- Email: [email protected]
Competition policy for health care provision in Germany
G Model
ARTICLE IN PRESS
HEAP-3660; No. of Pages 7
Health Policy xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Health Policy journal homepage: www.elsevier.com/locate/healthpol
Competition policy for health care provision in Germany Mathias Kifmann Universität Hamburg, Esplanade 36, 20354 Hamburg, Germany
a r t i c l e
i n f o
Article history: Received 15 June 2016 Received in revised form 24 October 2016 Accepted 16 November 2016 Keywords: Hospitals General practitioners Competition policy Germany Regulated competition Corporatism
a b s t r a c t Since the 1990s, Germany has introduced a number of competitive elements into its public health care system. Sickness funds were given some freedom to sign selective contracts with providers. Competition between ambulatory care providers and hospitals was introduced for certain diseases and services. As competition has become more intense, the importance of competition law has increased. This paper reviews these areas of competition policy. The problems of introducing competition into a corporatist system are discussed. Based on the scientific evidence on the effects of competition, key lessons and implications for future policy are formulated. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Traditionally, Germany’s social health insurance system is governed jointly by associations. The national association of sickness funds, the national and regional associations of physicians and the German Hospital Federation negotiate contract conditions and payment schemes for ambulatory and hospital care. In the 1990s, however, competitive elements were introduced in the system. The first major reform concerned sickness funds. Since 1996, almost all individuals insured in the social health insurance system can choose freely among sickness funds. This has fundamentally transformed the funds from administrative entities into service-oriented institutions. Initially, sickness funds had few possibilities to influence the provision of health care services as these were predominantly determined by collective negotiations and legal requirements. Inspired by managed care in the US, sickness funds were given some freedom to sign selective contracts with providers in the late 1990s. Today sickness funds offer Disease Management Programs, Integrated Care
E-mail address: [email protected]
Contracts and Gatekeeping Contracts. The vast majority of health care services, however, continue to be managed by the corporatist system. A prominent feature of the German health care system is the strong separation of the outpatient and inpatient sector. Hospitals are mainly restricted to inpatient care. Outpatient specialist care is provided by physicians with specialist training. In the last decade, some competition between the sectors was introduced. Hospitals were allowed to offer ambulatory care for specific diseases and for certain specialized services. Hospitals also opened medical treatment centers for patients seeking outpatient care. Both, hospitals and ambulatory specialists, offer ambulatory surgeries. As competition in German health care has become more intense, the importance of competition law has increased. Selective contracting by sickness funds is subject to competition law. The competition authority is also responsible for mergers of sickness funds and of hospitals. Between 2004 and 2014, 182 hospital mergers were approved and 7 prohibited. This report provides information on these areas of competition policy in Germany. In Section 2, a short overview of the German health care system is given. In Section 3, com-
http://dx.doi.org/10.1016/j.healthpol.2016.11.014 0168-8510/© 2016 Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Kifmann M. Competition policy for health care provision in Germany. Health Policy (2016), http://dx.doi.org/10.1016/j.healthpol.2016.11.014
G Model HEAP-3660; No. of Pages 7
ARTICLE IN PRESS M. Kifmann / Health Policy xxx (2016) xxx–xxx
2
petition policies are reviewed. Section 4 concludes with key lessons and implications for future policy. The focus is on competition between health care providers. Section 2 also provides summary information about health insurance competition. 2. Institutional set-up 2.1. Basic characteristics In 2014, Germany spent 11.2% of its GDP on health care [1]. Most health care expenditure (66%) was financed by “Statutory Health Insurance” (SHI, Gesetzliche Krankenversicherung), the public health insurance program in which sickness funds provide health insurance. A share of 9% was paid for by the private health insurance system which insures public servants, self-employed persons and individuals who opt out of SHI. The remaining share was financed by private households and non-profit organizations (13%), by other social insurance programs or from public budgets (7%) and by employers (4%). In the SHI system, currently 118 sickness funds insure about 87% of the population. Mergers have decreased the number of funds considerably. In 1990, there were still 1147 funds. Sickness funds are quasi-public corporations and