CHARACTERISTICS OF DRIVERS ARRESTED FOR DRIVING WHILE INTOXICATED. James Fell1, A. Scott Tippetts1, Robert Voas1, Susan Ferguson2, Allan Williams2, 1Pacific Institute for Research and Evaluation, Calverton, MD United States; 2Insurance Institute for Highway Safety, Arlington, VA United States
Past studies and surveys have indicated that about 11% of the drinking-and-driving public can be classified as problem drinkers. Problem drinkers were estimated at 27% of last year´s drinking drivers, accounting for about 46% of all drinking-driving trips reported in 2001 in that survey. A recent study of drivers killed in traffic crashes reported that anywhere from 21% to 61% of these fatally injured drivers who had very high blood alcohol concentrations (BACs) of .15 or more were considered problem drinkers depending on the criterion used. Only 1% to 7% of fatally injured drivers with BACs=.00 were considered problem drinkers in that sample of drivers.A study of more than 1,000 drivers arrested for driving while intoxicated (DWI) or driving under the influence (DUI) was conducted in Shelby County, Tennessee, and Montgomery County, Maryland, from 2001 to 2003. Drivers were interviewed by police immediately after their arrest in order to gather information on their alcohol consumption, drinking-and-driving frequency, and perception of enforcement. In addition, police provided data on prior DWI offenses for the arrested offenders, a BAC test result (if taken), their license status, and basic demographic information.Analysis of the resulting data permitted the classification of offenders into problem and nonproblem drinkers, and the relationship of this designation to BAC at the time of arrest and the 5+/4+ drinks in a session criterion for heavy episodic drinking. This classification also was stratified based on the source of arrest (crash investigation or regular patrol), prior driving record, age, gender, crash location, date, and time of day.The implications of the results for current DUI laws, particularly those related to assessing more severe penalties for high BAC offenders and requiring assessment and treatment for all multiple offenders is discussed. Also discussed is the relevance of the results to the allocation of criminal justice resources between efforts aimed at general deterrence of all drivers who drink compared to the resources that should be devoted to the small group of problem drinkers who are overrepresented in alcohol-related crashes.
SERVICE ECOLOGY OF JUVENILE OFFENDERS: INVESTIGATION OF THE NEED, RECEIPT, AND OUTCOMES OF COMMUNITY-BASED SERVICES FOR CHRONIC DELINQUENTS. David Lichtenstein1, J. Mark Eddy2, 1University of Oregon, Eugene, OR United States; 2Oregon Social Learning Center, Eugene, OR United States
Although mental health problems are increasingly common amongst juvenile offenders, mental health service use tends to be low. Little is known as to how factors such as insufficient access, the absence of need, or a lack of motivation contribute to this service-use deficit. Further, it is unclear whether service use predicts improved outcomes for juvenile offenders. In this study, we address three questions: First, to what extent do juvenile delinquents want, or need, mental health (and other) services? Second, to what extent are they receiving those services? Finally, do mental health services improve the long-term outcomes for those who receive them? Given the negative long-term outcomes associated with juvenile offending, it is important to identify promising (and inefficient) modes of service delivery.
This ten year study involved 79 adolescent boys who were referred by the juvenile court system for community placement. At baseline the participants averaged 14 previous criminal referrals and more than four previous felonies. For four years after baseline, participants were asked whether and how often they received 11 types of service, including therapy, medical care, and parenting assistance, and whether they desired any of these services. Outcome measures collected at termination included lifetime criminal charges, Brief Symptom Inventory (BSI) scores, and qualitative groupings of participants generated by study personnel based on global impressions of outcomes.
Results indicate that 95% of the sample desired services, 90% used at least one service type, and the average participant utilized 3 different types of service. Service utilization frequency was generally low; for instance, participants averaged 1 session per month of individual therapy, and 33% received no therapy. Far more participants desired school, housing, or employment assistance than actually received it. At termination, participants had a high incidence of substance abuse and moderately high incidence of depression and anxiety. The impact of services on outcomes appeared minimal. No significant group differences in service use were found. Although individual therapy was found to be significantly negatively correlated with lifetime criminal charges (p<.01), other services did not appear to be significantly related to lifetime charges or mental health (BSI score) at termination. Insignificant findings may be due to a “floor effect” (overall low levels of use), inadequacy, or inappropriateness of services. Future analyses will further investigate the impact of service use on developmental outcomes using mediational models. Implications of the findings for the juvenile justice and public health systems will be discussed.
