A. S.. & Weikhan. D. P. (1984). Changed lives: The effecis oi the PerryPreschool Program on youlhs through age 19. Ypsilami. Ml: High/
Scope Press.
Bien. N. Z.. SL Bry. B. H. (1980). An experimentally designed comparison of four inii... ..ies of school-based prevention programs for adolescents with adjustment problems. Journal of Communitv Psychology. 8. 110-116.
Bierman. K. L.. SL Furman. W (1984). The etfectsof social skills iramingand peer involvement in the social adjustment of preadolescents.
Child Development. 55 151-162.
Bierman. K. L. Miller. C. L.. SL Stabb. S. D. (1987). Improving the
social and peer acceptance of rejected boys: Effects of social skill
training with instructions and prohibitions. Journal of Consulting
and Clinical Psychology. 55. 194-200.
Biglan. A, SL Ary. D. V (1985). Current methodological issues in research on smoking prevention. In C. Bell SL R. J. Battjes (Eds ). Prevention research: Deterring drug abuse among children and adolescents (pp. 170-195) (NIDA Research Monograph No. 63). Washington. DC: U.S. Government Priming Office.
Biglan. A. Glasgow. R.. Ary, D, Thompson. R.. Severson. H . Lichenstein. E. Weissman. W. Faller. C. & Call ison.C.11987). How generalizable are the effectsof smoking prevention programs? Refusal skuii
training and parent messages in a teacher administered program.
Journal of Behavioral Medicine. 10. 613-627.
Biglan. A..Seveison. H.. Ary, D.. Faller. C. Gallison.C. Thompson. R..
Glasgow. R.. SL Lichtenstein. E. (1987) Do smoking prevention programsreallywork Attntion and the internal and external validity of
an evaluation of a refusal skills training program. Journal o' Behavioral Medicine. 10. 159-17 1.
1
1
J
0
�RISK A N D P R O T E C T I V E FACTORS FOR D R L G
Black. G S. (1989) T h e i H t i t u d i r u t l b a n i-i'drug use— /V.-}." andchaneme x::ini<ies toward drug u s e — / W f Reprirts irom the Media -{d\eriis;ne Partnership tor a Drug-Free Amenca. fnc Rochester. N Y. Gordon S Black.
Blechman. E . A . Taylor. C J.. & Schrader. S. M. i|9g 1). Familv problem
solving versus home notes as early iniervention wnh high-risk; children. Journal ni Consilium and Clinical Psvchofogx. -fv 9 19-926
Block. J. Block. J. H.. &. Keyes. S. il988V Longitudinally t'oretelling
drug usage in adolescence: Early childhood personality and env ironmental precursors. Child Developmem 59. 336-355.
Blose. J. O. & Holder. H. D. (1987) Liquor-bv-the-dnnk and alcoholrelated traffic crashes: A natural experiment using time-series analysis. Journal o/'Sludies on Alcohol. 48. 52-60.
Blum. K... Noble. E. P.. Sheridan. P. J.. Montgomery. A.. Ritchie. T.
Jagadeeswonan. P.. Nogami. H.. Briggs. A. H..& Cohen. J. B. (1990).
Allelic association of human dopamine D receptor gene in alcoholism. Journal of the American Medical Association. 263. 2094-2095.
Blum. R. (1987). Contemporary threats to adolescent health in the
United States. Journal oi ihe American Medical Association. 257.
3390-3395.
:
Bohman. M. (1978). Some genetic aspects of alcoholism and criminalily Archives o f O e n c s ! Psvchiatry. S * 9 - 2 7 6 .
r
1
Botvin. G. J. (1986). Substance abuse preven..
^ c ^ r c h . Tvecem developments and future directions. Journal of School Health. 50. 369374.
Botvin. G J. Baker. E.. Filazzola. A. D.. & Botvin. E. M. (1990). A
cognitive-behavioral approach to substance abuse prevention: Oneyear follow-up. Addictive Behaviors. 15. 47-63.
Botvm. G. 1.. Baker. E.. Renick. N. L.. Filazzola. A. D..& Botvin. E. M.
(1984). A cognitive-behavioral approach to substance abuse prevention Addictive Behaviors. 9. 137-147.
Botvin. G. J.. Batson. H. W. Witts-Vitale. S.. Bess. V. Baker. E.. & Dusenbury. L. (1989). A psychosocial approach to smoking prevention
for urban Black youth. Public Health Reports. 104. 573-582.
Botvin. G. J.. & Wills. T. A. (1985). Personal and social skills training:
Cognitive-behavioral approaches to substance abuse prevention. In
C. Bell & R. J. Battjes (Eds.). Prevention research: Deterring drug
ahuse among children and adolescenis (pp. 8-49) ( N I D A Research
^lono ..a^' No. 63). Washington. DC: US. Governm
P ' ..-.g
Office.
Bowlby. J (1969). Attachmenl and loss: Vol I. Ailachmenl. New York:
Basic Books.
Bowlby. J. (1973). Attachment and loss: Vol. 2. Separation-anxiety and
anger New York: Basic Books.
,
E
Braucht. G. N.. Kirby. M. W. 8L Berry. G. J. (1978). Psychosocial correlates of empirical types of multiple drug abusers. Journal ol 'Consulting and Clinical Psychology. 46. 1463-1475.
Bronson. M. B.. Pierson. D E. &. Tivnan. T. (1984). The effects of early
education on childrens competence in elementary school. Evaluation Review. 8. 615-629.
Brook. J S.. Brook. D. W. Gordon. A S.. Whiteman. M . & Cohen P.
(1990). The psychosocial etiology of adolescent drug use: A family
interactional approach. Genetic Social, and Ceneral Psychology
Monographs. 116 (Whole No. 2).
Brook. J. S. Gordon. A. S.. Whiteman. M.. &. Cohen. P (1986). Some
models and mechanisms for explaining the impact of maternal and
adolescent charaaeristics on adolescent stage of drug use. Developmental Psvchology. 22. 460-467.
Brook. J. S.. Lukoff. I. F. & Whiteman. M. (1980). Initiation into adolescent marijuana use. Journal o f Genetic Psychology. 127. 133-142.
Brook. J. S.. W h i t e m a n . M.. Gordon. A. S.. & Brook. D. W (1988). The
role of older brothers in younger brothers' drug use viewed in the
comext of parent and peer influences. Journal ot Genetic P .\ -hnlugy.
I ' I 59-"5.
c
00 3 3
PROBLEMS
Brophy. J.. A Good. T. L. (1986) Teacher behavior and siudent achievement. In M.C. W i n rock (Ed ). Handbook .." research or /,-4 v«. ':£r'3rd
ed.. pp. .-'8-375). New York: M a c M i l l a n .
,
>
Brunswick. V F.. & Bovie. J. M i l 9 7 9 i . Patterns of drug involvement:
Developr enial and secular influences on age ai initiation, 'ton: : -in-.:
Society, i i . 139-162.
1
Bry B. H. (19821. Reducing ihe incidence of adolescent problems
through preventive inlervenlion: One- and five-year followup. Amc-tcan Journal of Community Psycnoiogv. 10. 265-276.
Bry. B. H. (1983). Empirical foundationsof family-based approaches to
adolescent substance abuse. In T. J. G l y n n . C. G. Leukefeld. & J P.
Ludford (EdsJ. Preventing adolescent drugabuse: Imervennon uraiegies (pp. 154-17l-f N I D A Research Monograph No. 47). Washington. D C US. Government Printing Office.
Bry. B. H.. McKeon. P.. & Pandina. R. J. (1982). Extent of drug use asa
function of number of risk faaors. Journal ol'Abnormal Psychology.
91 273-279.
Bukoski. W J. (1986). School-based substance abuse prevention: A review-of program research. In S. Grisw-old-Ezekoye. K . L . Kumpfer. &
W J. Bukoski (Eds.). Childhood and chemical abuse (pp. 95-1 I5i.
New York: Haworth.
Bursik. R. J.. Jr. (in press). The use of contextual analysis in models of
criminal behavior. In J. D Haw-kins (Ed.). Some cur rc • theories m
delinquency and crime. New York: Springer-Verlag.
Bursik. R. J.. Jr.. & Webb. J. 11982). Communitv change and patterns of
delinquency. American Journal of Sociology. 88. 24-42.
Bush. P J.. Zuckerman. A. E.. Theiss. P K.. Taggan. V S.. Horowitz. C .
Sheridan. M. J.. S Walter. H. J. (1989). Cardiovascular risk factor
c
prevention in Black schoolchildren: Two-vear results of the "Know
Your Bodv" program. American Journal of Epidemiology. 129 466482.
Byram. O. W. & Fly. J. W (1984). Family structure, race and adolescent's alcohol use: A research note. American Journal of Drug ana
Alcohol Abuse. 10. 467-478.
Cadoret. R. J.. Cain. C. A.. & Grove. W M. (1980). Development of
alcoholism in adoptees raised apan from alcoholic biologic tziztives. Archives of General Psychiatry. ^S-. O b i - b O f
1 7
Cadoret. R. J.. & Gath. A. (1978). Inheritance of alcoholism m adoptees. British Journal of Addiction. 122. 252-258.
Cairns. R. B.. Cairns. B. D. Neckerman. H. J.. Gest. S. D. S Gairepy
c
J-L. (1988). Social networks and aggressive behavior: Peer suppon or
peer rejection'' Deveiopmemal Psychology 24. 815-S-3.
Catalano. R. F. Sc Hawkins. J. D. (1986). The social development model
A theory of antisocial behavior. Paper presented at the Safeco Lectureship on Crime and Delinquency. University o f Wash ington. Seattle.
Clayton. R. R. (in pressi. Legalization: A n idea whose time has noi
come. American Behavioral Scientist.
Cloninger. C. R.. Bohman. M.. Sigvardsson. S.. Sc von Knorring, A. L.
(1985). Psychopathology in adopted-out children of alcoholics: The
Stockholm Adoption Siudy. Recem Developments in Alcoholism. J.
37-51.
Cloninger. C. R- Sigvardsson. S.. & Bohman. M (1988) Childhood
personality predias alcohol abuse in young adults. Alcoholism. 12
494-503.
Coie. J. D. (1990). Towards a theory o f peer rejection. In S. R. Asher &
J. D. Coie ( E d ' ) . Peer rejection in childhood (365-398). New York.
Cambndge L iiversity Press.
Coie. J. D.. Sc Dodge. K. A. (1988). Multiple sources of data on social
behavior and social status in the school: A cross-age comparison
C h i l d Development. 59. 815-829.
Coie. J. D. Sc Krehbiel. G. (1984). Effects of academic tutoring on ihe
social siatus oflow-achieving. socially rejected chi Id ren. Child Deieiopmem. 55 1465-1478
•
�0036
D. H A W K I N S , R CATALANO. - . N D J
Comer. J. P. (19881 Educating poor minority children. V .cnnnc AmcriL-an :?<j 42-18.
Cook. P J.. & Tauchen. G. ! | 9 * 2 ) . The etTect of liquor ta.xes on heavy
drinking. Bell Journal :>' Economic IS. . " 0 . 3 9 0
Cook. P. J.. & Tauchen. G. (1984) The effect of m i n i m u m d r i n k i n g age
legislation on youthful auto fatalities. 1970-1977 Journal of Legal
Studies IJ. 169-190.
Cotton. N. S. (1979) The familial incidence of alcoholism: A review.
Journal 01 Sludies on Alcohol -it) 89- 1 16
Cowen. E.. & Work. W (1988). Resilient children, psychological wellness, and pnmary prevention. American Journal ol Communitv Psychology. 16. 591-607.
Decker. M. D. Graitcer. P. L . & Schaffner. W (1988). Reduction in
motor vehicle fatal ities associated wuh an increase in the m i n i m u m
drinking age. Journal o f the American Medical Association. 260.
3604-3610.
DeMarsh. J.. & Kumpfer. K. L. (1986) Family-oriented interventions
for the prevention o f chemical dependency in children and adolescents. InS. Gnswold-Ezekoye. K. L. Kumpfer. & W J Bukoski (Eds ).
Childhood and chemical abuse: Prevention and intervention (pp. I 17151). New York: H a w o n h .
Dembo. R.. Farrow D.. Schmeidler. J.. & Burgos. W (1979). Testing a
causal model of environmental influences on early drug invoivement of inner city junior high school youths. American Journal of
Drug and Alcohol Abuse. 6. 31 3-336.
DeVries. D. L . & Slavin. R. E. (1978). Teams-games-tournament
( T G T ) : Review o f ten classroom experiments. Journal of Research
and Deveiopment in Educaiion. 12. 28-38.
Dishion. T. J.. Kavanagh. K.. & Reid. J. B. (1989. November). Childrearing vs peer interventions in the reduction of risk for adolescenl
substance use and adjustment problems: A secondary prevention strategy. Paper presented at the Conference for the Advancement of Applied Behavior Therapy. Washington. DC.
Dolan. L.. Kellam. S.. & Biown. C. H. (1989). Short ierm impact o f a
mastery learning preventive intervention on eady risk behaviors. Baltimore: Johns Hopkins University.
Dwyer. J. H.. M a c K i n n o n . D. V. Pentz. M. A.. Flay. B. R.. Hansen.
W B.. Wang. E. Y. I.. & Johnson. C. A. (1989). Estimating interven.ion effects in longitudinal studies. American Journal of Epdemioiogy. 120. 781-795.
Ellickson. P. L.. & Bell. R. M. (1990) Drug prevention in junior high: A
multi-site longitudinal test. Science. 247 1299-1305.
Ellickson. P.. Bell. R. M . Thomas. M. A.. Robyn. A. E.. & Zellman.
G. L. (1988). Designing and implemennng Project A L E R T Santa
Monica. CA: The Rand Corporation.
Elliott. D. S.. Huizinga. D , &. Ageton. S. S (1985). Explaining delinquency and drug use. Beverly Hills. CA: Sage.
Elliott. D. S. Huizinga. D. &. Menard. S. (1989). Mulnple problem
youlh: Delinquency, substance use and mental health problems. New
York: Spnnger-Verlag.
Fagan. J. (1988). The sociai organization of drug use a n d drug dealing
among urban gangs. New York: John Jay College o f C r i m i n a l Justice.
Farrington. D. P.. Gallagher. B. Moriey. L . Ledger. R. J.. & West. D. J.
(1985). Cambridge study in delinqueni development Long term follow-up (First annual repon to the home office. August 31. 1985).
Cambridge. England: Cambndge University.
Farnngton. D. P. & Hawkins. J. D. (1991). Predicting panicipation.
early onset, and later persistence in officially recor led offending.
Criminal Behavior and Menial Health. I. 1-33.
Farrington. D. P.. Loeber. R.. Elliott. D. S. Hawkins. J. D. Kandel.
D. B.. Klein. M. W. McCord. J . Rowe. D. C . &. Tremblay. R. E.
(1990). Advancing knowledge aboui the onset of delinquency and
crime. In B. B. Lahc. & A. E. Kazdin (Eds.). Advances m clinical child
psychology-i Vol. 13. pp. 283-342). New York: Plenum Press.
MILLER
Feldhusen. J. F. Thurston. J. R.. SL Benmng. J. J ( I 9 " 3 ) . A longituomal
studv of delinquency and other aspects of children's behaviour. International Journal of Criminology and Penology I. 34 I - 35 I.
Felner. R. D. Si Adan. A. M. (1988> The school transitional environment project: An ecological intervention and evaluation. In R. H.
