Expansion or compression of care need in Germany between 2001 and 2009: a small-area study based on administrative health data
Daniel Kreft, Universität Rostock
Gabriele Doblhammer-Reiter, Universität Rostock
Background: A much-discussed question in public health research is whether the two health scenarios - expansion or compression of morbidity - are heterogeneous on the sub-national level. Further, we aim to detect if the trends in morbidity or in mortality are the decisive drivers of the CFLY and of the health scenarios. Methods: This study uses administrative census data of all beneficiaries in Germany from the Statutory Long-Term Care Insurance 2001-2009. We compute care need-free life years (CFLY) and life years with care need (CLY) at age 65+ for 412 counties. The CFLY and CLY gains are decomposed into the effects of survival and of the prevalence of care need and we investigate their linkages with the health scenarios by applying multinomial regression models. Results: We show an overall increase in CFLY, which is higher for men than for women and higher for severe than for any care need. However, spatial variation in CFLY and in CLY has increased. In terms of the health scenarios, a majority of counties show an expansion of any care need but a compression of severe care need. However, we detect expansion counties surrounding a compression county and vice versa. That high spatial heterogeneity is mainly caused by divergent trends in the prevalence. We show that mortality is responsible for the absolute changes in CFLY and CLY while morbidity is the decisive driver that determines the health scenarios. Conclusion: We combine a regionalized administrative data source and advanced statistical methods to get deeper insights into epidemiological processes. Our findings demonstrate a compression of life years with severe care need, which however, depends on the region of residence. To attenuate regional inequalities, more efforts are needed that improve health by medical and infrastructural interventions. In future research, the underlying mechanisms should be investigated in more detail.
Presented in Session 38: Health, well-being and morbidity