The Latest Research in Prevention

1
The Latest Research in Prevention

• William DeJong, Ph.D.
• Professor, Boston University School of Public
  Health
• Director, Center for College Health and Safety,
• Education Development Center, Inc.

April 15, 2005
2
Two Key Evaluations

• Social Norms Marketing Research Project
• A randomized trial of campus-based media
  campaigns to communicate facts about student
  drinking norms
• A Matter of Degree Program Evaluation
• A quasi-experimental trial of campus-community
  coalitions focused on environmental prevention
  strategies

3
Social Norms MarketingResearch Project

• National Institute on Alcohol Abuse and
  Alcoholism
• and
• U.S. Department of Education
• R01 AA 12471

4
SNMRPCurrent Staff

• William DeJong, PhD, Principal Investigator
• Laura Gomberg, MSPH, Project Director
• Shari Kessel Schneider, MSPH, Research Director
• Emily Doerr, Research Assistant
• Melissa Murphy, Research Assistant

5
Social Norms

• People hold beliefs about the behavior that is
  expected of them in particular social contexts
  (injunctive norms)
• These beliefs are partly based on information
  about others behavior (descriptive norms)
• Perceptions of social norms are often a good
  predictor of behavior

6
Misperception of Student Drinking Norms
Students have exaggerated views of how much
other students drink
Students perceive greater normative expectations
to drink
7
Use Campus-Based Media to Correct the
Misperception
Social Norms Marketing
Decrease in perceived normative expectations to
drink
Decrease in alcohol consumption
8
 Social Norms Marketing Campaign   Posters Newspa
per Ads Emails Group Trainings Contests  
  Awareness of Message ? Acceptance of
Message ? More Accurate Perception of Peer
Drinking
  Reduction in Perceived Normative Expectations
to Drink
  Increase in Behavioral Intentions to Reduce
Alcohol Consumption
  Reduction in Alcohol- Related
Problems   DUI Unsafe Sex Assaults Date
Rape Property Damage Academic Problems Injuries
  Reduction in Alcohol Consumption  
9
(No Transcript)
10
Early Case StudiesImpact on Heavy Drinking Rates

• University of Arizona (3 years)
• 5 drinks in one sitting in the past 2 weeks
• 40 31
• Hobart and William Smith Colleges (3½ years)
• Drinking 5 in a row 3-plus times in the last 2
  weeks
• 41 28

11
A Call to Action Changing the Culture of
Drinking at U.S. Colleges National Institute on
Alcohol Abuse and Alcoholism April 2002
12
Assessment of Social Norms Marketing
Tier 1 Evidence of effectiveness among college
students
Tier 2 Evidence of success with general
populations that could be applied
to college environments
Tier 3 Evidence of logical and theoretical
promise, but require more
comprehensive evaluation
Tier 4 Evidence of ineffectiveness
13
Social Norms Marketing Research Project
Findings for Study 1 Cohort A 18 Colleges and
Universities
14
Study Parameters

• Conduct a true experimental design (with random
  assignment by institution)
• Ensure that the social norms marketing campaigns
  meet minimum quality standards
• Assess alcohol availability (outlet density)
• Assess other current programs and policies (and
  changes over time)
• Conduct random sample surveys at each institution
• Analyze other indicator data, where available

15
Study Timeline Cohort A
Year 2 (2000-01)
Year 1 (1999-00)
Year 3 (2001-02)
Year 4 (2002-03)
Year 5 (2003-05)
All Schools
Survey of College Alcohol Norms and Behavior
Contextual data collection
Experimental Schools (9)
Just the Facts Implementation
Control Schools (9)
Just the Facts Implementation
16
Alcohol Use at BaselineExperimental vs. Control
Schools
Experimental and Control Schools Are Equivalent
at Baseline
Source SCANB 2000, Cohort A
17
Number of Drinks Per WeekActual Vs. Perceived
(Experimental Schools)
Misperception of Student Drinking Norms
Source SCANB 2000, Cohort A Experimental Group
Schools
18
Number of Drinks When Party Actual Vs.
Perceived (Experimental Schools)
Misperception of Student Drinking Norms
Source SCANB 2000, Cohort A Experimental Group
Schools
19
Intervention Principles