Price. E. L Cowen. R P. Lorion. & J. Ramos-McKay. (Eds.). /-<
Ounces 01 Prevention A CasenooK tor Pracnuonen ipp I 1 1-122).
Washington DC. American Psvcholoeical Association.
Felner. R. D. Adan. A. A.. & Evans. E (1987). Evaluation of schoolbased primary prevention programs Unpublished manuscript.
Felner. R. D.. Ginter. M. A.. & Primavera. J. (1982). Primary prevention during school transitions: Social suppon and environmental
struaure. American Journal of Communitv Psychology. 10. 227-24(1
Felner. R. D . Weissberg. R. P.. & Adan. A. M. (1987). Long-term followup of a school transition program. Unpublished manuscript.
Flasher. L. V. & Maisio. S. A. (1984). A review of theory and research
on drinking patterns among Jews. The Journal of 'Servous and Mental
Disease. 172. 596-603.
Flay. B. R. (1985). Psychosocial approaches to smoking prevention: A
review o f findings. Health Psychology 5. 449-188.
Flay. B. R.. Koepke. D.. Thomson. S. J.. Santi. S.. Best. J. A . & Brow n.
K. S. (1989). Six-year follow-up of the first Waterloo school smoking
,,,ev
on trial. American Journal of Public Health. 79. 1371-1376
Fleischman. M. J. (1981). A replication of Patterson's "Intervention for
Boys with Conduct Problems." Journal of Consulting and Clinical
Psychology. 49. 342-351.
Fleming. J. P.. Kellam. S. G . & Brown. C. H.(l982). Early predictors of
age at first use of alcohol, marijuana and cigarettes. D m g and Alcohol Dependence. 9. 285-303.
Fraser. M. W. Hawkins. J. D. SL Howard. M. O. (1988). Pareni iraining
Jor delinquency prevention. Btnghamion. NY: Hawonh Press.
Freiberg. H. J.. Brady. M. P.. Swank. P. R.. SL Taylor. R. D. (1989) Middle school interaction study of mainstreamed students. Journal of
Classroom Interaction. 24. 31^12.
Friedman. A. S. (1983. July). High school drug abuse clients. In Clinical
research notes. Rockville. M D : Division of Clinical Research. National Inslilute on Drug Abuse.
Garmezy. N.l 1985). Stress-resistant children: The search for protective
factcs. In '
""'evenson (Ed.). Recent research in developmenw
psychopathology ( p p . 2 i 3 - 233). Journal ofCh tld Psychology ..nd Psychiatry 4 (Book suppl.).
Gersick. K. E.. Grady. K.. & Snow. D. L. (1988). Social<ogniiive skill
development w i t h sixth graders and ils initial impact on substance
use. Journal o f Drug Education. 18. 55-70.
Gesten. E. L . Rains. M.. Rapkin. B. D. Weissberg. R P.. Flores de
Apodaca. R . Cowen. E. L..SL Bowen. R. (1982). Training children in
social problem-solving skills: A competence building approach,
first and second look. American Journal of Community Psychology.
10. 95-115.
Giltelman. R. S. Mannuzza. R. S.. & Bonagura. N. (1985). Hyperactive
boys almost grown up: I. Psychiatric status. Archives of General Psychiatry. 42. 937-947.
Goodwin. D. W (1985). Alcoholism and genetics: The sins o f the fathers. Archives o f General Psychiatry. 42. 17 I -174.
Goodwin. D. W. Schulsinger. F. Moller. N.. Hermansen. L.. Winokur
G . & Guze. S. B. (1974). Drinking problems in adopted and nonadopted sons o f alcoholics. Archives of General Psychiatry 21. 164169.
Goodwin. D. W. Schulsinger, F. Moller. N'.. Medmck. S.. SL Guze. S
(1977). Psychopathology in adopted and nonadopted daughters of
alcoholics. Archives of General Psvchiatry. 24. 1005-1007.
Gorsuch. R. L . & Butler. M. C. (1976) Initial drug abuse: A review of
predisposing social psychological factors. Psychological Bulletin S3.
120-137.
r
1
�RISK AND PROTECTIVE FACTORS FOR D*L'G PROBLEMS
Goul'redson. D. C. II9S6). An emoinca! lest ol'school-based ensironmental and individual interventions to reduce the risk ot'delinquent
behavior. Criminology. .'J ~ns_:}i
Gotlfredson. D. C (148S). .An evaluation of an organization deveiopment approach to reducing school disorder. Baiuauon Review: ! i
\W-"6--.
Gotlfredson. D C. & Cook. M. S iNSfjai. Incrcaung 7!.n>/ v/ci j ' t i v
and student deasummakme. Etieitot' v.rauvies lor reducing detinQuenc\ 'Unpublished manuscript Johns Hopkins L niversit>. Center
for the Social Organization of Schools. Baltimore.
Gotlfredson. D. C. & Cook. M. S. (1986bi. A test o f a school-based
program to reduce delinquency m increasing social integration. The
Milwood Alternative Education Project. Unpublished manuscript.
Johns Hopkins University. Cenier for the Social Organization of
Schools. Baltimore.
Gotlfredson. G. D. (1981). Schooling and delinquency. InS. E. Martin.
L. B. Sechrest. & R. Redner (Eds.). .Vew Directions m the rehabililaiion ol criminal offenders. Washington. DC: National Academv
Press.
Gotlfredson. G. D. (1988). Issues tn adolescenl drug use. Unpublished
final repon to the U S. Departmenl of Justice. Johns Hopkins University. Center for Research on Elementary and M iddle Schools. Baltimore.
Gotts. E. E. (I9&7I. HOPE revisited Preschool to graduation, retlections
on parenting and school-family relations Charleston. WV: Appalachia Educational Lab.
Grady. K.. Kelin. E. G.. & Boratynski. M (1985). Preparing parenis for
teenagers: A step in the prevention of adolescent substance abuse
Familv Relations. 34 54I-5J9.
Gurling. H. M. D. Clifford. L. A.. &. Murray R. M. (1981). Geneuc
contribution to alcohol dependence and us effects in brain function
In L. Gedder. P Pinsi. &(Nance/fw A^lEds.). Twin researchMpp.
? 77-87^New York: Alan R.Tlis?
Hansen, w; B.. Graham. J W. Sobel. J. L.. Shelton. D. R.. Flay. B. R..&.
Johnson. C. A. (1987). The consistency of peer and parent influences
on tobacco, alcohol, and marijuana use among young adolescents.
Journal ol Behavioral Medicine 10 559-579
Hansen. W B.. Johnson. C. A.. Flay B. R.. Graham. J. Wi. & Sobel. 1
,i .oo. Affective and social influences approaches to t --prevention
of multiple substance abuse among seventh grade students: Resulls
from ProjectSMART Preventive Medicine. 17. 135-154.
Harada. S.. Agarwal. D P.. Goedde. H. Wi. & Ishikawa. B. (1983). Aldehyde, dehydrogenase isozyme variation and alcoholism in Japan.
Pharmacology. Biochemisirv and Behavior. 18. 151-153.
Harford. T. C. (1985). Drinking patterns among Black and non^lack
adolescenis: Resulls of a national survey In R. Wright & T. D. Vans
( Eds.). Prevention of Black, alcoholism: Issues and strategies. Springfield. IL. Charles C Thomas.
Hanup. W W (1983). Peer relations. In P. H. Mussen (EdJ. Handbook
of child psychology: Vol. IV. Socialization, penonaluy. and social development. New York: Wiley.
Hawkins. J. D. Abbott. R. Catalano. R. F . & Gillmore. M. R. (1991).
Assessing effectiveness of drug abuse prevention: Long-term effects
and replication. In C. Leukjfeld & W Bukoski (Eds.). Drug abuse
prevention research: Methodological issues (NIDA Research Monograph Series No. 107. DHHS Publication No. ADM 91-1761. pp.
195-212). Washington. DC: US. Government Printing Office.
Hawkins. J. D. Catalano. R. F. Jones. G.. & Fine. D. N. (1987). Delinquency prevention through parent iraining: Resulls and issues from
work in progress. In J. G. W'ilson.&G. C. Loury (EdsJ. From children
to citizens: Families, schools, and delinquency prevention (Vol. 3. pp.
186-204). New York: Springer-Verlag.
Hawkins. J. D. Catalano. R. F. Morrison. D M.. O'Donnell. J.. Abbott.
R D.. & Dav. L. E. nn pressi. The Seattle Social Development ProjH
:
1
00 5 •
ect- Effecis of lhe tirst four vears on protective 'actors and proolcns
behaviors. In J McCord & R Tremblay lEds.i. The prcvcniio-i ,.,
umisoaal benavior m cnudren New York: Guilford Press.
Hawkins. J D. Doueck. H. J.. i Lishner. D M 1198.81. Changing tcacv
ing practices in mainstream classrooms to improve bonding j r , .
behav ior of low ac hie vers. Amencar. Educanonm Research J >t/"w.31-50.
Hawkins. J. D.. Jenson. J. M.. Catalano. R. F. & Lisnner. D. M i l ^ s )
Delinquency and drug abuse: Implications for social serv ices. S/'ciat
Service Review, o:. 258-284.
Hawkins. J. D.. & Lam. T 11987). Teacher practices, social Jeveiopmem. and delinquency. In J. D. Burchard & S. N. Burchard (Eds.).
The prevention of delinquent behavior {pp. 24 1 - 274). Newburv Park.
CA: Sage.
Hawkins. J. D.. Lishner. D.. &. Catalano. R F.(l985) Childhood predictors and the prevention of adolescent substance abuse In C. L. Jones
&. R. J. Battjes (Eds.). Etiology of drug abuse: Implications lor preven//oil NT DA Research Monograph No. 56. DHHS Publication A DM
85-1335. pp. 75-126). Washington. DC U.S. Government Pnnnng
Office.
Hawkins. J. D. Lishner. D. M.. Catalano. R. F. & Howard. M. O. (1986 >
.
Childhood predictors ofadolescent substance abuse: Towardan empirically grounded the:i v Journal ot Children inConiempif'irv '
ety. S I 1-48.
Hawkins. J. D. Lishner. D M.. Jenson. J. M.. & Catalano. R. F. (1987).
Delinquents and drugs: W hat the evidence suggests about prevention and treatment programming. In B. S. Brown & A. R. Mills
(Eds.). Youlh at high risk tor substance abusei DH HS Publication No
A DM 87-1537. pp Sl-133). Washington. DC- US. Governmenl
Printing Office.
Hawkins. J. D. V'on Cleve. E . & Catalano. R. F. (1991). Reducing earlv
childhood aggression: Results of a primary prevention program.
Journal of'the American Academy of Child and Adolescenl Psvchiatry.
30. 208-217.
Hawkins. J. D_& Weis. J. G. (1985). The social development model: An
integrated approach to delinquency prevention. Journal of Primarv
Prcvenlion. 6. 7 3-97
Hening. J. R.. &. Guesl. A M. (1985). Componemsofsaiisfaction wnh
local areas in . metropolis. The Sociological Quartern: 2t>. 99-! ; 5
Hetherirgton. E. M.. Con. M.. & Cox. R. (1979). Play and social interaction in children following divorce. Journal ol Social Issues. 35.
26-49
Hirschi. T. (1969). Causes of detmquencv. Berkeley: University of California Press.
Holder. H. D. & Blose. J. O. (1987). Impact of changes in distilled
spirits availability on apparent consumption: A time series analysis
of liquor-by-the-drink. British Journal of Addiction. 82. 623-631.
Holmberg, M B. (1985). Longitudinal sludies of drug abuse in a fifteen-year-old population: I . Drug career. Acta Psychiatnca Scandmavica. 71. 67-79.
Horacek. H. J. Ramey C. T. Campbell. F. A.. Hoffmann. K. P.. &
Fletcher. R. H. (1987) Predicting school failure and assessing earlv
intervention with high-risk children. Journal ol the American Academv of Child and Adolescent Psychiatry. 26. 758-763.
Hrubec. Z.. & Omenn. G. S. (1981). Evidence of genetic predisposition
to alcoholic cirrhosis and psychosis: Twin concordances for alcoholism and its biological end points by zygosity among male veterans.
Alcoholism. 5. 207-215.
Hundleby. J. D. & Mercer. G. Wi (1987). Family and friends as sociai
environments and their relalionship to young adolescents' use of
alcohol, tobacco, and marijuana. Journal of Clinical Psvcholog . J-'
I2.--1.W.
Jessor. R. (1976). Predictine time of onset of marijuana use- A deveioo
�00
D
HAW KIN'S. R C A T A L A N O . A N D J M I L L E R
Jessor. R.. Donovan. J. E.. i Wmdmer. K. 11980). PswiitHnaui ''actor-,
.n aaoU-icen: auvnoi ar.d drug uie The ! 9-80 Saiuwcl Sampie Stud::
and'he I 9 ~ J - ~ : ; Panei Studv Unpublished linal report. Universitv
ol'Colorado. Institute of Behavioral Science. Boulder.
grapn. D H H S Publication No. A D M S M - i T . PP- ! ' i - l 881 Wasnington. DC: U.S. Government Printing Office.
Kandel. D. B.. Kessler. R . C . i Margulies. R.S ( l 9 " S i . Antecedents of
adolescenl initiation into stages of drug use: A developmental
zmiy
sis. Journal of Youlh and Admescem e '. .5—10
Kandel. D. Simcha-Fagan. O. i i Davies. M. '19861 Risk factors for
Jessor. R . & Jessor. S L. (19""). Problem hehauor and purhosoetai
deveiopment 4 iongtiudtnal uudx ot youth San Diego. C.A: Academic Press.
delmguency and illicit drug use from adolescence to young adulthood. The Journal oi Drug Issues. /'"> 67-40.
Kandel. D B.. Single. E.. & Kessler. R (I9"6). The epidemiology of
Johnson. C. A.. Pentz. M. A.. Weber. M. D. Dwyer. J. H.. M a c K i n n o n .
D. P.. Flay. B. R . Baer. N. A.. & Hansen. W B (1989) The relative
effectiveness of comprehensive community programming for drug
abuse with risk and low nsk a^o/e.uen^Qep^rnenxihlLErevemior^
.Research^Unpublished manuscript. University o f Southern CaliforniaTPasadena.
drug use among New- York State High School students. Distribution.
mental stud) ot'hign school youth
Jnurr.ai>>'C>m\uiiiniiunai'iimcsi
Johnson. C. A.. & Solis. J. (1983). Comprehensive community programs for drug abuse prevention: Implications of the community
heart disease prevention programs for future research. National Institute on Drug Abuse Monograph Series. 47. 76-1 14.
Johnson. G. M.. Schoutz. F. C . & Locke. T. P. (1984). Relationships
beiween adolescent drug use and parental d r u g behaviors. Adolescence. 19. 295-299.
Johnston. L. D. (1985). The etiology and prevei . n of substance ase:
What can we learn from recent historical changes' In C. L. Jones. &
R. J. Battjes (Eds.). Etiology of drug abuse: Implications for prevention ( N I D A Research Monograph No. 56. D H H S Publication No.
A D M 85-1335. pp. 155-177). Washington. D C : U.S. Governmenl
Printing Office.
Johnston. L. D. (1991). Toward a theory of drug epidemics. In L. Donohew. H. E. Sypher. &. W J. Bukoski (Eds.). Persuasive communication
and drug abuse prevention (pp. 93-131). Hillsdale. NJ: Erlbaum.