non-for-profit institutions. A major characteristic of SHI in Germany is self-governance. Payers and providers are organized in “corporatist bodies” with mandatory membership and special legal status. In particular, this comprises the national association of sickness funds, the national and regional associations of SHI-physicians and the German Hospital Federation. These associations negotiate contract conditions and payment schemes for ambulatory and hospital care. The regulatory framework is provided by the Social Code Book V (Sozialgesetzbuch). The federal and state (Länder) authorities supervise the self-governance of the associations. The most important decision-making body is the Federal Joint Committee (Gemeinsamer Bundesausschuss) which consists of ten representatives of the associations and three neutral members. Five patient representatives can put topics on the agenda, but have no voting rights. To finance SHI, a contribution rate of 14.6% is levied upon labor income up to an income ceiling of D 4237.50 in 2016. This leads to a maximum levy of about D 619 per month. The contributions are transferred to a Central Reallocation Pool which is administrated by the Federal Insurance Authority (Bundesversicherungsamt). For each member, sickness funds receive risk-adjusted capitation payments based on age, sex, disability status and morbidity. In addition, sickness funds can charge an additional contribution rate on the same income base. It currently ranges from 0.3 to 1.7%. Information on prices and supplemental benefits, such as alternative health care or integrated care, is readily available (see, e.g., https://www.gesetzlichekrankenkassen.de/). However, there is no official publication of performance indicators. As the benefit package of sickness funds is to a large extent specified by law, competition between sickness funds focuses on price. A study by Bünnings et al. [2] with
data from 2007 to 2010 finds that price differences are the main factor for the choice between sickness funds. Service quality and supplemental benefits only seem to play a minor role. From 2009 to 2014, premium differences between sickness funds were expressed in absolute Euro values as opposed to percentage point differences in contribution rates. Schmitz and Ziebarth [3] show that this policy change strongly increased the willingness to switch funds. Demand elasticities increased by a factor of four. There is an ongoing debate about the risk adjustment scheme. Among the issues discussed are the number of diseases considered and the manipulation of diagnostic information. In a field experiment, Brosse and Himmel [4] find evidence for risk selection by sickness funds against elderly and those with chronic diseases. In contrast to sickness funds, private health insurers charge premiums related to risk. In 2015, about 11% of the population were covered by private health insurance. Of the 42 members of the association of private health insurers, 24 are for-profit companies and 18 are mutual insurance societies. The privately insured consist of active and retired civil servants, self-employed and employees who have opted out of the sickness fund system. Opting out requires that income exceeds a threshold (in 2016, a yearly income of D 56,250). Most physicians treat both SHI-patients and private patients in the same practice. Private patients pay according to a different fee schedule for outpatient services which is considerably more generous than payment for SHI-patients and not subject to volume restrictions. In hospitals, private patients often pay for additional services, e.g. for treatment by the head doctor. Payments by private patients are an additional and important source of income for both physicians and hospitals. According to a study by the Federal Statistical Office, private patients generated 28% of physicians’ income compared to 69% for SHIpatients [5]. Roll et al. [6] find that private patients obtain appointments earlier and remain shorter in the waiting room. Private health insurers compete on price and on the level of coverage, e.g., on the size of the deductible. There is also competition with sickness funds for high earners who have the possibility to opt out of the sickness fund system. Private health insurers typically offer lower premiums for young, single and healthy individuals. However, social health insurance is attractive for those with pre-existing medical conditions. In contrast to private health insurers, sickness funds also insure children and a non-working spouse without additional contributions. Private health insurers save part of the premium income in early years to dampen premium increases in old age. Until 2009, these old-age provisions were not transferable. This strongly restricted competition within the private health insurance system. On the other hand, this regulation avoided that individuals who remain healthy opt out of contracts, leaving insurers with an adverse selection of individuals. Since 2009, a share of old-age provision has become portable. Karlsson et al. [7] examine the data of a private insurer and find only a small effect of portability on
