Comprehensive Community Interventions to Promote Health: Implications for
College-Age Drinking Problems
RALPH W. HINGSON, Sc.D., and JONATHAN HOWLAND,
Ph.D. Social and Behavioral Sciences Department, School of Public Health, Boston
University, 715 Albany Street, T2W, Boston, Massachusetts 02118
ABSTRACT.Objective: This article reviews comprehensive community
interventions that sought to reduce (1) cardiovascular disease risks; (2) smoking;
(3) alcohol use disorders, alcohol-related injury and illicit drug use; or (4)
sexual risk taking that could lead to HIV infection, sexually transmitted disease
and pregnancy. Method: Comprehensive community programs typically involve
multiple city government agencies as well as private citizens and organizations
and use multiple intervention strategies such as school-based and public education
programs, media advocacy, community organizing, environmental policy changes
and heightened enforcement of existing policies. This review focused on English-language
papers published over the past several decades. Results: Some programs
in each of the four problem areas achieved their behavioral and health goals.
The most consistent benefits were found in programs targeting behaviors with
immediate health consequences such as alcohol misuse or sexual risk taking.
Results were less consistent when consequences of targeted behaviors were more
distant in time such as cardiovascular risks and smoking. Also, programs that
targeted youth to prevent them from starting new health-compromising behaviors
tended to be more successful than programs aimed at modifying preexisting habits
among adults. Programs that combined environmental and institutional policy
change with theory-based education programs were the most likely to be successful.
Finally, programs tailored to local conditions by the communities themselves
tended to achieve more behavior change than programs imported from the outside.
Conclusions: Comprehensive community intervention approaches may have
considerable potential to reduce college-age drinking problems, especially given
the success of these programs in reducing alcohol-related problems and in preventing
health-compromising behaviors among youth. (J. Stud. Alcohol, Supplement
No. 14: 226-240, 2002)
CITING THE SUCCESS of comprehensive multifactorial community interventions
with other public health problems, the National Academy of Sciences has recommended
this approach for reducing alcohol-related health problems (Institute of Medicine,
1989). This approach was initially used to reduce heart disease and cardiovascular
risks. More recently, it has targeted underage drinking, traffic and other unintended
injuries caused by alcohol use as well as unplanned pregnancy, infections with
HIV and other sexually transmitted diseases.
Traditional public health strategies attempt to identify
and intervene with specific subpopulations at high risk for
a targeted health problem. In contrast, comprehensive
multistrategy community programs attempt to involve the
total community and its constituent organizations, institutions
and individuals across demographic and risk spectrums.
Comprehensive community programs addressing
problems associated with alcohol have involved multiple
agencies in city government as well as private citizens and
private organizations. Such programs seek to stimulate behavior
change by influencing the normative environment in
which high-risk individuals live.
Comprehensive community interventions are also distinguished
by the use of multiple intervention strategies for
changing health-related behaviors. Although programs differ
with respect to content and behavioral targets, they typically
include some combination of: school-based education,
public information programs, medical screening and treatment,
media advocacy, organizing and mobilizing of different
community groups and populations, promotion of
environmental changes that can influence the price and availability
of products like alcohol and tobacco that affect health
and programs to publicize and enhance enforcement of existing
laws pertaining to the use of alcohol, tobacco or other
drugs. Programs can also provide social and entertainment
activities that are alcohol, drug or tobacco free or that promote
healthy lifestyles such as exercise and good nutrition.
Although comprehensive community interventions share
the common features of multiple participants and interventions,
they have varied on a number of dimensions, including
the primary target population, the geographic or
organizational setting, who initiated the program and the
location of the director. Primary target populations can include
all members of a community, youth, adults, elderly,
certain minorities or racial subgroups or groups with distinct
behaviors. Geographic and organizational settings can
range from cities to neighborhoods or counties and may or
may not reflect geopolitical units. Initiators of such programs
have varied from university researchers to members
of city government or private organizations within a
community.
To date, few comprehensive community interventions
have explicitly included colleges and universities in developing
and implementing interventions, and fewer still have
specifically evaluated program impact on college students.
Further, not all of these comprehensive community interventions
have produced the desired public health objectives.
This review of comprehensive community programs
was conducted to assess what program components or characteristics
might be useful to implement to reduce alcohol
misuse and related health and social problems among college
students.
Cardiovascular Risk Reduction Projects
Several comprehensive community intervention initiatives
to reduce cardiovascular risks have been reported in
the scientific literature. In the United States, these efforts
have focused much more on individual change than on the
political, regulatory or enforcement environments (Table 1).
The North Karelia Project (Puska et al., 1981; Vartiainen
et al., 1994, 1998), the first community demonstration program
to reduce cardiovascular disease risks, began in an
eastern providence of Finland in 1972, which at that time
had the highest coronary mortality rate in the world. A
steering committee and working groups on health education,
stroke registration and hypertension control were organized.
Community analyses were undertaken using
baseline survey, insurance, health systems and mortality
data. Education on smoking, diet and hypertension was conducted
through voluntary organizations, health and social
service organizations, radio and television. Leaflets and posters
on smoking, diet, hypertension, physical exercise and
advice to heart disease patients were distributed at clinics,
schools, voluntary organizations, banks, pharmacies and
other stores. Training was offered to health center personnel,
doctors, public health nurses, teachers and social workers
regarding the measurement and management of
cardiovascular risk factors, smoking cessation, nutrition,
hypertension and coronary disease. The program promoted
"no smoking" policies in health centers, schools, restaurants
and offices; the growing of vegetables; and production
of lower fat foods. Hypertension, stroke and heart
disease registries were started.
To evaluate the efficacy of these programs, North Karelia
was compared with the Kuopio Province in Finland. Surveys
of both areas were completed in 1972, 1977, 1982,
1987 and 1992. Blood pressure measurements and blood
samples were collected.
Serum cholesterol declined 16% in North Karelia and
12% in Kuopio from 1972 to 1992. Systolic blood pressure
declined 4% in Kuopio from 1972 to 1992. Declines in
diastolic blood pressure were similar in both areas: 8-10%
from 1972 to 1992. Smoking declined from 52% to 32%
among men in North Karelia, significantly more than among
men in Kuopio (from 49% to 37%). The greatest declines
occurred during the first 5 years of the program and helped
to stimulate interest in comprehensive community interventions
for coronary disease prevention in the United States.
