HRs for incident RA, and incident seropositive or seronegative RA phenotypes, according to time-varying Body Mass Index (BMI) in NHS and NHS II
| BMI | Nurses’ Health Study | Nurses’ Health Study II | Pooled HR (95% CI) | p Het† | ||||
|---|---|---|---|---|---|---|---|---|
| Number of cases /Person-years | Age-adjusted HR (95% CI) | Multivariable HR (95% CI)* | Number of cases/ person-years | Age-adjusted HR (95% CI) | Multivariable HR (95% CI)* | |||
| All RA cases | ||||||||
| All RA | ||||||||
| 18.5–24.9 | 407/1 490 667 | 1.00 | 1.00 | 128/1 040 272 | 1.00 | 1.00 | 1.00 | |
| 25.0–29.9 | 271/816 778 | 1.15 (0.98 to 1.34) | 1.16 (0.99 to 1.35) | 118/492 801 | 1.76 (1.37 to 2.27) | 1.68 (1.30 to 2.17) | 1.37 (0.95 to 1.98) | 0.015 |
| ≥30.0 | 148/457 750 | 1.11 (0.92 to 1.35) | 1.12 (0.92 to 1.37) | 109/401 445 | 1.92 (1.48 to 2.49) | 1.72 (1.31 to 2.45) | 1.37 (0.91 to 2.09) | 0.012 |
| p trend | 0.167 | 0.150 | <0.001 | <0.001 | 0.068 | |||
| Seropositive RA | ||||||||
| 18.5–24.9 | 257/1 390 405 | 1.00 | 1.00 | 84/1 019 688 | 1.00 | 1.00 | 1.00 | |
| 25.0–29.9 | 166/756 204 | 1.15 (0.94 to 1.40) | 1.15 (0.94 to 1.41) | 74/481 569 | 1.69 (1.24 to 2.32) | 1.63 (1.19 to 2.24) | 1.34 (0.96 to 1.88) | 0.071 |
| ≥30.0 | 87/421 741 | 1.09 (0.85 to 1.40) | 1.08 (0.84 to 1.39) | 74/390 262 | 1.98 (1.44 to 2.73) | 1.81 (1.30 to 2.52) | 1.38 (0.83 to 2.30) | 0.015 |
| p trend | 0.345 | 0.386 | <0.001 | <0.001 | 0.133 | |||
| Seronegative RA | ||||||||
| 18.5–24.9 | 150/1 484 299 | 1.00 | 1.00 | 44/1 038 235 | 1.00 | 1.00 | 1.00 | |
| 25.0–29.9 | 105/813 544 | 1.18 (0.91 to 1.51) | 1.19 (0.92 to 1.54) | 44/491 544 | 1.92 (1.26 to 2.93) | 1.80 (1.18 to 2.76) | 1.41 (0.95 to 2.11) | 0.102 |
| ≥30.0 | 61/456 258 | 1.21 (0.89 to 1.64) | 1.25 (0.91 to 1.70) | 35/400 415 | 1.84 (1.17 to 2.88) | 1.58 (0.99 to 2.51) | 1.34 (1.03 to 1.74) | 0.406 |
| p trend | 0.169 | 0.127 | 0.007 | 0.053 | 0.017 | |||
| RA onset ≤55 years of age | ||||||||
| All RA | ||||||||
| 18.5–24.9 | 170/802 579 | 1.00 | 1.00 | 112/999 985 | 1.00 | 1.00 | 1.00 | |
| 25.0–29.9 | 86/317 192 | 1.21 (0.93 to 1.58) | 1.22 (0.93 to 1.58) | 101/460 156 | 1.82 (1.39 to 2.39) | 1.72 (1.31 to 2.27) | 1.45 (1.03 to 2.03) | 0.072 |
| ≥30.0 | 56/166 957 | 1.51 (1.11 to 2.05) | 1.51 (1.10 to 2.06) | 93/370 245 | 2.03 (1.53 to 2.68) | 1.79 (1.34 to 2.39) | 1.65 (1.34 to 2.05) | 0.421 |
| p trend | 0.007 | 0.008 | <0.001 | <0.001 | <0.001 | |||
| Seropositive RA | ||||||||
| 18.5–24.9 | 108/739 670 | 1.00 | 1.00 | 72/979 647 | 1.00 | 1.00 | 1.00 | |
| 25.0–29.9 | 56/287 336 | 1.27 (0.92 to 1.76) | 1.27 (0.91 to 1.76) | 65/449 113 | 1.84 (1.31 to 2.57) | 1.76 (1.25 to 2.47) | 1.49 (1.08 2.04) | 0.178 |
| ≥30.0 | 36/149 759 | 1.57 (1.07 to 2.31) | 1.56 (1.05 to 2.31) | 61/359 345 | 2.07 (1.46 to 2.93) | 1.87 (1.31 to 2.68) | 1.72 (1.32 to 2.24) | 0.498 |
| p trend | 0.015 | 0.019 | <0.001 | < 0.001 | <0.001 | |||
| Seronegative RA | ||||||||
| 18.5–24.9 | 62/798 202 | 1.00 | 1.00 | 40/997 976 | 1.00 | 1.00 | 1.00 | |
| 25.0–29.9 | 30/315 436 | 1.16 (0.75 to 1.80) | 1.17 (0.75 to 1.82) | 36/458 922 | 1.83 (1.16 to 2.88) | 1.71 (1.08 to 2.70) | 1.41 (0.97 to 2.04) | 0.243 |
| ≥30.0 | 20/166 094 | 1.51 (0.91 to 1.52) | 1.53 (0.90 to 2.59) | 32/369 247 | 1.99 (1.24 to 3.18) | 1.69 (1.04 to 2.75) | 1.61 (1.13 to 2.31) | 0.782 |
| p trend | 0.115 | 0.113 | 0.004 | 0.037 | 0.009 | |||
Bold typeface indicates significant results.
*Adjusted for age, community median income, smoking (pack-years), alcohol use (<5.0, 5.0-<15.0, ≥15 g/day) and physical activity (0–3, 3–9, 9–18, 18–27, ≥27 METs/week). Additional adjustment for women's reproductive factors (menarche age, parity/breastfeeding, menopausal status and postmenopausal hormone use) did not change the results.
†p Value for test of heterogeneity (het) between the 2 cohorts. Risk estimates were pooled by using a DerSimonian and Laird random-effects model.
MET, metabolic equivalent; RA, rheumatoid arthritis.