Abstract
Purpose This study aimed to identify trajectories of sickness absence in workers on sick leave due to musculoskeletal disorders and explore the association between these trajectories and established prognostic factors for sickness absence. Methods We conducted a prospective cohort study of 549 workers (56% women, aged 18-67 years) on sick leave due to musculoskeletal disorders in Norway in 2018-2019. Sickness absence data were collected from the Norwegian sick leave registry and prognostic factors via self-reported baseline questionnaires. We used group-based trajectory modelling to define the different trajectories of sickness absence spanning a 1-year period. Multivariable multinomial logistic regression was used to estimate odds ratios and 95% confidence intervals for prognostic factors associated with the identified trajectory groups. Results We identified six distinct trajectories of sickness absence over 1 year: 'fast decrease' (27% of the cohort): 'moderate decrease' (22%); 'slow decrease' (12%); 'u-shape' (7%); 'persistent moderate' (13%); and 'persistent high' (18%). Prognostic factors, such as previous sickness absence days, return-to-work expectancy, workability, multisite pain, and health scores, differentiated between the sickness absence trajectories (all P < 0.05). Negative return-to-work expectancy was associated with the three trajectory groups with the highest number of sickness absence days ('slow decrease', 'persistent moderate', and 'persistent high'). Conclusions This is the first study to explore the association of return-to-work expectancy with trajectories of sickness absence. Our findings highlight different patterns of sickness absence and the complex range of prognostic factors. These findings have implications for secondary and tertiary prevention strategies for work absence in workers with musculoskeletal disorders.