Evaluating purchaser performance in promoting quality through budget reallocations
Niek Stadhouders | Radboudumc
Objective: It is unclear whether single payer systems outperform multiple payer systems in terms of value-for-money. Single payers may hold more leverage over providers, whereas multiple payers have stronger incentives due to competition on the purchaser side. One means for purchasers to increase value-for-money is to reallocate resources from low-performing to high performing providers. Purchasers may also reallocate budgets based on patient needs. The dynamics resulting from such reallocations can be seen as an indicator for purchaser performance. In the Netherlands, single and multiple payer purchasers appear alongside each other, enabling a comparison. The Dutch have 9 competing insurers purchasing primary and hospital care (multiple payers), and 32 non-competing regional care offices purchasing long-term care (single-payers). Our aim is to research how each purchasing system performs in terms of budget reallocations.
Data and methods: We construct a theoretical budget reallocation model based on purchasers aiming to maximize patient welfare under a principal agent relationship. In our model, budget reallocations increase bankruptcies and adversely affect continuity of care. As a result, reallocation outcomes depend on changes in patient needs, differences in provider quality and the relation between reallocations and continuity of care. Purchasers aim to increase market dynamics until loss of continuity of care through bankruptcies outweigh the positive effects of improved needs adherence and provider quality. In order to test the relative effect of these factors on market dynamics, we empirically estimate a linear fixed effects panel model.
To this aim, we collect annual provider financial statement data of all Dutch health providers, demographic data from the Dutch Statistical Bureau and publicly available quality data from the Dutch Quality Institute for the years 2007 to 2014. Correcting for mergers we track budget allocations over time. Next, we calculate the share of the total budget reallocated between providers over time in different purchaser systems. We document changes in national policy to isolate the effect of purchasing on market dynamics. Payer system comparisons may be biased as short term hospital care enables faster budgetary changes compared to long term care. To reduce bias, we distinguish long term trends in reallocation from short term fluctuations. Using market dynamics as dependent variable, we estimate the fixed effects model. We use purchaser characteristics, provider quality, patient needs and bankruptcy rate as independent variables. To check for robustness, we use different combinations of control variables.
Results: Current results are preliminary. Single payers reallocate about 3% of total long-term care resources between providers annually, while multi-payers reallocate about 4% for hospital care.
Policy implications: Governments have been aiming to improve purchasing, both in multiple payer systems as in single payer systems. The role of purchaser budget reallocations as a means of stimulating value-for-money is often overlooked. Further insights into effects of purchasing on market dynamics allow policy makers to promote purchaser efforts in obtaining high value care.