Parity, body mass index, smoking and risk of rheumatoid arthritis: data from the Australian Longitudinal Study on Women’s Health

Background Smoking, obesity and lower parity appear to be associated with higher risk of rheumatoid arthritis (RA) in European, American and some Asian studies, but no studies have tested these associations in the Australian context. Methods We studied an incident RA cohort of the prospectively collected Australian Longitudinal Study on Women’s Health (ALSWH) dataset. Smoking status, body mass index (BMI) and parity were assessed using survey data. Incident RA status was determined using the government medication dispensing database. We used multivariable logistic regression models to assess the impact of smoking, BMI and parity on RA risk, controlling for age cohort, education and alcohol intake. We performed sensitivity analysis by repeating analyses in the 1946-51 cohort alone. Results Among 40393 women we identified 332 incident RA cases. Age, ever smoking, BMI and parity were significantly associated with RA risk on univariate analyses. Lower parity was associated with higher odds of RA after adjustment for all covariates (OR 1.12, 95% CI 1.00-1.25, p = 0.05), and became stronger (OR 1.29, 95% CI 1.18–1.40, p < 0.01) when age cohort was excluded. Ever smoking (OR 1.29, 95% CI 1.02–1.64, p = 0.03) was associated with a higher risk of RA controlling for all factors except age. Each unit increase in BMI was significantly associated with higher odds of RA (OR 1.03, 95% CI 1.01–1.05, p < 0.01) when controlled for smoking, alcohol intake and education, but not once controlled for either age or parity. Conclusion Using a novel case finding method for RA that relies entirely on pre-collected administrative data, we show an increased risk for RA in Australian women with smoking and lower parity when adjusted for BMI, alcohol intake and education.