Johnston. L. D . Bachman. J. G.. & O'Malley. P. M . (1981). Student drug
use in America 1975-1981 ( D H H S Publication No. A D M 82-1208).
Washington. DC: U.S. Government Printing Office.
Johnston. L. D. O'Malley. P M.. & Bachman. J. G. (1985) Use of licit
and illicit drugs by America's high school students I'975-1984. Rockville. M D : Nalional Institute of Drug Abuse.
Johnston. L. D . O'Malley. P. M.. &. Bachman. J. G. (1988). Na'ional
trends in drug use and related factors among American hitth school
students and young adulls. Rockville. M D : Nalional Inslilute on
Drug Abuse.
Johnston. L. D . O'Malley. P M.. &. Bachman. J. G. (1989). Drug use.
drinking, and smoking National survey resulls f r o m high school, college, and young adull populations ( D H H S Publication No. A D M
89-1638). Washington. DC. U.S. Government Printing Office.
1
Joksch. H. C. (1988). The impact of severe penalties on drinking and
driving. Washington. D C : A A A Foundation for Traffic Safety.
Kaij. L. (I960). Alcoholism iniwins Studies on the etiology and sequelae
of abuse of alcohol. Stockholm. Sweden: Alonguist Sc Winkell.
Kandel. D. B. (1978). Homophily seleaion and socialization in adolescent friendships. American Journal of Sociology. 84. 427-436.
Kandel. D. B. (1982). Epidemiological and psychosocial perspeaives
on adolescent drug use. Journal of American Academic Clinical Psychiairv. 21. 328-347.
Kandel. D. B. (1986). Processes of peer influence in adolescence. In R.
Silberstein (Ed.). Developmem as action in context: Problem behavior
and normal vouih developmem (pp. 203-228). New York: SpringerVerlag.
Kandel. D. B..&Andrews. K.(I987). Processes ofadolescent socialization by parents and peers. Imemational Journal oflhe Addictions. 22.
319-342.
Kandel. D B.. & Davies. M. (1991). Cocaine use in a national sample o f
U.S. youlh ( N L S Y ) : Epidemiology, predictors, and ethnic patterns.
In Thecptdcniologvoicmatnetucundabuse! N I D A research m o n o
trends, and change in ratesof use. American Journal of Public Healih.
66 43-53
Karolv. P.. & Rosenthal. M. (1977). Training parents in behavior m o d i fication: Effects on perceptions of familv interaction and deviant
child behavior. Behavior Therapy 8. 406-410.
Kaufman. E.. S Kaufman. P. N. (1979) Familv therapy of drug and
c
alcohol abuse. New York: Gardner Press.
Kazdin. A. E. (1990. June). Prevention of conduct disorder Paper presented to the National Conference on Prevention Research. National Institute of Mental Health. Bethesda. Maryland.
Kellam. S. G.. S Brown. H. (1982). Social adaptanonaland psychologic
cal antecedents of adolescent psychopatholog) ten years later. Baltimore, .ohns Hopkins Universitv.
Kellam. S. C Ensminger. M. E.. & Simon. M. B. (1980). Menial health
in first grade and teenage drug, alcohol, and cigarette use. Drug and
Alcohol Dependence. 5.2''3-304.
Kelly. D. H..& Balch. R. W i ( i 9 7 [ ) . Social origins and school failure: A
reexamination of Cohen's theory of working-class delinquency. Pacific Social Review. 14. 413-430.
Kelchel. J. A.. Sc Bieger. G. (1989. April). The efficacy of a psvehosociallv based drug prevention program for young adolescents. Paper
presented at the New England Educational Research Corporation
Annual Meeting. Portsmouth. N H .
K i m . S. (1979). An evaluation o f ombudsman primary prevention program on studenl drugabuse. Charlotte. NC: Charlotte Drug Education Cenier.
K i m . S . . McLeod. J..&Palmgren. C. L 11989) The impact of t h e ' T m
Special" program on studenl substance abuse and other related student problem behavior. Journal o f Drug EStion 19. 83-95.
K l e i n . N. C . Alexander. J. F.. Sc Parsons. B. V (1977). I m p a a of family
systems imerveniion on recidivism and sibling delinquency: A
model of primary prevention and program evaluation. Journal oj
Consulting and Clinical Psychology. 45. 469-474.
Klepp. K.. Halper. A.. S Perry. C. L. (1986). The efficacy of peer
c
leaders in drug abuse prevention. Journal of School Health. 56. 4 0 7 411.
Knight. R. C . Sheposh. J. P.. Sc Bryson. J. B. (1974). College student
marijuana use and societal alienation. Journal of Health and Social
Behavior. 15. 28-35.
K n e g , T. L. (1982. December). Is raising the legal drinking age
watramed? The Police Chief pp. 32-34.
K r o h n . M. D. Sc Massey. J. L. (1980) Social control and delinquent
behavior: An examination of the elements of the social bond. Sociological Quarterly. 21. 529-543.
Krosnick. J. A . & Judd.C. M. (1982). Transitions in social influence at
adolescence: W h o induces cigarette smoking? Developmental Psvchology. 18. 359-368.
Kupersmidt. J. B.. Coie. J. D. Sc Dodge. K. A. (1990) The role of poor
peer relationships in the development of disorder In S. A Asher Sc
J. D. Coie. Peer rejection in childhood {pp. 274-305). Cambridge. England: Cambridge University Press.
Labouvie F. W. Sc McGee. C. R. (1986). Relation of personality to
alcohol and drug use in adolescence. Journal of Consulting and Clinical Psvchology. 54 289-293
�KISK A N D P R O T E C T I V E FACTORS FOR DRUG PROBLEMS
Laod. G. W . l ^ S n . Effectiveness of a social learning meihod lor enhancing children's social interaction i n d peer acceptance C h i l d Di'v e h i n m c n . : 2 I " I - I "S
f
Ladd. G. W. i Asher. S. R. (198?!. Sociai skill training and children's
peer relations Current issues m research and practice. In L. L
Abate.'i M. A. Milan (Eds.). Handb<*<k »i soaal sKdh iraimne ana
researcl; i pp. 219-2441. New Y x k v^'iie;..
Lazar. 1.. Darlington. R.. Murrav. H.. Ro>ce. J.. £ Snipper. A. (1982)
Lasting effects of early education: A repon from the consortium for
longitud inal studies. Monographs ot the Societv lor Research in C h i l d
Development. 47. (2-3. Serial No. 195).
Lerner. J V. &. Vicary. J. R. (1984). Difficult temperament and drug
use: Analyses from the New York longitudinal study. Journal ot Drug
Education. 14. 1-8.
Lettieri. D. J.. Savers. M.. & Pearson. H. W (1980). Theories on drug
abuse Selected contemporarv perspectives ( N I D A Research Monograph No. 30. DHHS). Washington. DC: U.S. Government Printing
Office.
Levy. D.. & Sheflin. N. (1985) The demand for alcoholic beverages: An
aggregate lime-series analysis. Journal of Public Poiicy and Marketing. 4 47-54.
Lewis. C. E.. Robins. L. N.. & Rice. J (1985). Association ofalcoholism
with antisocial personality m urban men. Journal ot Servous and
Mental Disease. 17}. 166-174
L i . T. K. (1977). Enzymology of human alcohol metabolism. In A.
M e i s t e r l E d ) . Enzymology ot human alcohol metabolism New ibrk:
Wiley.
Lochman. J. L. (1988. August) Cogninve-nehaviorai intervention with
aggressive boys: Three vear follow-up etiects Paper presenied at the
96ih Annual Convention of the American Psychological Association. Atlanta. GA.
Loeber. R. (1988). Natural historiesofconduct problems, delinquency,
and associated substance use: Evidence for developmental progressions. In B. B. Lahey & A. E. Kazdin (Eds.). Advances in clinical child
psychology (Vol. I I. pp. 73-124). New York: Plenum Press.
Loeber. R. T. &. Dishion. T. (1983). Early predictors of male delinquency: A review. Psychological Bulletin. 93 68-99.
Loeber. R . & Stouthamer-Loe^.. . M. S. (I 986). Family faciors as correlates and predictors o f juvenile conduct problems and delinquency.
In M. Tonry & N. Morris (Eds.). Cnme and justice (Vol. 7. pp. 29149). Chicago: University of Chicago Press.
Loney. J.. Kramer. J.. & M i l i c h . R. (1979). The hvperkinetic child grows
up- Predictors ot symptoms, deiinquencv. and u^hievemeni at followup. Paper presented to the American Association for the Advancement of Science. Houston. T X .
Maddahian. E.. Newcomb. M. D, & Bentler. P M. (1988). Adolescent
drug use and intention io use drugs: Concurrent and longitudinal
analyses of four ethnic groups. Addictive Behaviors. 13. 191-195.
Madden. N. A.. & Slavin. R. E. (1983). Cooperative learning and social
acceptance of mainstreamed academically handicapped students.
Journal of Special Education. 17. 171-182.
Marley. R J.. Miner. L. L . Wehner. J. M.. & Collins. A. C. (1986).
Differential effects of central nervous system depressants in longsleep and shon-sleep mice. Journal of Pharmacology Experimental
Therapeutics. 238. 1028-1033.
M a n i n . B. (1977). Brief family therapy intervention: Effectiveness and
the importance of including the father. Journal of Consulting and
Clinical Psvchology. 45. lOOI-lOlO.
McAlister. A. L. (1983). Social-psychological approaches. In T. J.
Glynn. C. G. Leukefeld. & J. P. Ludford (Eds). Preventing adolescenl
drugabuse: Intervention siraievies ( NI DA Research Monograph No.
47. pp. 36-50). Washington. DC: L'.S Government Printing Office.
McAlister. A L . Perry. C L.. K i l l e n . J . Slinkard. L. A . i Maccoby. N.
0039
:l980i. Pilot siudy of smoking, alcohol, ana drug aruse prevention.
American Journal <» Public Healih '0 " ! 9 - " 2 l .
.
McCord. J. 0 979). Some child-rearing antecedents of criminal bertiv ior in adult men Journal ot Persona::;': ar.d W K S ; PSVCIOIOI"
1477-1486
McCord. J 11981). Alcoholism and criminality Joumci •" Studies o,Alcohol, a. . ::.9-7ax.
1
McCord. W. & McCord. J. (1959). Origins •>•• , r:»v
Columbia University Press.
Montciair. NJ.
McDermott. D (1984). The relationship of parental drug use and parent's attitude concerning adolescent drug use to adolescent drug use.
Adolescence. 19 89-97.
McKiniay. S. M.. Stone. E. J.. & Zucker. D. M. (1989). Research design
and analysis issues. Health Education Quaneriy 16. 307-313
Moskowuz. J. M. (1989). The p n m a r v prevention of alcohol problems.
A critical review of the research literature. Journal of Studte- on
Alcohol. 50. 54-88
Moskowitz. J. M..&. Jones. R.(1988). Alcohol and drug problems in the
schools: Results of a nalional survey of school administrators. Journal of Sludies on Alcohol. 49 299-305
Murray. C. A. (1983). The physical environment and community control of crime. In J. Q. Wilson (Ed.). Crime and Pubiic Policy (pp.
107-122). San Francisco: Institute for Contemporarv Studies.
Murray. D. M.. Davis-Hearn. M.. Goldman. A. I.. Pine. P.. & Luepker.
R. V (1988). Four and five-year follow-up results from four seventhgrade smoising prevention strategies. Journal of Behauoral Medicine.
I I . 395-105. "
Murray, D. M.. & Hannan. P. J (1990). Planning for the appropriate
analysis in school-based drug-use prevention studies Journalt.>fConsulting ana Ciinieai Psvchology 5S. I - I 1.
Murray. D. M.. Hannan. P. J.. & Zucker. D. M.(|989). Analysis issues in
school-based health promotion studies. Health Education Quanerh.
16. 315-320.
Murray. D. M.. Johnson. C. A.. Luepker. R. V. & Mittelmark. M. B
(1984). The prevention of cigarette smoking in children: A comparison of four strategies. Journal o f Applied Sociai Psvchology. 14. 274288.
Murray D. M.. Pirie. P.. Luepker. R. V. & Pallonen. U. (1989). Five and
six-year follow-up results from four seventh-grade smoking prevention strategies. Journal of Behavioral Medicine. 12. 207-218.
Murray. D. M.. Richards. P. S.. Luepker. R. V. & Johnson. C. A (198").
The prevention of cigarette smoking in children: Two and ihree-year
follow-up comparisons of four prevention strategies. Journal of Behavioral Medicine. 10. 595-61 I.
Murray, R. M.. & Slabenau. J. R. (1982). Genetic factors in alcholism
predisposition. In Encyclopedic handbook oj alcoholism (pp. 135144). New York: Gardner Press.
Newcomb. M. D.. & Bentler. P. M. (1986). Substance use and eihnicity
Differential i m p a a of peer and adull models. Journal o f Psychology.
120. 83-95.
Newcomb. M. D. &. Bentler. P. M. (1988). Consequences ofadolescem
drug use: Impacl on the lives o f young adults. Newbury Park. C.A.
Sage.
Newcomb. M. D.. Maddahian. E.. & Bentler. P. M. (1986) Risk factors
for drug use among adolescents: Concurrent and longitudinal analyses. American Journal of Public Health. 76. 525-530.
Newcomb. M. D. Maddahian. E.. Skager. R.. &. Bentler. P. M. (1987,.
Substance abuse and psychosocial risk factors among teenagers: Associations with sex. age. ethnicity, and type of school. American
Journal of Drug and Alcohol Abuse. 13. 41 3-433.
Norem-Hebeisen. A.. Johnson. D Wi. Anderson. D.. & Johnson. R
(1984). Predictors and concomitants o('changes in drug use patterns
among teenagers. The Journal „ i S'.'ual Psvaiology. •2-' 43-50
�l)04()
0
HAWKINS. R CATALANO
Nviander. I iHTOi The developmem oi'amibocial behaviour in children .-it-w Paedopsvchiainca. •>'. " 1 - 8 0
O'Donnell. J. A.. &. Clavion. R. R. il9T9i. Determinants ol'early marijuana use. In G. M. Beschner & A S. Friedman lEds.i. Ymun druv
a h t w Problems, issues, ana ireaimemi pp. 6 5 - I ! 0 l . Lexington. M.A.
Lexington Books.
Olds. D. L . Henderson. C R .Chamberlin. R . i Tateibaum. R.i 19861.
Preventing child abuse ana neglect: A randomized trial ot' nurse
home visitation. Pediamcs. ~S o 5 - " 8 .
Parker. J. G.. & Asher. S. R. 11987). Peer relations and later personal
adjustment: Are low-accepted children at risk'' Psychological Bulletin. 102. 357-389.
Paton. S. M.. & Kandel. D. B. (1978). Psychological factors and adolescent illicit drug use: Ethnicity and sex difference. Adolescence. 13.
187-200.
Paton. S.. Kessler. R_ & Kandel. D. (1977). Depressive mood and adolescem drug use: A longitudinal analysis. Journal ot Oeneuc Psychology. 131. 267-289.
Patterson. G. R . Chamberlain. P.. &. Reid. J. B. (1982). A comparative
evaluation of a parent training program. Behavior Therapy. 13. 638650.
Patterson. G. R.. & Fleischman. M. J. (1979) Maintenance of treatment effects: Some considerations concerning family systems and
follow-up data. Behavior Therapv. 10. 168-185.
