Sickness absence in the Norwegian workplace doubled in the period 1993-2003. However, the extent to which the driving factors were medical or non-medical remains unclear, as does the extent to which the cause may be found in the composition of the workforce.
A differences-in-differences regression model was used to estimate the added sickness absence associated with major life events such as separation, death of spouse and pregnancy in the period 1993-2005. The data were obtained from administrative registers covering the entire Norwegian population, and include all absence periods of 16 days' duration or more reported by a doctor's medical certificate. The primary outcome measures were incidence (the proportion of absentees in a given time window) and absence (the proportion of sick days in a given time window). The level of absence among employees exposed to the specified life events was compared to control groups matched for gender, age, education and income.
In 1993, people in each of the three groups exposed to major life events had more frequent and longer periods of absence than people in the control groups. This added sickness absence increased between 1993 and 2005. The changes in added sickness absence were at times significant, particularly for pregnant women. While sickness absence among pregnant women in 1993 was 15.4 percentage points higher than in the control group, the difference had increased to 24.8 percentage points in 2005.
We find it improbable for the increase in added sickness absence to be caused by changes in the medical impact of life events or alterations in the workforce composition. We believe the increase is caused by changing attitudes among the working population and in the medical profession towards sickness absence on grounds that are not strictly medical, combined with improved social acceptance and diagnosis of mental health issues, and/or a medicalisation of natural health variations (pregnancy) and emotional distress (grief).