April 23-24, 1998 • Ramada Inn • Bethesda, Maryland

As the Working Group meeting drew to a close, NIAAA asked the members to sketch the rudiments of four potential models for intervening with risk drinking during pregnancy. Three were to be based on the target groups specified in the IOM report–universal, selective, and indicated. The fourth model was to cut across all three approaches. Time did not permit the full development of any one model; however, the outlines that follow suggest the directions that the Working Group members were following as they began the process of conceptualizing model research intervention programs.

Universal Model

The group proposed two separate models. Group members agreed that physicians do not provide adequate guidance for alcohol use by women of childbearing age. Although time constraints during office hours may account for this failure to some extent, a more compelling reason may be that physicians do not recognize the importance of providing such information and are unaware of the implications of alcohol use during pregnancy for the possible development of FAS, ARBD, and ARND.

The goal of the first model is to determine whether providing physicians with specific information on the levels of alcohol use among their patients will result in better counseling during office visits. Physicians will receive one of three levels of feedback to determine if there is a threshold of information that will result in a change in their clinical approach to women of childbearing age.

A group of medical practices will be selected and randomly assigned to experimental or control conditions. In the experimental group, all women of childbearing age will undergo a series of three assessments of their knowledge, attitudes and alcohol consumption practices during pregnancy subsequent to their regular office visits. Successive assessments will occur over a six-month period and will include increasing levels of detail. At the conclusion of the data collection phase, physicians will attend training that will describe the implications of alcohol use for FAS and provide them with a summary of what the screenings revealed about the aggregate drinking practices of their patients. Control group patients will receive the three assessments, but feedback will not be reported to their physicians.

As a result of this intervention, the workgroup hypothesized that the practices of experimental group physicians would be altered, while those of the control group would remain the same. Control group physicians, for example, would advise subsequent patients who use alcohol to employ realistic contraception plans.

The workgroup further hypothesized that the degree of practice change will be directly related to the level of detail about drinking status that is conveyed to physicians. There may not be a linear relationship, but rather a threshold beyond which knowledge and intentions change as a result of the specificity of the feedback. Physician training will focus only on information exchange; it will not include instruction in or encouragement to provide counseling to patients.

The second model proposed by the universal group focused on youth. There is evidence that pre-adolescents and adolescents are not receiving adequate information about the teratogenic properties of alcohol use during pregnancy. The goal of this project is to identify and recommend key biological concepts pertaining to alcohol’s toxic effects that are appropriate for youth of varying ages and should be incorporated in educational materials.

Focus groups will be the primary vehicle for collecting data/information and will occur in two stages during the project. Each stage will include four different age groups. In Stage I, currently existing materials (e.g., curricula, posters) will be presented to each age group and their reactions to them will be measured (e.g., appeal of materials, belief that materials were accurate). Based on the results from Stage I, new materials will be produced that are age specific in terms of cognitive complexity, developmental appropriateness, and appeal. These materials will be presented in Stage II, following the procedures devised for Stage I. An important objective of this model research project is to identify the age/developmental stage at which specific topics can be introduced and the level of detail that can be comprehended.

Selective Model

The selective workgroup’s goal was to describe the parameters of a cost-efficient and efficacious model for use in a variety of managed care settings (e.g., public health clinics, HMOs, planned parenthood, women’s health clinics, public schools) that would demonstrate positive outcomes and cost-savings sufficient for program sustainability. Women who are pregnant or likely to become pregnant and whose continued drinking during pregnancy places their fetus at risk, rather than women who are dependent on alcohol, will be the primary target of the intervention.

Women will be randomly assigned to several interventions and a control group (that will receive standard care). Three proven interventions–brief intervention, motivational interviewing, and an unspecified but more intense intervention–will be compared against the standard of care. Elements of special interest during the study will be the impact of varying intensities of treatment and the impact of the provider in changing behavior. Outcomes will include rate of pregnancy/contraceptive use; birth outcome; outcomes for children two years after birth; and endurance of abstention/reduction in problematic use of alcohol for women.

Expertise needed to implement the proposed study includes a multi-disciplinary team consisting of a physician, substance abuse specialist, developmental expert, and an experienced principal investigator to design, lead, monitor, and evaluate the program. A health care provider comfortable in delivering services in a managed care setting will also be required, as well as an economist to determine cost-effectiveness, an advocate for the program within the managed care entity, and an advocate for the female patients involved in the study. Sensitivity to cultural issues and literacy levels will be prerequisites.

The workgroup hypothesized that the model project will:

• Demonstrate that interventions can be conducted in managed care settings and sustained by managed care entities;
   
• Distinguish between differing levels of intervention in terms of efficacy;
   
• Prove that interventions produce cost-offsets (e.g., costs associated with the provision of expensive mental health and special education services are avoided by reducing the number of alcohol-affected births) and that it is less expensive to mitigate than treat problems;
   
• Result in improved birth outcomes and enhanced contraception for those choosing that approach; and
   
• Improve women’s overall health and their substance abuse and related problems.

Indicated Model

The indicated prevention group chose to design a Request for Applications (RFA) because they did not believe that a single study could provide the kind of information needed to improve efforts at guiding women who drink heavily to treatment. The purpose of the RFA is to prevent FAS, ARBD, ARND, and risk drinking during pregnancy. Potential subjects include women who are offspring of mothers who have borne children with FAS or who abused alcohol and drugs, and women whose partners drink heavily or abuse alcohol. Women will be recruited from the criminal justice system, emergency rooms, shelters, substance abuse treatment programs, and public and private medical practices, including obstetric and gynecologic practices.