Beginning in 1978, a school-based component of the
North Karelia program exposed seventh grade students to a
2-year five-session smoking prevention program using a
social influence approach (Vartiainen et al., 1998). Students
were taught about pressures to smoke exerted by peers,
adults and mass media and were trained to deal with these
pressures. Short- and long-term hazards of smoking were
discussed.
In 1993, mean lifetime cigarette consumption was 22%
lower among program participants than among students in
comparison schools. The mean prevalence of smoking was
30% among program participants and 41% in control subjects.
Investigators concluded that long-term smoking prevention
effects could be achieved if a school-based program
using a social influence model were combined with community
and mass media interventions (Vartiainen et al., 1998).
Comprehensive community interventions to reduce cardiovascular
risks among U.S. adults have not been as successful
as the North Karelia study, but school-based components have achieved some desired
behavior changes.
The Pawtucket Heart Health Program (Carleton et al.,
1995) sought to reduce elevated blood pressure, cigarette
smoking and physical inactivity. Phases of behavior change,
awareness and agenda setting were promoted, and training
was provided in behavioral skills and development of social
supports. A community activation component encouraged
individuals to participate in self-help courses, direct
risk-factor measurement and counseling. More than 500
community organizations and more than 3,500 volunteers
were involved, including public and private schools, supermarkets,
grocery stores, religious and social organizations,
larger work sites, restaurants and most departments of city
government. The program introduced grocery store labeling
of low fat food; a nutrition education program at the
public library; restaurant menus highlighting health foods;
and a heart health curriculum in elementary, middle and
high schools. At least 42,000 individuals participated in
one or more program.
In a quasi-experimental evaluation design, Pawtucket was compared with a southern
New England city with similar sociodemographic characteristics. Large-scale
cross-sectional surveys were conducted in each community at 2-year intervals.
Smoking declined 6.6% in the intervention group and 8% in the comparison city.
There was no significant difference between cities in changes in mean blood
pressure or blood cholesterol. Body mass indices increased in both cities but
significantly more so in the comparison city. The projected cardiovascular disease
rates were significantly less15% in Pawtucket during the intervention
programbut the difference declined to 8% after the program ended, a difference
that was no longer statistically significant. The Pawtucket Heart Health evaluators
concluded that they found very limited evidence that cardiovascular risk factor
behaviors and disease risk changed through a process of community activation
at the individual, group, organization and community levels.
The Stanford 5 City Project (Farquhar et al., 1990;
Winkleby et al., 1996) sought to reduce cardiovascular risk
factors, morbidity and mortality. The program used both
mass media and interpersonal education programs. Messages
were developed using social learning theory, social
marketing theory, persuasion theories and community development
strategies. Community organizing was undertaken
to create social and institutional support to sustain the
program's initiatives.
In both intervention and comparison cities, there were improvements over time
in cardiovascular disease knowledge, cholesterol and smoking, making between-city
group differences more difficult to observe. Six years after the start of the
program, there were significant improvements in the two program communities
relative to three matched comparison communities in knowledge about coronary
heart disease etiology and risk factors. These differences dissipated by the
9-year follow-up. There were no significant reductions in blood pressure at
6 years, but by the 9-year follow-up the intervention cities had significantly
greater declines in blood pressure than comparison areas. No significant differences
in smoking rates were seen between intervention and comparison areas at 6- or
9-year follow-ups. There were no significant differences in body mass indices
for women. However, for men, the control cities actually had more favorable
scores at the 9-year follow-up. Nine years after the start of the program, the
treatment cities had a significantly better mortality risk for women but not
men.
The Minnesota Heart Health Program (Luepker et al., 1994) involved nearly
400,000 people in six communities. Initiated in 1980, it sought to reduce cardiovascular
morbidity and mortality by 15% by reducing sedentary behaviors, cigarette smoking,
blood cholesterol levels and blood pressure levels. The intervention operated
at the individual, group and community levels and embraced a wide range of strategies
and theories, including social learning theory and persuasive communications
theory. The program alerted people to health issues and provided incentives
to adopt effective health promoting behavioral alternatives. Community leaders
were encouraged to foster environmental change to support risk reduction. Mass
media established awareness of the program, disseminated risk factor messages
and enhanced the salience of healthy lifestyles. Community health professionals
were involved through their local organizations and preventive practice advisory
committees. More than 60% of all adult residents received onsite riskfactor
screening, education and counseling. The adult education component provided
multiple-contact programs of intensive personal counseling on cardiovascular
risk reduction. School-based education discouraged health-compromising behaviors
in youth and their parents.
Three pairs of communities, matched on size and type,
were compared through cohort surveys over a 6- to 7-year
period. Against a backdrop of strong secular trends of increasing
health promotion and declining risk factors, the
overall program effects were modest in size and duration
and generally within chance levels. No significance between
group differences were seen in blood cholesterol, smoking
among men, blood pressure, body mass index or overall
coronary heart disease risk. Slight beneficial treatment effects
were seen for smoking among women and exercise in
the final wave of the study.
Two school-based educational components of the Minnesota Heart Health Program,
however, achieved more marked behavioral changes. First, the Class of 1989
Study (Perry et al., 1992), an intensive intervention to reduce or prevent
adolescent smoking, was offered to seventh grade students in the fall of 1984.
It addressed social and psychological factors that encourage smoking. Students
identified the short-term consequences of smoking, such as smelling like smoke
or having bad breath. Their expectations of how many of their peers smoke were
compared with actual data on smoking prevalence. Reasons people smoke were studied,
and positive alternatives were offered. Finally, students learned skills to
resist advertising, peer and adult pressures to smoke. Participants created
anti-tobacco advertisements, skits and role-playing scenarios. Sixth grade students
made public commitments not to smoke.
Annual surveys from 1983 to 1989 compared one of the
intervention communities to its matched pair. Although
smoking rates among students in the two communities were
comparable at baseline, the proportion of adolescents smoking
at the end of high school was significantly lower in the
intervention community, with 14.6% of students smoking
at least weekly in the intervention group compared with
24.1% in the control community. The authors concluded
that the study results permit optimism about the benefits to
youth of long-term school-based community-wide programs
for delaying onset of smoking.