Please cite this article in press as: Kifmann M. Competition policy for health care provision in Germany. Health Policy (2016), http://dx.doi.org/10.1016/j.healthpol.2016.11.014
G Model HEAP-3660; No. of Pages 7
ARTICLE IN PRESS M. Kifmann / Health Policy xxx (2016) xxx–xxx
switching behavior. However, internal switching between contracts offered by this insurer increased substantially. 2.2. Physicians in the social health insurance system About 98% of all ambulatory care physicians are affiliated with the SHI system. They are obligatory members of one of the 17 regional associations of SHI-physicians (Kassenärztliche Vereinigungen). The associations receive morbidity-adjusted capitations for each SHI-member from the sickness funds. In return, the associations guarantee the health care provision for SHI-members (Sicherstellungsauftrag). Total revenue is distributed by the associations to their members based on a complex catalog which specifies fees for cases as well as for specific services. Physicians face target budgets. If they exceed their budgets, the reimbursement for further cases is strongly reduced. Physicians are obliged to take part in external quality assurance measures and to implement internal quality management. Each year, the physician associations randomly select physicians and check whether specific medical services have been provided according to the quality standards. The Federal Joint Committee can define penalties if physicians do not meet their quality assurance obligations (see Ref. [8,pp. 81–82]). Most SHI-affiliated physicians work in private practices. In 2012, 46% were GPs (Hausärzte) while 54% were specialists (Fachärzte). Specialists who are employed by hospitals are not SHI-affiliated. In contrast to many other health care systems, patients can access specialists without first seeing a GP. This, however, applies only for ambulatory specialists and not for specialists in hospitals unless patients suffer from specific diseases (see Section 3.2). Patients also have the possibility to join Gatekeeping Contracts which require them to see a GP initially before seeing a specialist (see Section 3.1). 2.3. The hospital sector In 2014, the hospital sector was composed of a mix of public hospitals (29.7%), private non-profit hospitals (35.2%) and for-profit hospitals (35.1%). The total number was 1980 hospitals, down from 2166 ten years before. This number includes both single and multi-specialty hospitals. In particular, 273 hospitals only treat patients with psychiatric or neurological diseases. Between 2004 and 2014, take-overs and mergers increased the number of private for-profit hospital from 555 to 695 [9]. A considerable number of private for-profit hospitals are owned by chains. The two biggest are Helios with 111 and Asklepios with 102 hospitals. Hospitals are paid per case according to the German DRG-system. Relative weights of DRGs are determined at the national level by the German DRG Institute InEK (Institut für das Entgeltsystem im Krankenhaus) based on the cost data of selected hospitals. Multiplication with the statespecific case rate yields the per-case payment for a DRG. The state-specific case rates are negotiated by sickness funds and hospitals at the state level each year. In 2015, they ranged from D 3191 to D 3396. Furthermore, sickness funds and hospitals negotiate the kind and quantities of services with each hospital. German states also pay for
3
capital investments in hospitals. However, these payments have fallen considerably in the last decade. Between 1991 and 2014 they have almost halved in real terms [10]. A number of measures are implemented to secure hospital quality. Based on data reported by hospitals, federal or state quality-assurance institutions enter into a “structured dialogue” with a hospital if there are indications for low quality. The hospital must either explain the findings with special circumstances or show that the quality problem has been taken care of. Furthermore, hospitals are required to publish quality reports. These are lengthy documents which are not easily accessible for patients and are criticized for providing little valuable information [11]. There is also no official platform which allows patients to compare hospitals. However, some sickness funds and other institutions provide online guidance on hospital choice. Hospitals can participate in voluntary quality inspections and certification procedures ([8,p. 81]). Minimum service volumes are defined for several projectable services, e.g., transplants. The introduction of a pay-for-performance scheme is planned for the next years. Two studies have examined how active patients are in choosing different hospital services. Wübker et al. [12] find that voluntary publication of quality data attracts more patients. Furthermore, hospitals which publish data increase their market share if their quality is above average, while hospitals with lower than average quality lose market shares. Pilny and Mennicken [13] find that a significant share of patients undergoing a coronary artery bypass graft are willing to accept additional travel time to be treated in a hospital with better reputation.