• Campaign message and delivery venues tailored for
  each school
• A single key message for each campus, coupled
  with secondary messages
• Oversight by EDC staff
• Frequent monitoring phone calls and technical
  assistance
• Approval of marketing plans and campaign
  materials

20
SNM Message Guidelines

• All campaign messages must be targeted to the
  entire undergraduate population
• Campaign messages must
• Convey information about a behavior typical of a
  majority of students
• Correct an identified student misperception
• Be simple statements of fact (i.e., should not
  have a judgmental or moralistic tone)

21
SNM Message Guidelines (contd)

• Campaign materials must include
• Normative message
• Campaign logo
• Identification of the source of information
  (survey, date)
• Drink equivalency line (i.e., 1 drink 12 oz.
  beer 4 oz. wine 1 oz. liquor)
• Print advertisements should include an
  eye-catching photograph

22
Campaign Messages
Number of Schools
NAME students have 0-5 drinks when they
party. NAME students have 0-5 drinks a
week. NAME students have 0-4 drinks when they
party. NAME students have 0-4 drinks a week.
NAME students have 0-3 drinks when they
party. NAME students have 0-3 drinks a week.
1 0 5 1 1 1
Alternative wordings NAME students have
or fewer drinks when they party. NAME students
have or fewer drinks a week.
23
Media Plan

• Reach at least 50 of undergraduate students at
  least 2 times per month
• Deliver messages through
• Traditional print media venues (e.g., posters,
  flyers, newspaper ads)
• Other free and low-cost venues (e.g., electronic
  venues, presentations, newsletters)

24
Media Venues

• Newspaper Ads
• Radio and TV ads
• Flyers/Posters
• Billboards
• Table Tents
• Item Giveaways

• Talks/Presentations
• Student Orientation
• Reinforcement Activities
• Newsletters
• Press Releases
• Other

25
Training and Oversight

• In-person training
• 95-page guidebook
• Checkpoint schedule
• Weekly campaign updates and reminders via
  listserv
• Technical assistance calls
• Detailed progress reports each semester

26
Survey Response Rates
Response rates are comparable to those of other
national and school surveys of college alcohol use

• Year 1 (2000) 53.1 (Baseline)
• Year 2 (2001) 58.4
• Year 3 (2002) 57.7
• Year 4 (2003) 57.4
• 5,308 returned surveys
• Response rates at individual schools ranged from
  44.1 to 70.0

27
Campaign Exposure

• Over half of students (55.4) reported seeing JTF
  alcohol information during the current school
  year
• 35.8 reported this occurred a few times a month
  or more often
• The percent of students who reported seeing the
  JTF logo or hearing the JTF slogan during the
  current school year ranged from 25.5 to 88.0 at
  individual schools

28
Message Recall

• Overall, the percent of students correctly
  recalling their schools campaign message in
  2003 was 38.2 (ranging from 9.6 to 68.1 at
  individual schools)
• Of those reporting exposure to JTF, correct
  message recall was 62.0 (ranging from 20.4
  to 89.7 at individual schools)
• Defined as selecting exact message does not
  include underestimates
• Average of recall statistics, with all schools
  weighted equally

29
Data Analysis

• Data weighted to account for unequal response
  rates across schools
• Ensure that each school carries the same weight
  in the analysis
• Mean number of interviews per school for the
  entire study sample divided by the number of
  interviews obtained at a particular school
• Hierarchical linear regression to account for
  students clustered by school (multilevel model)
• Student level model
• School level model
• Cross level interaction

30
Perception Index Measures

• During the past 30 days, on how many occasions do
  you think the average or typical student at this
  school has used alcohol (beer, wine, liquor)?
• What do you think is the average number of drinks
  consumed by the average or typical student at
  this school in a week? for (a) male students and
  (b) female students.
• How many alcoholic drinks do you think students
  at this school have when they party? for (a)
  male students and (b) female students.

