Passive and active smoking. A problem for adults.

Abstract

Over the last 10 years, 14 epidemiological studies1 2 3 4 5 6 7 8 9 10 11 12 13 14 and experimental studies15 16 17 have consistently shown that passive smoking increases the risk of heart disease death. Combining these studies in a meta-analysis yields an estimate that passive smoking significantly increases the risk of death from heart disease by about 20% (RR, 1.2; 95% CI, 1.1 to 1.4). As with the evidence that passive smoking causes lung cancer,18 19 20 the tobacco industry has vigorously attacked these studies as too small, biased, or failing to control for confounding variables adequately. These arguments have become harder to make over time as the number and quality of epidemiological studies on both lung cancer and heart disease have improved, so the tobacco industry has started to emphasize a new explanation: publication bias. For example, in the 1994-1995 hearings held by the federal Occupational Safety and Health Administration (OSHA) on a proposed rule making workplaces smoke free,21 several of the tobacco industry's witnesses argued that the reason that there were so many studies in the literature showing an elevation in risk of death associated with passive smoking was that journals simply did not publish “negative” studies because of a bias on the part of editors against the tobacco industry despite the fact that formal studies of the question of publication bias demonstrated that at least in the case of passive smoking, no such bias existed. Faced with this evidence, the tobacco industry took the argument one step further and said that publication bias included people not doing research or analyzing data that might not support the view of the health establishment. In particular, both at the hearing and in print,22 tobacco industry consultants speculated that because of …