Sickness absence trajectories following labour market participation patterns in Catalonia, 2012-2014

Background

Working life is characterized by transitions between different employment status which could affect future health status. Previous studies on sickness absence (SA) have focused on risk factors in the workplace; however, there is scarce evidence regarding labour market participation (LMP) patterns. The aim of this study is to examine the association between prior LMP patterns and the course of SA.

Methods

Cohort study based on a sample of 11,968 salaried workers affiliated with the Spanish Social Security system, living in Catalonia, who accumulated more than 15 days on SA at least in one quarter during 2012-2014, from three working life cohorts according to the working life stage in 2002: early (18-25 years), middle (26-35 years) and late (36-45 years). Sequence analysis was used to identify LMP patterns (2002-2011). Latent class growth analysis was applied to identify SA trajectories (2012-2014). Finally, crude and adjusted odds ratios (aOR) were estimated using multinomial logistic regression models.

Results

Overall, four SA trajectories were identified: low stable (83%-88% of the workers), decreasing (5%-9%), increasing (5%-11%) and high stable (7%-16%) accumulated days on SA, for men and women. Similarly, seven LMP patterns were obtained: stable employment (63%-81%), increasing employment (5%-22%), delayed employment (7%-8%), decreasing employment (4%-10%), varying employment (13%-14%), steeply decreasing employment (9%), and steeply labour market exit (8%). No significant associations were observed among LMP patterns and SA trajectories, except for young men, where an increasing employment pattern was significantly associated with a lower risk to increase days on SA over time (aOR: 0.21 [95% CI: 0.04-0.96]).

Conclusions

A prior 10 years of LMP pattern does not seem to show an effect on the course of SA. A closer working life to the SA course could be considered to assess this relationship.

Funding: Grants FIS PI17/00220 and PI14/00057

Key messages

A longitudinally approach is warranted to evaluate the relationship between working life and sickness absence. Extended prior working lives are not related to the course of future sickness absence.

Employment and health of working parents: a joint effort to uncover hidden treasures in birth cohorts

Key messages

Birth cohorts hide a treasure of occupational and health information of young working parents. OMEGA-NET has plans to dig them out to investigate their interplay.

The effect of multimorbidity on sickness absence by specific diagnoses

BACKGROUND

As the world's population ages, the prevalence of multiple chronic and non-chronic health-related conditions is increasing. Research on multimorbidity, the co-occurrence of two or more health-related conditions, has mainly involved patient and older populations. Its effect in working populations, presumably younger and healthier, is not well known but could conceivably affect sickness absence (SA) and ability to return to work.

AIMS

To examine the effect of multimorbidity on the incidence and duration of SA episodes by frequent diagnostic groups.

METHODS

A prospective study (in 2006-2008) of workers in Spain. Information on health-related conditions was gathered with a standardized questionnaire and used to construct a sex-specific multidimensional multimorbidity score (MDMS). In order to estimate the effect of MDMS on incidence and duration of SA episodes due to cardiovascular diseases (CVD), musculoskeletal disorders (MSD) and mental health disorders (MHD), we fitted Cox models adjusted by age, occupational social class and number of prior SA episodes for both sexes.

RESULTS

The study population was 372370. Men with high MDMS showed a trend towards higher incidence risk for SA due to CVD and MSD [adjusted hazard ratio (aHR) = 2.03; 95% confidence interval (CI) 1.48-2.78 and aHR = 1.20; 95% CI 1.01-1.43, respectively]. Women showed a similar trend for MSD, but MHD had the strongest association (aHR = 4.78; 95% CI 1.97-11.62) for high MDMS. In both sexes, the effect of MDMS was strongest among those without a prior SA. No consistent associations with SA duration were observed.

CONCLUSIONS

Multimorbidity increased the risk of incident musculoskeletal, mental and cardiovascular SA episodes but not their duration.