To determine the relative importance of the variables of the MLR model, we used the χ2 values for both sets of variables: the variables included in and the variables excluded from the final model.
Adjusted regressions for the remaining variables of each model include the complete set of variables with both measures of psychopathology.
Table 1 Imputation models for different types of variables Type of variable The model used for imputation Continuous variable Linear regression Binary variable Logistic regression Ordinal variable Ordinal logistic regression Nominal variable Multinomial logistic regression.
To allow for a variable predicting "missingness", this variable needs to be included as either one of the explanatory or dependent variables of the mixed model.
Linear regression models with delta variables were corrected for baseline values of the variables of interest.
For each response variable is shown the included explanatory variables of the seven models and an indication for the direction of the correlation of each variable.
Multivariate (adjusted) models were used to identify variables predictive of malnutrition, controlling for all other variables in the model.
In all models of logistic regression, the independent variables were adjusted for the other variables in the model.
