Smoking and mortality in Eastern Europe: results from the PrivMort retrospective cohort study of 177,376 individuals.

Background: The estimated prevalence of smoking and proportion of deaths due to tobacco in Eastern European countries are among the highest in the world. Existing estimates of mortality attributable to smoking in the region are mostly indirect. The aim of this analysis was to calculate the proportion of tobacco-attributed deaths in three Eastern European countries using individual level cohort data. Methods: The PrivMort project established a cohort of relatives of participants in population sample surveys in Russia, Belarus and Hungary. Survey participants provided data on smoking habits and vital statistics of their close relatives between 1982 and 2013. Population attributable risk fractions (PARF) in men (n = 99 528) and women (n = 77 848) aged 40-79 years were calculated from the prevalence rates of smoking and hazard ratios of mortality for smokers versus non-smokers. Trends in PARF over four 8-year time periods (1982-1989, 1990-1997, 1998-2005, and 2006-2013) were examined. Results: In men in the most recent period (2006-2013), the proportions of deaths attributable to tobacco were 23% in Russia, 22% in Belarus, and 22% in Hungary. The respective estimates in women were lower (2%, 2%, and 13%), possibly due to underestimation of smoking prevalence. PARF estimates have declined slightly since the early 1990s in men but increased in women. Conclusions: Consistently with existing indirect estimates, our results based on individual level cohort data suggest that over one fifth of all deaths in men aged 40-79 years are attributable to tobacco. While these proportions are lower in women, the increasing trend is a major concern.