Prevalence and Trends of Dementia and Parkinson's Disease Based on Comprehensive German Health Claims Data

Gabriele Doblhammer, University of Rostock
Michael Nerius, Rostock Center for the Study of Demographic Change

Background: Dementia and Parkinson’s disease are the most common neurodegenerative diseases at old age. So far, estimations of prevalences and trends in Germany may be biased because (1) community-based surveys mostly cannot reach institutionalized and critically ill persons which leads to an underestimation and (2) studies based on health claims data from single health insurance funds lack representativeness which might also leads to under- or even overestimation of the respective disease.

Methods: We used newly available health claims data sets of the years 2009-2011 comprising all Germans insured in statutory health insurance funds reflecting 87.5% of the German population. The analysis population consisted of 14.5 million persons aged 65 to 100+ living in their community or in nursing homes. Age-standardized prevalences for dementia and Parkinson’s disease were calculated for both sexes. The time-trend over all ages combined was estimated by negative binomial regression using 2009 as the reference year.

Results: Over all ages, the yearly decline in dementia was 0.7% (RR=0.993, p-value=0.103, 95% CI=0.984-1.002) among men and 1.5% (RR=0.985, p-value=0.001, 95% CI=0.979-0.991) among women. Parkinson’s disease trends were not significant among both sexes (Men: RR=1.006, p-value=0.101, 95% CI=0.999-1.014; Women: RR=0.999, p-value=0.760, 95% CI=0.992-1.006).

Conclusions: Using newly available health claims data sets we were able to assess the actual, unbiased burden of dementia and Parkinson’s disease in Germany. For dementia, we could demonstrate a decline among both sexes, albeit starting at different ages. Parkinson’s disease trends were inconsistent over age and between the two sexes, and there was no significant general trend. Due to the relative short observation period from 2009-2011 secular trends should be derived cautiously. However, we will expand this period up to 2013 which gives us the possibility to present more robust insights on trends of dementia and Parkinson’s disease.

Presented in Session 6: Depression and Quality of Life