Summary, in English
The aim of this thesis is to study early life risk exposures in relation to adult health and mortality in Sweden during 250 years. A number of causal mechanisms by which exposure to diseases and stressful economic and social conditions early in life may lead to increased morbidity and mortality later in life are discussed (paper I). The early life exposures investigated are the foetal origins (nutrition) and the inflammation hypotheses. Longitudinal demographic and socioeconomic data for individuals and households from parish registers are combined with local area data on food costs (rye prices) and disease load (infant mortality rate, IMR) using a Cox regression framework to analyse the 55-80 year age group mortality based on a data material from four parishes in Scania, southern Sweden, in the late 18th and almost entire 19th centuries (The Scanian Economic Demographic Database) (papers II and III). Register based cohort data with individual demographic and socioeconomic characteristics and mortality from the Swedish Longitudinal Immigrant Database (SLI), including individuals in Sweden born in eleven countries in the inter-war period who were residents in Sweden in 1980-2001, is analysed in Cox regression analyses of adult mortality, including IMR and gross domestic product (GDP) per capita in the year and country of birth in the analyses (paper IV). A cross-sectional public health questionnaire survey in Scania conducted in 2008 based on respondents born in Sweden in 1927-1960 is used to analyse associations between individual self-reported data on socioeconomic early life conditions and IMR in the year of birth, and adult self-rated health (SRH) in logistic regression analyses (paper V). The results of paper II suggest that cohorts exposed to high IMR during the first year of life have high mortality in ages 55-80 specifically from airborne infectious diseases, while early life food costs have no such effect. Paper III concerns early life exposure of poor nutrition, disease load on mothers during pregnancy and IMR in infancy (including airborne infectious disease mortality) leading to subsequent cohort effects on old-age mortality. The results suggest that high IMR and a high disease load of particularly airborne infectious diseases in infancy have a strong impact on mortality in later life. In contrast, hypotheses concerning the access to nutrition and the disease load during pregnancy are not supported. Adult individual socioeconomic status is essentially not associated with adult morality in papers II and III. In contrast, in paper IV current adult socioeconomic conditions are more strongly and consistently associated with adult mortality than early life indicators IMR and GDP per capita in the year of birth in the country of birth. The results of paper V show significant associations between individual socioeconomic early life factors and adult SRH, but not between IMR in the year of birth (and the year after birth) and SRH. In conclusion, historical results suggest cohort effects of IMR on adult later life mortality, which supports the inflammation hypothesis, while modern results show significant associations between individual adult (paper IV) and early life (paper V) socioeconomic indicators and adult mortality and SRH, respectively.