EFFECTS OF A SCHOOL-BASED DRUG ABUSE PREVENTION PROGRAM FOR ADOLESCENTS ON HIV RISK BEHAVIORS AMONG YOUNG ADULTS. Kenneth Griffin1, Gilbert Botvin1, Nichols Tracy1, 1Cornell University Medical College, New York, NY United States
Background: Early onset of substance use has been found to be associated with later risky sexual behaviors. Because competence enhancement approaches to drug prevention target etiologic factors common to both drug abuse and HIV and teach skills that have a broad application, such programs may reduce later HIV risk. This study examined long-term follow-up data from a large randomized school-based drug prevention trial to determine effects on HIV risk behavior among young adults.
Methods: Self-report data were collected prior to the intervention from students in the 7th grade (in 1985) and were matched to follow-up surveys among 2,042 young adults (in 1998). Participants in the intervention condition received a 30-session drug prevention program (Life Skills Training) in 7th through 9th grades. The follow-up sample was 47% male, 91% were white, almost half were college graduates, 39% were married or cohabitating, and the median age was 24.6 years.
Results: As young adults, participants were considered at high risk for HIV infection if they reported having multiple sex partners, having intercourse when drunk or very high, and having used substances in the past month. Findings indicated that 10.3% of the intervention group engaged in these behaviors compared to 13.6% of the control group, and this difference was statistically significant (chi-square(1)=4.856, p<.033). A logistic regression analyses revealed that the intervention had a protective effect on HIV risk (OR= .72, 95% CI = .53, .97) after controlling for relevant covariates.
Conclusions: The behavioral effects of competence-enhancement drug abuse prevention programs can extend to risk behaviors that put one at high risk for HIV infection.
ON THE PRIMARY PREVENTION OF PARTNER VIOLENCE: DATING VIOLENCE PROGRAMS AND BEYOND. Daniel Whitaker1, 1Centers for Disease Control and Prevention (CDC), Atlanta, GA United States
Most prevention work regarding intimate partner violence (IPV) focuses on secondary and tertiary prevention, either working with perpetrators to prevent re-assault, or with victims on safety planning and recovery. Reviews regarding the impact of those intervention strategies on rates of victimization are not encouraging (Babcock, Green, & Robie, 2004; Wathen & McMillan, 2003). Primary prevention efforts for partner violence are needed but have received less attention. The goal of this paper is to review current primary prevention efforts for partner violence, and then to suggest new directions for the field.
We conducted a systematic review of primary prevention programs for partner violence. We found 11 studies published since 1990, all of which focused on dating violence prevention for youth. Two reviewers coded each study for sample, design, intervention, and outcome variables. There was little variance among the studies with regard to the intervention approach; all 11 studies utilized interventions with a specific dating violence curriculum. Ten of the 11 were universal offered to students in middle- or high-schools (the other targeted at-risk teens and was conducted in a community based setting). Study designs were generally strong with most using either experimental or quasi-experimental designs, but measurement was weak in that only 5 of 11 studies measured behavior (all via self-report) and follow-up periods were generally short. Of the five studies that measured behavior, only three reported data on behavioral outcomes, with all three showing a positive impact and two showing a longer term impact of one year or more. A formal meta-analysis was not conducted because of the limited number of studies.
More work is clearly needed in the area of primary prevention of partner violence, both to improve school-based dating violence prevention programs, and to expand prevention efforts beyond those types of programs. The presentation will conclude with the recommendation that the development of new primary prevention strategies for partner violence draw from the child and adolescent development literature and literature regarding the development of youth violence and other problem behaviors. Specifically, child development research offers guidance for strategies to intervene prior to middle school (i.e., sources, settings), and the problem behavior literature may point researchers to focus on partner violence as another type of teen problem behavior and intervene by promoting prosocial behaviors prior to the initiation of the specific problem behavior.
PRELIMINARY OUTCOMES OF BREAK THE CYCLE´S INTIMATE PARTNER VIOLENCE PREVENTION PROGRAM FOR LATINO YOUTH. Lisa Jaycox1, Daniel McCaffrey2, Elizabeth Eiseman3, Jessica Aronoff4, Gene Shelley5, Beverly W. Ocampo6, 1Rand Corporation, Arlington, VA United States; 2RAND Corporation, Pittsburgh, PA United States; 3Rand Corpotation, Arlington, VA United States; 4Break the Cycle, Los Angeles, CA United States; 5Centers for Disease Control and Prevention (CDC), Atlanta, GA United States; 6Rand Corporation, Santa Monica, CA United States
Intimate partner violence is a serious problem among adolescents. This project evaluates an innovative prevention and early intervention program for 9th graders that focuses on legal rights and responsibilities. The program was created and is implemented by Break the Cycle, a non-profit organization whose mission is to end domestic violence by working proactively with youth. Taught by attorneys, Break the Cycle´s curriculum, Ending Violence, focuses on legal aspects of domestic violence. The collaborative effort between the RAND Corporation, Break the Cycle, the CDC, and selected schools in the Los Angeles Unified School District, evaluates Ending Violence´s impact on Latino student attitudes, knowledge, and help-seeking in three successive cohorts of 9th graders. The study employs a randomized experimental design: tracks within year-round schools are randomly assigned to a wait-list control group or to the Ending Violence prevention curriculum; and 2-3 classrooms per track are randomly selected to participate. Preliminary data from the first cohort of data show that students receiving the Ending Violence curriculum have improved knowledge about dating violence after the program, controlling for several possible confounding variables. Moreover, this gain in knowledge is maintained 6 months later. In addition to changes in knowledge, we will present program outcomes related to attitudes about dating violence and help-seeking, and interpret our findings in light of existing prevention programs to improve them in the future.
