[Direct costs of diabetes mellitus in Germany: first estimation of the differences related to educational level]. / Direkte Krankheitskosten von Diabetes mellitus in Deutschland: erste Abschätzung der Unterschiede zwischen Bildungsgruppen.

OBJECTIVE: There are many studies on health inequalities, but these are rarely combined with cost-of-illness analyses. If the cost-of-illness were to be calculated for the individual status groups, it would be possible to assess the economic potential of preventive measures aimed specifically at people from low status groups. The objective of this article is to demonstrate for the first time the preventive potential by taking the example of diabetes mellitus (DM) from an economic perspective. METHODS: Based on a systematic literature review, the average direct costs per patient with DM were assessed. Then, the prevalence of DM among adults with different educational levels was estimated based on the nationwide survey 'German Health Update' (GEDA), conducted by the Robert Koch-Institute in Germany in 2009. Finally, the cost and prevalence data were used to calculate the direct costs for each educational level. RESULTS: The direct costs of DM amount to about 13.1 billion € per year; about 35% of these costs can be attributed to patients with a low educational level. Thus, their share of the total costs is about 67% higher than their share of the total population. If the prevalence in the group with 'low educational level' (14.8%) could be reduced to the prevalence in the group with 'middle educational level' (7.9%), this would save about 2.2 billion (about 16.5%) € of direct costs. DISCUSSION: The analysis provides a first estimate of the potential savings from an effective status specific prevention programme. However, the direct costs per patient used were only an average for all people with DM, as a breakdown by educational level was not available. Since education can also affect health behaviour and compliance, which are also determinants of cost, the analyses presented here are probably conservative.