Second, a peer-led physical activity program designed for eighth grade students
using social learning theory encouraged regular aerobic physical activity through
a 4-week community-wide competition (Kelder et al., 1993). Students in the intervention
community were challenged to exercise outside of school the equivalent of bicycling
250 miles. They received instruction on monitoring heart rates and choosing
aerobic activities. In tenth grade, a 10-lesson peer-led curriculum to promote
healthy eating and regular aerobic activity was introduced. The program used
environmental-level strategies by having same-age peers provide new opportunities
for healthier eating, physical activity and reducing barriers to aerobic activity
at school and by creating peer, family and school personnel support. Based on
annual surveys in the intervention and matched-pair community, females in the
intervention community reported at follow-up greater hours of exercise in all
but the eleventh grade. Males in the intervention community reported significantly
more exercise in the seventh and eleventh grade surveys. Overall, the most pronounced
differences were among females.
Community Interventions to Reduce Smoking
Comprehensive community programs have also attempted
with mixed effectiveness to reduce tobacco use among youth
and adult populations (Table 2).
The Tobacco Policy Options for Prevention Project (Forster
et al., 1998), a 32-month intervention, sought to reduce youth
access to tobacco by community mobilization to change local
ordinances, retailer and other adult practices regarding provision
of tobacco to youth and increased enforcement of age of
sales laws. Fourteen Minnesota communities were randomly
assigned to intervention or comparison conditions. Tobacco
purchase attempts by youth were tried at all tobacco outlets
in June 1993 and June 1996.
School surveys in 1993 and 1996 with more than 6,000
students indicated smoking by adolescents increased in both
sets of cities, but less in the intervention communities. The
intervention had little effect on perceptions of tobacco availability
through social sources, but it reduced perceived access
through commercial sources. Purchase attempts
declined significantly in the intervention communities. In
all communities, there was a sharp decrease in youth purchase
attempts that resulted in sales. This decline, however,
was not significantly greater in the intervention cities.
Rigotti et al. (1997) compared three Massachusetts communities
that increased enforcement of youth tobacco laws
with three matched comparison communities. Health departments
in the intervention communities began quarterly
compliance checks with underage purchase attempts. At
baseline, 68% of vendors sold to minors, with no statistical
difference between intervention and control community.
Two years later, only 18% of the vendors in the intervention
communities compared with 55% in the comparison
communities sold to minors. However, three annual surveys
with more than 17,600 respondents revealed only a
small drop in the ability of adolescents under age 18 to
purchase tobacco and no decline in its use.
Cummings et al. (1998) reported on a similar enforcement
program in Erie County, New York. Six pairs of communities
were matched on sociodemographic characteristics,
population and number of tobacco outlets. Underage purchase
compliance checks directed by local police were conducted
in 366 stores at baseline and 319 at follow-up. All
tobacco vendors were sent a letter about tobacco laws pertaining
to minors and a warning that compliance checks
were planned. There was a dramatic increase in compliance
with the law in both enforcement and nonenforcement
communities. However, compliance rates between the two
groups of communities did not vary because most vendors
in both areas knew about the enforcement program and the
perceived enforcement as more vigilant.
Gemson et al. (1998) reported results from a randomized trial of 15 tobacco
vendors in Central Harlem. Stores were randomly allocated to control, education
or enforcement conditions. Surveys of underage tobacco purchase compliance were
conducted in October 1993 and in April 1994. During both surveys, violators
in the enforcement group of stores were fined in accordance with the state law.
At baseline, 70% of stores sold loose cigarettes, and 98% sold either singles
or packs of cigarettes. At the 6- month follow-up, sales declined 16% among
control stores, 34% among education stores and 56% among enforcement stores.
The Community Intervention Trial for Smoking Cessation
(COMMIT) (COMMIT Research Group, 1995a,b) was
designed to help smokers, especially heavy smokers, achieve
and maintain cessation. Within each of 11 matched community
pairs, one community was randomly assigned to
receive the intervention from January 1989 to December
1992. COMMIT was a partnership with National Cancer
Institute staff, 11 participating research institutions and their
corresponding local communities. Each community formed
a community board. COMMIT fostered demand for cessation
services, using public education through media and
community-wide events, health care providers, work sites
and other organization and cessation resources. The protocol
mandated 58 activities to be carried out by local staff
or agencies.
Based on a longitudinal survey, the investigators found no significant changes
at follow-up among heavy smokers across the experimental groups. Quit rates
for light/moderate smokers at baseline were .306 in intervention communities
versus .275 in comparison communities, a significant between-group difference.
Random digit dial cross-sectional surveys of approximately 2,800 subjects per
city were conducted at baseline and follow-up. Overall, smoking prevalence decreased
3.5% in the program communities versus 3.2% in the comparison communities, a
nonsignificant difference.
Neighbors for a Smoke Free North Side (Fisher et al.,
1998) emphasized neighborhood-based governance and resident
involvement around the goal of nonsmoking. Wellness
councils including neighborhood volunteers and paid staff
members ran the program for 24 months. The program included
smoking cessation classes, billboards, door-to-door
campaigns and a gospel fest. Three predominately low-income
black neighborhoods in St. Louis received the intervention
and were compared with similar neighborhoods in
Kansas City. Random digit dial telephone surveys in 1990
and 1992 indicated smoking prevalence declined 7% in St.
Louis, significantly more than in Kansas City, where it declined
1%.
The evaluators of the Smoke Free North Side intervention
suggest this program had greater success than COMMIT
because COMMIT was centrally developed at the
national level. COMMIT was delivered to communities
rather than developed by the communities.
Comprehensive Community Programs Addressing
Alcohol-Related Problems
Several comprehensive community intervention programs
have addressed alcohol use or related problems with positive
results (Table 3). Two sought to reduce alcohol and
other substance use among elementary and middle school
students (Chou et al., 1998; Pentz et al., 1989; Perry et al.,
1996). One focused on reducing access to alcohol and drinking
among underage adolescents (Wagenaar et al., 2000a,b).
Two others targeted entire community populations, including
adolescents and young adults (Hingson et al., 1996;
Holder, 1997; Holder et al., 2000). Although none specifically
measured college drinking and related problems, these
studies nonetheless have the most direct implications for
those planning comprehensive community college interventions
to address alcohol-related health problems.
The Midwestern Prevention Project (Pentz et al., 1989)
attempted to prevent substance abuse among adolescents
ages 10-14 in Kansas City, Missouri, and Indianapolis, Indiana.
A quasi-experimental design in Kansas City and a
randomized experimental design in Indianapolis evaluated
the program. In Kansas City, from September 1984 to January
1986, students received a 10-session youth training program
on skills for resisting alcohol, tobacco and illicit drug
use; homework involving active interviews; and role plays
with parents and family. Most students interviewed parents
and family members about family rules regarding the use
of these substances and regarding successful techniques to
avoid their use and counteract media and community influences.