SCREENING AND INTERVENTION FOR PARTNER VIOLENCE IN RURAL HEATH CARE CLINICS IN SOUTH CAROLINA. Daniel Whitaker1, Ann Coker2, Vicki Flerx3, Page Hall Smith4, 1Centers for Disease Control and Prevention (CDC), Atlanta, GA United States; 2University of Texas Health Science Center at Houston, Houston, TX United States; 3University of South Carolina, Columbia, SC United States; 4University of North Carolina at Greensboro, Greensboro, NC United States
The question of whether health care providers should screen women for intimate partner violence has received much recent attention. Screening for partner violence could be an important point of intervention; two-thirds of women had sought medical care during the year prior to their murder (Sharps, 2001), and female murder victims are often not properly identified or referred as victims of partner violence in emergency room visits prior to their murder. Several national organizations have called for widespread domestic violence screening in health care clinics, but recently the U. S. Preventive Services Task Force failed to recommend IPV screening based on a lack of evidence for benefit or harm.
This study attempts to address two questions regarding screening to understand and maximize its benefits: what proportion of women will screen positive in primary health care clinics for different types of abuse, and what efficacious interventions can follow screening? In the current study, universal screening has been implemented in several primary health care clinics in rural South Carolina. All women are offered screening for partner violence annually, and are screened for both physical assault (one or more acts of physical or sexual assault) and battering (enduring abuse characterized by threat, fear, and control). Following screening, all women who screened positive for IPV are offered an intervention. Two interventions were developed and assigned to clinics in a 2 x 2 design. In the on-site specialist intervention, an advocate from the local domestic violence agency is present at the clinic to meet immediately with women who screen positive. In the empowerment intervention, clinic-based social workers and psychologist were trained to deliver a seven-session intervention designed to increase women´s decision-making and problem solving regarding their abuse. To date 5700 women were offered screening, and 3329 (73.4%) agreed to be screened. Of the women screened, 24.42% screened positive for some form of partner violence (19.7% physical abuse, 23.3% psychologically battered, and 17.1% both). Surprisingly, only 40% of women experiencing violence believed it to be a problem. Women experiencing physical violence vs. battering, and women experiencing violence in current vs. past relationships were more likely to report it to be a problem. Evaluation of intervention participation and impact is still on-going, but current analyses indicate that women who believed IPV to be a problem were more likely to report seeking help from one of the interventions.
SCHOOL-BASED HEALTH CENTERS:EFFECTIVENESS OF PREVENTION PROGRAMMING. Brenda Fenton1, Constance Catrone2, 1MATRIX Public Health, New Haven, CT United States; 2School-Based Health Centers, Bridgeport, CT United States
As providers of culturally competent care in a system serving largely inner-city high-risk minority youth, the Bridgeport school-based health centers provide an `appropriate´ access point through which to deliver effective prevention practices in the schools. The current universal and indicated evidence-based prevention interventions in city elementary (Promoting Alternative Thinking Strategies or PATHS) and high schools (Reconnecting Youth or RY) have significantly reduced high-risk behaviors (e.g., aggression and school violations) and improved positive behaviors (e.g., emotional and social competence). Students who received the PATHS program significantly improved their levels of social and emotional competence, concentration and attention, aggressive behavior, and behavioral regulation. Grade and gender variations in student change in these parameters were observed. Negative changes in the elementary school cohort that was not exposed to PATHS in kindergarten suggest that the effectiveness of PATHS in later grades may be tied to its delivery in the first year of school perhaps due to socialization issues. In Bridgeport´s high schools, school record data show significant reductions in the number of school violations (in-school suspensions and unexcused tardiness) in RY students, including those with the highest number of violations prior to participation in RY. Indicators of school performance remained steady in the pre-post period which is a positive finding, given the expected decline of unassisted high-risk students over time. Students report the strong positive experience of support and respect that they have in RY creates a “safe place” for them, a special bond to their teacher, and an increase in self-efficacy. Students report increases in the level of respect and restraint/anger management they show in their interactions with both teachers and students. These positive changes in comportment were supported by principal report. Based upon evaluation feedback regarding the need to increase the support of RY in the school environment, two items were recently created by the RY team to increase teacher understanding and backing, including rewarding “small changes” observed in individual students in their classes: 1) a student letter to his/her teachers announcing that he/she is working on making positive changes and 2) a teacher RY fact-sheet. These effectiveness trials involving largely minority youth expand the knowledge base of two evidence-based prevention practices, including the effect of school context and its amelioration. School-based prevention/interventions contribute to community well-being through the development of prosocial skills, reduction in reliance on violence, and prevention of youth mental health disorders.
