Childbirths and risk of female predominant and other autoimmune diseases in a population-based Danish cohort.

To evaluate the possible biological role of pregnancy on the risk of autoimmune diseases we assessed associations between reproductive history and subsequent risk of autoimmune diseases characterized by female predominance and other autoimmune diseases. Our study cohort comprised 4.6 million Danes born since 1935 for whom a complete record of childbirths was available. Cohort members were followed for hospital contacts for 31 autoimmune diseases from 1982 to 2008. Female predominant autoimmune diseases were those with a female:male sex ratio >2:1. Ratios of first hospitalization rates were calculated using Poisson regression, adjusting for potential confounding by age, birth cohort, calendar period and marital status. During 45.5 million person-years of follow-up 102,260 women were hospitalized with one or more autoimmune diseases. Overall, compared with childless women, women with children were at a relative risk of 1.04 (1.02-1.06) for any autoimmune diseases, 1.11 (1.08-1.14) for female predominant and 0.97 (0.95-1.00) for other autoimmune diseases. Possibly biologically related associations with parity were found for Hashimoto thyroiditis (1.11; 1.00-1.24), Graves' disease (1.19; 1.14-1.24), erythema nodosum (1.15; 1.01-1.32), psoriasis (1.08; 1.01-1.15), sarcoidosis (1.17; 1.06-1.28) and systemic lupus erythematosus (0.83; 0.74-0.93). Especially the one-year postpartum period was associated with an increased risk of Hashimoto thyroiditis, Graves' disease and sarcoidosis. Overall, parity was associated with an 11% increased risk of female predominant autoimmune diseases. Pregnancies resulting in liveborn children therefore seem to contribute only little to the general female predominance in autoimmune diseases. However, for a number of autoimmune diseases; especially autoimmune thyroid diseases, erythema nodosum and sarcoidosis parity might somehow be involved in disease development.