3. Review of competition policies Two elements of competition are very present in German SHI. First, sickness funds compete for SHI-members. Second, SHI-members have free choice between SHIaffiliated physicians (general practitioners as well as ambulatory specialists) and between hospitals. This creates competition for patients in areas with a high density of suppliers. In addition, the introduction of DRGs has intensified competition in the hospital sector. However, the German SHI system is mainly run by collective negotiations in which the scope of services and payments are determined. This runs counter to the idea of competition. A further obstacle for competition is that hospitals are mostly confined to inpatient care [14]. Since 2002, a number of policies have been introduced which potentially increase competition between providers. One approach was to allow and promote selective contracting by sickness funds with selected providers, in particular with ambulatory care providers. Some of these contracts provide an add-on to the services agreed in the collective negotiations, others substitute for these services. They will be reviewed in Section 3.1. In recent years, competition between ambulatory care providers and hospitals has also increased to some extent. This is the focus of Section 3.2. Finally, Section 3.3 reviews the application of competition law in the German health care sector.
Please cite this article in press as: Kifmann M. Competition policy for health care provision in Germany. Health Policy (2016), http://dx.doi.org/10.1016/j.healthpol.2016.11.014
G Model HEAP-3660; No. of Pages 7
ARTICLE IN PRESS M. Kifmann / Health Policy xxx (2016) xxx–xxx
4
3.1. Selective contracting Initially, sickness funds barely had possibilities to differentiate regarding the scope of services provision which was highly regulated at the level of self-governance. Competition between the funds was mainly limited to price competition. This motivated policymakers to allow selective contracting. It was hoped that quality competition between sickness funds would intensify and that health care providers would compete to contract with sickness funds beyond the collectively negotiated payments. Overall, the objective was to achieve a better coordination and cooperation in health care, resulting in higher quality, higher efficiency and the introduction of innovations [14]. In 2002, sickness funds were given the option to introduce Disease-Management Programs (DMPs) for chronic diseases (asthma, breast cancer, COPD, diabetes type I and II, ischemic heart disease). The aim was to organize care for the chronically ill according to evidence-based guidelines and to coordinate services among different providers, mostly at the ambulatory care level. Sickness funds recruit patients with participation being voluntary. Providers are reimbursed for the additional services they provide for the DMP. The design of reimbursement can be freely chosen and is determined in negotiations. Fee-for-service, capitations and pay-for-performance elements are used. However, little information is available on the combination of these elements as contracts are usually not disclosed. Sickness funds had strong financial incentives to set up DMPs because the risk-adjustment scheme considered membership in a DMP as an adjustor until 2008, leading to considerably higher risk-adjustment payments for DMP patients. Today, sickness funds receive an administratively set capitation for each member inscribed in a DMP. Furthermore, payments from the risk-adjustment scheme tend to be higher as DMP indications are used as risk adjustors. In 2012, more than 10,000 programs existed with 7.16 million participants [15]. Sickness funds are obliged to externally evaluate the DMPs and publish the results. However, a limitation is that most evaluations do not use a control group receiving routine care. In a review of evaluations until 2008, the Federal Insurance Authority concluded that most evaluations of DMPs for diabetes type II report positive effects on patient outcomes [16]. Considering only evaluations of DMPs for diabetes type II with control groups, Fuchs et al. [17] find that these DMPs have positive effects with respect to mortality and survival time. There are also improvements in process parameters, indicating a positive impact on the health of patients with diabetes. No clear effect emerges for other outcomes and economic parameters. In a controlled observational study using administrative data on a DMP for COPD, Achelrod et al. [18] find significant improvements in mortality, morbidity and process quality. Costs were found to be significantly higher. Another important form of selective contracts are Integrated Care Contracts (ICC, Integrierte Versorgungsverträge). These aim at overcoming intersectoral barriers to care and use managed care concepts from the US, e.g., case management and coordinated patient