Patterson. G. R.. & Reid. J. B. (1973). Intervention for families of aggressive boys: A replication study. Behavior Research The-apy. I I .
383-394.
Peed. S.. Roberts. M.. & Forehand. R. (1977). Evaluation of the effectiveness of a standardized parent training program in altering the
interaction of mothers and their noncompliant children. Behavior
Modification. I. 323-350.
Penning. M.. & Barnes. G. E. (1982). Adolescent marijuana use: A
review. Internaiional Journal of Addictions. 17. 749-791.
Pentz. M. A.. Brannon. B. R.. Charlin. V L.. Barrett. E. J.. M a c K i n n o n .
D P.. & Flay. B. R. (1989). The power of policy: Relationship of
smoking policy to adolescent smoking. American Journal of Public
Health. 79. 857-862.
Pentz. M A.. Dwyer. J. H . M a c K i n n o n . D. P.. Flay. B. R . Hansen.
W B.. Wang. E. Y. L. & Johnson. C. A. (1989). A multi-community
trial for primary prevention of adolescent drug abuse: Effeas on
drug use prevalence. Journal of the American Medical Associaiion.
261. 3259-3266.
Pentz. M. A.. M a c K i n n o n . D. P.. Flay. B. R.. Hansen. W B.. Johnson.
C A . & D w y e r . J H.(I989). Primary prevention o f chronic diseases
in adolescence: Effecis of the Midwesiern Prevention Projea on
tobacco use. American Journal of Epidemiology. 130. 713-724.
Perry. C. L. (1986). Community-wide health promotion and drugabuse
prevention. Journal of School Health. 56. 359-363.
Perry. C. L.. Crockett. S. J- <fc Pirie. P. (1987). Influencing parental
health behavior: Implications of community assessments. Health
Education. 18(5). 68-77.
Perry. C. L . Klepp. K. 1. & Shultz. J. M . (1988). P n m a r y prevention o f
cardiovascular disease: Communitywide strategies for youlh. Journal ol'Consulung and Clinical Psychology. 56. 358-364.
Perry. C. L.. Luepker. R. V. Murray. D. M., K u r t h . C . Mullis. R . Crockett. S. & Jacobs. D. R. (1988). Parent involvement with children's
health promotion: The Minnesota Home Team. American Journal
o f Public Health. 78. 1 156-1 160.
Pierson. D. E.. Bronson. M B.. Dromey. E.. Swanz. J. P.. Tivnan. T . &L
Walker. D. K. (1983). The i m p a a of early education as measured by
classroom observations and teacher ratings of chi Id ren in kindergarten. Evaluation Review '. 191-216.
Polich. J. M . Ellickson. P. L.. Reuter. P.. & Kahan. J. P. (1984). Strate-
AND J MILLER
gies tor cor.trnHme aaoieHem drug .<»«• Santa Monica. C.A: Raau
Corporation
Pollock. V E.. Volavka J.. & Goodw-.n. D W 1198}). The EEG after
alcohoi administration in men at risk ('or alcoholism j r e n i v a •>'•
General Psvchtatrv 41/ 857-86!
Porter. B i O'Leary K D i l 9 8 0 i . Marital discord and childhood
problems. Journal oi Abnormal Chad Pwcnoiogy S. 28"-295
Puska. P.. Vartianen. E.. Pallonen. L.. Salonen. J. T. Poyma. P.. Koskela. K . & McAlister. A. (1982) The N o n h Karelia Youth Project.
Evaluation of two years of intervention on health behavior and O ' D
risk faaors among 1 3- to 15-year-old children. Preventive Medicine
11. 550-570.
Rachal. J. V. Guess. L. L.. Hubbard. R. L.. Maisto. S. A.. Cavanaugh.
E. R- Waddel. R.. & Benrud. C. H. (1982). Facts for planning No. 4:
Alcohol misuse by adolescents. Alcohol Health and Research World
6(3). 61-68.
Ramey. C. T . Bryant. D. M.. Campbell. F. A.. Sparling. J. J.. & Wasik.
B. H. (1988). Early intervention for high-risk children: The Carolina
Early Intervention Program. In R. H. Price. E. L Cowen. R P. Lorion. Sc J. Ramos-McKay (Eds.). 14 ounces of prevention. A casebook
for practitioners (pp. 32-13). Washington. DC: American Psychological Association.
Raymond. M. R. (1987). Missing data in evaluation research. Toalualion and the Health Professions. 9. 395-420.
Reilly. D. M. (1979). Family factors in the etiology and treatment of
youthful drug abuse. Family Therapv l l . 149-171.
Reiss. A. J. (1986). Why are communities imponant in understanding
crime'? In A. J. Reiss S M. Tonry (Eds.). Communities and crime.
c
Chicago: University of Chicago Press.
Robins. L. N. (1978). Sturdy childhood predictors of adult anti-social
behavior: Replicationsfrom longitudmalsiudies. Psychological Medicine. 8. 611-622.
Robins. L. N. (1980). The natural history o f drug abuse. Acta Psychtatrica Scandinavia. 62(Suppl. 284). 7-20.
Robins. L. N. (1984). The natural history of adolescent drug use. American Journal of Public Health. 74. 656-657.
Robins. L. N'.. Sc Przybeck. T. R. (1985). Age of onset of drug use as a
factor in drug and other disorders. In C. L. Jones Sc R. J. Battjes
(EdsJ. Etiology of Drug Abuse: Implications for Prevention ( N I D A
Research Monograph No. 56. D H H S Publication No. A D M 851335. pp. 178-192). Washington. DC: U.S. Government Printine Office.
Robins, L. N'..& Ratcliff. K. S. (1979). Continuation ofanlisoc.al behavior into adulthood. International Journal of Mental Health. '. 96116.
Rolheram. M. J. (I982al. Asseniveness training in f o u n h and fifthgrade children. American Journal of Communitv Psvchology. 10.
567-582.
Rotheram. M. J. (1982b). Social skills training with underachievers.
disruptive, and exceptional children. Psvchology in the Schools. 19
532-539.
Rotheram. M_ Armstrong. M_ Sc Booraem. C. (1986). Asseniveness
training with elementary school children. American Journal o f Community Psychology. 10. 567-582.
Rutter. M. (1980). Changing youth in a changing society. Cambndge.
M A : Harvard University Press.
Rutter. M. (1985). Resilience in the face of adversity: Protective factors
and resistance to psychiatric disorder. British Journal of Psychiatry
147. 598-611.
Rutter. M.. Sc Giller. H. (1983). Juvenile delinquency: Trends and perspectives. New York: Penguin Books.
Saffer. H.. S Grossman. M. (1987). Beer taxes, the legal d r i n k i n g age.
c
�004 1
RISK AND PROTECTIVE FACTORS FOR DRLG PROBLEMS
and vouth motor •-ehioe lata I ities. journal of L a d Siuaws. If). 351Sampson. R J. i M j h i Crime in cities. The erfects ot formal and informal social control. In A. J Reiss A M Tonrv iEds.). Cnme and
lusuce in annual r^vie^. (U research U.'i $ Communilies ana crime
ipp. 2"1-3IM Chicaeo Universitv of Chicago Press.
Sampson. R. J.. Castellano. T. C. & Laub. J. H 1198 I). Ju\eniiecriminal
heha\ior and us reianon io ncmnporhood characteristics. Washington.
DC: Office of Juvenile Justice and Deiinquencv Prevention.
Schaie. K. w: (1965). A general model for the studv of developmental
problems. Psvcholoittcal Bulletin. 04. 92-107.
Schaps. E. Moskowitz. J. M.. Malvm. J. H . & Schaeffer. G. H. (1986).
Evaluation of seven school-based prevention programs: A final reporl on the Napa Project. International Journal ofthe Addictions. 21.
1081-1 1 12.
Schinke. S. P.. Bebel. M. Y.. Orlandi. M. A.. St Botvm. G. J. (1988).
Prevention strategies for vulnerable pupils: School social work practices to prevenl substance abuse. Urban Educaiion. 22. 5 10-5 19.
Schinke. S. P.. Botvin. G. J.. Trimble. J. E.. Orlandi. M. A.. Gilchrist.
L. D. & Locklear. V S. (1988). Preventing substance abuse among
American-Indian adolescenis: A bicultural competence skills approach. Journal ol Counseling Psychology. 35. 87-90.
Schlegel. R.. & Sanborn. M. (1979). Religious affiliation and adolescent drinking. Journal of Studies on Alcohol. 40. 693-703.
Schuckit. M. A. (1980). Self-rating alcohol intoxication by young men
with and without family histories of alcoholism. Journal of Studies
on Alcohol. 41. 242-249.
Schuckit. M. A (1987). Biological vulnerability to alcoholism. Journal
of Consulting and Clinical Psychology. 55. 301-309.
Schuckit. M. A.. Parker. D. C. & Rossman. L. R. (1983). Eihanol-related prolactin responsesand risk foralcoholism. Biological Psychiatry IS. I 153-1 159.
Schuckit. M. A.. &. Rayes. V (1979). Ethanol ingestion: Differences in
blood acetaldehyde concentrations in relatives of alcoholics and
controls. Science. 203. 54-55.
Seitz. V. Rosenbaum. L. K.. &. Apfel. N. H (1985). Effects of familysupport intervention: A ten-year follow-up. Child Development. 56.
376-391.
jw. G. Wi (1987). Parent-child relationships and single and two
parent families: Implications for substance usage. Journal of Drug
Education. 1 7. 315-326.
Shaw. C. R.. & McKay. H. D. (1969). Juvenile delinquency and urban
areas (Rev ed.). Chicago: University of Chicago Press.
Shedler. J.. & Block. J. (1990). Adolescent drug use and psychological
health: A longitudinal inquiry. Amencan Psychologist. 45. 612-630
Shure. M. B.. & Spivack. G. (1982). Interpersonal problem-solving ii
young children: A cognitive approach to prevention. American Journal oi' Communiiy Psychology. 10. 341-356.
Simcha-Fagan. O.. Gersten. J. C & Langner. T. (1986). Early precursors
and concurrent correlatesof illicil drug use in adolescents. Journal ol
Drug Issues. 16. 7-28.
Simcha-Fagan. O.. S Schwartz. J E. (1986). Neighborhood and delinc
quency: An assessment of contextual effects. Criminology 24. 667703.
Slavin. R. E.. Madden. N. .A_ Karweit. N. L Liverman. B. J.. S Dolan.
c
L. (1990). Success for all: First year outcomes of a comprehensive
pian for reforming urban education. American Educational Research Journal. 27. 255-278.
Smith. G. M.. S Fogg. C. P. (1978). Psychological predictors ol'early
c
use. late use. and non-use of marijuana among teenage students. In
D. B. Kandel (Ed.). Longitudinal research on drug use: Empirical
findings and methodological issues ipp. 101-1 13). Washington. DC:
Hemisphere-W'iley
1
r
Spivack. G. (1983). Hign ;sk eariv behaviors indicating •••uineraruits to
delinquency in the communitv and school National Institute of Juvenile Justice and Delinquency Prevention. Otfice of Juvenile Justice
and Delinquency Prevention. Law Enforcement Assistance Administration. Washington. DC: U.S. Government Printing Office.
Spivack. G . S Shure. M. B. 119 4). Socta: adjustment ot voungcnudren
c
San Francisco Josses-Bass.
Stanton. M. D. (1979). The client as familv member: Aspects of continuing treatment. In B. S. Brown i Ed.). Addicts and altercare Community integration of the former drug user (pp. 81-102). Beverly Hills.
C.A: Sage.
Stuart. R. B. (1974) Teaching facts about drugs: Pushing or preventing'.' Journal of Educational Psvchology. 37. 198-201.
Surgeon general. (1988). The health consequences of smoking. Mconne
addiction. A report oflhe surgeon general Rockville. MD: U.S. Depanment of Health and Human Services.
Suwaki. H.. S Ohara. H. (1985). Alcohol induced facial Hushing and
c
drinking behavior in Japanese men. Journal of Studies on Alcohoi.
46 196-198.
Swift. C. F. (1988). Stopping the violence: Prevention strategies for
families. In L. A. Bond & B. M. W-'agner (Eds.). Families in transition,
primary prevention programs that work: Primary prevention ot ps\ •
chopuihology[Vo\. 1 I. pp. 252-285). Newburv Park. C.A: Sage.
Tabakoff. B.. S Hoffman. P. L. (1988). Geneticsand biological markers
c
of risk for alcoholism. Public Healih Reports. 103. 690-698
Tec. N. (1974). Parent-child drug abuse: Generational continuity or
adolescent deviancy'' Adolescence. 9. 350-364
Tobler. N. S. (1986). Meta-analysis of 143 adolescent drug prevention
programs: Quantitative outcome resulls of program panicipants
compared to a control or comparison group. Journal ot Drug Issues.
16. 537-567.
Tonry. M.. Ohlin. L. E.. S Farrington. D. P. (1991). Human development
c
and criminal behavior. New York: Springer-Verlag.
Tremblay, R. (1988. September). Peers and the onset of delinquency
Paper prepared for ihe Onset Working Group Program on Human
Development and Criminal Behavior. Casune. ME.
Tremblay. R. E.. McCord. J.. Boileau. H.. LeBlanc. M.. Gagnon. C.
Charlebois. P.. S Lanvee. S. (1990. November). The Montreal prevenc
tion experimem School adjustment and self-reported delinquenc:
after three years oi follow-up Paper presented at the 42nd Annual
Meeting of the American Society of Criminology. Baltimore.
Vaillam. G. (1983). Thenatural history of alcoholism. Cambridge. M.A:
Harvard University Press.
Vaniainen. E.. Pallonen. U . McAlister. A.. Koskela. K.. S Puska. P.
c
(1983). ElTectof two years of educational intervention on adolescenl
smoking (The Nonh Karelia Youth Project). Bulletin of inc Uorid
Health Organization. 61. 529-532.
Vaniainen. E.. Pallonen. U . McAlister. A.. Koskela. K . S Puska. P.
c
(1986). Four-year follow-up results of the smoking prevention program in the Nonh Karelia Youlh Project. Preventive Medicine. 15.
692-698.
Vaniainen. E. Pallonen. LL McAlister. A.. & Puska. P. (1990) Eightyear follow-up results of an adolescent smoking prevention program: The Nonh Karelia Youth Project. Amencan Journal of Public
Health. SO. 78-79.
von Knorring. L.. Oreland. L.. S von Knorring. A. L. (1987). Personalc
ity traits and platelet MAO activity in alcohol and drug abusing teenage boys. Acta Psychiatnca Scandinavica. 75. 307-314.
Wallerslein. J. S.. & Kelly. J. B.(1980). Effeas of divorce on the visiting
father-child relationship. American Journal ot' Psvchiatry 1 j " . i : 341539.
Walters. H. 1.. S Gilmore. S. K. (1973). Placebo versus social learning
c
effecis on parental iraining procedures designed io alter the behav ior of aggressive boy s. Behavior Therapy 4 5 i 1-57"
7
1
�0042
D H A W K I N S . R. C A T A L A N O . A N D J. M I L L E R
Wans. R K . i Rabow. J. 11983). Alcohol availabilily and alcohol-relaied problems in 213 Californian cities. Alcoholism - Clinical and
"'• i. expenmemal research.
47-58.
Weaver. S. C. £ Tennant. F S. (1973). Effectiveness of drug education
programs for secondary school students. American Journal oi Psycho
a m HO. 812-814.
