Using descriptive statistics, data were presented as the number, percentage (%), mean, and standard deviation (SD). All sociodemographic and medical parameters were compared between groups. We used the chi-square test for nominal and categorical variables (e.g., gender, county of residence, diagnosis, and condition at the end of sickness absence) and an independent-samples t-test for numeric variables (e.g., age and number of days of sick leave).
The strength and the direction of the relationship between different variables was studied by correlation analysis. Multiple linear regression was used for the continuous dependent variables (e.g. length of sick leave) and multiple logistic regression for dichotomous dependent variables (e.g. return to work or not).
The statistical significance (p-value) was established at 0.05, as is the convention, and parameters were estimated for a 95% confidence interval. The data were processed with PSPP software.
Do you have any questions about this protocol?
Post your question to gather feedback from the community. We will also invite the authors of this article to respond.