Keele Research Repository

OBJECTIVE: To establish if gravidity and parity associate with the development of rheumatoid arthritis (RA), and to establish if this effect is influenced by the time elapsed since pregnancy/childbirth, the number of pregnancies/childbirths, and serological status, through systematically reviewing the literature and undertaking a meta-analysis. METHODS: We searched Medline/EMBASE (from 1946 to 2018) using the terms "rheumatoid arthritis.mp" or "arthritis, rheumatoid/" and "pregnancy.mp" or "pregnancy/" or "parity.mp" or "parity/" or "gravidity.mp" or "gravidity/" (observational study filter applied). Case-control/cohort studies that examined the relationship between parity/gravidity and the risk of RA in women were included. Studies reporting effect size data for RA in ever vs. never parous/gravid women as ORs/RRs with 95% confidence intervals were included in a meta-analysis. Other relationships (i.e. risk by pregnancy/childbirth numbers) were analysed descriptively. RESULTS: Twenty studies (from 626 articles) met our inclusion criteria, comprising 14 case-control (4799 cases; 11,941 controls) and 6 cohort studies (8575 cases; 2,368,439 individuals). No significant association was observed in the meta-analysis of studies reporting the risk of RA in ever vs. never parous women (OR 0.91; 95% CI 0.80-1.04) and ever vs. never gravid women (OR 0.86; 95% CI 0.46-1.62). No consistent evidence of a relationship between the number of pregnancies/childbirths and RA risk was seen. No significant association was observed between being pregnant, or in the immediate post-partum period, and the risk of developing RA. CONCLUSION: Our systematic review does not support the concept that gravidity and parity are associated with the risk of RA development.