A backward multiple linear regression model was constructed to establish the relationship between each quality of life domain and the variables related to adherence, controlling for covariates.
The variables of education, spouse's occupation, sufficiency of income for expenses, crowding index, primary support source, and ethnicity entered the backward multivariable linear regression model.
Groups were compared by using t tests, and associations were tested by using univariate linear regression (p value and standardized coefficient shown) and in a backward multiple linear regression model with the least significant variables sequentially removed, according to the log-likelihood test.
Indeed, in our backward stepwise linear regression models, we found that DM loses its predictivity on AD when BMI was introduced as cofactor.
The final backward multivariate linear regression models for the relationship between selection factors and demographic characteristics with academic performance throughout the course are outlined in Table 5.
In order to elucidate the predictive value of each variable on AD level we drew different backward stepwise linear regression models (Table 3).
In the last step, we tested different multivariate backward stepwise linear regression models with AD as dependent variable and age, gender, ACC/AHA HF stage, BMI, NT-proBNP, HDL cholesterol and DM as independent variables, to test their independent predictive value.
In the linear regression model, after backward elimination, only water source remained in the model.
All variables found to be statistically significant at a 10% level in the simple model were then included in a multiple linear regression model using backward elimination.
In addition, univariate and multivariate analyses (stepwise selection method for the linear regression model or backward elimination method for the logistic regression model) were performed to determine which of the independent variables were related to the secondary outcome variables.
