Skip to main content
. 2016 Nov 29;18:277. doi: 10.1186/s13075-016-1171-2

Table 2.

Associations between overweight or obesity and risk of development of rheumatoid arthritis

All Men Women
OR (95% CI) OR (95% CI) OR (95% CI)
All BMI (OR per 5 kg/m2 increase in BMI) 1.13 (1.00, 1.28) 1.17 (0.90, 1.50) 1.13 (0.98, 1.30)
BMI category vs. low-normal weight
All Obesity 1.45 (1.07, 1.95) 1.78 (1.01, 3.12) 1.37 (0.96, 1.95)
Overweight 1.21 (0.96, 1.52) 1.29 (0.84, 1.96) 1.21 (0.92, 1.59)
Normal-underweight Reference Reference Reference
Population divided by the age of the case at RA symptom onset:
≤50 years Obesity 1.89 (1.02, 3.52) 4.63 (1.14, 18.8) 1.39 (0.68, 2.87)
Overweight 1.43 (0.89, 2.29) 2.90 (1.08, 7.76) 1.20 (0.68, 2.10)
Normal-underweight Reference Reference Reference
51–59 years Obesity 1.37 (0.76, 2.47) 1.93 (0.68, 5.50) 1.16 (0.56, 2.40)
Overweight 1.10 (0.71, 1.72) 1.02 (0.47, 2.25) 1.23 (0.72, 2.13)
Normal-underweight Reference Reference Reference
60 − 65 years Obesity 1.62 (0.91, 2.89) 1.06 (0.35, 3.23) 1.96 (0.98, 3.92)
Overweight 1.08 (0.68, 1.71) 0.60 (0.25, 1.46) 1.28 (0.73, 2.25)
Normal-underweight Reference Reference Reference
≥66 years Obesity 1.14 (0.61, 2.14) 1.15 (0.32, 4.16) 1.20 (0.58, 2.48)
Overweight 1.29 (0.81, 2.01) 1.68 (0.67, 4.21) 1.17 (0.67, 2.05)
Normal-underweight Reference Reference Reference
Population divided by the serologic status for the case after RA diagnosis:
Seropositive RA Obesity 1.40 (1.00, 1.96) 1.33 (0.71, 2.50) 1.47 (0.99, 2.18)
Overweight 1.27 (0.99, 1.63) 1.15 (0.73, 1.84) 1.34 (0.99, 1.81)
Normal-underweight Reference Reference Reference
Seronegative RA Obesity 1.76 (0.88, 3.52) 12.3 (2.14, 71.1) 1.07 (0.47, 2.46)
Overweight 1.93 (0.51, 2.15) 3.44 (0.82, 14.5) 0.59 (0.27, 1.30)
Normal-underweight Reference Reference Reference

Analyses of the risk of development of rheumatoid arthritis (RA) associated with body mass index (BMI) for every 5 kg/m2 increase, or BMI categories of overweight or obesity compared with normal-underweight. Results are presented in subjects overall, and divided by sex. Results from conditional logistic regression, adjusted for education and smoking, in case–control sets matched for age, sex and year of examination. Analyses of BMI categories were also performed after dividing the case–control sets by quartiles of the age of the case at onset of RA symptoms, or whether the case subsequently developed seropositive or seronegative RA disease. OR odds ratio