cited for this, including competing or poorly coordinated funding agency priorities, inadequate research methodologies, lack of communication across disciplines, and an insufficiently coordinated data base. In addition, most investigators focused only on single causes of smoking (such as physical dependence on nicotine) or made use of single communication channels (such as mass media or physician office-based programs) for prevention.
Faced with this scattershot approach, NCI launched in 1982 the Smoking, Tobacco, and Cancer Program (STCP), a major planning and research effort to coordinate smoking prevention trials and develop large-scale comprehensive community interventions (Glynn, 1991). Lacking a consensus on how best to persuade people to quit or not begin smoking, the STCP mounted a well-planned, carefully phased three-phase campaign.
First, program administrators consulted with hundreds of experts to identify areas in which significant gains could be expected. They then called for and funded research to develop and evaluate prevention and cessation interventions that would be effective, cost efficient, durable, generalizable, and widely applicable. Between 1984 and 1987, NCI began 49 large trials, most of which were to last five years, at a total cost of $82 million.
Reflecting the multifactorial emphasis, these trials covered eight areas. First, there were school-based interventions with adolescents. These recognized that whatever can be done to prevent smoking by young people is doubly important, not only because it minimizes tissue damage during youth but also because it minimizes the hazard of addiction (few people start smoking after age 20, and those who do so may be less prone to addiction). There were self-help programs, based on the observation that 90 to 95 percent of all people who have stopped smoking claim they quit on their own. There were interventions conducted by physicians and dentists, who were considered a greatly underutilized resource. There were mass media interventions, which adopted many of the concepts developed in the Stanford cardiovascular risk programs. And there were interventions focused on four special populations: African-Americans, Hispanics, women, and smokeless tobacco users. At the time these trials began, the attributable risk of smoking (the rate of a disease or other outcome in exposed individuals that can be attributed to the exposure) had been fairly well established through epidemiological studies, although data were stronger for lung cancer than cardiovascular disease. There were few data, however, on the level of effect that the various types of interventions might have on smoking rates.