Poor Cognition : Early-life Socioeconomic Status and Cognitive Abilities in Adulthood : The Helsinki Birth Cohort Study 1934–1939

Introduction. Cognitive abilities are strongly associated with well-being in many aspects of life: more cognition translates to a longer, healthier life with more education and income. Differences in cognitive abilities have been associated with childhood socioeconomic status (SES). This thesis examines, how early-life SES is related to differences in cognitive ability in early adulthood in 473 men born in Helsinki in 1934–1939. In this context SES likely translated into considerable differences in the developmental environments of children, who experienced both the long shadow of the Great Depression and the shock of the Second World War. Methods. Linear regression was used to explore the relationship between childhood parental SES and cognitive abilities at the average age of 20.3 years, as measured by the Finnish Defence Forces Basic Ability Test. The problems of over-control bias and collider bias with linear regression were noted. Early-life data came from the Helsinki municipal tax records as well as from birth, child welfare clinic and school records. SES was measured by parental income and father’s occupational status. In addition, pre-test education and nutritional status proxied by anthropometrics at 2, 7 and 20 years of age were analyzed as possible mediators, along with some other covariates. Both the models without the mediators and including them were examined, in order to get clues into whether a variable is likely to be a mediator. Results. Childhood income was consistently related to adult cognitive abilities in all of the regression models. As expected, the association attenuated when the mediators were added. Father’s senior clerical occupational status was also consistently related to the cognitive test scores, while junior clerical status was not statistically significant, when the mediators were included. In comparable standard deviation units, a one SD higher income predicted about 0.26 SD higher cognitive scores without the other variables. When father’s occupational status was included, income predicted 0.16 SD higher scores, while father’s senior clerical status compared to manual worker status predicted 0.25 SD higher scores and junior clerical status 0.15 SD higher scores. When the other covariates apart from anthropometrics were added, income predicted 0.13 SD and father’s senior clerical occupational status 0.17 SD higher scores. Of the mediators, having at least upper secondary education predicted 0.39 SD higher scores. Anthropometrics predicted about 0.06–0.15 SD higher scores for one SD increase in each measure, when individually added to the models. Conclusion. This thesis shows that early-life socioeconomic status consistently predicts adult cognitive abilities in a cohort of men born in Helsinki in the 1930s. Education and nutritional status are some possible mediators of this relationship