The Contribution of Managed Care to the Performance of Healthcare Systems - Evidence from Three Countries

Schoder, Johannes; Zweifel, Peter Article The Contribution of Managed Care to the Performance of Healthcare Systems - Evidence from Three Countries Swiss Journal of Economics and Statistics Provided in Cooperation with: Swiss Society of Economics and Statistics, Zurich Suggested Citation: Schoder, Johannes; Zweifel, Peter (2008) : The Contribution of Managed Care to the Performance of Healthcare Systems - Evidence from Three Countries, Swiss Journal of Economics and Statistics, ISSN 2235-6282, Springer, Heidelberg, Vol. 144, Iss. 3, pp. 477-493, https://doi.org/10.1007/BF03399263 This Version is available at: https://hdl.handle.net/10419/185897 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ © Swiss Journal of Economics and Statistics 2008, Vol. 144 (3) 477–493 a University of Zurich, Socioeconomic Institute, Hottingerstr. 10, CH-8032 Zurich, Switzerland. E-mail: (J. Schoder) [email protected], (P. Zweifel) [email protected]. The authors thank Philippe Widmer, Boris Krey, Maria Trottmann, Karolin Becker, and participants of the 2008 Annual Meeting of the SSES for helpful comments. The support of the Association af Pharma Importers in Switzerland (VIPS) is gratefully acknowledged. 1 In Switzerland 58.5 percent, in Germany 76.9 percent, and in the Netherlands 62.5 percent of HCE are paid by the government (Who Statistics 2005). The Contribution of Managed Care to the Performance of Healthcare Systems – Evidence from Three Countries Johannes Schoder and Peter Zweifela JEL-Classification: I11 Keywords: healthcare expenditure, managed care, performance, principal-agent relationship 1. Introduction Healthcare expenditure (HCE) continues to increase at a faster rate than GDP in almost all industrialized countries. Governments have tried to alleviate the pressure on their budget1 mainly in two different ways. One has been to limit HCE by regulation, the other, to introduce competition in an attempt to increase efficiency (Cutler 2002). Since the first option has not proved too successful, more and more countries are seeking ways to enhance competition, among them by fostering Managed Care (MC). By vertically integrating health insurance and healthcare provision, MC may improve the allocation of resources in healthcare while limiting HCE. Indeed, most of the current literature on MC focuses on its impact on HCE. Based on the Rand Health Insurance Experiment, Manning et al. (1984) studied the effect of MC on the utilization of healthcare services and on the level of HCE. They had randomly assigned a group of 1,580 persons to receive care free of charge from either a fee-for-service physician of their choice (representing conventional care) or a physician participating in a Health Maintenance Organization (HMO, representing MC). In addition, a group of 733 individuals, already enrolled in a HMO, constituted a control group. The crucial innovation of this study was that participants were assigned to plans, which served to avoid risk selection effects, causing healthier individuals to enrol in MC plans. 478 Schoder / Zweifel Both groups enrolled in the MC plan had 40 percent fewer inpatient admission levels than those assigned to the conventional insurance plan. Their total HCE was about 25 percent lower than under conventional care. Cutler, McClellan and Newhouse (2000) analyzed the effect of MC on the cost of treatment of one particular disease. They compared the treatment of heart disease in Health Maintenance Organizations (HMOs) and traditional insurance plans using two datasets from Massachusetts. For the HMOs they found 30 to 40 percent lower HCE than for traditional plans, mainly due to differences in unit prices. They concluded that MC may yield substantial increases in measured productivity relative to traditional insurance. Using Swiss panel data, Lehmann and Zweifel (2004) were able to distinguish cost savings due to risk selection and due to innovation effects. They found some evidence of risk selection effects, which however, accounted for only one-third of the cost advantage in the case of HMOs, with the remainder attributed to innovation effects. This paper follows a more comprehensive (but more descriptive) approach by assessing the contribution of MC to the performance of an entire healthcare system. Performance is measured using five standard criteria developed for the assessment of an economy. They are (1) matching of consumer preferences, (2) technical efficiency, (3) adaptive capacity, (4) dynamic efficiency, and (5) a rentfree distribution of income that provides incentives for producers to attain criteria (1) through (4). These criteria are applied to the three contractual relationships typically characterizing a healthcare system, viz. (a) between the insured and patients and health insurers (the government as it were in the case of National Health Service-type systems); (b) between insurers (the government, respectively) and healthcare providers; and (c) between the insured and patients and healthcare providers. The countries to be analyzed according to these five criteria and three contractual relationships are Germany, the Netherlands, and Switzerland. This choice can be justified for