pathways. ICCs can integrate different providers at the same system level (horizontal integration, e.g., within the ambulatory care sector) or integrate providers across sectors (vertical integration, e.g., ambulatory care and inpatient care). Indication-based contracts focus on case management for a certain indication, e.g. hip replacements or stroke patients. Population-based contracts aim at providing care for a certain population in a broad field of indications. Sickness funds negotiate ICCs with single providers or a network of providers. In addition to physicians and hospitals, rehabilitative care providers and other health professionals can participate. The design of reimbursement is determined in negotiations between the contracting parties and can involve any scheme like fee-for-service, service-based or capitation-based lumpsums [14]. The participation of patients in ICC is voluntary. With the enrolment in a program, patients are in principle committed to the contracted health care providers. In practice, however, sickness funds do not penalize patients for seeking care from different providers. Initiatives to foster Integrated Care Contracts already started in the 1990s. However, several barriers made such contracts unattractive for sickness funds, in particular the requirement that a regional physician association needed to be a contractual partner and that spending should not be increased. These barriers were removed in 2004. Furthermore, between 2004 and 2008, one percent of the resources for ambulatory physicians were earmarked for ICCs. By 2008, about 6400 contracts were in place. This number remained roughly constant until 2011 but the number of participants increased from 1.66m to 1.92m [19]. In contrast to DMPs, services provided within ICCs can substitute for services provided in the collectively negotiated contract. A correction of the transfers from sickness funds to the regional physician associations becomes necessary in order to avoid that sickness funds pay twice. This has been a controversial topic as physician associations resisted the loss of revenue. Only in 2009, the parties agreed on a procedure to correct the transfers from sickness funds. The majority of ICCs are indication-based. Among the population-based contracts, the project “Gesundes Kinzigtal” which provides integrated care for a valley in the Black Forest has received particular attention. A health management company, two large sickness funds in cooperation with a physicians’ network, hospitals and several other providers have rearranged the organization of health care, incentivized by sharing of financial gains. Reaching half of the population of the region, they improved patient pathways and fostered cooperation and communication between providers. Evaluations point to financial gains for all partners and improved health outcomes for the population [20]. Overall, little is known about the cost-effectiveness of ICCs as there is no obligation to evaluate the contracts. The fact that over 6000 contracts are in place could be interpreted as a sign of cost-effectiveness because otherwise the contractual parties would not continue these contracts. This view, however, neglects that selective contracts may also be useful instruments to influence the coding of patients to generate higher payments from the risk-adjustment scheme.
Please cite this article in press as: Kifmann M. Competition policy for health care provision in Germany. Health Policy (2016), http://dx.doi.org/10.1016/j.healthpol.2016.11.014
G Model HEAP-3660; No. of Pages 7
ARTICLE IN PRESS M. Kifmann / Health Policy xxx (2016) xxx–xxx
Gatekeeping Contracts (Hausarztzentrierte Versorgung) are another example for selective contracting. In 2004, sickness funds were given the option of signing such contracts with GPs. The idea was to strengthen the position of GPs in the health care system and to improve the coordination of health care received from different providers, potentially saving costs. Initially, sickness funds were very reluctant to conclude such contracts because they faced additional expenditure for GPs while it was not clear whether these programs would lead to cost reduction in other areas. In addition, there was uncertainty about the correction of their payments to the physician associations. The law was changed in 2007 obliging sickness funds to conclude such contracts with a contracting partner representing more than 50% of GPs in an area. This gave the members of the German Association of Family Physicians (Deutscher Hausärzteverband) a very strong position as a contracting partner. However, sickness funds can also conclude “add-on” contracts in cooperation with a regional physician association, embedding Gatekeeping Contracts within the traditional corporatist system. Contracts with members of the German Association of Family Physicians, by contrast, replace parts of the services provided within the corporatist system. As in the case of ICCs, a correction of the transfers from sickness funds to the regional physician associations becomes necessary. Currently, 91 such contracts are active [21]. Overall, almost 4m SHI-members are inscribed and over 16,500 GPs participate [22]. Reimbursement of the contracts with members of the Deutsche Hausärzteverband consists of a mix of lumpsum payments, surcharges and fees for specific services. An exemplary contract pays a GP a lump-sum of D 60 Euro annually for each patient who is inscribed in the contract. If the patient visits the patient, a lump-sum of D 40 is paid per quarter with a maximum of D 120 per year. For patients with chronic diseases, the GP receives a