MOVE WORKSHOP on PROSPECTIVE PAYMENT AND HEALTHCARE PROVISION
20th and 21ST November 2009, Casa Convalescència (Barcelona)
EXECUTIVE SUMMARY
Policy and incentive implications of budgets for gatekeeping general practicesLuigi Siciliani (University of York)
The speaker presents his analysis, joint with Hugh Gravelle, of the incentives arising from introducing budgets for general practitioners, a policy called Commissioning (PBC). The effect of this policy in England is the reduction of the number of referrals if hospitals are paid according to an activity-based funding mechanism of the DRG pricing type. Otherwise, if hospitals are paid according to a fixed budget, the introduction of budgets for general practitioners has no effect on the number of referrals. The result is that the optimal price paid by the GP for a referral is different from the price paid to the hospital. Mathias Kifmann, from the University of Ausburg, in his discussion of the paper argues that, although the model provides a possible justification for PBC, the complexities involved in computing the optimal price could make its application problematic.  |
Determinants of hospital expenditure for general practice patients: implications for capitation based practice budgetsHugh Gravelle (University of York)
Discussant: Pedro Pita Barros (Universidade Nova de Lisboa) The speaker presents his analysis, joint with Mark Dusheiko, Steve Martin, Nigel Rice, Peter Smith, and Jennifer Dixon of the new English system of budgets for covering hospital expenditures. These authors propose a procedure to set these budgets for general practices using the characteristics of the patients, mainly by distinguishing them between chronic and non-chronic patients. This procedure allows for payments according to health care needs and is independent on supply factors. The authors come to the conclusion that actual allocation differs considerably from those based on measures of need derived from the previous expenditure models. Pedro Pita Barros (Universidade Nova de Lisboa) is the discussant of the paper. He shows some concern on whether it is actually possible to predict individual expenditure of non-chronic patients. |
Optimal cost reimbursement of health insurers to reduce risk selectionMathias Kifmann (University of Ausburg) The speaker presents a solution to the trade-off between cost-based payments and prospective payments in the health system. Cost-based payment eliminates the practice of insurers to select patients according to their risk type but causes insurers not to operate cost-efficiently. Prospective payments make insurers efficient in terms of costs but provide incentives to select patients if risk adjustment is not possible. In his analysis, joint with Normann Lorenz, they find a reimbursement scheme that provides the optimal balance between risk selection and cost efficiency in community-rated health insurance markets. They show that the distribution of costs for each risk type is of key importance. Furthermore, they demonstrate that outlier risk sharing is generally not optimal. The discussant (William Encinosa, from AHRQ and Georgetown University) suggests that the mechanism could be improved by considering price regulation as well as cost-side regulation. |
For what should primary care practitioners be paid? The Primary Care Activity Level (PCAL)Randall Ellis (Boston University)
The speaker remarked the importance of creating incentives for U.S. Primary Care Practitioners (PCP) so they will provide high quality services. He proposes a Patient-Centered Medical Home (PCMH) that implies an organizational change in health structures, practices that are team-oriented, bundled payments, and incentives that imply pay-for-performance. This examined how to calculate risk-based payments adjusted to the expected activity level for each practice. After showing his empirical model based on U.S. data, he concluded that PCMH is an attractive organizational model that will work well with good risk adjustment. The discussant, Manuel Garcia-Goñi, argues that integration is important and it is needed to see the system as a whole. He proposes that in future studies, the variation in clinical outcomes should be taken into account in order to talk of risk adjustment.  |
SURVEY 1: An overview of the links between social choice theory and the problem of risk adjustmentErik Schokkaert (CORE, Université catholique de Louvain and Department of Economics, KULeuven) The speaker proposes to apply some results in social choice theory to the issue of risk adjustment. First, the regulator or health authority must decide which risk adjusters are legitimate or illegitimate, in the sense that the latter should not be used in adjusting the capitation rates. He then shows how one should go from the estimating costs to computing the capitation formula. Second, he argues that the theoretical literature on the so-called claims (or estate division) problem offers interesting insights about how to adjust the risk-adjusted capitations in situations with a binding budget constraint. |
SURVEY 2: Health care payment systems and risk-adjustment in Catalonia: evolution and complementaritiesPere Ibern (Universidad Pompeu Fabra) Professor Ibern presents a historical overview of the Catalan Health System since 1996. He describes some characteristics of the system (hospital, health areas, expenditure…) and he introduces the present hospital payment system. He also discusses the heterogeneity among hospitals and describes the new assessment of payment systems, exposing its positive and controversial elements and also its goals. He also describes the evolution and the improvement on the quality obtained in the last years. Finally, he talks about the current debates, as whether the new system could bring excessive activity. His conclusions included the need to redefine individual provider contracts and to delimitate the “rules of the game” (dynamic issues, institutional design, and performance measurement). |
ROUND TABLE 1. Capitation Experiences in CataloniaChairperson: Pere Ibern (Universitat Pompeu Fabra) Vicente Ortún (Universidad Pompeu Fabra) and Josep Maria Argimon (Servei Català de la Salut) This round table was based on the relationship that exists between incentives and capitation. Josep Maria Argimon describes the future of per capita financing in Catalonia under the goal of improve healthcare equity, continuity and efficiency through the promotion of territorial alliances among providers. Vicente Ortún discussed the good and bad elements of the Catalan health system. The former included an excellent financial protection and a high level of satisfaction among users. The latter included lack of control in utilization and institutional deterioration. He proposes that risk adjustment formulae should be complemented with ex-post payments based on quality assessment.  |