the following reasons. First, all three are insurance-based, which facilitates the comparison. Second, elements of MC were introduced in all three countries during the last few years. Third, the Netherlands underwent an important reform of their healthcare system in 2006, which allows to test the hypothesis that the contribution of MC to system performance depends on the institutional framework. This paper is organized as follows. Section 2 contains the definition of MC and an explanation of the criteria for measuring performance. In section 3, these criteria are applied to the three contractual relationships of a healthcare system before the introduction of MC. Finally, the contribution of MC to the performance of the three healthcare systems is assessed in section 4 by applying the The Contribution of Managed Care to the Performance of Healthcare Systems 479 criteria to the contractual relationships after the introduction of MC. The last section presents a conclusion and suggestions for future work. 2. Analytical Framework 2.1 The Common Building Block of Healthcare Systems and the Scope of Managed Care The common building block of all healthcare systems is the relationship between the patient (the principal in the economic theory of contract) and the physician (the agent). Patients experience a significant informational disadvantage, causing them to delegate decision-making authority to the physician. In particular, they may at best observe the outcome of a treatment, but not physician effort. For the physician, however, additional effort is costly, at the very least in terms of leisure forgone. This fact alone prevents physicians from being a perfect agent of their patients. Generally, physicians will set their effort at a level they consider optimal from their own point of view. Since it is in general impossible for the patient to find the payment function inducing the optimal treatment effort by the physician, there is scope for complementary agents who promise to mend the physician-patient relationship (Zweifel, 1998). In Germany, the Netherlands, and Switzerland health insurers (in Germany medical associations as well) represent the dominant complementary agents. However, complementary agents induce new information asymmetries leading to moral hazard and adverse selection effects. Specifically, insured patients tend to consume more healthcare services than medically necessary. The objective of MC is therefore to rearrange the relationship between these three players in order to mitigate information asymmetries and enhance efficiency as well as to optimize the allocation of the healthcare resources used (Finsterwald, 2004). Different forms of MC exist, including Health Maintenance Organizations (HMOs), Preferred Provider Organizations (PPOs), and Independent Practice Associations (IPAs). They integrate insurance and provision of healthcare services, however to a different extent. This integration is designed to reduce moral hazard effects between healthcare service provider and insurer, to optimize the use of healthcare resources (e.g. by avoiding double visits), and to better structure treatment processes. To this end, MC organizations listed above apply different instruments, such as selective contracting, gatekeeping, and disease management (Amelung, 2007; Felder, 2003). In the following we will analyze the impact of MC on the performance of healthcare systems. The criteria used for evaluation are presented below. 480 Schoder / Zweifel 2.2 Evaluation Criteria The five criteria listed below were originally developed for the assessment of an economy in general. Zweifel (2006) adapted them to the healthcare sector in the following way. Matching of consumer preferences: Healthcare services should match the preferences of the insured, who are assumed to decide about the types of medical care that should be covered by insurance before they are ill. Technical efficiency: The healthcare services that are provided according to criterion (1) should be produced at least cost. Adaptive capacity: The insured as well as insurers and healthcare providers should adapt e.g. to population ageing or to medical technical change. Dynamic efficiency: The healthcare sector should have an optimal mix of product innovation (goods with changed characteristics that may fetch a higher price) and process innovation (unchanged goods at lower cost and price). In general, insurance-induced moral hazard biases this mix in favor of product innovation (Zweifel and Breyer, 1997). Income distribution according to performance: Providers of healthcare services should not be able to enjoy monopolistic rents (e.g. incomes that exceed the amount that is necessary to keep them in their current activity). Rents jeopardize the attainment of the other four criteria because providers lack the incentive to make the pertinent efforts. These criteria are applied to the three contractual relationships characterizing a healthcare system defined in section 2.1. Each time, the issue is whether MC contributes to the improvement of the contractual relationship in the light of the five criteria. 