Obese women who regained weight after successful weight reduction (relapsers, n = 44); formerly obese, average-weight women who maintained weight loss (maintainers, n = 30); and women who had always remained at the same average, nonobese weight (control subjects, n = 34) were interviewed. Most maintainers (90%) and control subjects (82%) exercised regularly, were conscious of their behaviors, used available social support (70% and 80%, respectively), confronted problems directly (95% and 60%, respectively), and used personally developed strategies to help themselves. Few relapsers exercised (34%), most ate unconsciously in response to emotions (70%), few used available social support (38%), and few confronted problems directly (10%). These findings suggest the advisability of development and prospective evaluation of individualized treatment programs designed to enhance exercise, coping skills, and social support.

IntrodudionA substantial number of research reports evaluating programs designed to prevent adolescent smoking have been published in the last 2 decades. A very important set of questions deals with (1) how successful these programs have been in general, (2) whether programs of different orientation show different levels of success, and (3) what variables, in addition to program orientation, may account for different levels of success. Nonquantitative reviews'" of this literature devote little effort to answering these three questions; rather, they focus on pervasive methodological issues encountered by much ofthe published evaluation research in this area. None of the four reviews employed meta-analytic techniques for assessing program outcomes. The methodological critiques provided by these reviews'-4 centered on (1) the use of appropriate comparison groups; (2) the use of appropriate pretests, posttests, and follow-up assessments; (3) the control of research attrition; (4) the validity of dependent variable assessment; and (5) entations. These authors" note that such reviews should (1) comprehensively cover a defined set of evaluation studies spanning a specified time period; (2) systematically screen out studies with weaker research methodology; (3) systematically code the program orientation of the studies analyzed; (4) employ a defensible meta-analytic strategy for categorizing study effect size; and (5) employ appropriate statistical techniques for the quantitative meta-analysis. None of the three quantitative reviews6--8 meets these five criteria, and thus the concerns of the critics regarding the apparent superiority of newer programs cannot be allayed. The present paper presents a meta-analysis of studies published in the 1970s and 1980s that evaluated school-based programs designed to deter adolescents from smoking tobacco products. The meta-analysis was

Tobacco and alcohol use among adolescents continue at historically high rates, and school-based interventions designed to deter students from smoking and drinking are increasingly being implemented. This study reports a meta-analysis of 47 smoking and 29 alcohol school-based intervention programs published after 1970. Results indicate that, in general, smoking and alcohol interventions have equally modest effects on immediate behavioral outcomes. Smoking interventions, however, have been more successful than alcohol interventions at altering students' long term behavior. All of the alcohol programs and all but one of the smoking programs reviewed successfully increased knowledge regarding the risks of these behaviors. Attitude change appears to be more difficult to achieve. Twenty-nine of 33 smoking studies and only 19 of 31 alcohol studies successfully changed students' attitudes. Finally, the data indicate that for immediate smoking outcomes and long-term alcohol outcomes innovative interventions relying upon social reinforcement, social norms, and developmental behavioral models are more effective than traditional "awareness" programs designed to inform adolescents about the health risks associated with tobacco and alcohol use. The implications of these findings for future of school-based health promotion programs are discussed.

No abstract