further D 20 per quarter. This is complemented by a number of surcharges and fees which can be related to the physician’s performance [23]. Only few Gatekeeping Contracts have been evaluated. These evaluations point to improvements in a number of dimensions, e.g., there are fewer avoidable hospital cases and the number of dangerous drug combinations has been reduced. Patients also show a higher level of satisfaction. Yet, there are also surprising results, such as more contacts with specialists and higher pharmaceutical expenditure [24]. The cost effects are not clear. While two evaluations of contracts with the Deutsche Hausärzteverband point to cost reductions, the experience with add-on contracts points to cost increases, at least in the short run. 3.2. Competition between ambulatory care providers and hospitals The German health care system is characterized by a strong separation between outpatient and inpatient care. Sickness funds negotiate separately with ambulatory physicians and hospitals. Services are reimbursed in different ways (see Section 2). Also different rules apply for the use of new technologies. In the outpatient sector, these
5
need to be approved by the Federal Joint Committee before they can be used. In the inpatient sector, hospitals can use them as long as the Federal Joint Committee does not explicitly exclude them. Reimbursement is on top of DRG payments after approval from the German DRG Institute ([8,p. 73]). Traditionally, hospitals are mainly restricted to inpatient care. Specialists in the ambulatory sector strongly oppose that hospitals offer outpatient specialist services. Still, there are areas where hospitals can offer such services and where competition has increased in the last decade. Since 2004, hospitals can offer ambulatory care for patients with certain rare diseases and special forms of disease progression, as well as highly specialized services. In particular, this covers the treatment of cancer patients. First, it was up to the sickness funds to conclude such contracts. After 2007, the states decided whether a hospital can offer specialized services. Typically, larger hospitals offer these services. Today, the Federal Joint Committee defines the diseases and the requirements. Specialized physicians and hospitals can apply to provide these services. For the treatment of cancer patients, ambulatory physicians and hospitals are required to enter into cooperation agreements. In 2014, payments to hospitals for these services amounted to D 174.2m, equivalent to 0.25% of total expenditure on hospital care. An indication for strong competition is that 32.7% of hospitals reported conflicts with ambulatory physicians and regional physician associations in a questionnaire study ([19,p. 256]). Another area of competition are ambulatory surgeries. Both, specialists in the ambulatory sector and hospitals, can offer such surgeries. Since 2006, such surgeries have become particularly attractive for the first group because budgetary restrictions were lifted. In 2014, spending amounted to D 1.8bn for ambulatory surgeries by specialists and D 0.6bn for ambulatory surgeries by hospitals. Studies show high levels of satisfaction for patients who were operated upon by ambulatory specialists [25,26]. However, there are no studies examining how competition has affected outcomes. One conjecture is that the ambulatory surgeries by specialists have substituted for more expensive surgeries in hospitals. However, it has been criticized that the lifting of budgetary restrictions has created incentives for ambulatory specialists to induce demand ([19,p. 290]). In 2004, a new form of outpatient treatment was introduced. Medical treatment centers (Medizinische Versorgungszentren) are headed by a physician and employ physicians and other health care professionals. The idea was to foster cooperation across medical disciplines. Two studies find that patients judge quality to be higher in medical treatment centers compared to standard outpatient care [27,28]. However, both studies only elicited the preferences of patients who obtained treatment in these centers. Since hospitals are allowed to run these centers, medical treatment centers can intensify competition between ambulatory care providers and hospitals. Strategically, opening a center can be attractive for hospitals. It can be used to prepare patients for inpatient stay and provide follow-up treatment. Furthermore, it can be a means to channel-in patients. In 2014, 2073 medical treatment cen-
Please cite this article in press as: Kifmann M. Competition policy for health care provision in Germany. Health Policy (2016), http://dx.doi.org/10.1016/j.healthpol.2016.11.014
G Model HEAP-3660; No. of Pages 7
ARTICLE IN PRESS M. Kifmann / Health Policy xxx (2016) xxx–xxx
6
