Table 3. Multiple logistic regression models predicting impact on impact on physical activity levels and increased sitting time due to COVID-19.
| Variables | β (EP) | OR | 95%CI | p value |
|---|---|---|---|---|
| Impact on PA level* | ||||
| Age, years | -0.013 (0.004) | 0.98 | 0.97-0.99 | 0.001 |
| Physically active, yes as reference | 0.790 (0.108) | 2.20 | 1.78-2.72 | <0.001 |
| Chronic diseases, no s reference | 0.259 (0.117) | 1.29 | 1.03-1.63 | 0.027 |
| Overweight, no s reference | 0.297 (0.105) | 1.34 | 1.09-1.65 | 0.005 |
| Impact on sitting time† | ||||
| Age, years old | -0.029 (0.005) | 0.97 | 0.96-0.98 | <0.001 |
| Physically active, yes s reference | 0.414 (0.134) | 1.51 | 1.16-1.96 | 0.002 |
| Chronic diseases, no s reference | 0.504 (0.150) | 1.65 | 1.23-2.22 | 0.001 |
| Social isolation, days | 0.013 (0.004) | 1.01 | 1.00-1.02 | 0.002 |
| Schooling, low‡ as reference | 0.629 (0.201) | 1.87 | 1.26-2.78 | 0.002 |
* Hosmer-Lemeshow test: χ2 11.231, p=0.189; † Hosmer-Lemeshow test: χ2 6.825, p=0.556; ‡ elementary or high school. β (EP): regression coefficient (error-standard); OR: odds ratio; 95%CI: 95% confidence interval; PA: physical activity.