Table 2 |
|||
|
Odds ratios [95% CI] for educational attainment associated with four knee OA outcomes |
|||
|
Men |
Women |
Postmenopausal women |
|
|
|
|||
|
Radiographic knee OA |
n = 1,053 |
n = 1,574 |
n = 1,330 |
|
Unadjusted |
1.65 [1.25, 2.17] |
2.30 [1.84, 2.86] |
2.18 [1.72, 2.76] |
|
Adjusted age/race |
1.12 [0.83, 1.53] |
1.61 [1.26, 2.05] |
1.55 [1.19, 2.01] |
|
+ Lifestyle, clinical |
1.23 [0.89, 1.70] |
1.39 [1.08, 1.80] |
1.35 [1.02, 1.78] |
|
+ Occupation |
1.21 [0.85, 1.71] |
1.48 [1.12, 1.96] |
1.34 [0.99, 1.81] |
|
+ HRT |
N/A |
N/A |
1.33 [0.98, 1.79] |
|
|
|||
|
Bilateral radiographic knee OA |
n = 908 |
n = 1,338 |
n = 1,132 |
|
Unadjusted |
1.81 [1.25, 2.61] |
2.77 [2.09, 3.67] |
2.59 [1.92, 3.49] |
|
Adjusted age/race |
1.05 [0.69, 1.59] |
1.72 [1.26, 2.35] |
1.68 [1.21, 2.34] |
|
+ Lifestyle/clinical |
1.17 [0.76, 1.80] |
1.45 [1.03, 2.04] |
1.44 [1.01, 2.07] |
|
+ Occupation |
1.06 [0.67, 1.68] |
1.62 [1.12, 2.36] |
1.55 [1.05, 2.30] |
|
+ HRT |
N/A |
N/A |
1.53 [1.03, 2.27] |
|
|
|||
|
Symptomatic knee OA |
n = 1,052 |
n = 1,571 |
n = 1,327 |
|
Unadjusted |
2.03 [1.43, 2.86] |
2.69 [2.06, 3.5] |
2.39 [1.81, 3.16] |
|
Adjusted age/ethnicity |
1.55 [1.06, 2.26] |
1.90 [1.42, 2.54] |
1.74 [1.28, 2.36] |
|
+ Lifestyle, clinical |
1.71 [1.14, 2.56] |
1.66 [1.21, 2.27] |
1.51 [1.08, 2.11] |
|
+ Occupation |
1.86 [1.20, 2.87] |
1.64 [1.16, 2.31] |
1.46 [1.02, 2.10] |
|
+ HRT |
N/A |
N/A |
1.46 [1.01, 2.09] |
|
|
|||
|
Bilateral symptomatic knee OA |
n = 943 |
n = 1,403 |
n = 1,172 |
|
Unadjusted |
1.50 [0.82, 2.76] |
3.11 [2.08, 4.64] |
2.86 [1.86, 4.41] |
|
Adjusted age/ethnicity |
0.94 [0.48, 1.84] |
2.07 [1.33, 3.20] |
1.99 [1.25, 3.19] |
|
+ Lifestyle, clinical |
1.07 [0.53, 2.13] |
1.75 [1.10, 2.81] |
1.76 [1.07, 2.91] |
|
+ Occupation |
0.90 [0.43, 1.88] |
1.87 [1.13, 3.11] |
1.87 [1.09, 3.20] |
|
+ HRT |
N/A |
N/A |
1.84 [1.07, 3.16] |
|
|
|||
|
Odds ratios compare low (<12 years) with high (≥ 12 years) educational attainment for having a knee OA outcome. For all models, data are restricted to individuals having a complete set of nonmissing values for their explanatory variables, and covariates are added (+) cumulatively. Adjusted age/race, adjusted for age and race. +Lifestyle/Clinical, additionally adjusted for alcohol and tobacco use, BMI, and knee injury. +Occupation, additionally adjusted for occupational category, physical activity score. +HRT, additionally adjusted for use of hormone replacement therapy. |
|||
|
Callahan et al. Arthritis Research & Therapy 2010 12:R46 doi:10.1186/ar2956 |
|||