Mass media coverage focused on psychosocial consequences
of substance abuse; correction of perceptions
about the prevalence of peer drug use; recognition of adult
media and community influences on substance use; peer
and environmental pressure resistance; and statements of
public commitments to avoid alcohol, tobacco and other
drug use. Modeling and role playing of resistance skills,
feedback with peer reinforcement, peer leader facilitation
and discussion of homework results were part of the program.
Of the 42 schools that participated, 4 were randomly
assigned to the program condition and 4 to the control condition.
The remaining 34 were assigned on the basis of
schools' willingness to accept the program; 20 were willing,
and 14 were not. This willingness may have reflected
higher salience of substance abuse in those schools. The 20
willing schools were assigned to the program, raising the
total number of intervention schools to 24; there were 18
control schools.
Although cigarette, alcohol and marijuana use increased
in both groups, increases for all substances were significantly
lower in the intervention group 2 years after the
program. When students in the 24 intervention schools were
compared at 1-year follow-up with students in comparison
schools, prevalence of use of all three drugs was lower in
the intervention schools: 11% versus 16% for alcohol use,
17% versus 24% for cigarette use and 7% versus 16% for
marijuana use.
In a study by Chou et al. (1998), investigators tracked
1,904 students exposed to the program in Indianapolis. They
were compared with 1,508 students in the control group.
Schools were randomly assigned to groups, and students
were followed at 6 months, 1.5 years, 2.5 years and 3.5
years after baseline. After analytically controlling for
ethnicity, gender, socioeconomic status, father's occupation
and school type and grade, the researchers found that, among
subjects using alcohol, tobacco or other drugs at baseline,
secondary prevention effects reducing alcohol use were
found at the 6-month and 1.5-year follow-up and for tobacco
use at 6-month follow-up. Results for marijuana use
were inconsistent over time.
Project Northland (Perry et al., 1996) in Minnesota was
designed to reduce alcohol use among young adolescents.
Sixth, seventh and eighth graders were exposed to a 3-year
behavioral curriculum, with peer leadership, parental involvement
and community task force activities. Students
learned how to resist peer influence and normative expectations
about alcohol and methods to bring about community
social, political and institutional change in
alcohol-related programs and policies. Students interviewed
parents, local government officials, law enforcement personnel,
retail alcohol merchants, schoolteachers and administrators
about their beliefs and activities concerning
adolescent alcohol use. A "town meeting" conducted by
students made recommendations for community action for
alcohol use prevention.
Community task forces including government officials,
law enforcement personnel, school representatives, health
professionals, youth workers, parents, concerned citizens
and adolescents stimulated passage ordinances to prevent
alcohol sales to minors and intoxicated patrons. Businesses
provided discounts to students who pledged to be alcohol
and drug free.
At baseline, 2,351 students were surveyed. Two-year
follow-up rates greater than 80% were achieved in intervention
and comparison groups. A higher percentage of
students in the intervention group were alcohol users at
baseline, whereas at follow-up the percentages that used
alcohol in the past week and past month were lower in the
intervention group. Differences were greatest and significant
among those who did not use alcohol at baseline. No
significant follow-up differences between groups were found
on measures of cigarette smoking or marijuana use.
In the Communities Mobilizing for Change project
(Wagenaar et al., 2000a), 15 communities were randomly
allocated to intervention or comparison groups. A community
organizing intervention sought to reduce the number
of alcohol outlets selling to youth under the legal drinking
age and the availability of alcohol to youth from noncommercial
sources such as parents, siblings and older peers.
Action was encouraged through city councils, school and
enforcement agencies, alcohol merchants, business associations
and the media. A leadership strategy team changed
numerous policies, procedures and practices in the Communities
Mobilizing for Change group. In 1992 and 1995,
approximately 4,500 twelfth graders were surveyed. A telephone
survey of 3,095 18- to 20-year olds was conducted
in 1992 and repeated in 1995. Response rates were greater
than 90%. Compliance check surveys of sales to study confederates
who appeared underage were conducted at more
than 25 off-sale outlets in 1992 and 1995.
Relative to the comparison communities, the intervention
communities achieved a 17% increase in checking age identification of youthful-appearing
alcohol purchasers and
a 24% decline in sales to potential underage purchasers by
bars and restaurants. There was a 25% decrease in the proportion
of 18- to 20-year olds seeking to buy alcohol, a
17% decline in the proportion of older teens who provided
alcohol to younger teens and a 7% decrease in the percentage
of respondents under age 21 who drank in the last 30
days. Among 18- to 20-year olds, there was also a significant
decline in arrests for driving under the influence
(Wagenaar et al., 2000b).
The Community Prevention Trial Program (Holder, 1997;
Holder et al., 2000) was a 5-year initiative designed to reduce
alcohol-involved injuries and death in three experimental
communities. The program used five reinforcing
components to change individual behavior by altering the
environmental, social and structural contexts of alcohol use.
First, community mobilization stimulated public policy interventions
by increasing general awareness and concern
about alcohol-related trauma. Program initiatives jointly
planned by project organizers and local residents were
implemented by the residents. Media, mobilization and intervention
activities had specific behavioral objectives tailored
to each site. Second, a "responsible beverage server"
component sought to reduce sales to intoxicated patrons
and to increase local enforcement of alcohol laws by working
with restaurants, bar and hotel associations, beverage
wholesalers and the Alcohol Beverage Control Commission.
Third, a driving-while-intoxicated (DWI) component
sought to increase the number of DWI arrests by a combination
of officer training, deployment of passive alcohol
sensors and media publicized checkpoints. Fourth, media
advocacy focused news attention on underage drinking; enforcement
of underage sales laws was increased; and clerks,
owners and managers were trained to prevent sales of alcohol
to minors. Fifth, local zoning powers regarding alcohol
outlet density were used to reduce availability of alcohol.
A quasi-experimental evaluation of intervention and comparison
communities documented the effects of each
intervention as well as the overall project effects on
alcohol-related injuries. During the study, local regulation
of alcohol outlets and public sites for drinking changed in
all three experimental communities. Compliance checks at
150 outlets where underage persons attempted to purchase
alcohol revealed a significant reduction in alcohol purchase
by youth.
