Life course determinants of women’s health : from reproductive age to menopause

Abstract: Over the past four decades, growing evidence has indicated that characteristics such as birth weight and length of gestation are not only key indicators for infant’s health, but also predictors of adult health and disease risk. These findings lend support to the developmental origins of health and disease theory. However, evidence remains inconclusive in terms of female hormone-related disorders, including endometriosis and perimenopausal disorders. Also, little is known about the social patterning of these female health burdens from population-based studies. Furthermore, the psychological health of women with endometriosis has not been adequately explored in longitudinal studies. Based on identified knowledge gaps and taking advantage of Swedish high-quality population-based registers, this thesis aims to study how factors operating during early life, such as parental and birth characteristics, adult socioeconomic and reproductive factors are associated with subsequent risks of endometriosis and perimenopausal disorders. Further, it explores whether women with endometriosis have higher risk of psychiatric comorbidity. Study I explored the associations of early life social and health characteristics with risk of endometriosis in a cohort of the second generation of women from the Uppsala Birth Cohort Multigenerational Study. Lower birth weight-for-gestational age, fewer births, and previous infertility disorder were found to be associated with an increased risk of endometriosis. Nevertheless, the inverse association between low birth weight-for-gestational age and endometriosis could not be explained by women’s lower number of live births in adulthood. Study II replicated the original findings in Study I in a nationwide population-based cohort of females born in Sweden between 1973 and 1987. This study confirmed the inverse association between fetal growth rate and risk of endometriosis, and expanded Study I by showing associations of maternal smoking during pregnancy and lower maternal education with endometriosis risk in early to mid-adulthood. The study also found a part of the association between maternal smoking and risk of early-onset of endometriosis was due to slow fetal growth. Study III focused on the psychological health of women with endometriosis by assessing the bi-directional associations of endometriosis with all psychiatric disorders, as well as the role of familial confounding, in a nationwide cohort of all women born in Sweden in 1973-1990. Statistically significant bi-directional associations were found for endometriosis with many different types of psychiatric disorders, including affective psychotic disorders, depressive, anxiety and stress-related disorders, eating disorders, alcohol/drug dependence, personality disorders, and attention-deficit hyperactivity disorder. These bi-directional associations observed at the population level largely remained in comparisons between exposed and unexposed sisters, suggesting that shared familial liability may not fully explain these associations. Study IV investigated the developmental origins of three subtypes of perimenopausal disorders using the same cohort as Study I. Positive association between birth weight and a clinical diagnosis of menopausal and climacteric states was found. Higher risk of being diagnosed with other perimenopausal disorders (e.g., atrophic vaginitis) was observed among women born with shorter gestational age. This study also documented that women with higher parental and own educational level in adulthood were more likely to be diagnosed with perimenopausal disorders. Taken together, this thesis supports the developmental origins of two important female hormone-related disorders during reproductive age and menopause, namely endometriosis and perimenopausal disorders. Our findings highlight the importance of intrauterine environment in shaping the developmental adaptations of metabolism and organ function. In addition to the developmental origins, these female health burdens were associated with a range of socioeconomic and reproductive factors as well as mental health in earlier life. It is therefore important to take a life-course perspective for a greater understanding of the etiology of hormone-related health outcomes and consider potential targeting of the high-risk groups for earlier public health intervention.