Wechsler. H.. & McFadden. T. (1979). Patterns of alcohol consumption
among the young: High school, college and general population studies. In H. J. Blane & M . E. Chafez (EdsJ. Youlh. alcohol, and sociai
poiicy {pp. 39-58). New York: Plenum Press.
Weissberg. R. P.. & Allen. J. P. (1985). Promoting children's social skills
and adaptive interpersonal behavior. In B. A. Edelstem & L. Michelson (Eds.). Handbook of prevention (pp. 153-175). New York: Plenum Press.
Weissberg. R. P.. & Caplan. M. (1989. April). A follow-up studv of a
school-based competitive promotion program for young adolescenis
Paper presented at the Biennial Meeting of the Society for Research
in Child Development. Kansas City. MO.
Weissberg. R. P.. Gesten. E. L.. Carnrike. C. L . Toro. P. A.. Rapkin.
B. D. Davidson. E.. & Cowen. E. L. (1981). Social problem solving
skills training: A competence-building intervention with second- to
fonh-grade children. American Journal ofCommunity Psychology. 9.
411-423.
Werner. E. E. (1989). High-risk children in yc-.^g adulthood: A longitudinal study from birth to 32 years. American Journal o f Orthopsychiatry. 59. 72-81.
Werner. E. E.. & Smith. R. S. (1982). Vulnerable but invincible New
York: McGraw-Hill
Wiison. J. Q.. & Herrnsiein. R J II9S5) Cnme and human naiure New
York: Simon & Schuster.
Zabin. L. S.. Hardy J. B.. Smith. E. A.. & Hirsch. M. B. il98c.i. Suhstance use and its relation to sexual activity among mner-city adolescents. Journal ol Adolescent Health Care. ' 520-331.
Ziegler. S (1981). The effectiveness of cooperative learning teams for
increasing cross-ethnic friendship, additional ev idence. Human Organizanon. 40 264-268.
Ziegler-Driscoll. G. (1979). The similarities in families of drug dependents and alcoholics. In E. KLaufman & P. Kaufman. Family therapy
ol drug ana alcohol abuse (pp. 19-39). New York. Gardner Press.
Zucker. D M. (in press). An analysis of variance pitfall. The lixec
effects analysis in a nested design. American Educalional Research
Journal.
Zucker. R. A. (1979). Developmental aspects of d r i n k i n g through the
young adult years. In H. T. Blane Sc M. E. Chafetz (Eds.). Youth,
alcohol and social poticyipp. 91-146). New York: Plenum Press.
Zucker. R. A.. & Harford. T. C. (1983). National study of thedemography ofadolescent d r i n k i n g practices in 1980. Journal of Studies on
Alcohol. 44 974-985.
Zuckerman. M. (1987). Biological conneaion between sensation seeking and drug abuse. In J. Engel & L. Oreland <Eds.). Brain reward
...stems and abuse (pp. 165-176). New York: Raven E I L .
Received N o v e m b e r 29. 1989
R e v i s i o n received A u g u s t 7.199 I
A c c e p t e d August 11. 1991
•
�The Search f o r E f f e c t i v e Prevention Programs: What We Learned
Along t h e Way. Richard H. Price, e t a l .
Prevention models t h a t have been i d e n t i f i e d by the American
Psychological Association have shown t h a t prevention programs can
be e f f e c t i v e . They have also demonstrated t h a t successful
programs have several common features:
They are t a r g e t e d t o the problems of a s p e c i f i c group.
They are aimed a t long-term changes i n l i f e s t y l e .
They seek t o strengthen the support of f a m i l y and
communities i n a d d i t i o n t o the targeted i n d i v i d u a l s .
They also found t h a t most prevention programs t a r g e t c h i l d r e n and
t h a t most programs don't attempt t o evaluate t h e i r e f f e c t i v e n e s s .
�48
Amer
PIONEER PROGRAMS
Governments. (1988). Focus on lhe first sixty
(Eds ). Evaluating family programs (pp. 3-36).
months: The next sups: A guide lo implememation.
Hawthorne. NY: Aldine.
Washington, D.C: Author.
Weiss, H.. & Halpem. R. (1988). Whal is known
Osofsky, J.D. (1986). Percpectives on infant menial
about the effectiveness offamity orienled early childhealih. In M. Kessler A S. Goldslon (Eds.), A dehood intervention programs? Paper prepared for the
cade of progress in primary prevention (pp.
Cenier for the Sludy of Social Policy, Baltimore.
181-201). Hanover. NH: University Press of New
Weiss, H . A Hite. S. (1986). Evaluation: Who's doEngland.
ing it and how? A repon from a national program
Pfannensliel, J.C, * Sellzer. D.A. (1985). Evaluasurvey conducted by the Harvard Family Research
tion repon: New Parents As Teachers Projecl. JefProjecl. FRC Repon. (3), pp. 4-7.
ferson City: Missouri Departmenl of Elementary and
Weiss, H., Resnick, G.. & Hausman, B. (1987). 77ie
Secondary Education.
place of family support and educaiion programs on
Tableman. B. (1986). Statewide prevention programs:
the social policy agenda. Unpublished manuscript
"Die polilics of the possible. In M. Kessler & S.
prepared for the Rockefeller Foundation. Harvard
Goldslon (Eds.), A decade of progress in primary
Family Research Project.
prevention (pp. 33J-34J). Hanover, NH. UniverWeissbourd. B. (1987). Design, stafllng, and funding
sity Presa of New England.
of family support programs. In S. Kagan, D. Powell,
Weiss. C. (1983). Ideology, interests, and informaB. Weissbourd, & E. Zigler (Eds ), ^ w r i r a ' j famtion: The basis of policy positions. In D. Callahan
ily suppon programs (pp. 245-268). New Haven:
* B. Jennings (Eds ), Ethics, lhe social sciences,
Yale Univenity Press.
and policy analysis (pp. 213-245). New Yoric: PleWeissbourd, B. & Kagan. S.L. (1989). Family supnum.
port progranu: Catalysu for change. American JourWeiss, H. (1986. February 25). Testimony lo Ihe U.S.
nal of Orthopsychiatry, 59, 20-31.
House of Representatives Select Committee on ChilZigler. E.. t Weiss. H. (1985). Family support sysdren. Youth, and Families, hearings on Family
tems: An ecological approach to child development.
Strengths.
In R. Rapoport (Ed ), Children, youth, and /amiWeiss. H B. (1988). Family support and educaiion
lies: The action-research relalionship {pp. 166-205).
prognmi: Working through ecological theories of
Cambridge: Cambridge University Press.
human development. In 11 Weiss & F. Jacobs
J. Orthopsychial
59(1). January 1989
THE SEARCH FOR
EFFECTIVE PREVENTION PROGRAMS:
What We Learned Along the Way
Richard H. Price, Emory L. Cowen, Raymond P. Lorion, Julia Ramos-McKay
Efforts by an American Psychological Association task force to identify model
prevention programs for high-risk groups throughout the life span are summarized. Criteria for selection and program content are described, and implications
for the construction, implementation, and evaluation of effective programs are
discussed.
I
n 1984, the National Institute of Mental
Health (Eaton & Kessler, 1985) reported that 43 million people in the United
Sutes, 19% of the nation's population, were
suffering from some form of psychological
disorder or distress. There will never be
enough mental health professionals to provide help for such widespread distress. For
that reason, prevention advocates have argued that we must mount large-scale preventive efforts, as public health pioneers
did in dealing wilh diseases such as polio,
small pox, and measles.
Although a growing number of researchers agree that a wide range of psychological
and health problems are preventable, the
logic of prevention must be turned into concrete reality. One approach is to identify
model programs that work. Identifying effective programs has two aspects. First, research evidence for claims of program effectiveness must be examined carefully.
Many preventive measures (hat look as if
they should work actually may^ not. Second, once identified, model programs can
serve as examples to be repealed in other
settings such as schools, communities, or
work organizations. It was (his idea that led
an American Psychological Association task
force to launch a major search for effec(ive
model prevention programs (Price. Cowen.
Lorion. A Ramos-McKay. 1988).
Unfortunately, not everything called a
prevention program is, in fact, a real prevention program. Practitioners aboul to s(art
a prevention program must know how to
evaluate ils promise of real success. To do
so, (hey must consider not only ils content,
the appropriateness and ease of ils application, and how well specified ils (argel groups
are, but also the quality of theresearchupon
which a program stands. For those reasons,
lhe Task Force concluded lhal il could besl
contribute lo (he field by conducting a search
for model preventive programs with well
Based on a reporl of the American Psychological Associaiion Task Force on Promotion, Prevention and
Iniervention Alternatives. The Task Force consisted of the authors, who are at: University of Michigan (Price):
University of Rochester (Cowen): University of Maryland ILorion): and Hartford (Conn.) Police Department
(Ramos-McKay): as well as Beverly Hitchins of APA. Research was supported by APA and by a Pew Memorial
Trusl gram lo APA and the Nalional Council of Community Mental Health Centers.
© 1989 American Orthopsychialric Association, Inc
49
�50
EFFECTIVE PREVENTION PROGRAMS
documented fealures and solid outcome Findings.
Before beginning the search for model
programs, the Task Force identified several
benchmarks to be used in its mission. First,
we hoped to find programs relevant to all
stages of the life span. Prevention is clearly
important not only for children bul for adults
and the elderly as well. In addition, we
hoped to Find programs that were being delivered in many different community settings, including hospitals, schools, service
agencies, and the workplace. Finally, we
hoped that the exemplary prevention programs identified would be aimed at a wide
range of different health, menial healih, and
criminal justice outcomes.
The Task Force contacted experts
throughout the country whom we believed
were knowledgeable about prevention programs. We received many replies describing a variety of prevention efforts and then
sel aboul examining Ihese programs, searching for evidence of effectiveness. In Ihe
end, we ideniified a number of programs
that we believe could serve as models. Our
search was certainly not exhaustive, there
are surely many effective programs lhal we
were unable to identify. Nevertheless, what
follows is a brief sketch of the search itself,
the criteria we developed, the procedures
we used to evaluate candidates for model
programs, and some information about the
kinds of program submissions we received.
havioral condition to be prevented; b) a statement of a rationale for Ihe intervention including its liming, duration, and sequencing;
c) a description of Ihe actual iniervention;
d) a description of Ihe skills necessary lo
conduct the intervention; e) a specification
of Ihe program steps taken to recmil intervention participants; f ) a specification of
observable and measurable program objectives; g) a description of Ihe program evaluation, monitoring, and follow-up data; h)
a description of how the program relates lo
community groups, organizations, and agencies; i) consideration of ethical issues;/) Ihe
transferability of the intervention to other
sellings; and k) roles of professional and
nonprofessional caregiver resources.
The Task Force invited submissions of
model programs from approximately 900
prevention researchers and practitioners,
through direct written requests as well as
announcements in professional publications and newsletters. Each of 300 submissions was reviewed by Iwo Task Force members and rated along several five-point
continua for: overall promise for more detailed scrutiny, importance of the problem,
and probable quality of the evalualion. Inlerrater agreement was examined for each
program description. In cases of marked
disagreement, a third or fourth rater was
used.
Fifty-two programs emerged from this
process as especially promising. The developers of ihese 52 programs were Ihen asked
PROGRAM CRITERIA
to submit detailed program materials, manOur screening efforts involved the con- uals, and research reports and other indicasideration of program descriptions using tors of program effectiveness. These malemultiple criteria. These criteriareflectedfour rials were examined by Task Force members
aspects of the intervenlions: /) the problem for Iheir informative and scientific quality,
addressed; 2) the program's targeted goals; resulting ultimately in the selection of 14
2) procedures followed in reaching ihose model programs. Although Ihese 14 varied
goals; and 4) evidence documenting ihe at- somewhat in their reported evaluation findtainment of program goals. Program de- ings, each had specifically documented the
scriptions were examined by members of achievement of intended outcomes. Morethe Task Force who sought specific infor- over, each was judged to be an "exportmation about the following characteristics able" program, (hat is one that had a deof Ihe intervention: a) a clear description of fined, replicable set of procedures (ha( could
the group at risk and the emolional or be- be adopted by practitioners. Finally, in se-
PRICE ET AL
51
lecling the model programs, the Task Force case, is shaped by at least a preliminary
anempted to provide a( leas( reasonable cov- understanding of the risks and problems enerage of target groups across the life span, countered by the target group. Second, all
service se((ing, and possible prevemion ou(- programs are designed to alter the life trajectory of Iheir participants. They are aimed
comes.
TABL£ 1 provides a brief description of at long-term change, setting individuals on
the final 14 prevention programs identified a new developmental course, opening opby the Task Force. The first column of the portunities, changing life circumstances, or
table identifies the primary program devel-providing support. Many of these programs
opers and makes references to reports of provide people with new skills to cope more
theirresearch;the balance of the table sum- effectively or provide social support in the
marizes target groups, major objectives, ma-face of stressful life conditions. Another
jor intervention methods, and program out- common denominator of successful procomes. More detailed descriptions of these grams is that they strengthen the natural
programs are set forth in Price et al. (1988). support from family, community, or school
settings. Finally, although it may seem obvious, successful programs have managed
IMPLICATIONS FOR
to collectrigorousresearch evidence to docPOLICY, PRACTICE, AND RESEARCH
ument Iheir success. Indeed, in a variety of
In the process of sifting evidence, examdifferent ways, each of the model programs
ining program protocols, and conferring wilh
has provided evaluation evidence of its
model program developers, the Task Force
effectiveness.
identified a number of issues and future
Programs for adults and the elderly are
needs thatrequirethe attention of researchers, practitioners, and policy makers. In underrepresented. Examination of the iniwhat follows we discuss some of them, both tial pool of submitted programs provided
to qualify our findings and to point to issues useful insights into the profile of preventive
intervenlions available at the time of the
for further research and policy action.
Prevention programs can be effective. initial Task Force survey. In general, Ihese
The programs described in TABLE 1 involve preventive interventions were targeted prea wide range of target groups across the life dominantly to children and adolescents;
span, have been implemented in a variety fewer than 20% focused on adults or the
of settings, and address a number of differ- elderly. Demographically, therecipientsof
ent preventive goals. Even though all 14 preventive interventions tended disproporare identified as model programs, the per- tionately to be poor, minority group memsuasiveness of the evidence they present for bers, and otherwise disenfranchised. Proeffectiveness varies. Nevertheless, the ex- grams were, for the mosl part, provided
istence of Ihese programs underscores a fun- through schools and social service agendamental point. Preventive strategies can cies. Most prevention service providers were
be effective and represent an important al- nonprofessional community caregivers, edternative to the frequent assumption that ucators, or members of participants' famitreatment orremediationis the only way to lies. Although we have no way of knowing
deal wilh mental health, health, develop- whether this sample of programs reflects
the actual distribution of preventive efforts
mental, or psychosocial problems.
Effective programs share a number of across the life span, we suspeel lhat prefeatures. Although Ihe programs presenied ventive programs for adulls and Ihe elderly
in TABLE 1 are quite diverse, they share are indeed underrepresented. Preventive efseveral common fealures lhal can guide fu- forts are obviously not only for the young,
ture prevention efforts. First, these pro- but are needed across the entire life span.
Rigorous evaluations of preventive programs are targeted. Their focus, in each
�52
EFFECTIVE PREVENTION PROGRAMS
PRICE ET AL
53
Table 1
Table 1
MODEL PREVENTION PROGRAMS
PRIMARY
AUTHORS
TARGET
GROUP
Bernard L.
