Missing values were estimated by using multivariable regression models that were conditional on relevant predictors.
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 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 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.
Using each participant as the unit of analysis, we estimated the associations between first episode of kidney stone and outcomes by using multivariable Cox proportional hazards models adjusted for all variables presented in table 1.
The associations of coffee intake in categories (described above) and continuously (per cup increase) with risk of HCC were evaluated by using multivariable conditional logistic regression models, accounting for matching factors (as described above) with additional adjustment for a priori chosen covariates mainly based on previous knowledge about liver cancer risk factors.
26 27 28 In sensitivity analyses, we examined associations between overall mortality and baseline characteristics by using multivariable Cox proportional hazards regression models restricted to patients with radiographic osteoarthritis defined by a more stringent Kellgren-Lawrence grade of at least 2 and restricted to patients with complete information on covariates.
We used multivariable modelling to estimate the association of patient-level costs by intent, firearm type and payer source, controlling for covariates and hospital characteristics.
By using a multivariable model with an interaction term, we were able to estimate the effect of residing in a Q fever-affected area in 2008 2010, and not the already existing differences before the outbreak.
A logistic regression model showed that GLPS >−15 gave a hazards ratio for death at 90 days of 2.5, but the result was not statistically significant (95% CI 0.66, 9.46), and no predictive value was gained by using a multivariable model with LVEF, é and NT-proBNP in addition to global strain.