The DWI reduction component resulted in an increase in
news coverage of DWI, additional police enforcement and
greater use of breath analyzer equipment. Telephone surveys
revealed a significant increase in perceived likelihood of DWI
arrest and a reduction in self-reported frequency of driving
after drinking. Roadside surveys also revealed reduction in
driving after drinking. Alcohol-related crash involvement as
measured by single vehicle night crashes declined 10-11%
more in the program group than the comparison communities.
Alcohol-related trauma visits to emergency departments
declined 43% (Holder et al., 2000).
The Massachusetts Saving Lives Program (Hingson et
al., 1996) was a 5-year (1988-93) comprehensive community
intervention designed to reduce alcohol-impaired driving
and related traffic deaths. Six program communities
were selected based on a competitive proposal process.
These were compared with five matched communities whose
applications also satisfied selection criteria but were not
funded. The rest of Massachusetts also served as a comparison.
Outcome data were collected for the period 5 years
before and 5 years after the intervention.
In each program community, a full-time coordinator from
the mayor's office organized a task force of concerned private
citizens and organizations and officials representing
various city departments (e.g., school, health, police and
recreation). Each community received approximately $1 per
inhabitant per year in program funds. Half the funds were
spent to hire the coordinator and the balance was for increased
police enforcement and other program activities and
educational materials. Voluntary activity was also encouraged.
Active task force membership ranged from 20 to 100
persons in each community. An average of 50 organizations
participated in each city.
Most initiatives were developed by the communities. The
program sought to reduce drunk driving and related risky
behaviors such as speeding, running red lights, failure to
yield to pedestrians in crosswalks and failure to wear safety
belts. To reduce drunk driving and speeding, communities
introduced media campaigns, checkpoints, business information
programs, speeding and drunk-driving awareness
days, speed watch telephone hotlines, police training, high
school peer-led education, Students Against Drunk Driving
chapters, college prevention programs, alcohol-free prom
nights, beer keg registration and increased liquor outlet surveillance
by police to reduce underage alcohol purchase.
To increase pedestrian safety and safety belt use, program
communities conducted media campaigns and police checkpoints,
posted crosswalk signs warning motorists of fines
for failure to yield to pedestrians, added crosswalk guards
and offered preschool education programs and training for
hospital and prenatal staff. Coordinators engaged in numerous
media advocacy activities to explain trends in local
traffic safety problems and strategies the communities were
implementing to reduce traffic injury and death. The proportion
of drivers under age 20 who reported driving after
drinking in random digit dial telephone surveys declined
from 19% during the first year of the program to 9% in
subsequent years. There was little change in comparison
areas. The proportion of vehicles observed speeding through
use of radar was cut in half; there was also little change in
comparison cities. There was a 7% increase in safety belt
use, a significantly greater increase than found in the comparison
area.
Fatal crashes declined from 178 during the 5 preprogram
years to 120 during the 5 program years. This was a
25% greater reduction than in the rest of Massachusetts.
Fatal crashes involving alcohol declined 42%, and the number
of fatally injured drivers with positive blood alcohol
levels declined 47% relative to the rest of Massachusetts
(90% of fatally injured drivers in Massachusetts are tested
annually for alcohol). Visible injuries per 100 crashes declined
5% more in the program than the rest of the state
during the program period. The fatal crash declines were
greater in program cities, particularly among younger drivers
15- to 25-years old. All six program cities had greater
declines in fatal and alcohol-related fatal crashes than did
comparison cities or the rest of the state. Interventions varied
somewhat by community. This suggests that organizing
the community program and combining environmental
policy changes and enforcement with theory-based school
education programs was more important than any specific
initiative in contributing to program success.
In addition to the programs reviewed above, several large
community intervention initiatives addressing alcohol and
other drug use have been launched. The Center for Substance
Abuse Prevention provided 251 cities with 5-year
grants from 1990 to 1996 to organize community antidrug
coalitions.
Awarded approximately $350,000 per year, each program
developed a steering committee, mobilized and trained
volunteers, undertook a needs assessment of prevention services
and developed a comprehensive prevention plan. Each
also implemented media campaigns, community school and
cultural events, alternate youth recreation activities, parent
and family programs and employment and workplace programs.
Policy and regulatory initiatives varied by community
but included drug-free and smoke-free school,
workplace and other location policies; heightened penalties
for drug use possession; lower legal blood alcohol limits
for adult and youth drivers; fines for selling alcohol and
tobacco to youth; and an evaluation component to assess
program implementation and impact on substance use.
Summary evaluations (Kaftarian, 2000; Yin et al., 1997) randomly selected 24
partnerships from a total of 251 and compared them with 24 nonpartnership communities
matched on demographic characteristics. Repeat cross-sectional surveys were
conducted with 83,473 randomly selected adults, tenth graders and eighth graders.
Adults were surveyed at home, and youth were surveyed at school. Substance use
rates were compared over an 18-month interval from 1994/1995 to 1996. In the
partnership communities, male substance use rates were 3-5% lower at follow-up
on five of six outcome measures of regular alcohol and illicit drug use.
When responses of males and females were combined,
only one of six outcomes significantly favored the partnerships.
This finding persisted with regressions controlling
for age, gender and race. However, of the remaining 11
outcomes, all but the smallest were in the predicted direction
favoring the community partnership program. When
individual partnership-comparison communities were examined,
8 of 24 partnership communities showed some statistically
significant reduction in substance use relative to their
comparison communities.
Comprehensive Community Interventions to Reduce
Behavioral Sexual Risks (Table 4)
The Mpowerment Project (Kegeles et al., 1996) was undertaken
to reduce sexual risk taking among young gay
men. A core group and a community advisory board ran
the program. Young male peer outreach workers diffused
safer sex messages developed by other volunteers at bars
and community and special events and recruited other men
into the project, who in turn conducted program outreach
and education.
The teams also tried to create new events that would
attract young gay men at which safer sex could be promoted.
For example, the Mpowerment center offered weekly
small group meetings; video, dance and open house parties;
rap groups; drop-in hours; picnics; hikes; and bicycle
rides. At least 500 men in the study community of Eugene,
Oregon, attended these events. These group sessions were
presented as a fun way for young gay men to meet other
young gay men. Efforts were made to recruit 75-80% of
young gay men in the community into the groups. Publicity
included articles and advertisements in the gay newspaper.
Eugene was compared with Santa Barbara, California.
Following the intervention, the proportion of men engaging
in any unprotected anal intercourse decreased from 41% to
30%. It decreased from 70% to 11% with nonprimary partners
and from 59% to 45% with boyfriends. No significant
changes occurred in the comparison community.
