*3.4. Correlation between Comorbidity Index (CI), Working Days Lost, and Sociodemographic Variables*

The association between the individual health status measured by the CI, and the absenteeism measured in working days lost, as well as the sociodemographic variables were analyzed through bivariate correlation. Higher CI scores indicate a level of increased vulnerability to individuals' health and may influence work absences. Table 5 shows the result of this analysis through Pearson's coefficient.

The CI scores were positively and significantly associated with the work absences of the servants at a 1% level (r = 0.254, *p*-value < 0.01), thus an increase in the level of vulnerability of the individual's health due to chronic diseases is associated with an increase in the working days lost. Moreover, the CI was positively and significantly associated with both the age of the servants at a 5% level (r = 0.116, *p*-value < 0.05) and the length of service (seniority) at a 1% level (r = 0.133, *p*-value < 0.01). Therefore, an increase in the age of the servants and their length of service is correlated with an increase in the CI score, which can lead to a longer time away from work. It should also be noted that age and seniority were statistically highly significant at a 1% level (r = 0.762, *p*-value < 0.01). Thus, the greater length of time on the job implies an increase in the servants' age, which represents a favorable situation for higher CI scores.


**Table 5.** Correlations between CI, working days lost, and sociodemographic variables.

\* The correlation is significant at the 0.05 level (2 extremities). \*\* The correlation is significant at the 0.01 level (2 extremities).

Sickness absenteeism, as already mentioned, is a multifactorial phenomenon; that is, it consists of a variable dependent on several other variables that act as predictors of work absences motivated by health problems. However, the CI proved to be a possible predictor of relative importance, considering the moderate correlation (r = 0.254) with the working days lost by the legislative servants from all LHs.

Regarding the prevalence of comorbidities according to the gender of the participants, no significant differences between genders were observed through the independent t-test (t 445 = 0.039; *p*-value = 0.969). There may be qualitative differences once male and female servants present differences in terms of the type of diseases or health problems.
