Lessons from the history of tobacco harm reduction: The National Cancer Institute's Smoking and Health Program and the "less hazardous cigarette".

Scientists and public health practitioners are sharply divided today over the risks and benefits of tobacco harm-reduction strategies. At the same time, a range of novel tobacco products is being marketed with claims of reduced exposure or risk. Current scientific efforts to study tobacco products and harm reduction should be informed by past experience. During the 1960s and 1970s, there was substantial support within government and academia, as well as among voluntary health organizations, for efforts to modify tobacco products to reduce harm. This paper analyzes the former National Cancer Institute (NCI) Smoking and Health Program, which, between 1968 and 1980, pursued the development of "less hazardous" cigarettes as its primary goal. During this period, the program spent over dollar 50 million on contract research, of which 74% went toward biological and chemical analysis of modified cigarettes, 9.6% to epidemiological studies of risk factors, and only 1.4% to studies evaluating smoking cessation or prevention programs. NCI officials predicted during the mid-1970s that new "low-tar" cigarette brands would substantially reduce smoking-related mortality, but by 1978 the research agenda began to change in response to a reorganization of NCI research activities, modification of government antismoking efforts, and an emerging understanding of nicotine addiction that challenged key scientific assumptions. In retrospect, the program suffered from significant weaknesses that severely limited the likelihood that it would generate knowledge beneficial to public health, including a research agenda that failed to include surveillance and behavioral research, tobacco industry influence of the research agenda, and a lack of access to information about the characteristics of products on the market. There exists today a need for a public health-oriented research agenda on tobacco products and harm reduction, but current efforts should include input from a diverse range of disciplines, collect data on users' behavior, and limit the involvement of industry scientists.