The CDC AIDS Community Demonstration Projects
(CDC AIDS Community Demonstration Projects Research
Group, 1999) sought to reduce the risk of infection among
active injection drug users, female sex partners of male
injection drug users, female commercial sex workers, other
women who trade sex for money or drugs, youth in highrisk
situations, men who have sex with men and residents
of census tracts with high rates of sexually transmitted diseases.
All sites used a common intervention based on behavioral
theories, study community ethnographic research
and interventions in prior community studies. The intervention
sought to mobilize community members to distribute
and verbally reinforce prevention messages, create small
media materials featuring theory-based prevention messages
in role-model stories and increase availability of condoms
and bleach kits. One thousand persons from targeted at-risk
communities, other local residents and area business personnel
who had regular contact with the targeted population
were recruited and trained to distribute intervention
materials from July 1991 to June 1994. A total of 585,000
small media materials were distributed containing authentic
stories about people from the community who were
changing behaviors. The stories emphasized stages of behavior
change based on the transtheoretical model. A quasiexperimental
cross-sectional design compared randomly
allocated intervention and comparison areas. Anonymous
field interviews were conducted in 10 cross-sectional waves
from 1991 through 1994. Data from 15,205 interviews were
analyzed.
By the end of the intervention, more than half the population
had been reached by the intervention at least once in
the prior 3 months. Significant increases in the intervention
relative to the comparison communities were achieved in
condom use with a main partner, in condom use with a
nonmain partner and in the carrying of condoms.
The School/Community Program for Sexual Risk Reduction
among Teens (Vincent et al., 1987) was initiated in the
western portion of a South Carolina county in 1982 to reduce
unintended pregnancies among never-married teens
and preteens. It promoted postponement of initial voluntary
sexual intercourse and consistent use of effective contraception.
Intervention strategies included increasing decisionmaking
skills; improving interpersonal communication
skills; enhancing self-esteem; aligning personal values with
those of the family, church and community; and increasing
knowledge of human reproductive anatomy, physiology and
conception.
Adults in the community were trained initially. Two-thirds of district teachers,
administrative staff and special service personnel completed at least one university
course related to facets of sex education. The trained teachers assisted the
project staff in implementing sex education in all grades (kindergarten-12),
beginning in 1983. Teachers integrated units of instruction within their biology
science and social studies classes. Clergy, church leaders and parents were
recruited to attend five-session minicourses with much the same content as given
to teachers. Local newspapers and radio promoted program messages. Messages
on alcohol, drug abuse, nutrition and smoking were also integrated into the
program.
The intervention county was compared with four other
counties with similar demographics. The rate of pregnancies
per 1,000 females was recorded among females ages
14-17 in each community. Rates prior to the intervention
from 1981-1982 were compared with intervention years
1984-1985. There was a sharp reduction during the program
period in pregnancies in the target relative to comparison
areas: from 60 in 1,000 females to 25 in 1,000 in
1984-1985.
The AIDS Prevention for Pediatric Life Enrichment
Project (Santelli et al., 1995) was a community-based program
to prevent prenatal HIV infection in women that used
street outreach and targeted small media to promote condom
use. Based on social learning theory, small media publicized
HIV risk-reduction messages (e.g., condom use) in
the form of role model stories. Stories drawn from focus
groups on experiences of persons in the target audience
were put into comic books, newsletters, pamphlets and condom
envelopes. Three full-time paid street outreach workers
and volunteers contacted community residents on street
corners, in local shopping areas and through community
agencies. In the quasi-experimental time series design, 500
or more face-to-face interviews using a modified street intercept
approach were conducted annually in intervention
and comparison communities.
Between October 1990 and May 1992, 26,461 street outreach
contacts, 26,020 media materials and 65,217 condoms
were distributed. The program's name was known by 40%
of the respondents by 1992; 36% had contact with street
workers. Condom use increased significantly more (from
30% to 40%) in intervention communities than in the comparison
communities (from 22% to 27%). No differences,
however, occurred over time with respect to inquiring about
a sexual partner's history of sexually transmitted disease,
rejecting sex for fear of a disease or avoiding sexual contacts
when usual partners were not available.
Conclusions
From this review of comprehensive community interventions,
several conclusions can be drawn that may have
relevance for college interventions to reduce alcohol-related
problems. First, these studies indicate this approach can be
effectively applied to a variety of public health problems.
Comprehensive community programs consistently yielded
significant reductions in alcohol use or related problems.
Second, these studies indicate that health-related behavioral
change can be accomplished among college-age adolescents
and young adults. The majority of programs
(although not all) targeting youth achieved changes in the
desired direction. Programs targeting middle-age adults (e.g.,
the cardiovascular risk reduction programs) had the least
differential success, but were implemented in a context of
widespread secular changes on program outcomes as were
seen in comparison communities. Indeed, in some respects,
young people who are in the process of adopting behaviors
and lifestyles may be more receptive to intervention than
more habituated adults. Moreover, adolescents and young
adults may be more sensitive to shifts in behavioral norms
because their primary social units involve friends and acquaintances
rather than spouses and children. Those strategies
aimed at norm change and diffusion of innovation may
be more effective in college settings than in more diverse
communities because students at a given campus will share
a range of characteristics and attributes. Accordingly, the
community intervention approach might be specifically appropriate
for college intervention programs.
Third, programs are more likely to succeed if they combine environmental and
institutional change with theory-based education interventions designed to change
individual behavior. This is particularly pertinent to colleges because the
administration has influence over aspects of the campus' physical and social
environments and many colleges have researchers who are familiar with theory-based
educational strategies to promote behavior change.
Fourth, programs that involve community ownership appear
to succeed more often than programs imported into
the community from the outside. Community development
requires a matrix of local organizations and institutions,
both public and private. Although these entities are part of
most municipalities, colleges also have extensive networks
of campus-based student organizations concerned with student
life. Accordingly, college administrators desiring collaboration
with students around alcohol interventions should
have no problem identifying and tapping into student organizations
willing to contribute.
Comprehensive community interventions to reduce college
alcohol-related problems have not yet been evaluated.
A multicomponent college program, which was implemented
in the 1980s, achieved little discernible impact (Kraft, 1988),
but that program did not involve surrounding communities,
public officials or private citizens (Hingson et al., 1997).
Without involvement of the communities that surround colleges,
intensive multifaceted efforts by colleges may drive
alcohol use into the community or may be undermined by
alcohol availability and promotion in the community.
