Several risk factors for transition from long-term sickness absence into disability pension were identified. The finding that spells of sickness absence with duration up to seven months did not imply increased risk of disability during the first three years may have implications for interventions aimed at long-term sickness absentees.

Besides socio-demographic risk factors, the sick-leave diagnoses constitute an important both medium and long-term predictor of disability pension among both men and women on long-term sickness absence.

In addition to previously known socio-demographic predictors, medical variables were important in identifying sickness absentees with an increased risk of DP.

Annual cumulative incidence of long-term sickness absence was 6.5% for women and 4.9% for men. During follow-up, 12.4% of the women and 12.6% of the men received disability pension. Among full-time employed women only 10.3% had become disability pensioners, while the corresponding proportion for women working part-time was 15.5%. For men the figures were 12.1% (full-time) and 18.1% (part-time). In the logistic regression of the whole sample the female odds ratio was insignificant. The dominant predictive factors for disability pension were age and duration of the sickness spells. Working part-time also increased the risk. Higher levels of education and having children below 7 years reduced the probability for disability pension. Separate regressions for men and women showed that the 'protective' effect of having small children only remained for women.