Essays on the Interrelation Between Sickness Allowance, Disability, and Mortality in Finland: National and Ethnolinguistic Perspectives

Disability pension is an important pillar of the modern welfare state. As it is almost always a permanent state, it is important to understand risk factors preceding it, as well as the consequences which disability receipt has for its recipients. Much research has already been conducted in this area: Sickness allowance receipt has been found to be a major risk factor for transitioning into disability pension, and disability pension receipt is linked to a heightened mortality risk, both in working ages and during retirement. However, the aspect of time has so far not been considered sufficiently. Finland has ideal conditions for in-detail analyses for the matter at hand, as it has a reliable population registry which allows for large-scale longitudinal research. The first two research questions, which this thesis seeks to answer, are: 1) How does the association of sickness allowance and disability pension change over the course of 22 years? and 2) Does the negative association of disability pension with mortality extend into the early retirement ages also in Finland? For research in this area, Finland offers an interesting additional angle. About five percent of the population has Swedish as their mother tongue, and, unlike most ethnic minorities, the Swedish-speaking Finns are not marginalized. Instead, they are in a largely comparable position as the Finnish-speaking majority population, yet the Swedish-speaking Finns show health advantages for all three indicators of interest, i.e. sickness allowance, disability pension, and mortality. The second set of research questions this thesis seeks to answer are concerned with whether the observable ethnolinguistic differences in health outcomes also extend to the transition from one indicator of ill health to an indicator of more severe ill health: 3) Are there ethnolinguistic differences in the association of sickness allowance and disability pension? And 4) Are there ethnolinguistic differences in the association of disability pension and mortality in the early years of retirement?

The four research questions are answered by utilizing a sample from the Finnish population register, which encompass five percent of the Finnish-speaking population and 20 percent of the Swedish speaking population and cover the years 1987–2011. The risk for receiving disability pension according to previous sickness allowance receipt in ages 16–60 is analysed for the years 1989–2010. The risk for dying in ages 65–70 is analysed according to labour market status in ages 50–64 for the years 1987–2011. The method of analysis for all research questions is Survival Analysis.

The results show that the risk for receiving disability pension is highly dependent on previous sickness allowance receipt, and the association remains strong even after two decades. The length of sickness allowance receipt is an important factor within the first decade, as those who received sickness allowance for longer than two months face a much higher risk than those who received sickness allowance for a shorter period. However, after about ten years, the risk of both groups converges. By and large, this association is the same for Finnish- and Swedish-speaking Finns, as well as for men and women. Concerning the association of labour market status with mortality in the early years of retirement, previous disability pensioners are the only ones who face a higher risk for retirement mortality. This association is the same for Finnish- and Swedish-speaking Finns, as well as for men and women.

In conclusion, there is considerable variation over time in the association of sickness allowance and disability pension, and the firmly established gradient according to length of sickness absence weakens over time. Disability pension receipt is linked to a heightened mortality risk, and the situation in Finland is similar to what is found in other Western countries. The absence of ethnolinguistic differences in either of the associations discussed above, despite marked differences for being in either state, shows the ethnolinguistic differences for being in either state of ill health are likely to be related to the culture associated with speaking the respective language and that the Finnish welfare system does an adequate job in equalising the adverse outcomes when ill health has had an onset.