Sickness absence and disability pension trajectories among individuals on sickness absence due to stress-related disorders. Two prospective population-based cohorts with 13-month follow-up

Abstract

Background Stress-related disorders are common diagnoses for sickness absence (SA) and disability pension (DP) in many Western countries. Knowledge on future SA/DP trajectories among those starting such a SA spell is limited. The aims were to identify future SA/DP days trajectories among individuals starting an SA spell due to stress-related disorder and investigate socio-demographic and morbidity characteristics associated with specific trajectories. Methods Using microdata from nationwide registers, we established two cohorts of all living in Sweden who started a new SA spell >14 days due to stress-related disorder in 2011 (N = 32,417) or in 2018 (N = 65,511), respectively. Group-based trajectory models were used to identify trajectories of monthly average SA/DP days during the following 13 months, separate for each cohort. We used multinomial logistic regression to investigate the associations between sociodemographic and morbidity-related predictors and trajectory membership. Results We identified six SA/DP trajectories in the two cohorts: steep drop (30.6% and 35.9% of all included in 2018 and 2011); constant fluctuating (8.7%, 11.2%); fast decrease (25.5%, 24.4%); medium decrease (18.1%, 13.1%); slow decrease (10.8%, 7.3%), and constant high (6.2%, 8.0%). The distributions of sociodemographic factors, multi-morbidity, and history of SA/DP differed between the trajectory groups. For example, compared to the steep drop trajectory, individuals in the other trajectories were more likely to be a woman, older, having had prior SA/DP or specialized outpatient healthcare visits. Conclusions In these two explorative, population-wide cohorts, we identified six different trajectories of SA/DP days among all with a new SA spell with stress-related disorders. The trajectory groups differed regarding both sociodemographic and health-related covariates.