Hospital payment and volume of care

Objectives: We examine the effect of hospital payment on volumes of care based on a reform of hospital financing in Germany. The focus is on treatments with a high degree of regional variation where induced demand is most suspected, such as cataract surgery, C-sections, hip replacements, etc.

Background: According to the theory of physician-induced demand (Evans 1974) physicians weigh the benefits of adhering to ethical and medical standards against the benefits of higher revenues. This theory predicts that physicians will induce additional demand if they face declining prices. Whether medical providers induce demand – and to what extent- is a controversial question in health economics. Previous studies have mostly focused on the relationship between the regional concentration of health-care providers and the volume of medical treatments (Douven et al. 2011, Dranove and Wehner 1994, Gruber and Owings 1996). An essential difficulty with this approach is that location decisions of health care providers also follow market demand. As an alternative approach we directly examine the effect of price changes on volumes of treatment.

Institutional Setting: Starting from the year 2003 hospital payment for publicly insured patients in Germany was transformed from a system of negotiated budgets to a system where hospital reimbursement is based on the main diagnosis of each patient. A particular aspect of the German reform which sets it apart from similar reforms in other countries is that payment changes were introduced gradually. At the start of the reform, payment rates for the same diagnosis varied widely between hospitals according to historical costs. Between the years 2005 and 2009 payment rates were gradually equalized across hospitals. Thus, payment rates increased for some hospitals, and they decreased for other hospitals. This variation in payment rates allows us to estimate the causal effect of hospital payment on volumes of care.

Methods: Starting point for our empirical analysis are long-difference regression models which regress changes in treatment volumes on changes in payment rates over alternative time periods. One concern is that hospitals’ volumes of care will be affected by the decisions of neighboring hospitals. We take this concern into account by aggregating patient numbers by county of residence and regressing changes in aggregate treatment volumes on changes in average payment rates at the county level. We also examine if volumes of care react asymmetrically to price increases vs. price decreases.

Data: The analysis is based on administrative data provided by the German statistical office for on-site access. These data include detailed information on all admissions to German hospitals for the period from 2000 to 2009, merged with information on hospital specific payment rates (Basisfallwerte).

Results: We find that a 1% decrease in payment rates leads to an increase in the number of hospital admissions by 0.35% for the period from 2004 until 2007, and to an increase of 0.26% for the period from 2004 until 2009.  This suggests that hospitals were able to partially offset the effects of lower payment rates by increasing volumes of care.