Multiple regression models for SF-36 subscale scores showing the relationship with poor/fair Self-Rated Heath (SF-1) (ORs and 95% CI)
| Model 1 n=20803 | Model 2 n=20803 | Model 3 n=10809 | Model 4 n=10044 | Model 5 n=6822 | Model 6 n=1674 | |||||||
| SF-36 subscales* | ||||||||||||
| Physical Function | 3.68 | 3.27 to 4.14 | 3.43 | 3.05 to 3.86 | 3.40 | 2.80 to 4.09 | 3.48 | 2.98 to 4.05 | 3.27 | 2.70 to 3.96 | 3.75 | 2.63 to 5.35 |
| Role Physical | 2.25 | 2.01 to 2.52 | 2.19 | 1.99 to 2.45 | 2.26 | 1.87 to 2.73 | 2.15 | 1.86 to 2.48 | 2.18 | 1.83 to 2.59 | 1.48 | 1.03 to 2.11 |
| Bodily Pain | 1.94 | 1.75 to 2.15 | 1.82 | 1.64 to 2.02 | 1.79 | 1.51 to 2.13 | 1.86 | 1.63 to 2.12 | 1.66 | 1.42 to 1.95 | 1.96 | 1.39 to 2.76 |
| Mental Health | 1.36 | 1.22 to 1.52 | 1.36 | 1.22 to 1.52 | 1.39 | 1.16 to 1.69 | 1.32 | 1.15 to 1.52 | 1.42 | 1.22 to 1.67 | 1.69 | 1.02 to 2.83 |
| Role Emotional | 1.11 | 1.03 to 1.24 | 1.13 | 1.02 to 1.26 | 1.07 | 0.88 to 1.29 | 1.13 | 0.99 to 1.29 | 1.06 | 0.91 to 1.25 | 1.24 | 0.87 to 1.76 |
| Social Function | 1.89 | 1.00 to 1.24 | 1.83 | 1.63 to 2.04 | 1.73 | 1.43 to 2.09 | 1.89 | 1.64 to 2.17 | 1.82 | 1.55 to 2.15 | 2.32 | 1.52 to 3.57 |
| Vitality | 3.64 | 3.24 to 4.09 | 3.59 | 3.19 to 4.03 | 3.71 | 3.05 to 4.51 | 3.53 | 3.05 to 4.08 | 3.65 | 3.08 to 4.32 | 3.90 | 2.48 to 6.16 |
| Medical Conditions | 1.72 | 1.56 to 1.89 | ||||||||||
| Depression | 1.19 | 1.06 to 1.34 | ||||||||||
Reference: excellent/very good/good Self-Rated Heath. The models show the effects of stepwise adjustments for: Model 1—age, sex and social class; Model 2—age, sex, social class, medical conditions and depression; Model 3—age, sex and social class and excluding all those with medical conditions or depression; Model 4—age, sex and social class including only those with medical conditions or depression; Model 5—age, sex and social class and including only those with medical conditions; Model 6—age, sex and social class, and including only those with depression.
↵* SF-36 subscale scores dichotomised above and below median.