We aimed to assess the relationships between anthropometric measures and risk of rheumatoid arthritis (RA). The E3N cohort included 98,995 women (aged 40−65 years at the recruitment) who completed mailed questionnaires on reproductive factors, lifestyle, and health-related information, including anthropometric measures, every 2−3 years. Cox proportional hazards regression models with age as the time scale and adjusted on known RA risk factors were used to estimate hazard ratios (HRs) and 95% confidence intervals for the risk of incident RA in the overall population (n = 78,452) and after stratification on smoking exposure. Incident RA diagnosis was validated in 698 women. Abdominal obesity (waist circumference >88 cm) was associated with RA (HR = 1.2 (1.0−1.5)), independent of BMI; whereas obesity, defined as BMI ≥ 30 kg/m2, was marginally associated with RA (HR = 1.26 (0.9−1.5), ptrend = 0.0559). Taking lean body shape (BS) as reference, medium BS at puberty (HR = 1.3 (1.0−1.7)) and medium-large BS at perimenopausal period (HR = 1.5 (1.1−1.9)) were associated with the risk of RA among never-smoker women, independent of BMI. Regarding BS trajectory, taking constantly lean BS as reference, constantly large BS from puberty to perimenopause was associated with RA among non-smokers (HR = 2.10 (1.2−3.6)), independent of BMI.
Keywords: anthropometric measures; body shape trajectories; prospective cohort; rheumatoid arthritis; risk factor.