31
Perception Index Results
32
(No Transcript)
33
Experimental Condition x Time Interactions

• Greatest number of drinks on one occasion in past
  two weeks (n4,603)
• p .0018
• BAC for greatest number of drinks on one occasion
  in past two weeks (n3,877)
• p .0178
• Drinks consumed when students party (n4,622)
• p .01
• Drinks consumed per week (n4,793)
• p .05
• Number of occasions alcohol was consumed in the
  past 30 days (n4,595)
• p .94

34
Stay Tuned

• Study 2 Replication
• Cohort B 14 Colleges and Universities
• Integrated Analysis
• 32 Colleges and Universities
• Including Alcohol and Policy Environment
• (Richard Scribner and Colleagues, LSU)
• Hypothesis SNM campaigns will be less effective
  in alcohol-intensive environments

35
A Matter of DegreeProgram Evaluation
Elissa R. Weitzman and Colleagues Harvard School
of Public Health American Journal of Preventive
Medicine 2004

• Robert Wood Johnson Foundation

36
Environmental ManagementGrounded in the Public
Health Approach
Health Education Treatment
Individual factors Peer factors
Institutional factors Community factors Public
policy
Environmental Management
37
Contributing Factors in the Campus Environment

• Many students arrive at college believing that
  heavy, even dangerous drinking is the norm, which
  can be reinforced by campus life
• Students are bombarded by alcohol advertising,
  which glorifies the student drinking culture and
  promotes high-risk consumption

38
Contributing Factors in the Campus
Environment(contd)

• It is easy for college students to obtain free or
  inexpensive alcohol, either from acquaintances or
  local retail outlets
• Many students have a great deal of free,
  unstructured time with few social and
  recreational options that do not involve alcohol
  consumption

39
Contributing Factors in the Campus Environment
(contd)

• Campus rules and local and state laws are not
  consistently and vigorously enforced

40
Building a Prevention Infrastructure

• Environmental management
• Campus
• Community
• State/Federal Policy
• Prevention infrastructure
• Permanent campus task forces
• Campus/community coalitions
• Regional and state initiatives

41
Environmental Strategies

• Create an environment that supports
  health-promoting norms
• Restrict marketing and promotion of alcoholic
  beverages
• Limit alcohol availability and access
• Offer social, recreational, public service, and
  other extracurricular options
• Develop and enforce campus policies and local,
  state, and federal laws

42
A Matter of Degree

• A National Effort to Reduce High-Risk Drinking
  Among College Students
• National Program Office American Medical
  Association
• 8.6 million, 7-year program
• Designed to foster campus-community collaboration

43
Campus-Community Partnerships

• Florida State University Tallahassee, FL
• Georgia Institute of Technology Atlanta, GA
• Lehigh University Bethlehem, PA
• Louisiana State University Baton Rouge, LA
• University of Colorado Boulder, C O
• University of Delaware Newark, DE
• University of Iowa Iowa City, IA
• University of Nebraska at Lincoln Lincoln, NE
• University of Vermont Burlington, VT
• University of Wisconsin Madison, WI

44
AMOD EvaluationHarvard School of Public Health

• Case studies, with repeated cross-sectional
  student surveys (1997-2001)
• Alcohol consumption
• Alcohol-related harms
• Secondhand effects of others alcohol use
• Quasi-experimental study of program effects
• 10 AMOD sites
• 32 comparison sites from College Alcohol Study

45
Hypotheses

• Coalitions will change alcohol environment on
  campus and in the community
• 10 AMOD sites will see more improvement than the
  32 comparison sites
• Degree of improvement will depend on the extent
  to which environmental strategies are implemented
• Compare high (n5) and low (n5) AMOD sites