Many questions remain about how best to implement
comprehensive college/community interventions to reduce
drinking and alcohol-related problems. For example, it needs
to be determined whether the behavioral focus should be
on drinking practices only or the multiple problems
alcohol-using college students pose for themselves and others.
The literature on comprehensive community interventions
seems to show the greatest impact when the behavioral
targets are more focused rather than multifaceted, as were
the cardiovascular risk reduction projects.
The locus of decision making and responsibility can vary
from the community to colleges or both, and the implementation
of this decision-making locus may influence the
success of the program. The balance between governmental
versus private organizational control or direction could
also influence program success, but this has not been
studied.
Identifying optimal strategies to mobilize communities for action warrants
research attention. Wagenaar et al. (1999) outlined a specific process to mobilize
communities that included (1) assessing community interests, (2) building a
core base of support in the community, (3) expanding the base, (4) developing
a plan of action, (5) implementing the plan, (6) maintaining the effort and
institutionalizing it and (7) evaluating and disseminating results. Whether
that plausible sequence is the optimal one to follow can be systematically tested.
Similarly, the speed and breadth of recruitment into community task forces can
be studied. Potential tradeoffs between smaller but more motivated and cohesive
task forces versus larger more broadly representative but perhaps less unified
work groups can be explored.
The role of evaluators in identifying program interventions needs to be assessed.
Many of the projects reviewed in this article had evaluators who actively engaged
in problem identification and feedback to the communities being studied about
their progress (or lack there of) in meeting program objectives. How directive
program evaluators should be in the process of selecting or modifying program
initiatives warrants study. On the one hand, evaluators may have greater access
to scientific evaluations of interventions. On the other hand, evaluators may
not have as clear an understanding as community members regarding the feasibility
of adoption and implementation of specific interventions. Also community ownership
in selecting interventions may yield greater motivation to implement them vigorously.
The role of students in formulating and implementing
comprehensive programs also deserves research attention.
Peer-led educational interventions have shown consistent
success among middle and high school students. For example,
in Project Northland, peers who participated in planning
alcohol-free social events reported less drinking than
did nonparticipants (Komro et al., 1996). Whether similar
results will apply among colleges and universities has yet
to be determined.
Some interventions, such as Project Northland, involved
youth in policy-setting activities. Whether students will be
more supportive and compliant with policies they had a
role in defining warrants research attention. Whether they
will arrive at different and perhaps more effective policy
decisions than would city officials or campus administrators
also is unknown.
Legal and environmental interventions that influence access
to alcohol and enforce laws governing behavior after
drinking such as drunk-driving laws have had an influence
on college-age persons. Interventions targeting individual
knowledge and behavior change have also produced behavior
changes, particularly among youth. Comprehensive
community college intervention programs may want to include
both types of activities. The best balance has yet to
be determined. Whether they will produce additive or interactive
effects needs to be studied.
References
Carleton, R.A., Lasater, T.M., Assaf, A.R., Feldman, H.A. and McKinlay, S.
The Pawtucket Heart Health Program: Community changes in cardiovascular risk
factors and projected disease risk. Amer. J. Publ. Hlth 85: 777-785,
1995.
CDC Aids Community Demonstration Projects Research Group. Community-level HIV
intervention in 5 cities: Final outcome data from the CDC AIDS community demonstration
projects. Amer. J. Publ. Hlth 89: 336-345, 1999.
Chou, C.P., Montgomery, S., Pentz, M., Rohrbach, L.A., Johnson, C.A., Flay,
B.R. and MacKinnon, D.P. Effects of a community-based prevention program on
decreasing drug use in high-risk adolescents. Amer. J. Publ. Hlth 88:
944-948, 1998.
COMMIT Research Group. Community Intervention Trial for Smoking Cessation (COMMIT):
I. Cohort results from a four-year community intervention. Amer. J. Publ. Hlth
85: 183-192, 1995a.
COMMIT Research Group. Community Intervention Trial for Smoking Cessation (COMMIT):
II. Changes in adult cigarette smoking prevalence. Amer. J. Publ. Hlth 85:
193-200, 1995b.
Cummings, K.M., Hyland, A., Saunders-Martin, T., Perla, J., Coppola, P.R. and
Pechacek, T.F. Evolution of an enforcement program to reduce tobacco sales to
minors. Amer. J. Publ. Hlth 88: 932-936, 1998.
Farquhar, J.W., Fortmann, S.P., Flora, J.A., Taylor, C.B., Haskell, W.L., Williams,
P.T., Maccoby, N. and Wood, P.D. Effects of communitywide education on cardiovascular
disease risk factors: The Stanford Five-City Project. JAMA 264: 359-365,
1990.
Fisher, E.B., Auslander, W.F., Munro, J.F., Arfken, C.L., Brownson, R.C. and
Owens, N.W. Neighbors for a smoke-free north side: Evaluation of a community
organization approach to promoting smoking cessation among African Americans.
Amer. J. Publ. Hlth 88: 1658- 1663, 1998.
Forster, J.L., Murray, D.M., Wolfson, M., Blaine, T.M., Wagenaar, A.C. and
Hennrikus, D.J. The effects of community policies to reduce youth access to
tobacco. Amer. J. Publ. Hlth 88: 1193-1198, 1998.
Gemson, D.H., Moats, H.L., Watkins, B.X., Ganz, M.L., Robinson, S. and Healton,
E. Laying down the law: Reducing illegal tobacco sales to minors in central
Harlem. Amer. J. Publ. Hlth 88: 936-939, 1998.
Hingson, R., Berson, J. and Dowley, K. Intervention to reduce college student
drinking and related health and social problems. In: Plant, M. and Single, E.
(Eds.) Alcohol: Minimising the Harm: What Works? New York: Free Association
Books, 1997, pp. 143-170.
Hingson, R., McGovern, T., Howland, J. and Hereen, T. Reducing alcohol-impaired
driving in Massachusetts: The Saving Lives Program. Amer. J. Publ. Hlth 86:
791-797, 1996.
Holder, H. (Ed.) A Community Prevention Trial to Reduce Alcohol-Involved Trauma.
Addiction 92 (Suppl. No. 2): S155-S301, 1997.
Holder, H.D., Gruenewald, P.J., Ponicki, W.R., Treno, A.J., Grube, J.W., Saltz,
R.F., Voas, R.B., Reynolds, R., Davis, J., Sanchez, L., Gaumont, G. and Roeper,
P. Effects of community-based interventions on high-risk driving and alcohol-related
injuries. JAMA 284: 2341-2347, 2000.
