Case mix in the Dutch health care system: Is case mix reflected in the negotiated price of a DTC?

Summary Introduction In 2006, the Dutch health care system has shifted towards a more market-oriented health care system for specialist medical care. With the introduction of the Diagnosis Treatment Combination (DTC) system, specialist medical care has become a predefined interchangeable product. Negotiations on price, volume and quality are expected between hospitals and health care insurers. Each DTC can be priced differently per hospital and health care insurer. Price differences can be caused by differences in patient population, based on clinical severity and patient characteristics, which influences care consumption and resource use. According to the literature, there is a need to adjust for case mix to avoid selection and to create an appropriate reimbursement. Therefore, this study investigated whether case mix is discussed in the negotiations and whether case mix is reflected in the negotiated price of a DTC. This study used age, gender, continuation of indication, referral, chronic illness, multiple DTCs, diabetes and socioeconomic status (SES) as parameters of case mix. The following diagnoses were involved in this study: knee osteoarthritis, cataract, meniscus lesion and hernia of the nucleus pulposus (HNP). Methods This study used a mixed method approach. Quantitative data were obtained from the database of a Dutch health care insurer. This database contained information of the case mix parameters and negotiated prices over the years 2007 to 2009 for the six DTCs that were involved in this study. Although the original dataset consisted of observations per insured, the data were aggregated on hospital level. The relationship between the case mix parameters and the negotiated price of a DTC was tested with Pearson correlation tests, multiple hierarchical regressions, which adjusted for the effect of type of institution, and fixed-or random-effects regressions. In the correlation tests and hierarchical regressions, a weight for hospital size was included. For the qualitative analyses, data were obtained from semi-structured interviews which were conducted with seven respondents from hospitals and from a health care insurer who were involved in the negotiations between hospitals and health care insurers. Important subjects of the interview were the respondents‟ thoughts about case mix and the role of case mix, the necessity to adjust for case mix and the practical implications of the results of the quantitative analyses. The interviews were analyzed by coding and categorizing the transcribed text. Results The results of the hierarchical multiple regressions, which were performed for each year separately and adjusted for type of institution, did not show a clear relationship between the case mix parameters and the negotiated price of the DTC except for the diagnosis cataract. Over the years, the adjusted R squares range from 9.8% to 16.1%. Significant parameters were gender, multiple diagnoses, referral and chronic illness. The variable for SES2 showed a significant negative contribution which was in line with the results of the correlation test. The results of the panel data regressions, which combined the data of all years, showed that the case mix parameters explained a significant part of the variance in the price for the diagnoses cataract (Adj. R2=0.478) and meniscus lesion (Adj. R2=0.131). However, for meniscus lesion the hierarchical multiple regressions only showed a significant relationship between the case mix parameters and the price in 2009 (Adj R2=0.07) and not in 2008. Therefore, it can be concluded that case mix differences are only reflected in the price of cataract. Based on the qualitative analyses, it can be concluded that case mix played a limited role during the negotiations. In addition, case mix was not directly translated in the price of a DTC, but only indirectly by differentiating between types of institutions or by encouraging hospitals to use their cost prices. Some respondents mentioned that a direct translation would be desirable if a relationship between case mix and costs could be proved. However, practical problems as the lack of clear criteria and a uniform definition of case mix impede this. Discussion In this study, case mix is defined very broadly based on the literature. However, not all case mix parameters from literature were used in the qualitative and quantitative analyses since this study focused on case mix differences within a DTC. Although during the interviews respondents mentioned that comorbidity and age were important parameters of case mix, the currently performed quantitative analyses gave no clear evidence for a relationship between case mix parameters and the price. This lack of significant results can be caused by the use of negotiated prices instead of cost prices or can be due to other causes for price differences, like quality of care. Another possible explanation, which is indicated by the literature, descriptive statistics and results of the interviews, is that case mix and price differences only exist between types of institutions, while this study adjusted for the influence of type of institution. Besides, insignificant results may be due to the fact that currently, case mix does not play a role during price negotiations. For cataract, case mix seems to be reflected in the DTC price which can be explained by the clear structure of the DTC, the non-complex nature of the treatment and the higher competition level of ITCs on the market. An additional conclusion that can be drawn, based on the insignificant quantitative results, is that the aim of the Risk Equalization Fund, which is reducing the incentive for selection, does not seem to be achieved. Recommendations for further research are the development of general and disease specific parameters of case mix, the development of a uniform general definition of case mix, the use of case mix parameters from hospitals, the use of cost prices and the investigation of the way case mix is expressed in the DOT system.