Vortragssitzung
Analysis of Health Care Systems 1
Vorträge
The Role of Incomplete Administrative Health Care Data: The Case of a Coverage Expansion in Switzerland
Andreas Kohler, ZHAW / tarifsuisse
Einleitung / Introduction
We explore the role of administrative data in analyzing health care expenditures when the data at hand is incomplete in important dimensions. To do this, we analyze the causal effect of a policy change in Switzerland that extended the coverage of mandatory health insurance towards complementary medicine.
Methode / Method
Using two different administrative data sources and a difference-indifference framework, we estimate the effects of coverage extention on the costs of general practitioners who practice complementary medicine.
Ergebnisse / Results
First, we find that that costs (+7%) per patient increased significantly when looking only at mandatory health insurance. Second, with more complete data, we are able to show that the policy change had no real impact on these physicians’ health care provision and merely shifted expenditures across payers.
Zusammenfassung / Conclusion
We conclude that administrative data are no panacea for empirical health economics and data completeness is crucial to provide meaningful answers to policy questions.
AutorInnen
Andreas Kohler, ZHAW / tarifsuisse
Boris Kaiser, BSS Volkswirtschaftliche Beratung
Christian P.R. Schmid, CSS Institute for Empirical Health Economics
Concentration of hospital capacities and patients’ access to care
Anne Mensen, RWI - Leibniz-Institut für Wirtschaftsforschung
Einleitung / Introduction
Concentration of hospital capacities usually involves closures of smaller hospital sides. While advocates of hospital concentrations emphasize increased quality of care and cost savings, the population fears insufficient care and longer travel distances. In this paper, I explore recent hospital closures in Germany and analyze its effect on travel time and hospitalization rates. By doing so, I analyze the effects for the population that is affected most by the closure, i.e., the people for whom the hospital was the nearest one in their surroundings. Using rich patient-level data, I show that travel time to the nearest hospital increase only slightly for most of the population indicating that concentrations of hospital capacities do not jeopardize accessibility. Nevertheless, their hospitalization rate is decreasing slightly with the closure indicating that patient care seems to be affected.
AutorInnen
Anne Mensen, RWI - Leibniz-Institut für Wirtschaftsforschung
To risk-equalize or not? An analysis of administrative costs in health insurance across five countries
Jürgen Wasem, Universität Duisburg-Essen, Deutschland
Einleitung / Introduction
After more than 30 years of research the performance of risk adjustment for medical expenditures has been substantially improved. This raises the question: assuming we have a ‘perfect’ prediction of everyones’ future medical claims, is then the goal of risk-equalization achieved? The answer is no because in addition to medical claims enrollees also have to pay a loading fee. The loading fee is the excess of the premium above the expected medical claims to be paid by the insurer. If insurers need to charge a higher loading fee for a high-risk than for a low-risk enrollee, incentives for risk selection remain present. Therefore, the question is whether the loading fee should also be (partly) risk adjusted? In this paper we focus on administrative insurance costs which is a clearly demarcated cost category and which is the main cost component of the loading fee.
Methode / Method
We gathered data for the insurers in six markets of five countries: Australia, Germany, the Netherlands, Switzerland and two insurance markets in the US, the small group and the individual market for various years. We constructed an (ex ante) risk-score for each insurer, using the reported risk-equalization payments. As an alternative approach we constructed an (ex post) risk-score by using actual expenditures of the insured. We run a fixed effects panel regression for each of the six markets, with the administrative costs per life year as dependent variable, regressed by an insurers risk score, insurer and year fixed effects, insurers population size and its inverse, and an error term.
Ergebnisse / Results
The share of administrative costs in total expenditures of insurers in this multiple year analysis range from 3.7 % (Netherlands) to 16.0 % (US Small Group Market), with Germany (5%) being at the “lower end”. In the regressions we found for the majority of the markets a positive coefficient for the risk-score for both specifications of the risk score, suggesting that insurers with a high-risk population have higher administrative costs per capita. Using the results from the regressions, a formula for risk adjustment of administrative costs can be calculated. For instance, for Germany according to this formula risk adjustment should be for 59% of the administrative costs of insurers.
Zusammenfassung / Conclusion
In this multi-country comparison we have studied the role of administrative costs in risk adjustment between competing health insurance carriers. Our analysis shows that administrative costs increase with the risk score of the health insurers. As a consequence, to avoid incentives for risk selection administrative costs should partly be risk-adjusted. This is actually done in Germany and the US, whereas it is not done in the Netherlands, Switzerland and Australia, which leaves room for risk selection in these three countries.