We estimated direct and productivity-loss costs and examined associated characteristics by using multivariable linear regression with backward selection.
Multivariable associations between baseline clinical variables and baseline MDHAQ pain score were examined by using multivariable linear regression.
The associations between baseline clinical predictors and MDHAQ pain at baseline and 1 year were assessed by using multivariable linear regression.
Estimated mean and 95% CI were calculated by using multivariable linear regression model adjusted for age, gender, body mass index, regular medication, daily coffee drinking (none, < 1 cup, 1 2 cups, ≥ 3 cups) and smoking (none, < 0.5 pack, 0.5-1 pack, > 1 packs) amounts.
Estimated mean and 95% CI were calculated by using multivariable linear regression model adjusted for age, gender, body mass index, regular medication, daily alcohol (none, M < 60 g or F < 20 g, M ≥ 60 g or F ≥ 20 g) and coffee (none, < 1 cup, 1 2 cups, ≥ 3 cups) drinking amounts.
Estimated mean and 95% CI were calculated by using multivariable linear regression model adjusted for age, gender, body mass index, regular medication, daily alcohol drinking (none, M < 60 g or F < 20 g, M ≥ 60 g or F ≥ 20 g) and smoking (none, < 0.5 pack, 0.5-1 pack, > 1 packs) amounts.
Comparative analyses of health behaviour and adiposity by income quintile and LICO groups were performed using multivariable linear regression models.
Table 3 presents the results of modeling odds of significant physical impairment at baseline by type of abuse exposure, reported at baseline, using multivariable linear regression.
Finally, data were analysed using multivariable linear regression models.
Determinants of 24UV were identified using multivariable linear regression models.
We determined associations using multivariable linear and logistic regression models.
