Variations in the risk of disability pension in Norway 1970-99. A gender-specific age-period-cohort analysis

AIMS: A study was undertaken to investigate whether cohort or period effects could explain the varying and generally increasing incidence of disability pension in Norway between 1970 and 1999. METHODS: The study used data from a complete national register of new disability cases in Norway, including all cases of disability pension in the 16-60 age group categorized according to age and gender for each year from 1970 to 1999. The population at risk was defined for each year from census data and number of individuals already receiving disability pension. Data were organized in five-year age groups, five-year time periods and corresponding overlapping nine-year birth cohorts. Age- and gender-specific rates were displayed graphically for periods and cohorts. Separate Poisson regression models were fitted for age periods and age cohorts. Finally a combined age, period, and cohort model was applied. RESULTS: The overall incidence was 7.4/1,000 non-disabled persons per year for women and 6.0/1,000 for men. For women 52.1% of the cases were in the 51-60 age group, whereas the corresponding figure for men was 57.6%. Statistical analysis showed an increasing trend for both genders, more pronounced for women than men. All time periods deviated significantly from the trend, either upwards or downwards. Age-cohort models showed less variation, but recent cohorts had higher than expected rates, especially for men. CONCLUSIONS: Further studies should investigate why Norwegian women were more affected by the period effects than men. An increasing incidence of disability pension among recent cohorts is a major challenge for the Norwegian welfare system.