46

• Intervention High Low Selected Examples
• Availability 26 5 Keg registration
• Mandatory RBS training
• Over-service enforcement
• Legal Sanction 21 4 Increasing penalties and
• sanctioning policies
• Campus-community police
• collaboration
• Physical Context 8 2 Substance-free housing
• Enforcement of bar capacity

47

• Intervention High Low Selected Examples
• Advertising 7 4 Advertising bans
• Promotion Ban on alcohol-related
• items in bookstore
• Key Influencers 16 8 Parental notification
• Peer intervention teams
• Outreach to faculty
• Sociocultural 79 23 Alcohol-free programming
• Context Faculty senate resolution

48
Summary of Findings

• No statistically significant change found for the
  10 AMOD sites overall
• Statistically significant change found for the 5
  programs that most closely implemented the AMOD
  model of environmental change
• Reduced alcohol consumption
• Fewer drinking-related harms
• Fewer secondhand effects from others drinking

49
Results for Key Measures

• Measures (1997?2001) High Low
  Control
• Frequent heavy drinking 34.6 30.7
  28.1
• 
  ? ? ?
• 
  34.4 31.3 31.1
• 5 alcohol-related problems 31.6 25.0
  25.2
• 
  ? ? ?
• 
  26.0 24.7 24.7

50
Results for Key Measures

• Measures (1997?2001) High Low
  Control
• Drove after 5 drinks 19.0
  16.6 15.3
• 
  ? ? ?
• 
  15.9 12.6 16.6
• 3 secondhand effects 60.8
  56.0 56.9
• 
  ? ? ?
• 
  54.5 52.1 52.6

51
www.hsph.harvard.edu/cas

• Weitzman ER, Nelson TF, Lee H, Wechsler H.
  Reducing drinking and related harms in college
  Evaluation of the A Matter of Degree program.
  Am J Prev Medicine, 21(3) 187-196, 2004.
• Nelson TF, Weitzman ER, Wechsler H. The effect of
  a campus-community environmental alcohol
  prevention initiative on student drinking and
  driving. Results from the "A Matter of Degree"
  program evaluation. Traffic Injury Prevention, In
  press.
• Weitzman ER, Nelson TF. College student binge
  drinking and the 'prevention paradox'
  Implications for prevention and harms reduction,
  J Drug Education, 34(3)247-266, 2004.

52
Another Role for Social Norms Marketing

• Setting the Stage for
• Environmental Management Interventions

53
Survey of College Alcohol Norms and Behavior
(SCANB) 2001-2002

• 32 colleges and universities across the U.S.
• 300 students per campus, less exclusions
• 2 or more pieces of undeliverable mail
• Student no longer enrolled
• Total sample 9,270
• Mailed survey, with extensive follow-up
• Response rate 56.2
• Range 45.3 to 71.4
• Total number of respondents 5,210

54
Survey Questions(32 Colleges and Universities)

• To what extent do you support or oppose the
  following possible policies or procedures?
• To what extent do you think other students at
  this school support or oppose the following
  possible policies or procedures?

55
Perceived Support Strict Enforcement( of
Respondents)

•  

56
Perceived vs. Actual Support Strict Enforcement
( of Respondents)
57
Perceived vs. Actual Support Campus Management
( of Respondents)
58
Perceived vs. Actual Support Alcohol
Availability ( of Respondents)
59
Student Majorities (No. of Colleges)

• Use stricter disciplinary sanctions for students
  who engage in alcohol-related violence 32
• Conduct undercover operations at bars,
  restaurants, and liquor stores to increase
  compliance with underage laws 12
• Eliminate low-price bar and liquor store
  promotions targeted to college students 8
• Require more early morning and Friday classes to
  discourage alcohol use during the week 0

60
Conclusion

• A majority of students support environmental
  management policies, especially stricter
  enforcement
• College administrators and community officials
  should not assume a lack of student support for
  policy change
• Students can be effective partners in
  community-based prevention focused on
  environmental management

61
(No Transcript)