3. The Contractual Relationships Prior to Managed Care in Germany, the Netherlands, and Switzerland First, the healthcare system of Germany, the Netherlands, and Switzerland will be described prior to MC using criteria (1) to (5). This is somewhat difficult for the Netherlands since some MC elements have been part of the healthcare system for a long time. The Contribution of Managed Care to the Performance of Healthcare Systems 481 2 Since more than 90 percent of the population is covered by the social insurance, the analysis performed below does not relate to the private health insurance sector. 3 Medical associations and health insurer associations distribute the budgeted amount proportionally according to billed activity between the primary care physicians of a given land. 3.1 The Contractual Relationship between Consumers and Health Insurers In all the three countries, consumers can choose among different health insurers. However, insurance policies are highly regulated. The government not only limits the range of admissible premiums (contribution rates, respectively in Germany), but also the treatments to be covered. Only in the Netherlands, cost-effective- ness constitutes a criterion for the inclusion or exclusion of benefits (Schreyögg, Stargardt, Velasco-Garrido and Busse, 2005). Dutch and Swiss health insurers have more freedom to launch different types of insurance policies than their German counterparts (Schut and van de Ven, 2005; Becker, Brändle and Zweifel, 2007). In Germany, only private health insurers have the right to differentiate their products2 (Jacobs and Schulze, 2006). Clearly, the obligation for health insurers to offer largely uniform insurance policies makes it difficult to match consumer preferences, to quickly adapt to changes of the economic environment, and to sustain technical and dynamic efficiency. Therefore, criteria (1) through (4) are violated (see Table 1). However, competition for consumers has been enough in the three countries to prevent the creation of rents [criterion (5)] (Zweifel, 2006). 3.2 The Contractual Relationship between Health Insurers and Healthcare Providers In Germany and Switzerland, health insurers are subject to an “any-willing-pro- vider” clause, i.e. they are forced by law to contract with every approved physician. In the Netherlands, selective contracting has been possible since 1994; however, health insurers have been making very limited use of this right up to the present (Baur, Heimer and Wieseler, 2001). Health insurers also lack the right of negotiating differentiated, incentive-compatible modes of physician remuneration in the countries analyzed. In Germany, the association of social health insurers and contract physicians negotiate both the global medical budget and the nationwide fee schedule3 (Busse, 2000). In the Netherlands, insurers have the choice between paying (primary care) physicians either fee-for-service or using capitation (i.e. a fixed amount per enlisted patient). Swiss health insurers must apply Tarmed (Tarif médical), a nationwide fee schedule. For hospital 482 Schoder / Zweifel services, German health insurers are subject to a nationwide fee schedule (Pflegesatzverordnung). Dutch insurers have some negotiating leeway, which is however constrained in several ways. In Switzerland, they are confronted with cantonal hospital associations. For pharmaceuticals, all three countries impose a national benefit list along with regulated prices. Obviously, collective contracting and uniform payment schedules in Switzerland and Germany violate criteria (1), (2), and (5). However, criterion (2) is satisfied to a higher degree in the Netherlands due to more flexibility with regard to modes of payments. 3.3 The Contractual Relationship between Consumers and Healthcare Providers In Germany and Switzerland, patients can choose their preferred physicians without any limitation. In the Netherlands, they are obliged to see a primary care physician first, which may not be in accordance with consumer preferences [criterion (1)]. Table 1 summarizes the rough overall assessment of the German, Dutch, and Swiss healthcare system prior to MC. Table 1: Main Violations of Performance Criteria Prior to MC Consumers-Insurers Insurers-Providers Consumers-Providers Germany criteria (1) through (4) criteria (1), (2), and (5) Criterion (2) Netherlands criteria (1) through (4) criteria (1) and (5) Criterion (1) Switzerland criteria (1) through (4) criteria (1), (2), and (5) Criterion (2) 4. Assessing the Contribution of Managed Care to the Performance of the German, Dutch, and Swiss Healthcare Systems This section is devoted to an assessment of the contribution MC makes to the performance of the German, Dutch, and Swiss healthcare systems. The scales used will be 2 points if MC fully contributes to the attainment of the criterion, 1 point if MC partially contributes, and 0 points if it does not contribute to the attainment of the criterion. Points will be simply added to obtain a total score. For each country, the assessment focuses on the MC element that is most prominent, e.g. Disease Management Programs in the case of Germany. The Contribution of Managed Care to the Performance of Healthcare Systems 483 4 Therefore, MC is known as integrated care (integrierte Versorgung) in Germany. 5 In Germany statutory sickness funds act as social health insurers. 6 The risk adjustment scheme is currently based on age, sex, gender, and the four DMPs officially implemented. 