Physicians’ views on the role of smoking in smoking-related diseases: findings from cross-sectional studies from 1982–2014 in Estonia

BACKGROUND Previous reports have shown that physicians who smoke underestimate the effects of smoking on health and this influences their practice. This study was designed to investigate the views of Estonian physicians on the role of smoking in smoking-related diseases. METHODS Cross-sectional postal surveys were sent to all practising physicians in Estonia in 1982, 2002 and 2014 providing data from 3504, 2694, and 2865 physicians respectively. Data analysis involved calculation of the prevalence of smoking with 95% confidence intervals and calculation of the prevalence of agreement with a causal role of smoking in ischaemic heart disease (IHD), lung cancer and chronic bronchitis. Logistic regression was used to analyse associations between agreement with the statements that smoking is a cause of IHD, lung cancer, chronic bronchitis and study year, smoking status, age group and ethnicity. Fully adjusted odds ratios with 95% confidence intervals were calculated. RESULTS The age-standardized prevalence of current smoking among men decreased from 39.6% in 1982 to 14.2% in 2014, and among women from 12.4 to 5.1%, respectively. Compared with 1982, the proportion of physicians agreeing with statements that smoking is a major cause or one of the causes of IHD, lung cancer, or chronic bronchitis was significantly higher in 2002 and 2014. Past and never smokers were more likely to admit a causal role of smoking in lung cancer than smokers. Agreement with a causal role of smoking in IHD and chronic bronchitis was significantly higher among never and past smoking women only. Compared with non-Estonians, the odds ratio indicating agreement with all three statements was significantly higher among Estonians. CONCLUSION From 1982 to 2014, physicians' attitudes towards the health risks of smoking improved in Estonia. However, their assessment of a causal role of smoking in smoking-related diseases was related to their own smoking habits and ethnicity. A further decline in smoking among Estonian physicians would require special efforts targeted at physicians. Societal pressure from a national policy could support a further decline in the social acceptability of smoking in Estonia and developments in medical education could provide continuing evidence-based information about the effects of smoking to Estonian physicians.