Bloom, William
F. Hodges, U.
ol Cotorado.
Boulder, ffitoom
els'., 1985;
Newly uparated persons,
Gilbert Bolvin.
American
Health Founda
lion, NY.
IBoMn ai al.,
1984)
Junior high
school students,
William S
Davidson.
Michigan Slate
U. (Davidson el
al., 1985)
Youth
charged with
person, property, or status
offenses and
referred by
court referee.
Robert Felner.
U. ol Illinois.
(Felner al al.,
1982)
Young,
low-income
adolescents
enlering high
school
Dale L. Johnson. Depl. of
Psychology, u.
of Houston
(Johnson t
Walker, in
press)
Low-income
MenlcanAmerican
families with
one-yearold child,
a
Continued
OBJECTIVES
METHODOLOGIES
To provide social support and fadtale oonpetence building in socializalion. child rearing
end single pareming,
career planning and
employmenl. legal and
finandal Issues, housing and homemaVing.
6-montti program provided by a paraprofessional and subject matter experts in the form
ol Individual and group
oonsultfitlon. upon demand, on topics Identified in the progrem ob/ectives
Provide students with
skills lo resist pressures
to smoke, drink and use
druos. he*) develop selfesleem. help to cope
with sodal anxiety, and
increase knowledge of
Immediate negative
consequences ot substance use.
To provide an Intervention for delinquent
youlhs outside the criminal justice system
which will reduce the
likelihood of recidivism.
Reducing predictable
negative effects ol the
crisis of Iransitlon to
high school.
To enhance school performance and lo reduce
Ihe incidence of behavior problems in schoolage children.
Life skills training in a
school-based 12-unlt
curriculum delivered by
classroom teachers or
older peer leaders.
Booster sessions are
added in subsequent
years.
Trained, selected, college student volunteers
work one-on one with
yoult) for 18 weeks, 610
8 hours per week. Specific Intervention condiNons included behavioral contracting, relationship building, youth
edvocacy within the
family.
Increasing peer and
teacher support, minimizing environmental
flux and complexity.
Mothers are visited 25
Kmes/year by paraprofessionals In yaar 1 and
given Information on
betoycare.creetingastmulaling home environment, emotional development, and coping with
stress. Families attend
many weekend sessions and mothers parllcipate In English
classes. In year 2. chHdren partidpate In nursery school while mothers partidpate In child
management classes at
Center.
OUTCOMES
Intervention group members were significantly
higher in adjustment, had
fewer separation-related
problems, and reported
signiricanily greater
separation-related benelils lhan controls. Positive
program effects still evident after four years.
Significant reduction In
new smoking In progrem
students based both on
sell-report and saliva
tests. Additional etiects
were observed on smoking, psychosocial and advertising knowledge, and
on sodal anxiety and suggeslabllllty.
PRIMARY
AUTHORS
OBJECTIVES
METHODOLOGIES
OUTCOMES
Nathan Maccoby, Dept. ot
Communication. Stanford
U. Heart Disease Prevention Program.
fMeyer ef af.,
1990)
Entire communities end.
in particular,
residenls
who ara
overweight,
smoke, practice poor nutrition, and do
nol exercise.
To slimulale and mainlain changes In lifestyle lhal will result In a
community-wide reduction In risk lor cardiovascular diseases
A community education program aimed at
smoking, nutrition, exercise, hypertension,
end obesity. Mass media, community orgsniletion, and social marketing ol health promotion programs.
Increase in knowledge
and modilicallon ol behavioral and physiological indicators ol risk, particularly when mass
media campaigns were
supplemented wilh face
lo-face Instruction.
David Olds, U.
of Rochester
Medical
School (Olds
etal.. 1986)
Sodally disedvanlaged
primaparas
and their children (women
who are either teenagers, unmarried, or poor
bearfng their
tirst child).
Improve prenatal
health habits and behavior. Inlormal sodal
support, use of community services; reduce
low birthweight. Improve Infant heafth and
development. Improve
maternal school and
occupational achievement, reduce repeat
pregnancy and wellare
dependence, reduce
child abuse and neglect.
Pre- end post-natal
nurse home visitation,
transportation for
health care, sensory
end developmentel
screening
Nurse-visited women during pregnancy made better use ol community
services, experienced
grealer social support.
Improved their diets, and
reduced number ol dgareltes smoked: Improvements In birthweight and
length of gestation for
young adolescenis and
smokers* Nurse-vislled
mothers al highest sodal
risk (poor, unmarried
teenagers) had fewer verified cases ol ebuse and
neglect during lirst 2
years postpartum.
Donald E. Pierson. Deborah
K. Walker.
Terrene* Tinvan. Brookline
Early Education
Project.
(Pierson et at,
1983)
Families of
preschool
children.
To reduce learning dilAculttes in preschool
children and to develop effective parentschool communlcallon
links
Parent education and
support, dlagnosllc
monitoring, periodic
health and develop
menial exema lor children Irom 6 months
and, beginning at age
2 years, weekly playgroups followed al
ages 3 and 4 by a dally
morning pre-klndergar.
ten program.
Structured observallon ol
classroom behavior
showed program children lo have less learning difficulty and lewer
reading problems In second grade then comparison children. Parents ol
program children
more relevant Interests
with their child s second
grade teacher as well.
Cost effectiveness analyses showed lhal more
intensive versions ot Ihe
program were more effective for children v
parenis are nol highly
educated.
Craig T. Ramey, Frank
Porter Graham.
Child Development Center,
U. ol North
Carolina.
fRamey 4
Campbell,
1984)
Disadvantaged rural
black preschool children al risk
for mlkj mental retardation.
To provide a learning
environment lo develop children's communlcallon, language,
motor, and social skills.
Child-centered prevention program delivered
in a day care setting
Irom Infancy to age 5.
emphasizes language,
cognitive perceptual
motor and sodal developmem
Beginning al age 18 mos.
end Intervals Ihereatter
lo 54 months, program
children scored signiticamly higher than controls on a range ol mental ability lesls. wilh
experimentals exceeding nalional averages
while controls declined.
(Continued)
Slgnlflcanlty lower levels
ol recidivism as measured
by court petitions 2.5
years after intervention.
Number ol police contacts were also lower tor
Interventions conducted
outside the court system.
After 1 year the experimental group had higher
grades and better school
attendance. Experimental group has less negalive self-concepts and
perceptions of school environment lhan controls.
Al 4-year follow-up. experimental group has better
grades, lewer absences
and lower dropout rates.
TARGET
GROUP
At five lo eight years postprogram.partldpaling children show tewer aggressive, acting-out behavtora
end are less hostile and
more considerate than
controls.
�54
EFFECTIVE PREVENTION PROGRAMS
PRICE ET AL
55
Table 1
Table 1
Continued
Continued
PRIMARY
AUTHORS
Maiy Jans Rotheram, Columbia U. (Romtmm el el.,
1981)
Phyllis Stverman. Inst, of
Health Professions. Massachusetts Qeneral Hospital.
Boston. fS*vemtn, 1988;
Vachon el el.,
1980)
Qeorge Sp*
vack, Myma
Shure. (Slture
t Splvtck.
198!)
Clporah S. Tadmor, Rambam
Medical Center,
HaHa. Israel.
(Tadmore al al.,
1988)
TARGET
GROUP
Fourth and
fifth grade
children.
Recerttty widowed women.
4-6 year old
urban preschool and
kindergarten/
first grade
children.
Pregnant
women undergoing cesarean birth.
METHODOLOGIES
OBJECTIVES
To Improve social
skills, assertiveness,
and Interpersonal oompetance In 4th and Sth
grade children.
To provide sodal support, mutual help to
newly widowed women
to reduce psychological distress.
To teach children Interpersonal problem-sotvIng skills In order to promote positive social
behavior and decrease/
prevenl high-risk negative behavior.
Mobilize natural and organized supports, provide information, share
decision-making processes, and develop
task-oriented activity lo
enhance emotional,
cognitive, and behavioral control and prevent emotional dysfunction.
OUTCOMES
Group-based social
skills and assertiveness
training 2 hours per
week for 12 weeks locused on training nonvertul behavioral skills.
Interpersonal problemsoMng. and emotional
self-oontrol In role play
oontert.
Widows contacted
newly
bereaved
women, provided onetoone support, located
community resources,
mede supportive telephone calls, and led
small group meetings.
Format 12-week training programs and associated procedures for
use throughout the day
(one for preschool and
one lor kindergarten/
first graders) both enhanced Ihe ability ol
children lo generate ellernatlve solutions lo
peer and adull problems and anticipate potential consequences of
Interpersonal acts.
Anticipatory guidance
session familiarizes
couple with medical setting «nd personnel. Provide detailed Information on birth process,
anesthesia, anticipated
reactions, pain, duration. Cesarian birth supPort group provides support. guidance and help
during hospital stay.
Discharge planning occurs before release.
Teacher rated adjustment (or behavior),
achievement and peer
popularity were all superior In asserttveness'skills
group. Grsde-pdnt averages were higher lor experimental group 1 year
post-Intervention.
Experimental group members evidenced Improved
mood, lower anxiety,
made more friends, and
began more aclivilies.
Overall, experlmenlal
group women progressed more rapidly In the
course ol adaptation, reduction ol Internal distress, and resodallzatlon.
Experimentals acquired
higher levels of problemsolving skills than conlrols, enhanced positive
sodal behavior, and decreased Impulsive and inhibited behavior. Effects
endured over time, the Inddence of new high-risk
cases was diminished,
and linkages between
cognitive and adjustlve
gains were shown
Experimental mothers released Irom hospital
sooner than controls. Initiated Independent care ol
the baby sooner and continued nursing longer. Alter day 1, experlmenlal
mothers requested less
medication than controls
and experlmenlal lathers
showed closer attachmenl to babies than oontrol fathers. Experimental
molhers had speedier
psychological recovery.
PRIMARY
AUTHORS
David Weikart.
Lawrence
Schwelnhart.
High Scope
Educalional
Research
Foundation,
Ypsllantl, Ml.
(BerruefaClemenl el el..
1984)
TARGET
GROUP
Black children, ages
^ - 4 , from
families ol
low sodoeconomic stalus
who were al
risk ol laillng
in school.
METHOD
OLOGIES
OBJECTIVES
To Implement a high
quality preschool curriculum, involve parents,
wilh coordinated stall.
adnUnlstration. and parerrt Invotvement lor preschool children.
grams are scarce. If so few programs survived the initial screening, a major reason
is that only one-third had attempted to evaluate their efforts systematically. Moreover,
a significant number of these lacked program manuals or other means of documenting their efforts. Programs that had only
vaguely defined procedures had little chance
to be replicated successfully in olher settings. For these reasons, it is misleading to
use total number of submissions as a framework for assessing the efficacy of existing
prevention efforts. A more accurate indicator of the state of the field might be the 52
programs that the Task Force considered in
detail. Of these, however, nearly 25%
lacked follow-up information considered essential by Task Force members to document the enduring achievement of preventive goals. Many others had evaluation
designs or results that cast doubt on program oulcome statements or lacked the documentation needed to permit program replication. Although we do not represent the
final group of 14 programs as flawless in
these respects, they nevertheless provide a
reasonable estimate of (he profile of currently "mature" preventive intervention
strategies.
High quality, early
ctiikffiood education for
2 years. 2'A hours per
school dsy lor T / i
months per year. Children participated In
cognltively-orf ented
curriculum. Weekly
home visits were conduded.
OUTCOMES
Significant cognitive
gains. Improved scholastic placement snd
achievement during
school years, decresses
In crime and delinquency, use of welfare
assislance at age 19.
Experimentals also had
better high school graduation rates, post-secondary enrollments, and
higher employmenl than
controls. Analyses show
benefits exceed costs
seven-fold. Findings persist through age 19.
Many of the remaining programs submilted represented interesting, potentially
promising ideas which had not yet evolved
into fully developed preventive interventions. For some, too little lime had passed
to assess the preventive strategy in terms
of the desired outcome. Others simply
lacked systematic program evaluations. Task
Force members wpre nevertheless persuaded, by the range and substance of the
submittals, that evidence for effective prevention programs will increase significantly over the coming years as necessary
outcome data are obtained using longitudinal evaluations. We therefore view the results of the original survey optimistically
and look forward to continued healthy
growth in the development of effective preventive interventions.
Benefit-cost assessments are possible but
seldom attempted. Benefit-cost analysis is
a critical consideration for policy makers
and the general public in evaluating preventive program models of the kind reported
here. Both technical and value issues must
be addressed before benefit-cost evaluation
methodology can be applied thoughtfully to
preventive programs.
ft is worth noting that some preventive
�56
programs may never be tmly cost beneficial in the narrow sense. Even so, as a society we may choose to develop and implement such programs because they reduce
human suffering, increase human dignity,
or otherwise reflect deep human values. Preventive programs obviously should not be
rejected simply because the " b o t t o m l i n e "
does not indicate lhat the program saves
money in the short term. That much said, it
is inevitable that, as sophistication in conducting research and evaluation on preventive programs grows, benefit-cost questions w i l l be raised and researchers w i l l
attempt to do benefit-cost analyses on preventive programs.
A n example of such a pioneering effort
is the Perry Preschool Program described
by Bemieta-Clement, Schweinhart, Barnett, Epstein, and Weikart (1984). They documented the costs of early education and
the long-term benefits resulting from the
positive program outcomes. A s the authors
observed, "changes in economic success,
self-sufficiency, and social responsibility can
be predicted quantitatively from observed
effects at age 19" ( p . 89). In commenting
on these findings, Gramlich (1984) made
several important points that policy makers
and prevention researchers should consider. First, benefits from prevention programs may increase over time. Short-term
evaluations of benefits may show that they
are either small or nonexistent, but benefits
may accrue over time as children are, for
example, engaged in less crime, depend on
welfare less, or begin to reap the benefits of
higher levels of educalional achievemeni.
Anolher critical issue is that sensitive
benefit-cost analyses should identify gainers and losers in society. For example,
Gramlich's (1984) analysis identified net
social benefits received by participants in
the program, by taxpayers, and by potential
victims o f crime. As Gramlich observed,
elected officials are concerned about who
gains and who loses, as well as how big the
overall gain or loss actually is. For example, benefit-cost analyses of the Perry Pre-
EFFECTIVE PREVENTION PROGRAMS
school Program suggest that the total net
benefits lo each preschool participant were
approximately $5,000. On the other hand,
by the time the program recipients are 19
years o l d , total net benefit to taxpayers and
potential crime victims is estimated at
around $23,000 for each year of preschool
experience.
Although preventive programs are lime
consuming and costly to develop and evaluate, their cost is trivial compared to the
social costs of drug abuse, school dropout,
depression, or delinquency. As health care
costs and state expenditures continue to soar,
we are becoming more aware that, for every problem of this k i n d , someone is paying the bill in tax dollars, insurance premiums, or productivity losses to employers.
Systematic knowledge f o r implementing
and sustaining preventive efforts is lacking.
Even highly effective model prevention programs are not always easily or automatically transformed into local program replications that w i l l produce the intended
preventive effects. Successful implementation of a model prevention program is a
form of organizational reinvention (Rice &
Rogers. 1980). The term "organizational
reinvention" well describes this process because (he implementation of a model prevention program is inherendy organizational in nature. This process involves Ihe
orchestration of both internal and external
organizational resources; scanning the organizational and community environment;
focusing program goals and objectives; and,
finally, implementing the program and monitoring it for fidelity lo the original model.
