Table 2. Association between change in leisure-time physical activity behavior and TV viewing (independent variables) and chronic musculoskeletal pain (dependent variable) after 24 months in schoolteachers, Londrina/PR, Brazil, 2012–2014.
(N = 527).
| Variables | Modela adjusted for sex and age | Previous model plus adjusted for BMI | Previous model plus adjusted for depression |
|---|---|---|---|
| Change in LTPA time (60 minutes/week)b | 0.938 (0.886, 0.992)* | 0.938 (0.886, 0.992)* | 0.938 (0.887, 0.993)* |
| Change in TV-viewing time (30 minutes/day)b | 1.054 (1.005, 1.106)* | 1.054 (1.005, 1.106)* | 1.051 (1.001, 1.102)* |
| Sex (female/male) | 1.575 (1.108, 2.239)* | 1.575 (1.108, 2.239)* | 1.498 (1.055, 2.127)* |
| Age (y) | 1.025 (1.008, 1.041)* | 1.025 (1.008, 1.041)* | 1.024 (1.007, 1.041)* |
| Change in BMI (kg/m2) | 1.038 (1.003, 1.075)* | 1.036 (1.002, 1.072)* | |
| Change in depression status (yes/no) | 1.873 (1.272, 2.757)* | ||
| Constant | 0.022 (0.006, 0.085)* | 0.022 (0.006, 0.085)* | 0.024 (0.006, 0.091)* |
| Wald Chi2, Prob >chi2 | 26.09, <0.001 | 29.92, <0.001 | 36.19, <0.001 |
*p-value <0.05
a Values are the odds ratios (95% confidence interval) estimated though the generalized estimating equation (GEE) regression model (parameters used: family: binomial; link: logit; correlation: independent; variance estimators: robust).
b Both variables were simultaneously included in the same model.