Dissecting epidemiological associations in alcohol drinking and anorexia nervosa

Background. Biases threaten the validity of practically every epidemiological study. In this dissertation, potential sources of bias in psychiatric epidemiology were tackled with systematic, population-based cohort studies in the context of anorexia nervosa and alcohol drinking. Multiple imputation was used to reduce selection bias, and previously overlooked potential confounders were examined with traditional methods and using a natural experiment, the discordant-twin design. Aims and methods. The population-based FinnTwin16 cohort (studies I and II) and the population-based Older Finnish Twin Cohort (study III) were used. In study I (n = 2639), the association of fathers’, mothers’ and women’s religiosity with lifetime anorexia nervosa (n = 91) was examined, and multiple imputation was used to reduce selection bias. In study II (1235 men and 1461 women assessed in early adulthood), it was examined whether potential confounders identified from the literature could explain the association between parents’ and their adult children’s problem drinking. In study III (n = 14787), the discordant-twin design was used to examine the potential confounding effects of genetic factors and childhood family environment in the association of alcohol drinking with all-cause mortality (2203 deaths). Findings. First, in a systematic study, in which selection bias was tried to be minimized with multiple imputation, parental and personal religiosity did not seem to be major risk factors for anorexia nervosa. This underscores the importance of systematic evidence, as many case reports suggest the opposite. Second, the association between parents’ and their children’s problem drinking did not appear to be attributable to the examined confounding factors (area of residence, family structure, and fathers’ and mothers’ education, religiosity and one relevant dimension of personality). Nevertheless, causality cannot be inferred, as the effect of genetic predisposition on problem drinking could not be excluded. Third, the confounding effects of genetic factors and shared childhood environment could not explain the associations of high total alcohol consumption (at least 259 grams per month corresponding to more than about 5 drinks per week) and alcohol-induced blackouts (at least twice a year) with all-cause mortality.