*2.3. Data Analysis*

The Shapiro–Wilk test was used to check for normality in the data, which was performed by the first author. The purpose of checking for normality in the data is to ensure that the data are normally distributed. A normally distributed dataset is essential for the validity of statistical tests. The categorical patient sociodemographic data were analyzed using frequency statistics. The Mann–Whitney U test was used to compare the RTW and non-RTW groups, while Pearson's correlation was used to analyze the relationship between the free variables and both the dependent and independent variables (r). Free variables are variables that are not controlled by the researcher but can affect the outcome of the study. The Pearson's correlation (r) was used to assess the strength of the relationship between the independent and dependent variables. r ranged from −1 to 1, with negative correlation indicating that one variable decreases as the other increases, and positive correlation indicating both variables increasing or decreasing together. The strength of the relationship was categorized as negligible (r < 0.2), low (r = 0.2–0.49), moderate (r = 0.5–0.69), high (r = 0.7–0.85), or very high (r = 0.86–1.00), with higher values indicating stronger linear relationships.

For the continuous data, we used independent sample *t*-tests (t), and for the ordinal data, we utilized Mann–Whitney tests. The Chi-square tests were used in order to investigate the differences in the categorical data. Multivariate logistic regression was used to investigate the association between several predictor factors and the outcome variable. To address the research question of this study in analyzing the relationship between the work ability index and quality of life between the RTW and non-RTW participants, multivariate logistic regression is the most suitable approach since it controls for the effects of other variables.

The study used logistic regression to assess the association between independent and dependent variables while controlling for other factors. The aim of the analysis was to examine the relationships between various factors and the outcome of interest, taking into account the influence of other variables. The independent and dependent variables were determined using a questionnaire that included sociodemographic and disability-related questions as well as assessments of QOL and WAI using validated questionnaires. The logistic regression was performed using SPSS version 26.0, with a *p*-value of less than 0.05 and a 95% confidence interval.
