There exists a body of literature which individually identifies job characteristics determining longer sick absences. To date, however, these factors have not been considered simultaneously to draw trajectories on individuals’ employment perspectives. This study uses an exhaustive dataset compiling Belgian social security data which identifies individuals on sick leave and their potential transitions to other (in)active statuses. Advanced survival methods are used to identify workers more at risk of long-term work incapacity and to draw trajectories depending on different exit routes (i.e. (un)employment, career break, retirement, etc.). The authors complement the findings by analysing job prospects through an event study. The evidence shows that workers with relative higher job stability and those working for employers with higher reallocation capabilities returned to work more quickly. Conversely, workers in the health care and education sectors stayed away longer from the labour market. This extends previous findings on how sick presenteeism (going to work despite feeling ill) and the difficulty to spare workload can aggravate sicknesses. The final contribution of this paper was to find evidence that workers reduce their volume of work and revenue after a period of at least 6 months on sick leave. Diverging effects were found depending on the type of worker: low-skilled and part-time workers were less affected.