Vortragssitzung

Health Insurance

Vorträge

Falling sick makes the insurance tick? The effect of illness on administrative costs of health insurance
Lukas Kauer, CSS Institut für empirische Gesundheitsökonomie

Einleitung / Introduction

In a private health insurance market, the premium an insurer charges contains a fee to cover the administrative costs of supplying insurance. We study variation in administrative costs within the basic health insurance for a Swiss insurer and investigate whether a customer’s morbidity causes higher administrative costs. We show that there is a causal relationship between the morbidity of a customer and the number of administrative customer contacts (phone, visits, email etc.) with the insurer. We find a positive and concave relationship between a customer’s health care expenditures and her administrative contacts, with a local expenditure elasticity around the average of 0.27. Using a back-of-the-envelope analysis we measure the costs of an administrative contact and find that administrative cost differences related to differences in the morbidity composition of an insurer’s population are small. For example, the difference in total administrative costs between two insurers, one with a very ill and the other with a very healthy population is about 5 CHF (about 5 €) per customer.


AutorInnen
Lukas Kauer, CSS Institut für empirische Gesundheitsökonomie
Access to novel treatment pathways of oral antidiabetics by health insurance status in Germany
Eva Goetjes, Universität Duisburg-Essen

Einleitung / Introduction

We study equal access to pharmaceutical care for the treatment of type 2 diabetes by health insurance status in Germany (statutory vs. private). Evidence suggests that patients with statutory health insurance receive new drugs significantly slower and that pharmaceuticals without an added benefit are prescribed at a much higher volume. Previous evidence further suggests that dual practice has negative impacts on patient health caused by prioritization of financial interests, skimming of patients by profitability, or over-provision of health care. In the German context, comparisons between patients with private or statutory health insurance have focused on waiting times. This study aims to identify if patients with different insurance status have equal access to pharmaceutical antidiabetic care.

Methode / Method

We used a nationally representative prescriber panel of 1,928 ambulatory care physicians, including all prescriptions of oral antidiabetics (ATC groups A10, excluding insulins and insulin devices, i.e. A10C, A10D, A10E). In total, we observe 322,156 patients (about 6% with private health insurance) and 2,172,801 Mio prescriptions from 2011 to 2014. The underlying hypothesis is that patients with private health insurance receive a more novel oral antidiabetic treatment (mixed therapy of two active ingredients in one tablet (metformin-mix) compared to combination-therapy of two active ingredients in different tablets. To identify the causal effect of a patient’s insurance status on the prescription of metformin-mix, we performed a confounder control study considering physician and time fixed effects and accounting for patient characteristics.

Ergebnisse / Results

Metformin-mix makes up about 15% of the total antidiabetic prescriptions. Of the patients with statutory health insurance, about 13% (vs. 18% of the patients with private health insurance) receive metformin-mix. The analysis shows a significant (p< 0.001) odds ratio (OR) of 1.51 for patients with private health insurance receiving metformin-mix. This translates to a 60% probability of receiving metformin-mix as a privately insured patient. This effect is robust (OR 1.42 - 1.54 (p< 0.001)) when accounting for confounding by time, patient, and physician characteristics.

Zusammenfassung / Conclusion

We provide evidence of differences in prescription of oral antidiabetics based on the patient’s insurance status. We show that patients with private insurance have a higher possibility of receiving metformin-mix than their statutory insured counterparts. As metformin-mix is a more novel one-tablet treatment, this could have an impact on patients’ adherence as well as resulting therapy costs.


AutorInnen
Eva Goetjes, Universität Duisburg-Essen
Katharina Blankart, Universität Duisburg-Essen
Long-Term Health Insurance: Theory Meets Evidence
Nicolas Ziebarth, Cornell University

Einleitung / Introduction

Long-term health insurance contracts can insure policyholders against contemporaneous health expenditure shocks and reclassification risk, constituting an alternative to the community-rated spot contracts implemented in the ACA Exchanges. This paper builds on a recent theoretical literature on the welfare effects of long-term contracts to study the workings of the private insurance market in Germany, where 8 million people buy long-term insurance contracts. Germany provides a particularly appealing test ground for the theory as two of the main theoretical assumptions are met, namely (1) plans constitute a purely financial contract (provider network and provider charges are uniform across all plans), and (2) there is one-sided commitment (individuals can lapse at any time, but contracts have unlimited duration, often until death as Medicare does not exist in Germany).

Methode / Method

We start by building on existing dynamic insurance models, most notably Handel, Hendel and Whinston (2017), to derive the optimal long-term contracts given the health risk evolutions and income dynamics among the enrollee population for Germany. For this purpose, we combine (a) claim data from a unique panel of 400,000 individual long-term health insurance contracts from one of the biggest German private insurers spanning seven years from 2005 to 2011 with (b) income dynamics from the representative German Socio-Economic Panel Study. We analyze welfare under different counterfactual insurance arrangements and compare them to the first-best, the theoretically optimal contracts with one-sided commitment, and the real-world contracts.

Ergebnisse / Results

Although, by design, the GLTHI differs substantially from the optimal dynamic contract, we only find modest welfare differences between the two. Moreover, we show that a simple modification to the GLTHI contract would further close this welfare gap. Finally, we conduct counterfactual policy experiments to illustrate the welfare consequences of introducing GLTHI into a system with a “Medicare-like” public insurance that covers people above 65. Finally, we analyze welfare under different counterfactual insurance arrangements and compare them to the first-best, the theoretically optimal contracts with one-sided commitment, and the real-world contracts.

Zusammenfassung / Conclusion

We find that the simple GLTHI design generates only small welfare losses compared to the optimal contract of GHHW. Under various parameterizations and scenarios, replacing the GLTHI contract with the optimal contract would increase welfare by between zero and seven percent. compared to the optimal contract, the GLTHI contract entails less consumption smoothing over the lifecycle, but also less reclassification risk.


AutorInnen
Martin Karlsson
JP Atal
Haming Fang