Project Towards No Tobacco Use: implementation, process and post-test knowledge evaluation.

This paper describes the curricula contents, and presents data to evaluate the implementation, process and immediate post-test knowledge of Project Towards No Tobacco Use (Project TNT). Four different school-based tobacco use prevention curricula were developed to counteract the effects of three types of tobacco use acquisition variables typically addressed within a comprehensive social influences program: (1) peer approval for using tobacco (normative social influence), (2) incorrect social informational provided about tobacco use (information social influence) and (3) lack of knowledge or misperceptions about physical consequences resulting from tobacco use. Three curricula were designed to counteract the effects of single acquisition variables, whereas a fourth curriculum was designed to counteract the effects of combined social and physical consequences-related influences. These curricula were delivered to seventh grade students by trained project health educators to maximize implementation. 'Program' schools, those schools that received one of these curricula, were compared to 'control' schools that provided asystematic health education delivered by school personnel. A total of five conditions were contrasted through use of a randomized experiment involving 48 southern California junior high schools. This paper documents high levels of implementation in all program conditions. Also, favorable process ratings were obtained across the four program conditions, using multiple measures and sources of ratings (students, health educators and classroom teachers who observed curricula delivery). Finally, knowledge item sets completed by the students demonstrated discriminant validity across all five conditions. Because the program conditions were discriminable, yet were quite similar in implementation and process ratings, planned future study of behavioral outcomes can be interpreted as relatively uncontaminated by delivery or credibility confounds.