Institute Of Medicine. Prevention and Treatment of Alcohol Problems: Research
Opportunities, Washington, DC, National Academy Press, 1989.
Kaftarian, S. Prevention Works through Community Partnerships: Findings from
SAMHSA/CSAP's National Evaluation, DHHS Publication No. (SMA) 00-3374, Bethesda,
MD: Department of Health and Human Services, 2000.
Kegeles, S.M., Hays, R.B. and Coates, T.J. The Mpowerment Project: A community-level
HIV prevention intervention for young gay men. Amer. J. Publ. Hlth 86:
1129-1136, 1996.
Kelder, S.H., Perry, C.L. and Klepp, K.I. Community-wide youth exercise promotion:
Long-term outcomes of the Minnesota Heart Health Program and the Class of 1989
Study. J. School Hlth 63: 218-223, 1993.
Komro, K.A., Perry, C.L., Murray, D.M., Veblen-Mortenson, S., Williams, C.L.
and Anstine, P.S. Peer-planned social activities for preventing alcohol use
among young adolescents. J. School Hlth 66: 328-334, 1996.
Kraft, D.P. The prevention and treatment of alcohol problems on a college campus.
J. Alcohol Drug Educ. 34 (1): 37-51, 1988.
Luepker, R.V., Murray, D.M., Jacobs, D.R., Jr., Mittelmark, M., Bracht, N.,
Carlaw, R., Crow, R., Elmer, P., Finnegan, J. and Folsom, A.R. Community education
for cardiovascular disease prevention: Risk factor changes in the Minnesota
Heart Health Program. Amer. J. Publ. Hlth 84: 1383-1393, 1994.
Pentz, M.A., Dwyer, J.H., MacKinnon, D.P., Flay, B.R., Hansen, W.B., Wang,
E. and Johnson, C.A. Multicommunity trial for primary prevention of adolescent
drug abuse. JAMA 261: 3259-3266, 1989.
Perry, C.L., Kedler, S.H., Murray, D.M. and Klepp, K.I. Communitywide smoking
prevention: Long-term outcomes of the Minnesota Heart Health Program and the
Class of 1989 Study. Amer. J. Publ. Hlth 82: 1210-1216, 1992.
Perry, C.L., Williams, C.L., Veblen-Mortenson, S., Toomey, T.L., Komro, K.A.,
Anstine, P.S., McGovern, P.G., Finnegan, J.R., Forster, J.L., Wagenaar, A.C.
and Wolfson, M. Project Northland: Outcomes of a communitywide alcohol use prevention
program during early adolescence. Amer. J. Publ. Hlth 86: 956-965, 1996.
Puska, P., Tumiokhto, J., Salonen, J., Nissien, A., Virtamo, J., Bjorkquist,
S., Kloskela, K., Neittaa Naki, L., Takalo, T., Kottke, T., Maki, S.,
Sipila, P. and Varvikko, P.
Community Control of Cardiovascular Diseases: The North Karelia Project National
Public Health Laboratory of Finland, Geneva: World Health Organization Regional
Office for Europe, 1981.
Reynolds, R.I., Holder, H.D. and Gruenewald, P.J. Community prevention and
alcohol retail access. Addiction 92 (Suppl. No. 2): S261- S272, 1997.
Rigotti, N.A., DiFranza, J.R., Chang, Y., Tisdale, T., Kemp, B. and Singer,
D.E. The effect of enforcing tobacco-sales laws on adolescents, access to tobacco
and smoking behavior. New Eng. J. Med. 337: 1044-1051, 1997.
Saltz, R.F. and Stanghetta, P. A community-wide responsible beverage service
program in three communities: Early findings. Addiction 92 (Suppl. No.
2): S237-S249, 1997.
Santelli, J.S., Celentano, D.D., Rozsenich, C., Crump, A.D., Davis, M.V., Polacsek,
M., Augustyn, M., Rolf, F., McAlister, A.L. and Burwell, L. Interim outcomes
for a community-based program to prevent perinatal HIV transmission. AIDS Educ.
Prev. 7: 210-220, 1995.
Vartiainen, E., Paavola, M., McAlister, A. and Puska, P. Fifteen-year follow-up
of smoking prevention effects in the North Karelia Youth Project. Amer. J. Publ.
Hlth 88: 81-85, 1998.
Vartiainen, E., Puska, P., Jousilahti, P., Korhonen, H., Tuomilehto, J. And
Nissinen, A. Twenty-year trends in coronary risk factors in North Karelia and
in other areas of Finland. Int. J. Epidemiol. 23: 495-504, 1994.
Vincent, M.L., Clearie, A.F. and Schluchter, M.D. Reducing adolescent pregnancy
through school and community-based education. JAMA 257: 3382-3386, 1987.
Voas, R.B. Drinking and driving prevention in the community: Program planning
and implementation. Addiction 92 (Suppl. No. 2): S201-S219, 1997.
Wagenaar, A.C., Gehan, J.P., Jones-Webb, R., Toomey, T.L., Forster, J.L., Wolfson,
M. and Murray, D.M. Communities mobilizing for change on alcohol: Lessons and
results from a 15-community randomized trial. J. Comm. Psychol. 27: 315-326,
1999.
Wagenaar, A.C., Murray, D.M., Gehan, J.P., Wolfson, M., Forster, J., Toomey,
T.L., Perry, C.L. and Jones-Webb, R. Communities mobilizing for change on alcohol:
Outcomes from a randomized community trial. J. Stud. Alcohol 61: 85-94,
2000a.
Wagenaar, A.C., Murray, D.M. and Toomey, T.L. Communities mobilizing for change
on alcohol (CMCA): Effects of a randomized trial on arrests and traffic crashes.
Addiction 95: 209-217, 2000b.
Winkleby, M.A., Taylor, C.B., Jatulis, D. and Fortmann, S.P. The longterm effects
of a cardiovascular disease prevention trial: The Stanford Five-City Project.
Amer. J. Publ. Hlth 86: 1773-1779, 1996.
Yin, R., Kafterian, S., Ding, Y. and Jansen, M. Outcomes from CSAP's Community
Partnership Program: Findings from the National Cross Site Evaluation. Eval.
Prog. Plan. 20: 345-355, 1997.
Correspondence may be sent to Ralph W. Hingson
at the above address or via email at: rhingson@bu.edu.