POSTERS: EFFICACY TRIALS
EFFICACY OF AN ALCOHOL MISUSE PREVENTION PROGRAM FOR U.S. MARINE CORPS AVIATION PERSONNEL. Cynthia Simon-Arndt1, Suzanne Hurtado1, Robyn Highfill-Mcroy1, 1Naval Health Research Center, San Diego, CA United States
The purpose of this study was to develop a two-part alcohol misuse prevention program for Marine Corps aviation personnel, and to evaluate its impact on reducing drinking levels and alcohol-related negative consequences. The intervention was composed of a Web-based self-assessment of high-risk drinking behavior and a facilitated small-group training program. An evaluation was conducted using a repeated-measures design with two experimental groups and a control group. One experimental group received both components of the two-part program, while the other received only the Web-based self-assessment. A total of 455 subjects participated in the evaluation (control, n=307; computer-only, n=96; full intervention, n=52). Participants in the group receiving the full intervention reported drinking fewer beers per month (p<.05) and decreased productivity loss events at postintervention compared with the control group (p<.05). Additionally, they scored higher on readiness to change (p<.05) and tended to drink fewer total drinks per month at postintervention compared with the control group (p=.09). Results on alcohol-related knowledge, Alcohol Use Disorders Identification Test score, and other alcohol-related problems and measures of drinking indicated that, in general, the two levels of intervention tested in this study yielded drinking levels and alcohol-related incident rates that did not differ from those of the control group. While the participants who received both the small-group training and the Web-based alcohol misuse prevention program showed improvement in some areas, the absence of strong program effects reiterates the need for a variety of strategies to counter the social and environmental factors that influence risky alcohol use behavior among Marines.
TRANSTHEORETICAL MODEL VARIABLES AS OUTCOME MEASURES IN A SMOKING INTERVENTION STUDY. Anja Schumann1, Judith Stein2, Jodie Ullman3, 1University Greifswald, Institute of Epidemiology and Social Medicine, Greifswald, Germany Germany; 2University of California, Los Angeles, Los Angeles, CA United States; 3California State University, San Bernardino, San Bernardino, CA United States
This study reports the outcome of a smoking cessation intervention based on the Transtheoretical Model (TTM). The TTM is a health behavior change model. It incorporates the stages of change (different phases of readiness to quit smoking), the processes of change (activities that people use to progress through the stages), decisional balance (pros and cons of changing a behavior), and self-efficacy (confidence to refrain from the problem behavior). To evaluate the efficacy of a TTM-based intervention, traditional or alternative outcome measures may be employed. Traditional outcome measures are behavioral measures such as short-term or prolonged abstinence rates. Alternative outcome measures are changes in dependent TTM-variables. Here, we focus on analyzing alternative outcome measures.
The smoking intervention study was a randomized controlled trial with a TTM-based expert-system intervention condition (a special software producing individualized feedback letters, and additional stage-tailored self-help booklets) and an assessment-only control condition. Subjects were recruited from a representative general population survey in northern Germany (“Study of Health in Pomerania”, SHIP). Baseline (n = 507 smokers), 6-month (n = 411), and 12-month (n = 375) follow-up data were collected. A longitudinal predictive model for the TTM outcome variables was analyzed using structural equation modeling (SEM).
We applied a structured mean multiple group analysis to compare differences in the TTM outcome variables over time between the intervention and control group, controlling for preexisting differences. The increase in stage of change progress was significantly higher for the intervention than the control group. Although subjects of the intervention group showed significantly lower readiness to quit smoking at baseline, they had a significantly higher increase in readiness from baseline to 6-month. In addition, we examined whether the means of the TTM outcome variables changed over time, regardless of group membership. We found significant increases in all TTM outcome variables over time for both study groups.
We conclude that the intervention led to more stage of change progress. However, changes in other TTM outcome variables could not be established. Both study groups showed improvements in the TTM-variables over time. This might reflect historical or societal changes in attitudes towards smoking or selective attrition over the course of the intervention study. It might also be due to employing comprehensive smoking assessments in both study groups. The assessments may have caused smokers, including those of the control group, to be very attentive to the topic of smoking, thus fostering cognitive, affective, and behavioral changes with regard to smoking.