The success o f (hese activities depends
heavily on how receptive the host organization, whether a school, hospital, or social
service agency, is to the new program and
the degree to which the implementer is able
to work collaboratively with local institutional and community members.
But, even successful collaboration with
local community or service organization
members raises its own dilemmas. A d i lemma facing all practitioners in implement-
57
PRICE ET AL
ing a model prcvenlion program concerns
(he potential conflict between fidelity and
adaptation. Stolz (1984) noted that a major
controversy in the field of knowledge diffusion and utilization has to do with whether
a model program should be used as close to
the original form as possible to insure fidelity, or whether we should encourage organizations to modify and adapt the innovation and therefore enhance the likelihood
of local acceptance at the risk o f altering
critical features of the program. Future research must distinguish between " c o r e " program features iha( are its effective i n gredients and should not be altered, and
"adaptive" features that must be modified to
suil local circumstances. Although the developers of model programs discuss this and
related implementation issues in the context
of their own programs (Price et a l . , 1988),
these critical organizational issues remain at
the " c r a f t " level of understanding. T o date,
few generalizable principles are available to
guide practitioners wishing to replicate model
programs in other settings.
CONCLUSION
The A P A Task Force on Promotion, Prevention, and Intervention Alternatives set
out to identify effective prevention programs across (he life span (hat could serve
as models for practitioners to replicate in
local settings. I n so doing, the Task Force
Ieamed many other things about the prevailing state of the art.
Perhaps most importantly, we conclude
that prevention efforts can be effective and
that, while still scarce, new and promising
programs continue to emerge. Additional
research support can accelerate that process. Successful programs have a number
of common features including careful targeting of the population, the capacity to
alter life trajectory, the provision of social
support and the teaching of social skills, the
strengthening of existing family and community supports, and rigorous evaluations
of effectiveness. We also conclude that programs for adults and the elderly are under-
represented, that rigorous evaluations are
extremely scarce, that estimates of benefits
and costs are rare, and that knowledge to
implement and sustain programs effectively has not yel been systematically developed.
The promise o f prevention is clear
enough. The need to support systematic research and development to expand current
knowledge and skill is equally clear. The
Task Force's findings should be encouraging to policy makers and should stimulate
efforts to deliver—more speedily and systematically—on the promise of prevention.
REFERENCES
Benueu-Clement, LR . Schweinhart. L.J.. Bametl.
W.S.. Epstein, A S.. A Weikart, D P. (1984).
Changed lives: The effects of the Perry Preschool
Program on youths through age 19 (Monographs on
the High/Scope Educational Research Foundation.
No. 8). Ypsilami. M I : High/Scope. ,
Bloom. B.L.. Hodges, W.F.. Kern. M B., & McFaddin, S.C. (1985). A preventive intervention program for the newly separated. American Journal of
Orthopsychiatry. 55. 9-26.
Bolvin. G.J.. Baker, E., Renick. N.L.. Filaiwjla,
A . D . . & Botvin, E . M . (1984). A cognitivebehavioral approach to substance abuse prcvenlion.
Addictive Behaviors, 9, 137-147.
Davidson. W.S., Blakely, C H . Redner. R . Mitchell. C M . , ft Emshoff, J G. (1985). Diversion of
juvenile offenders: An experimental comparison. East
Lansing, M I : Ecological Psychology Program. Michigan Sute Univetsity.
Eaton. W . W . . & Kessler, L.G. (1985). Epidemiologic field methods in psychiatry: The NIMH Epidemiologic Catchment Area Program. Orlando, FL:
Academic Press.
Felner, R.D., Ginter. M . , A Primavera, J. (1982).
Primary prevention during school transitions: Social
support and environmental structure. American Journal of Community Psychology, 10, 277-290.
Gramlich, E.M. (1984). Commentary on Changed
lives. In J R. Berniela-Clemenl. L.J. Schweinhart.
W.S. Barnett. A S. Epstein, and D P. Weikart
(Eds). Changed lives: The effecis oflhe Perry Preschool Program on youlhs through age 19 (Monographs of Ihe High/Scope Educational Research
Foundation, No. 8, pp. 200-203). Ypsilami. M l :
HigWScope.
Johnson. D.L. A Walker, T. (in press). The pnmary
prevention of behavior problems in MexicanAmerican children. American Journal of Commu
niiy Psychology.
Meyer, A.J., Nash. J.D., McAlister, A . L . , Maccoby,
N , A Farquhar, J. (1980). Skills training in a cardiovascular healih education campaign. Journal of
Consulting and Clinical Psychology. 4», 129-142.
�58
EFFECTIVE PREVENTION PROGRAMS
Olds, D.L.. Henderson. C.R., & Tateibaum. R.
(1986). Prevenling child abuse and neglect: A randomized trial of nurse home visitation. Ptdiairics
78. 65-78.
Pier»n. D.E.. Bronson, M B.. Dromey. E.. Swartz
J.P . Tivnan. T.. ft Walker. D.K. (1983). The impact of early educaiion: Measured by classroom observalions and teacherratingsof children in kindergarten. Evalualion Rtview, 7. 191-216.
Price. R.H., Cowen. E.. Lorion. R.P.. ft RamosMcKay. J. (Eds.) (1988). Fourteen ounces of prevention: A casebook for practitioners. Washington.
DC: American Psychological Associaiion.
Ramey, C.T., ft Campbell. F A. (1984). Preventive
education for high risk children: Cogniiive consequences of the Carolina Abecedarian Projecl.
American Journal of Menial Deficiency. SS.
515-523.
Rice. R.E.. A Rogers. E.M. (1980). Reinvention in
the innovation process. Knowledge: Creation. Diffusion. Uliliialion. I . 499-514.
Rotheram, M.J., Armstrong. M . . ft Booraem. C
(1982). Assertiveness training in fourth and fifth
grade children. American Journal of Psycholoev
10. 567-582.
Shure. M B.. A Spivack. G. (1982). Interpersonal
Amer. J. Orthopsychial. 59(1). January 1989
RESEARCH
problem-solving in young children. A cognitive approach to prevention. American Journal ofCommunity Psychology. 10. 341-356.
Silverman. P. (1986). Widow io widow New York:
Springer.
Stolz. S B. (1984). Dissemination of standardized human service models In S C. Paine. C T. Bellamy.
A B. Wilcox (Eds ), Human services lhat work:
From innovation to standard practice (pp. 235-245)
Baltimore: Paul H. Brooks.
Tadmor.C.S.. Brandts. J.M , A Hofman, J.E. (1988).
Preventive intervention for a Caesarean birth population. Journal of Preventive Psychiatry, 3(3)
Vachon, M.L.S.. Lyall. W.A.L.. Rogers. R.N.,
Freedman-Utofsky, K. A Freeman. S.J.J. (1980).
A controlled study of self help intervention for widows. American Journal of Psvchiatry 137
1380-1384.
SUICIDAL BEHAVIOR IN "NORMAL" ADOLESCENTS:
Risk and Protective Factors
Judith L. Rubensleln, Ph.D., Timothy Heeren, Ph.D., Donna Housman, Ed.D.,
Carol Rubin, Ph.D., Gerald Stechler, Ph.D.
Risk and protective factors were examined in suicidal and nonsuicidal public
high school students. With life stress and depression as independent risk factors,
family cohesion was found to offset the effect of stress, and friendships to have a
more indirect effect. Differential effects of ten sources of stress were analyzed
from a developmental perspective, and the probability of suicidal behavior
associated with clusters of factors was estimated for the general population.
C
ompared with their nonsuicidal counterparts, suicidal adolescents have experienced more life stress in the form of
family disruption through separations, parental divorce or death, parental emotional
disorder, or severe family discord; cumulative negative life events of Ihe soil found
in life event inventories; and, in some cases,
troubled peerrelationshipsin the form of
social isolation or the acute loss of a boyfriend or girlfriend (Hirshfeld A Blumcnthal. 1986; Petzel A Cline. 1978). While
many suicidal adolescents have experienced major life stress, most individuals
who experience similar stress are not suicidal. Clearly, the known correlates of suicidal behavior are insufficient to account
for the phenomenon.
Although Ihe literature contains several
studies of risk factors for suicidal behavior
in this age group, most of Ihese studies have
Forreprim,:
Rich^d
H Pri«.
„, P ,
i c t > o ] o f l y
. ,
,
f o f
j ^ , ,
o
f
^
^
^
been conducted on patient populations, either inpatients in a psychiatric facility, individuals who present themselves to the
emergency room of a general hospital, or
individuals in outpatient psychiatric treatment (Hirshfeld A Blumenthal, 1986). Risk
and protective factors from population surveys of high-school-age adolescents have
not been clearly identified. A further problem with existing studies is lhat most have
examined correlates of adolescenl suicidal
behavior on a univariate basis, not controlling for the interrelationships among the variables. Thus Ihe independence of each risk
or protective faclor has not been established. A clearer understanding of adolescent suicidality may be derived from
studying this phenomenon in the general
population, using multivariate techniques to
examine lhe contribution of each factor with
Ihe others controlled.
Based on a presentation to the Society for Research in Adolescence. March 1988. Research was supported in
part by the Jack Spivack Child Development Fund. Boston University Medical School.
Authors are at: Departmenl of Child Psychiatry and Child Development, Boston University Medical School
fRubenstein, Housman, Stechler); Departmenl of Epidemiology and Biostatistics, Boston University School of
Public Health (Heeren); Departmenl of Psychiatry, Harvard Medical School IRubin).
© 1989 American Orthopsychialric Associaiion, Inc.
59
�14 Ounces of Prevention: A Casebook f o r P r a c t i t i o n e r s , eds.
Richard H. P r i c e , e t a l .
This book i s a s e l e c t i o n o f case studies of 14 substance abuse
and mental h e a l t h prevention programs. I t contains programs
targeted t o i n f a n t s and young c h i l d r e n , adolescents, and a d u l t s
and the e l d e r l y . The case studies show many d i f f e r e n t successful
models, and are presented f o r p r a c t i t i o n e r s who may be i n t e r e s t e d
i n r e p l i c a t i n g them.
�I
•• vi-'i;-;
7»T
BiSSS
4
S.f^
"ft
3^
^ >
Mi-
�R i c h a r d H . P r i c e , E m o r y L . C o w e n , R a y m o n d P. L o r i o n , and J u l i a R a m o s - M c K a y
Introduction
This book is dedicated to the idea that good psychological science is a critical ingredient in programmatic efforts to prevent human suffering,
maladaptation, and illness. With this book, we
intend to provide psychologists, and indeed citizens at large, with examples of successful prevention programs that can be adopted in local communities and be made part of the practice of
psychologists nationwide.
In 1984, the National Institute of . Mental
Health (NIMH) reported that 43 million people
in the United States, 19% of the nation's population, were suffering from some form of psychological disorder or distress. There w i l l never be
enough mental health professionals to provide
help for such widespread distress. For that reason, prevention advocates have argued that we
must mount large-scale preventive efforts, as
public health pioneers did in dealing with diseases such as polio, smallpox, and measles.
Although a growing number of psychologists
agree that a wide range of psychological and
health problems are preventable, the logic of prevention must be turned into concrete reality. To
do that, we must identify model programs that
work. Identifying effective programs has two aspects. First, research evidence for claims of program effectiveness must be examined carefully.
Many preventive measures that look as i f they
should work actually may not. Second, once identified, model programs can serve as examples to
be repeated i n other settings such as schools,
communities, or work organizations. This idea
led the American Psychological Association's
(APA) Task Force on Prevention, Promotion, and
Intervention Alternatives in Psychology to
launch a major effort to search for effective model
prevention programs. This book represents the
end product of the Task Force's efforts.
1
In this introduction, we describe the activities
of the Task Force in searching for model prevention programs in the context of a larger history
of resurgent interest in prevention and the criteria the Task Force used to select model programs.
Brief previews of the programs described in the
volume are also included here.
The Task Force:
Mandate and History
In winter 1978, the APA's Policy and Planning
Board appointed an ad hoc Task Force chaired by
Edward S. Bordin to consider the emerging roles
and activities for psychologists which, in the light
of changing times and problem definitions, might
help to balance the practices of current delivery
systems. The Task Force held two meetings in
early 1979, solicited the views of experts, and
established a set of criteria to be used in evaluating alternative models. The Task Force's first
and strongest recommendation was that APA establish a Continuing Committee on Promotion,
Prevention, and Alternative Interventions i n
Psychology. The Task Force also proposed eight
specific activities for the new committee to
undertake.
Based on those recommendations, the APA's
Board of Directors appointed a Task Force on Promotion, Prevention, and Intervention Alternatives i n 1981 to report through the Committee on
Professional Practice of the Board of Professional
Affairs. The Board of Directors instructed the
'The Task Force included psychologists Emory L. Cowen,
Raymond P. Lorion, Richard H. Price, and Julia RamosMcKay, as well as Beverly Hitchins of the American Psychological Association.
INTRODUCTION 1
�new task force to focus specifically on psychological approaches to the promotion of health and
prevention of the range of physical and mental
health problems such as those associated with
coping skills, stress reduction, and acquisition of
healthy behaviors and life-styles.
The mission statement of the new task force
included specific recommendations that (a) the
delivery settings and agencies of programs be
studied, and proposals be made for developing
and evaluating new approaches i n settings with
diverse characteristics; (b) legislation and other
initiatives affecting these new approaches be
monitored and assessed at the federal, state, and
local levels; and (c) liaison be maintained with
other professional and scientific organizations interested i n prevention and promotion.
Although the Task Force had its first meeting
in November, 1981, personnel changes limited its
early operation and planning efforts. By September, 1982, however, the four-member Task Force
was reconstituted. The first agenda item was to
identify those elements i n its broad, complex
charge that could be realistically addressed
within the constraints of limited resources. An
early decision was made to focus on what was
seen as the Task Force's single most central
charge, that is to identify and support the development of newly evolving and needed approaches
in psychology including the promotion of health,
the prevention of disorder, and emergent intervention alternatives that maximize growth and
minimize dysfunction.
The key words identify and support implied two
distinct but important processes: (1) to identify
innovative, well-documented "model" programs,
exemplifying the categories of promotion, prevention, and emergent intervention alternatives; and
(2) to spearhead the preparation of a volume describing these models i n practitioner-oriented
ways. The longer-term benefit that the Task Force
hoped would accrue from these steps was to facilitate more widespread field applications of
demonstrably effective preventive programs as
alternatives to traditional, after-the-fact repair
services.
The Emergence of Prevention
as a National Priority
APA's appointment of the original Task Force on
emerging roles in 1978 took place shortly after
the submittal of the Final Report of the 1978 U.S.
President's Commission on Mental Health.
Among that report's distinguishing features were
2
FOURTEEN OUNCES OF PREVENTION
its strong emphases on the concept of prevention,
including the recommendation that N I M H be
designated as a lead agency in furthering prevention; and the need for effective new service-delivery models for the unserved and underserved.
Those conclusions put into bold relief a set of
views that had been slowly gaining momentum
over a long time period which were in response
to perceived insufficiencies in the existing mental health and other service delivery systems.
The growing clarity of these problems and the
heavy human and social costs of health and mental health problems for vulnerable populations
stimulated consideration of conceptual alternatives. I f even the best and most sophisticated attempts to cure psychological or physical illness
can account for only a minor fraction of the prevalence of illness, might i t not be easier, farther
reaching, more humane, and more effective to
seek prevention strategies from the start? That
very real question stimulated a search for viable
alternatives.
