Supporting Ageing Parents and Changes in Quality of Life in Sweden and Denmark

Emily Grundy, Institute for Social and Economic Research, University of Essex
Thijs van den Broek, London School of Economics and Political Science

Denmark and Sweden are often grouped together in welfare state typologies and presented as countries where the coverage of care services is much more generous than elsewhere in Europe. In recent decades, however, reforms have resulted in a marked divergence between the long-term care arrangements of Denmark and Sweden, with coverage remaining at higher levels in the former than in the latter country. We investigate whether these emergent country differences in the organization of long-term care have (1) resulted in country differences in the likelihood of adult children becoming providers of care to ageing parents, and (2) led to differences between Denmark and Sweden, and between time periods, in the quality of life of adult children who become caregivers for parents. Logistic regression analyses of four waves of SHARE data did not provide support for the hypothesis that Swedish adult children were more likely to become caregivers than their Danish counterparts. Fixed effects regression analyses showed that taking on the caregiver role was more detrimental for quality of life in Sweden than in Denmark. This finding may be related to support policies for caregivers being more extensive in Denmark than in Sweden.

Supporting ageing parents and changes in quality of life inSweden and Denmark

Denmark and Sweden are often grouped together in welfarestate typologies and presented as countries where the coverage of care servicesis much more generous than elsewhere in Europe. In recent decades, however,reforms have resulted in a marked divergence between the long-term carearrangements of Denmark and Sweden, with coverage remaining at higher levels inthe former than in the latter country. We investigate whether these emergentcountry differences in the organization of long-term care have (1) resulted incountry differences in the likelihood of adult children becoming providers ofcare to ageing parents, and (2) led to differences between Denmark and Sweden,and between time periods, in the quality of life of adult children who become caregiversfor parents.

Dataand methods

Weuse data from wavesI, II, V and VI of the Swedish and Danish components of the Survey of Health,Aging and Retirement in Europe (SHARE). We restricted our sample to men andwomen aged 50-75 for whom data were available in waves I and II or waves V andVI, who had at least one living parent at baseline and had not provided supportto parents in the 12 months prior to the baseline interview. These selectioncriteria resulted in a sample of 1,924 observations nested in 961 women andmen. Multiple imputation was used to deal with missing information on variablesof interest.

Wefirst estimated logistic regression models to predict providing care to a parent withhealth limitations at follow-up. In a second set of models, we use fixedeffects linear regression models to study intra-individual change in midlifepersons’ quality of life. We used a difference-in-difference design. To testwhether the association between providing support to ageing parents and qualityof life was more strongly negative in Sweden than in Denmark, we estimated amodel in which we allowed the effect of the transition to the role of careprovider to vary by country. In our most complex model, the country differencesin the coefficient estimates of providing care to ageing parents were, in turn,allowed to differ between the 2004-2007 period (Waves I and II) and the2013-2015 period (Waves V and VI).  This enabled us to test our hypothesis thatthe difference between Sweden and Denmark in the effect of providing support toageing parents on quality of life increased over the studied period.

Results

Our logistic regression analyses did not provide support forour hypothesis that Swedish adult children were more likely than their Danishcounterparts to become caregivers at follow-up. Table 1 shows the results ofour fixed effect regressionanalyses of quality of life, measured with the CASP-12 instrument. The first model shows thatstarting to provide support to ageing parents is associated with a reduction ofquality of life. In the second model, we allowed the effect of supportprovision to differ between Denmark and Sweden by adding an interaction term.The model shows a marked negative effect of support provision on quality oflife in Sweden. As hypothesized, the negative effect of support provision was substantiallyweaker in Denmark than in Sweden.  The effect of support provision was also notstatistically significant in Denmark. In a third model (not shown in Table 1),we tested whether the country difference in the effect of support provision hasbecome larger over time. No support was found for this hypothesis.

Discussion

Inrecent decades, coverage of long-term care services has declined more stronglyin Sweden than in Denmark. Arguably, this has led to a stronger appeal on adultchildren to provide care to ageing parents in the former country.  Although wedid not find that Swedish adult children were more likely than their Danishcounterparts to take on care tasks, our analyses showed that taking on thecaregiver role was more detrimental for quality of life in Sweden than inDenmark. This finding may be related to support policies for caregivers beingmore extensive in Denmark than in Sweden.

Presented in Session 1123: Ageing and Intergenerational Relations