Gender and sex independently associate with common somatic symptoms and lifetime prevalence of chronic disease

Sex and gender influence health differently. Associations between sex and health have been extensively studied, but gender (i.e. psychosocial sex) has been largely neglected, partly due to the absence of gender measures in cohort studies. Therefore, our objective was to test the unique associations of gender and sex with common somatic symptoms and chronic diseases, using a gender index created from existing cohort data. We applied LASSO logistic regression to identify, out of 153 unique variables, psychosocial variables that were predictive of sex (i.e. gender-related) in the Dutch LifeLines Cohort Study. These psychosocial variables covered gender roles and institutionalized gender. Using the estimated coefficients, gender indexes were calculated for each adult participant in the study (n = 152,728; 58.5% female; mean age 44.6 (13.1) years). We applied multiple ordinal and logistic regression to test the unique associations of the gender index and sex, and their interactions, with common somatic symptoms assessed by the SCL-90 SOM and self-reported lifetime prevalence of chronic diseases, respectively. We found that in 10.1% of the participants the gender index was not in line with participants' sex: 12.5% of men and 8.4% of women showed a discrepancy between gender index and sex. Feminine gender characteristics are associated with increased common somatic symptoms and chronic diseases, especially in men. Female sex is associated with a higher common somatic symptom burden, but not with a higher prevalence of chronic diseases. The study shows that gender and sex uniquely impact health, and should be considered in epidemiological studies. Our methodology shows that consideration of gender measures in studies is necessary and feasible, based on data generally present in cohort studies.