Reporting and prediction of work-related sickness absence by general practitioners

Background Information on sickness absence (SA) duration in general practice is difficult to record. The duration of absence certified by general practitioners (GPs) can be viewed as a prognosis for return to work. The Health and Occupation Research network in General Practice (THOR-GP) collects SA information from GPs associated with cases of work-related ill-health. A sample of these cases is followed up 1 year retrospectively to gather information on the duration of absence. Aims To examine the extent of the underestimation of SA in routinely reported data and to investigate how well GPs predict patients' return to work. Methods THOR-GPs submit case and SA information using a web-based form. GPs who submitted selected cases were asked about the total number of days of SA and whether the patient had returned to work. Results THOR-GPs' routine SA data collection underestimated absence duration by 61%. According to the retrospective data, a much larger proportion of periods of absence due to work-related mental ill-health developed into long-term SA (60%) than episodes attributed to musculoskeletal disorders (32%). In over half the reported cases, the return to work was longer than the GP initially predicted. Conclusions THOR-GP prospectively reported SA data underestimated the total length of absence; however, these data can examine the episodic rates of absence within different groups. More accurate longitudinal data can be collected retrospectively. GPs' ability to predict the length of time a patient will be away from work is important to enable treatment and rehabilitation planning in order to decrease the likelihood of a patient falling into long-term SA.