Labour market participation after sickness absence due to cancer: a dynamic cohort study in Catalonia (Spain)

Ayala García, Amaya; Benavides, Fernando G. (Fernando García); Serra Saurina, Laura

doi:http://dx.doi.org/10.1186/s12889-023-17321-z

Labour market participation after sickness absence due to cancer: a dynamic cohort study in Catalonia (Spain)

License

Citation

Ayala-Garcia A, Benavides FG, Serra L. Labour market participation after sickness absence due to cancer: a dynamic cohort study in Catalonia (Spain). BMC Public Health. 2023 Dec 11;23(1):2477. DOI: 10.1186/s12889-023-17321-z

Abstract

Background: The consequences of cancer on working until retirement age remain unclear. This study aimed to analyse working life considering all possible labour market states in a sample of workers after sickness absence (SA) due to cancer and to compare their working life paths to those of a sample of workers without SA and with an SA due to other diseases. Methods: This was a retrospective dynamic cohort study among social security affiliates in Catalonia from 2012-2018. Cases consisted of workers with an SA due to cancer between 2012-2015 (N = 516) and were individually age- and sex-matched with those of affiliates with an SA due to other diagnoses and workers without an SA. All workers (N = 1,548, 56% women) were followed up from entry into the cohort until the end of 2018 to characterise nine possible weekly labour states. Sequence analysis, optimal matching, and multinomial logistic regression were used to identify and assess the probability of future labour market participation patterns (LMPPs). All analyses were stratified by sex. Results: Compared with workers with an SA due to cancer, male workers with no SA and SA due to other causes showed a lower probability of being in the LMPP of death (aRRR 0.02, 95% CI: 0.00‒0.16; aRRR 0.17, 95% CI: 0.06‒0.46, respectively) and, among women, a lower probability of permanent disability and death (aRRR 0.24, 95% CI: 0.10‒0.57; aRRR 0.39, 95% CI: 0.19‒0.83, respectively). Compared to workers with SA due to cancer, the risk of early retirement was lower among workers with no SA (women, aRRR 0.60, 95% CI: 0.22‒1.65; men, aRRR 0.64, 95% CI: 0.27‒1.52), although these results were not statistically significant. Conclusions: Workplaces, many of which have policies common to all diagnoses, should be modified to the needs of cancer survivors to prevent an increasing frequency of early retirement and permanent disability when possible. Future studies should assess the impact of cancer on premature exit from the labour market among survivors, depending on cancer localisation and type of treatment.