Missing data
The total number of missing data values for the analytical sample
including 1098 participants was 419 out of 15372 (2.5%). The percentage
of missing values varied from 0 to 16% between the variables. The data
was missing due to the invalid and missing measurements as well as
unclear or incomplete questionnaire response. Thus, missing data were
assumed to occur at random. Multiple imputation was used to create and
analyze 50 multiply imputed data sets, and the model parameters were
estimated separately for each data set. The used number of iterations
for chained equations21 was 50. Multiple imputation
and pooling of the model estimates were carried out in
R22 using the standard settings of the “mice”
package.21 For comparison, we also performed complete
case analysis, but the results were not notably different.