Studies included under the RFA will identify the sociocultural and intrapsychic context of heavy drinking, clarify the role of co-morbid medical and psychiatric conditions, and develop innovative and effective treatment approaches, including family-centered treatment and community support. Studies will also attempt to identify the occupational and personal attributes of both paraprofessional and professional providers that affect positive outcomes, components of a stepped approach to treatment, and treatment outcomes.

The workgroup hypothesized that results from the studies conducted under the RFA will:

• Lead to the identification of effective and appropriate treatment interventions;
   
• Generate further testable hypotheses;
   
• Inform the development of best practice models for obstetric, medical, and specialty settings; and
   
• Influence public policy.
   

A multidisciplinary approach and input from the target population will be essential to carry out the studies envisioned in the RFA.

Cross-Cutting Model

The goal of the cross-cutting model is to reduce prenatal drinking and promote responsible alcohol use, including no use. A key objective of this effort is to adjust community-wide expectancies about alcohol so they more accurately reflect its impact on health and behavior. The multiple targets for this initiative will include males and females across the age spectrum from elementary school to post-college age. High-risk men and women also will be included.

Two communities matched on most variables will serve as the experimental and control sites. The experimental group will receive appropriate universal warnings depending on age and life stage (e.g., facts on alcohol in elementary school, announcements about expectancies related to alcohol targeted at high school/college students); the control community will not receive any universal messages. As a result of the messages conveyed to various segments in the experimental community, the workgroup hypothesized that knowledge, attitudes, beliefs, expectancies, and drinking will change.

In order to implement this extremely complex, cross-cutting model, valid measures of alcohol consumption will have to be developed and prognostic indicators of intention to continue use will have to be identified. A multidisciplinary team comprised of educators, economists, statisticians, and community studies experts, as well as alcohol researchers and treatment providers, will also need to be assembled in order to carry out the numerous tasks required by a cross-cutting model.

Appendix B. References for Meta-Analysis of Validated Cessation in Randomized Controlled Trials

1. Sexton M, Hebel JR, 1984. A clinical trial of change in maternal smoking and its effect on birth weight. JAMA 251:911-915.
    
2. Windsor RA, Cutter G, Morris J, Reese J, Manzella B, Bartlett EE, Samuelson C, Spanos D, 1985. The effectiveness of smoking cessation methods for smokers in public health maternity clinics: A randomized trial. Am J Public Health 75:1389-1392.
    
3. Ershoff DH, Mullen PD, Quinn VP, 1989. A randomized trial of a serialized self-help smoking cessation program for pregnant women in an HMO. Am J Public Health 79:182-187.
    
4. Burling TA, Bigelow GE, Robinson JC, Mead AM, 1991. Smoking during pregnancy: Reduction via objective assessment and directive advice. Behav Ther 22:31-40.
    
5. Hjalmarson AIM, Hahn L, Svanberg B, 1991. Stopping smoking in pregnancy: Effect of a self-help manual in controlled trial. Br J Obstet Gynecol 98:260-264.
    
6. Price JH, Desmond SM, Roberts SM, Krol RA, Losh DP, Snyder SF, 1991. Comparison of three antismoking interventions among pregnant women in an urban setting: A randomized trial. Psychol Rep 68:595-604.
    
7. O’Connor AM, Davies BL, Dulberg CS, Buhler PL, Nadon C, McBride BH, Benzie RJ, 1992. Effectiveness of a pregnancy smoking cessation program. J Obstet Gynecol Neonatal Nurs 21:385-392.
    
8. Petersen L, Handel J, Kotch J, Podedworny T, Rosen A, 1992. Smoking reduction during pregnancy by a program of self-help and clinical support. Obstet Gynecol 79:924-930.
    
9. Windsor RA, Li CQ, Lowe JB, Perkins LL, Ershoff D, Glynn T, 1993. Health education for pregnant smokers: Its behavioral impact and cost benefit. Am J Public Health 83:201-206.
    
10. Secker-Walker RH, Solomon LJ, Flynn BS, Skelly JM, LePage SS, Goodwin GD, Mead PB, 1994. Individualized smoking cessation counseling during prenatal and early postnatal care. Am J Obstet Gynecol 171(5):1347-1355.
     
11. Kendrick JS, Zahniser SC, Miller N, Salas N, Stine J, Gargiullo PM, Floyd L, Spierto FW, Sexton M, Metzger RW, Stockbauer JW, Hannon, Dalmat E, 1995. Integrating smoking cessation into routine public prenatal care: The Smoking Cessation in Pregnancy Project. Am J Public Health 85(2):217-222.
     
12. Hartmann KE, Thorp JM, Pahel-Short L, Koch MA, 1996. A randomized controlled trial of smoking cessation intervention in pregnancy in an academic clinic. Obstet Gynecol 87(4):621-626.
     
13. Walsh RA, Redman S, Brinsmead MW, Byrne JM, Melmeth A, 1997. A smoking cessation program at a public antenatal clinic. Am J Public Health 87(7): 1201-1204.
     
14. Gielen AC, Windsor R, Faden RR, O’Campo P, Davis M, in press. Evaluation of a smoking cessation intervention for pregnant women in an urban clinic. Health Education Research:Theory and Practice.