4.1 The Contribution of Disease Management Programs to the Performance of the German Healthcare System The Laws on Health Insurance of 2000 and 2004 paved the way for MC in Germany (Bmj, 2006). The governments’ objective is to foster the integration of hitherto strictly separated ambulatory and hospital care.4 In the MC setting, sickness funds5 are allowed to selectively contract with physicians without the involvement of medical associations. Moreover, alternative forms of payment, including capitation can be implemented. The government promotes three different types of MC in particular, viz. Medical Care Centers (Medizinische Versorgungszentren), Independent Practice Associations (Hausarztmodelle), and Disease Management Programs (Strukturierte Behandlungsprogramme) (Greiner, 2005; Busse, 2004). The effects of MC will be illustrated for the Disease Management Programs (DMPs). They have been developed to improve quality and cost-effectiveness of treatment received by the chronically ill. So far the government has defined DMPs for four chronic diseases, diabetes, breast cancer, asthma, and coronary heart disease. The sickness funds receive payments out of the risk adjustment scheme6 for every individual enrolled in a DMP. It was hoped that connecting DMPs with the risk adjustment scheme would provide a stimulus for sickness funds to attract chronically ill people rather than eschewing them as high risks (Bmg, 2007; Wiechmann, 2003). Relationship between Consumers and Insurers Preferences of patients are not considered in the definition of DMPs. However, participation in DMPs is not mandatory but offers an additional choice. Therefore matching of consumer preferences is slightly improved. There is little reason to expect that the chronically ill will obtain their treatment at lower cost because the DMPs do not provide incentives to health insurers or providers for a better coordination of care. Therefore, static efficiency is not enhanced. The Government determines the design of DMPs, e.g. it decides (using lengthy procedures) which chronic diseases are included. This does not improve adaptive capacity of the system. However, sickness funds have incentives to support cost-reducing process innovation because the payment of the risk adjustment scheme is fixed, 484 Schoder / Zweifel putting them at risk for exclusive cost of treatment. This serves to redress the balance between product and process innovation somewhat. These considerations may justify the entries in the second column of Table 2. Relationship between Insurers and Providers Thanks to selective contracting, sickness funds are supposed to become prudent purchasers on behalf of their clients. One would expect them to contract only with those physicians exhibiting a favorable cost-benefit ratio in treatment of chronic illness. However, physicians participating in DMPs lose their autonomy in medical decision-making. Beside many other regulations, they must follow treatment guidelines and document the whole treatment process electronically. On the whole, DMPs are unattractive to physicians, who continue to have the option of billing fee-for-service. Therefore the DMPs do little to increase the degree of competition between healthcare providers and hence to improve the matching of consumer preferences, technical efficiency, adaptive capacity, or the avoidance of monopolistic rents. At least, the DMP guidelines may induce providers to focus more on process innovation, motivating the entry + 1 for criterion (4) in the second column of Table 2. Relationship between Consumers and Physicians In general, German consumers have a free choice of physicians but are expected to visit the hospital recommended by their primary care physician. However, usually primary care physicians place no restrictions on the hospital choice (Bmj, 2006). In contrast, DMP-patients are constrained to select a participating physician. This limitation is hardly compensated; in particular, there is almost no reduction in the rate of contribution. But at least consumers are not forced to participate, justifying the zero entry for criterion (1) in the third column of Table 2. Likewise, attainment of criteria (2) through (5) remains unchanged. In sum, the contribution of DMPs to the performance of the German healthcare system remains limited (3 out of 30 points, see the last column of Table 2). The main violations of the criteria prior to the introduction of MC (see the shaded fields in Table 2) could not be offset, except for the criteria (1) and (4) relating to the relationship between consumers and insurers. This limited contribution is mainly due to comprehensive and uniform regulation stifling any innovative action by sickness funds. For the same reason, selective contracting does not produce the expected benefits for consumers. Thus, DMPs fail their main objective, viz. improved coordination and quality of treatment provided to the chronically ill. The Contribution of Managed Care to the Performance of Healthcare Systems 491 Finally, the findings suggest that MC may depend on the institutional setting. The more freedom to contract between consumers, health insurers, and healthcare service providers, the greater the contribution of MC to the