Is compulsory restriction of tar yield of cigarettes a worthwhile public health policy?

Although the notion of a "safe cigarette" has proved an illusion, while people continue to smoke, an issue remains of whether compulsory restriction of tar yield is a worthwhile public health policy.

The study group was comprised of a random population sample of middle-aged Scottish men and women-3464 smokers of cigarettes with known tar yield at baseline. Tar yields were classified into low (< 10 mg), low/middle (10-14.99 mg), and middle or high (> or = 15 mg), according to standard groupings. Deaths within the study cohort were recorded over a period of 13 years.

Among low, low/middle, and middle or high tar smokers, 55 (10%), 178 (16%), and 276 (16%), respectively, died. In a comparison group of lifetime never-smokers, 178 (6%) died. After adjustment for daily cigarette dose, duration of smoking, age, gender, social class, type A personality, body mass index, urinary potassium, and antioxidant vitamin consumption, hazard ratios (95% confidence intervals) for all-cause mortality comparing low-tar smokers to low/middle-tar smokers and to middle- or high-tar smokers were 1.64 (1.04-2.58) and 1.46 (0.95-2.26), respectively. Corresponding results for cardiovascular disease were 1.48 (0.74-2.96) and 1.35 (0.79-2.60). For lung cancer, after adjusting for known confounding factors, corresponding hazard ratios were 2.82 (0.98-8.15) and 2.30 (0.81-6.49).

Smoking any type of cigarette is far more risky than abstaining altogether. Although the results obtained comparing tar levels are limited because of lack of repeat measures of smoking during follow-up, potential residual confounding factors, statistical imprecision, and failure to show a dose-response relationship, results do support the hypothesis that persistent smokers will reduce their tobacco-induced health risk if they smoke cigarettes with < 10 mg of tar. However, the data do not permit quantification of the benefits of switching to lower-tar cigarettes. Worldwide legislation to limit tar levels to 10 mg as an absolute maximum is recommended, while recognizing that this must be part of an integrated approach to tobacco control.