Prevention is one such option and is a sweeping
alternative, which challenges pivotal assumptions of the health and mental health systems,
not simply their practices. Rather than trying to
undo established pathology, the goal of prevention orientation is to develop initiatives to forestall problems in people at risk and to build
strengths and competencies to advance their wellbeing.
Primary prevention efforts cluster around two
broad strategies: (1) system-directed approaches
such as social policy development and modification of social environments designed to reduce
sources of stress and to enhance life opportunities; and (2) person-centered strategies such as
educational programs to impart adaptive skills
and competencies as well as preventive interventions for risk groups such as divorcing parents
and children of divorce. These two strategies have
produced programs that reflect a healthy diversity. A core quality that links prevention efforts
and sets them apart from past practice is their
intentional targeting to well people. In that regard, these prevention efforts reflect the popular
wisdom that an ounce of prevention is worth a
pound of cure.
There have been important signposts of the
growth, vitality, and promise of the prevention
movement beyond the President's Commission
Report. Although these developments have been
energetic and well motivated, they have also been
somewhat uncritical. The question, Which of
many preferred prevention contributions are enduring rather than ephemeral?, guided the Task
�Force's interpretation of its mandate and subsequent activities. From its inception, the overarching goal of the Task Force was to identify a diverse range of documented, effective prevention
programs—the best our f i e l d can offer at this
time—and to build bridges around those programs to span the sometimes wide gulf between
theory and research on the one side and application and dissemination on the other. How might
the potential benefits of these special prevention
programs best be harnessed to affect not just the
lives of the few people who were part of them,
but the large unidentified masses who stand to
benefit from exemplary model programs? To Task
Force members, an answer to this question was a
way of bringing the goal of prevention closer to
fruition.
The Need for a
Practitioner Focus
In the past, most mental health practitioners who
saw themselves as working " i n the trenches"
tended to ignore prevention because they viewed
it, i n part, as a luxury for those with surplus time
and money. Because of the pressure to meet the
great need for psychological services, many providers, who otherwise might have worked in prevention programs, attached a low priority to prevention programs and research within their own
work settings. In their experience, the number of
people needing services continued to grow while
resources for providing such services diminished.
Moreover, prevention programs being developed
by researchers did not necessarily address the
needs of service providers. More recently, however, progress in the field of prevention research
and a clamor by providers for usable prevention
programs has led to the development of programs
that can be applied in a variety of settings.
The recent expansion of work sites in which
psychologists practice, including industrial settings, employee assistance programs, law enforcement agencies, and sports programs, requires
providers of psychological services to be knowledgeable about appropriate ameliorative interventions and appropriate prevention techniques.
For example, a police psychologist may develop
programs to prevent or minimize stress and negative reactions in police officers, in addition to
treating officers who are experiencing symptoms
of psychological distress. A sports team that employs a psychologist may be interested not only
in treatment for team members but also in ways
to prevent such problems as substance abuse and
physical injuries. Prevention research and programming can no longer be seen as a luxury to
be carried out by others; employers of psychological-service providers have begun to see prevention programming as a critical skill.
The recent proliferation of prevention programs
is a response to real and urgent needs. Programs
to provide specific interventions for children who
are at risk for particular problems, programs to
teach children interpersonal skills, and programs
to teach children to say "No" to drugs and to
recognize and react to "good touches" and "bad
touches" are now available. Many of these programs appear to be designed for easy use and
require little training on the part of the practitioner. The busy practitioner, pressed for time because of service provision needs, takes these programs at face value and assumes that they have
been developed through the appropriate research.
Unfortunately, all "prevention" programs are
not real prevention programs. Practitioners must
know how to evaluate each program. To do so,
they must consider not only its appropriateness
and ease of application but also the quality of the
research upon which a program stands, how well
specified its target groups are, the actual nature
of the intervention, and whether expected outcomes have been demonstrated. For those reasons, the Task Force concluded that i t could best
contribute to the field by conducting a search for
model preventive programs with well-documented program features and solid, supporting
program-outcome findings.
The Task Force's Search
for Model Programs
Before beginning our search for model programs,
the Task Force identified several benchmarks to
be used in its search. First, we hoped to find
programs relevant to a l l stages of the life span.
Prevention is important not only for children but
for adults and the elderly as well. I n addition, we
hoped to find programs that could be delivered in
many different community settings including
hospitals, schools, service agencies, and the workplace. Finally, we hoped that the exemplary prevention programs identified would be aimed at a
wide range of different health, mental health,
and criminal justice outcomes.
The Task Force contacted over 900 experts
throughout the country who we believed were
knowledgeable about prevention programs. I n response to our inquiries, we received nearly 300
replies describing a wide range of prevention efINTRODUCTION 3
�forts. The Task Force then set about examining
these programs and searching for evidence of effectiveness. In the end, we identified 14 programs
that we believe can serve as models. There are
surely many effective programs that we were unable to identify as our search was certainly not
exhaustive. What follows is a brief sketch of the
search itself, the criteria we developed, the procedures we used to evaluate candidates for model
programs, and some information about the kinds
of program submissions we received.
Criteria for the Selection
of Model Programs
Task Force members discussed at length the diverse ways in which the concept of a model program could be realized. The prime concern was
how to determine whether an intervention
worked. At one extreme, we could rely entirely
upon the availability of rigorous empirical data
to decide this question. From this perspective, a
model would represent an intervention strategy
with irrefutable scientific evidence of its capacity
to reduce the incidence of identifiable disorders.
At the other extreme, one might rely upon community acceptance. From this perspective, a
model would represent an intervention strategy
that appeared useful and adoptable across multiple settings. Evidence that the strategy worked
would be reflected in the positive responses of
those who provided and received the intervention. Less emphasis would be placed upon the
outcomes of formal evaluations on the assumption that those charged w i t h the allocation of
scarce community resources would have examined program outcomes according to local norms.
Members of the Task Force concluded that the
selection of model interventions should be based
on quantitative outcomes. However, we felt that
consideration must also be given to the diversity
of targets, desired outcomes, and setting conditions for which preventive interventions are designed. To reflect that diversity, a three-dimensional matrix was created within which to locate
identified interventions.
The first dimension related to the life stage of
intervention recipients. Prevention strategies can
be focused at all points across the life span from
the prenatal period to old age. Thus, a life-span
continuum was created which included the following major developmental periods: prenatal
through preschool (0 to 4 years); elementary
school (5 to 12 years); adolescence (13 to 21 years);
adulthood (22 to 65 years); and retirement (65
4
FOURTEEN OUNCES OF PREVENTION
and beyond). We also considered program outcome. Initially, this dimension focused on distinguishing among mental-health, substance-abuse,
physical-health, criminal-justice and educational
outcomes. We classified each program in terms of
its specific preventive goals. We focused on the
setting i n which the program activities occurred.
We considered (among others) educational, occupational, medical, social service, and home settings as well as a combination of these settings.
Within this three-dimensional framework, we
classified interventions which successfully completed the initial program screening. Of the approximately 300 program descriptions received in
response to more than 900 practitioner surveys
distributed, 52 were considered sufficiently
promising to warrant detailed examination of
program materials and evaluation findings.
Eventually, 14 of those 52 were selected as exemplary programs. .
In screening, we considered program descriptions in terms of multiple criteria. Conceptually,
these criteria addressed four aspects of the interventions: (a) the problem addressed, (b) the program's targeted goals, (c) the procedures followed
in reaching those goals, and (d) the evidence documenting the successful attainment of program
goals. Specifically, members of the Task Force
examined program descriptions and sought information about the characteristics of the intervention. We used the following selection criteria for
model prevention programs.
• Clear description of the emotional or behavioral condition to be prevented and the group at
risk
• Description of the skills necessary to conduct
the intervention
• Statement of a rationale for the intervention
• Description of the actual intervention
• Rationale for the timing of the delivery, duration, and chronology of intervention activities
• Specification of the program steps taken to
recruit intervention recipients
• Specification of the observable and measurable program objectives
• Description of the evaluation, follow-up data,
and program monitoring
• Description of how the program relates to
other groups, organizations, and agencies in the
community
• Recognition and resolution of ethical issues
�• Transferability of the intervention to other
settings
• Roles of psychologists and other professional
and nonprofessional caregiver resources
As noted earlier, 52 of the approximately 300 program descriptions submitted provided sufficient
information on issues to warrant further examination. Each of the 300 submissions was reviewed by two Task Force members and rated
along a 5-point continuum in terms of overall
promise for more detailed scrutiny, importance of
the problem, and probable quality of the evaluation. Interrater agreements were examined for
each program description. In cases of marked disagreement, a third or fourth rater was used.
The developers of these 52 programs submitted
detailed program materials, manuals, research
reports, and other indicators of effectiveness.
Task Force members examined the informative
and scientific quality of these materials and selected 14 model programs. Although these 14 programs varied in their reported evaluation findings, each had specifically docuiriented the
achievement of intended outcomes. Moreover,
each was judged to be exportable, that is, a program that had a defined, replicable set of procedures that could be adopted by practitioners. Finally, in selecting the final set of programs, the
Task Force attempted to provide at least reasonable coverage of the three-dimensional matrix of
the life span, service settings, and prevention
outcomes.
Characteristics of Programs
Submitted
Examination of the initial pool of submitted programs provides useful insight into the profile of
preventive interventions at the time of the Task
Force's survey (1983-1984). In general, these preventive interventions were targeted predominantly to children and adolescents. Less than 20%
of the programs submitted focused on adults or
the elderly. Demographically, recipients of preventive interventions tended to be disproportionately poor, minority-group members, and otherwise disenfranchised groups. Programs were
provided for the most part within schools and
social service agencies, and most prevention service providers were nonprofessional community
caregivers, educators, and members of program
recipients' families.
Some readers may be surprised by the fact that
relatively few programs survived the i n i t i a l
screening. One reason for that is that only one
third of the programs systematically evaluated
their efforts, and a significant number of these
lacked program manuals or other means of documenting their efforts. Because these programs
had only vaguely defined procedures, they had
little chance to be replicated successfully in other
settings. With such high attrition, i t is misleading to use the total number of submissions as a
framework for assessing the efficacy of existing
prevention programs. A more accurate indicator
of the state of the field might be the 52 programs
that we examined i n detail. Of these, nearly 25%
lacked follow-up information considered essential
by Task Force members to document the enduring achievement of preventive goals. A number
of others appeared to have methodologically
flawed evaluations that cast a shadow on pro
gram outcome findings or insufficient documentation to permit program replication. Although
the final group of 14 programs are not flawless,
they may provide a reasonable estimate of the
profile of currently mature preventive intervention strategies.
Many of the programs not selected represented
interesting, potentially promising ideas which
had not yet evolved into fully developed preventive interventions. For some, too little time had
passed to assess the preventive strategy's impact
on the desired outcome. Others simply lacked
systematic program evaluations. In some of those
cases, the strategy's value as a preventive intervention was accepted without question by its adherents, that is, the need to evaluate the program
was not perceived.
The range and substance of the submittals persuaded the Task Force members to believe that
effective prevention programs will increase significantly over the coming years as necessary
outcome data are obtained using longitudinal
evaluations. We view the results of the original
survey optimistically and look forward to continued healthy growth in the development of viable
preventive interventions.
Overview of the Model Programs
This volume is divided into three sections describing programs focused on infancy and early
childhood, school-age children and adolescents,
and prevention programs for youth and adults.
The targets of these programs are diverse. They
include preschool minority children, poor teenage
INTRODUCTION 5
�mothers, school-age children in health classes,
young people who have just entered the juvenile
justice system, grade school children experiencing a transition to high/School, newly separated
couples, mothers about to undergo caesarean
birth, bereaved widows, and members of entire
communities at risk of cardiovascular disease.
Authors were asked to follow a uniform structure in their presentations and to aim their presentation at practitioners interested in replicating these programs in their home communities.
We proposed a uniform structure in which all
authors begin with a brief discussion of the program's issues, problems, and goals; describe how
V
6 FOURTEEN OUNCES OF PREVENTION
the program actually works; discuss briefly the
research evidence for program effectiveness;
highlight the program's limitations as well as
positive aspects; and offer practical suggestions
for starting a new program.
What follows, we feel, are descriptions of prevention programs that are the product of considerable ingenuity focused on the goal of prevention. Although there is a need for many more
such model programs, as well as for their adoption in the work of practicing psychologists, the
present volume is, nevertheless, an exciting
beginning.
�
Dublin Core
The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/.
Title
A name given to the resource
Health Care Reform
Identifier
An unambiguous reference to the resource within a given context
2006-0810-F
Description
An account of the resource
<p>This collection consists of records related to Hillary Rodham Clinton's Health Care Reform Files, 1993-1996. First Lady Hillary Rodham Clinton served as the Chair of the President's Task Force on National Health Care Reform. The files contain reports, memoranda, correspondence, schedules, and news clippings. These materials discuss topics such as the proposed health care plan, the need for health care reform, benefits packages, Medicare, Medicaid, events in support of the Administration's plan, and other health care reform proposals. Furthermore, this material includes draft reports from the White House Health Care Interdepartmental Working Group, formed to advise the Health Care Task Force on the reform plan.</p>
<p>This collection is divided into two seperate segments. Click here for records from:<br /><a href="http://clinton.presidentiallibraries.us/items/browse?advanced%5B0%5D%5Belement_id%5D=43&advanced%5B0%5D%5Btype%5D=is+exactly&advanced%5B0%5D%5Bterms%5D=2006-0810-F+Segment+1"><strong>Segment One</strong></a> <br /><a href="http://clinton.presidentiallibraries.us/items/browse?advanced%5B0%5D%5Belement_id%5D=43&advanced%5B0%5D%5Btype%5D=is+exactly&advanced%5B0%5D%5Bterms%5D=2006-0810-F+Segment+2"><strong>Segment Two</strong></a></p>
Provenance
A statement of any changes in ownership and custody of the resource since its creation that are significant for its authenticity, integrity, and interpretation. The statement may include a description of any changes successive custodians made to the resource.
Clinton Presidential Records
Publisher
An entity responsible for making the resource available
William J. Clinton Presidential Library & Museum
Text
A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text.
Original Format
The type of object, such as painting, sculpture, paper, photo, and additional data
Paper
Dublin Core
The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/.
Title
A name given to the resource
Population Based Prevention and Public Health [8]
Creator
An entity primarily responsible for making the resource
Health Care Task Force
General Files
Identifier
An unambiguous reference to the resource within a given context
2006-0810-F Segment 1
Is Part Of
A related resource in which the described resource is physically or logically included.
Box 64
<a href="http://clinton.presidentiallibraries.us/items/show/36144" target="_blank">Collection Finding Aid</a>
<a href="https://catalog.archives.gov/id/12090749" target="_blank">National Archives Catalog Description</a>
Provenance
A statement of any changes in ownership and custody of the resource since its creation that are significant for its authenticity, integrity, and interpretation. The statement may include a description of any changes successive custodians made to the resource.
Clinton Presidential Records: White House Staff and Office Files
Publisher
An entity responsible for making the resource available
William J. Clinton Presidential Library & Museum
Format
The file format, physical medium, or dimensions of the resource
Adobe Acrobat Document
Medium
The material or physical carrier of the resource.
Preservation-Reproduction-Reference
Date Created
Date of creation of the resource.
5/5/2015
Source
A related resource from which the described resource is derived
42-t-2194630-20060810F-Seg1-064-011-2015
12090749