Diabetes is a widespread disease and poses major challenges for statutory health insurance providers due to its costly treatment and rising prevalence rates. However, the starting points of individual health insurance providers vary considerably, as recent figures from INSIGHT Health on outpatient drug therapy with antidiabetics in general and insulin in particular show.

> Among 20- to 79-year-olds, the proportion of those affected by type 2 diabetes mellitus is almost nine percent, according to the recently published prevalence figures from the fourth IDF Diabetes Atlas. According to IDF calculations, the national prevalence will rise significantly to over ten percent in the next 20 years (see www.diabetesatlas.org). Diabetes not only severely limits the quality of life of those affected, but also entails enormous costs. Diabetes is considered the most expensive chronic disease in Germany: The direct costs for the diagnosis and treatment of diabetes and its complications are estimated at €14 billion annually (see ABDA press release: One in eight pharmacy customers has elevated blood sugar levels, November 13, 2009), with a large portion borne by statutory health insurance. The treatment of diabetics therefore represents a significant expenditure for health and social insurance systems.
There are significant differences between the individual statutory health insurance providers in the outpatient care of patients with antidiabetic medications. This will be illustrated by the prescriptions of the AOKs, the two largest substitute health insurance funds, BARMER (BEK) and Techniker (TK), as well as the Knappschaft (Miners' Welfare Fund).
The selected sample thus encompasses almost 39 million of the approximately 70 million people with statutory insurance. Of these, 24.0 million are insured by the AOKs, 6.8 million by the BEK, 6.3 million by the TK, and 1.7 million by the Knappschaft (miners' insurance fund) (source: dfg / INSIGHT Health). Due to its unique insured structure, the Knappschaft supplements the sample with 59.5 percent pensioners. By comparison, the average proportion of pensioners among substitute health insurance funds is just 24.2 percent, while the AOKs have a pensioner share of 31.9 percent (source: BMG, 2009).
As a first step, the number of insured persons treated with antidiabetics (ATC group A10 antidiabetics) in 2008 will be compared. This will be based on an extrapolation based on patient tracking by INSIGHT Health. The extrapolation is based on approximately 10 percent of all billed statutory health insurance prescriptions, so the sample can be assumed to be sufficiently representative for the extrapolation. In addition, the antidiabetic prescriptions recorded as part of a comprehensive survey, including the associated expenditures, will be analyzed based on pharmacy selling prices (source: National prescription information from INSIGHT Health).

Unequal distribution of diabetics among health insurance companies
The proportion of insured persons treated with antidiabetics varies significantly between health insurance funds: While 16 percent of Knappschaft insured persons received at least one antidiabetic medication in 2008, the figure was just three percent for TK insured persons. At AOK, just over one in ten insured persons received an antidiabetic medication, while at BEK, just under seven percent received an antidiabetic medication prescription.
The analyzed gross expenditures – excluding discount agreements, co-payments, etc. – amount to 1.2 billion euros. However, a balanced distribution of expenditures between the health insurance funds is by no means evident. While a TK insured person accounts for the relatively lowest pharmaceutical expenditures for antidiabetics, at around 15 euros on average, this picture is reversed when considering the costs per treated insured person: The gross expenditure per treated TK insured person, at 458 euros, is approximately 140 euros higher than that of an AOK insured person (see Figure 1). This is due both to the higher expenditure per prescription (60 euros vs. 54 euros) and the higher number of prescriptions (7.7 vs. 5.9 prescriptions per year). This indicates that while fewer TK insured persons suffer from diabetes, those affected incur significantly higher costs for outpatient drug therapy.
Analyzing the insulin subgroup (ATC group A10C according to EphMRA), the sample includes approximately 1.4 million insured persons and 7.9 million prescriptions, with gross expenditures of approximately €756 million. This results in a pattern almost analogous to that for antidiabetics overall, i.e., higher prevalence rates among the Knappschaft (mining insurance funds) and the AOK (a public health insurance fund), and lower ones among the substitute health insurance funds. Approximately 39 percent of insured persons treated with antidiabetics received insulin at least once in 2008.
It also shows that while the morbidity of TK insured persons is lower—only one percent are treated with insulin—insulin prescriptions are, on average, more expensive (TK: €103 vs. AOK: €95) and also more frequent (TK: 7.5 vs. AOK: 5.4 prescriptions per year). The average gross annual expenditure for insulin for insured persons treated with insulin amounts to over €776 at TK, €264 more than at AOK. The Knappschaft (Miners' Insurance Fund) has the highest proportion of insured persons treated with insulin: 6.4 percent of insured persons received insulin at least once as part of outpatient treatment in 2008.
Influence socio-demo-
graphic features the supply?
The Allgemeine Ortskrankenkassen (AOKs) and the Knappschaft (German Social Insurance Fund), with their averagely older insured population, have a higher proportion of diabetics. Other health insurance funds, such as the substitute funds BEK and TK examined, have a different insured structure and relatively fewer diabetic patients among their insured. Nevertheless, the costs per patient for outpatient treatment at AOK and Knappschaft are comparatively low, as a look at the higher figures at TK reveals.
One possible explanation for TK's increased expenditures for its diabetes patients could be that the average TK patient is generally "healthier" and thus cost-neutral. However, if they become ill, they have higher expectations of their treatment and higher treatment compliance – possibly due to a higher average level of education – and thus incur comparatively higher expenditures. Further health services research analyses could address this hypothesis.

Health insurance companies also differ in their diabetes programs
An analysis of the proportion of pensioners among the insured per health insurance fund provides an initial sense of the sociodemographic and morbidity-related factors that are partly relevant. In addition to age (measured here by the proportion of pensioners), numerous other variables are associated with diabetes. In addition to hereditary factors and the discussed educational level, weight (body mass index, BMI) and lifestyle (e.g., diet and physical activity) play a role. The latter factors are at least partially modifiable and thus provide starting points for preventive measures.
In this context, health insurance companies choose different ways of accessing their diabetes patients: In addition to large-scale offers such as structured treatment programs (Disease Management Programs, DMP), the BEK, for example, is pursuing patient access via more than 3,000 priority
pharmacies. TK, on the other hand, is involved with an online offering such as the "TK Diabetes Coach" in addition to the "TK Plus" DMP offering. Furthermore, TK is currently conducting a scientific study on diabetes prevention in cooperation with the German Diabetes Center.

by: Christian Bensing /
Dr. André Kleinfeld*