healthcare system. However, further research is necessary to test this hypothesis. References Amelung, Volker E. (2007), Managed Care: Neue Wege im Gesundheitsmanagement, Wiesbaden. Bag, Bundesamt für Gesundheit (2007), Statistik der obligatorischen Krankenversicherung 2006, Bern. Baur, Rita, Heimer, Andreas and Silvia Wieseler (2001), “Health Care Systems and Reform Efforts, an International Comparison”, in: Jan Böcken, Martin Butzlaff and Andreas Esche (eds.), Reforming the Health Sector, Gütersloh, pp. 23–136. Becker, Karolin, Brändle, Angelika and Peter Zweifel (2007), „Das Discrete-Choice-Experiment“, in: Jan Böcken, Katharina Janus, Uwe Schwenk and Peter Zweifel (eds.), Neue Versorgungsmodelle im Gesundheitswesen, Gütersloh, pp. 38–151. Bmj, Bundesministerium der Justiz (2006), Soziales fünftes Gesetzbuch – Gesetzliche Krankenversicherung, available at: http://www.gesetze-im-internet. de (April 2008). Bmg, Bundesministerium für Gesundheit (2007), Die Gesundheitsreform: Infographiken, available at: http://www.diegesundheitsreform.de (May 2008). Busse, Rheinardt (2004), “Disease Management Programs in Germany’s Statutory Health Insurance System – A Gordian Solution to the Adverse Selection of Chronically Ill People in Competitive Markets?”, Health Affairs, 23 (3), pp. 56–67. Cutler, David M., McClellan, mark and joseph P. Newhouse (2000), “How Does Managed Care Do It?”, Rand Journal of Economics, 31 (3), pp. 526–548. Cutler, David M. (2002), “Equality, Efficiency, and Market Fundamental: The Dynamics of International Medical-care Reform”, Journal of Economic Literature, 40 (3), pp. 881–906. Douven, Rudy, Mot, Esther and Marc Pomp (2007), “Health Care Reform in the Netherlands”, Die Volkswirtschaft, 3, pp. 31–33. 492 Schoder / Zweifel Exter, André den, Hermans, Herbert, Dosljak, Milena and Reinhardt Busse (2004), Health Care Systems in Transition: Netherlands, Copenhagen, WHO Regional Office for Europe on behalf of the European Observatory on Health systems and Policy. Felder, Stefan (2003), “Managed Care: Low Reputation but most Effective”, DICE-Report Journal for Institutional Comparisons, 1 (3), pp. 15–19. Finsterwald, Daniel (2004), „Managed Care – Pionierland Schweiz“, Schriftenreihe der Schweizerischen Gesellschaft für Gesundheitspolitik, 75. Greiner, Wolfgang (2005), “Health Economic Evaluation of Disease Management Programs – the German Example”, European Journal of Health Economics, 6 (3), pp. 191–196. Jacobs, Klaus and Sabine Schulze (2006), „Vertragswettbewerb: Mehr Freiheit bei der Partnerwahl“, Gesundheit und Gesellschaft, 7, pp. 36–41. Lehmann, Hans-Jörg and Peter Zweifel (2004), “Innovation and Risk Selection in Deregulated Social Health Insurance”, Journal of Health Economics, 23, pp. 997–1012. Manning, Willard G., Leibowitz, Arleen, Goldberg, George A., Rogers, William H. and Joseph P. Newhouse (1984), “A Controlled Trial of the Effect of a Prepaid Group Practice on the Use of Services”, New England Journal of Medicine, 310 (23), pp. 1505–1510. Schreyögg, Jonas, Stargardt, Tom, Velasco-Garrido, Marcial and Reinhardt Busse (2005), “Defining the Health Benefit Basket in Nine European Countries”, European Journal of Health Economics, 6 (S1), pp. 2–10. Schut, Frederik T. and Wynand P. M. M. van de Ven (2005), “Rationing and Competition in the Dutch Health Care System”, Health Economics, 14 (S1), pp. 59–74. Who (2005), Statistical Information System, Statistics 2005. Wiechmann, Michael (2003), Managed Care: Grundlagen, internationale Erfahrungen und Umsetzung im deutschen Gesundheitswesen, Wiesbaden. Zweifel, Peter and Friedrich Breyer (1997), Health Economics, New York. Zweifel, Peter (1998), “Managed Care in Germany and Switzerland: Two Approaches to a Common Problem”, Pharmacoeconomics, 14 (S1), pp. 1–8. Zweifel, Peter (2006), “Building a Competitive Insurance System: Switzerland’s Strategy for Managed-Care Health Care”, Pharmacoeconomics, 24 (S2), pp. 10 9 –117. Zweifel, Peter, Telser, Harry, and Stefan Vaterlaus (2006), “Consumer Resistance Against Regulation: The Case of Health Care”, Journal of Regulatory Economics, 29, pp. 319–332. The Contribution of Managed Care to the Performance of Healthcare Systems 493 SUMMARY This paper applies the five standard economic performance criteria to gauge the contribution of Managed Care (MC) to the performance of three healthcare systems, viz. Germany, the Netherlands, and Switzerland. The criteria are (1) matching of consumer preferences, (2) technical efficiency, (3) adaptive capacity, (4) dynamic efficiency, and (5) a rent-free distribution of income that provides incentives for producers to attain criteria (1) through (4). Being insurance-based, the German, Dutch, and Swiss healthcare systems comprise three contractual relationships that can be judged in the light of these criteria. The maximum contribution of MC to the performance of the healthcare system is found for the Netherlands followed by Switzerland. The Independent Practice Associations representing MC in the Netherlands, and the Health Maintenance Organizations representing MC in Switzerland score 15 respectively 6 out of 30 points. By way of contrast, the contribution of the Disease Management Programs to the performance of the German healthcare system remains limited (3 out of 30 points).