Cost efficient risk adjustments to deter creaming and skimpingWilliam Encinosa (AHRQ and Georgetown University) In the health insurance market one usual assumption is that there are constant returns to scale, which then leads to a zero profit condition. In this paper the author analyses the case of economies of scale, and looks for an equilibrium whit free entry in the industry, and where there is no creaming, nor skimping. In order to prevent creaming and skimping he finds that the payment cannot just depend on the average cost, but must rather be adjusted by the average incremental cost of doing the effort for the low risk. This also implies that the payment is now differentiated for the low and high risk, making the payment for the low risk to be below the usual average, and the payment for the high risk above it. In the discussion it was highlighted that efforts are considered to be substitutes, and moreover that a more precise interpretation is needed for the efforts. Also the quality, which is usually not contractible, needs to be contractible for its minimum. This arose the question of what would happen if this were not the case.  |
Risk sharing and risk adjustment strategies to deal with health plan selection and efficiencyManuel García Goñi (Universidad Complutense de Madrid) Health cost escalation has lead to more managed care through capitation, which in turn has lead to more efficiency savings, but also to more selection of patients. This paper makes a welfare analysis by comparing different reimbursement schemes that the regulator may choose, always assuming that the health plans have better information than regulators. Using simulations the author compares (i) payment schemes with no risk adjustment, (ii) conventional risk adjustment, and (iii) minimum welfare loss risk adjustment. He finds that the latter is the one that performs best, and that the more information the regulator has; the larger is the advantage of the third procedure over the other two. In the discussion, the question of the constraints of the regulator arose, as these are not considered in the model. For instance, this player should be subject to some budget constraint. Another possible suggestion for improvement was to consider a different market structure for the HMOs.  |
ROUND TABLE 2. Capitation Experiences in MadridChairperson: Xavier Martinez-Giralt (Universitat Autònoma de Barcelona) Félix Lobo (Universidad Carlos III de Madrid), Carlos Ostalé (Mensor-Servicios de Salud), Elena Miravalles (BRsalud, link to slides), Manuel García Goñi (Universidad Complutense de Madrid, link to slides) Faced with the pressures to contain health expenditure, recent experience in Madrid could constitute a possible alternative to static National Health System. Innovations have been made both in the management and in the organization of new hospitals in the area. One the one hand, the contractual model defines a financing system based on a pure capitation part (adjusted to the CPI) and an inter-centre billing part. A fixed annual payment has also been implemented. On the other hand, the new hospitals share a central laboratory that performs all non-urgent diagnostic tests (radiological, analytical). This allows for economies of scale, increased efficiency, and less disutility for patients. The main criticism to this system is the fact that capitation is not targeting quantity. Problems will arise in the long run since the hospitals’ activity and therefore demand for testing at the centralized laboratory will rise. Moreover, the budgets for hospitals have not bee reduced despite the fact that a good part of clinical tests have been derived to the central lab. |
Upcoding in a NHS - Evidence from PortugalPedro Pita Barros (Universidade Nova de Lisboa) In this presentation, the author discusses empirically the possibility of the existence of the practice of upcoding by health professionals. The idea is that patients may be misclassified in order to obtain higher budgets. To test the hypothesis that misclassification exists, the author takes DRGs with and without complications, and tests whether an exogenous change in the law implied a change in the relative profit of these two different classifications. He finds that there is evidence that Upcoding exists, and moreover that it is not different between hospitals managed under traditional public administration and private-like management rules in public hospitals. In the discussion it was noted that the change in prices may also have some real effect through a change on the rationing systems. Another point made by the discussant is that there could also have been a change in the way coding is done, namely on its quality, which does not imply upcoding, but could be captured by making some adjustments in the model. There was also a suggestion to make the analysis on an individual probability level, rather that at the hospital level. |
ROUND TABLE 3. Capitation Experiences in PortugalChairperson: Luigi Siciliani (University of York) In this roundtable, the payment schemes for Local Health Units in place in Portugal were analyzed, focusing on their design under different central management schemes. On a very first stage Local Health Units (LHU) were financed by the sum of the costs in Primary Health Centers (PHC) and the payment schemes of Hospitals (which included the case mix, incentives for quality and production lines). Risk-adjusted capitation was generalized in 2009. The new payment scheme revised for 2010 depends on adjusted capitation that is composed of demographic characteristics, health needs and health provision. Also the mortality rates are taken into account when calculating this index, as they work as a proxy for health needs in each area. Problems that still remain include how to give incentives, and how to measure the integration between PHC and Hospitals. The discussion went around issues like the importance of governance over the payment scheme, the incentives to reduce consumption of medication and the substitution caused by medical visits, and the integration between different levels of health care. |
 |