ters were registered, 38.4% of which were run by hospitals [29]. 3.3. Application of competition law in the German health care sector Collective contracts between sickness funds and associations of health care providers are the backbone of the German public health care system. The Social Code Book obliges the different associations to conclude such contracts. For this reason, collective contracts are exempted from German competition law. However, the Social Code Book states explicitly that competition law with respect to “abuse control” applies to selective contracting by sickness funds. Furthermore, drug discount contracts between sickness funds and pharmaceutical companies are subject to competition law. The competition authority is also responsible for mergers of sickness funds and of hospitals. With respect to selective contracts, only one case is documented in which the competition authority (Bundeskartellamt) intervened. In 2013, it started a case against a selective contract between a sickness funds and an association of ophthalmologists in Brandenburg which banned ophthalmologists located in other states from joining the association and therefore the contract. As a result, the contract had to be opened for other ophthalmologists. Hospital mergers are subject to German Antitrust law. Mergers should be prohibited if the merged firm obtains more than 40% market share or if the merger leads to significant market concentration (joint dominance with 3 or less firms with 50%, or of 5 or less firms with 66% market share). If hospitals intend to merge, the competition authority defines the market according to comparable services from the perspective of patients, e.g., the market for acute care or rehabilitative care. Only hospitals which can be reached by patients within a certain time are included. Between 2004 and 2014, the competition authority approved 182 mergers and prohibited 7 mergers [30]. Mergers were prohibited to prevent significant market power in the relevant regional market. The biggest merger approved yet was the take-over of 40 hospitals and 13 medical treatment centers of Rhön-Klinikum AG by Helios Kliniken, both private hospital chains. Initially, a take-over of 43 hospitals and 15 medical treatment centers was planned. The number was reduced after the competition authority raised concerns about four hospital markets [31]. Little is known about the economic effects of hospital mergers. In a report for the association of sickness funds, Preusker et al. [32] find that hospital mergers played only a minor role for hospital closures. 4. Key lessons and policy implications In the last decades, Germany has introduced a number of competitive elements into its corporatist public health insurance system. Selective contracting between sickness funds and providers was regarded as a way to improve quality and efficiency by encouraging better coordination and cooperation in health care. However, it took some time before sickness funds took initiative. An important lesson
is that contracting only got started when either further financial incentives were given, as in the case of DMPs and Integrated Care Contracts, or when sickness funds were forced to conclude contracts as in the case of Gatekeeping Contracts. The likely explanation for this phenomenon is that sickness funds are reluctant to provide special care options because these can attract high-risk individuals. This, in turn, may result in a competitive disadvantage for a sickness fund. This problem was particularly relevant until 2008 when the risk-adjustment scheme did not yet use morbidity-related data. Even with improved risk adjustment, however, it is not clear whether the problem is resolved. The scheme is restricted to 80 diseases and only compensates average cost. The incentives to set up selective contracts which improve quality at higher cost remain low. However, selective contracting can also be a means to benefit from risk adjustment if sickness funds can influence coding in their favor. The experience of Germany also demonstrates that introducing competition into a corporatist system produces frictions. The corporatist stakeholders resist change. A controversial topic has been how payments for the collective systems should be reduced if services are provided via a selective contract. Some competition between hospitals and outpatient care providers has also been introduced in recent years. Hospitals can provide selected services for outpatient services, physicians can perform ambulatory surgeries. Whether competition has improved care for certain groups of patients and reduced costs still needs to be corroborated with empirical data on outcomes and costs. Overall, it is difficult to judge to what extent the competitive reforms have achieved their objectives since only few studies evaluate the effects on both outcomes and costs with good data. In a corporatist system in which each group strongly guards its interest, such data is hard to obtain. For selective contracting, there are either no evaluation requirements as in the case of Integrated Care Contracts or few methodological requirements as in the case of Disease Management Programs. Nevertheless, the few high-quality studies indicate improvements in the quality of care. The effects on costs are not yet clear. The evidence rather points to cost increases, at least in the short run.
Acknowledgments This research is supported by the Health Foundation, an independent charity committed to bringing about better health and health care for people in the UK. The views expressed in this paper and in the case studies are those of the authors and do not necessarily reflect the views of the funders.
References [1] Bundesamt für Statistik. Gesundheitsausgaben. Wiesbaden; 2016. [2] Bünnings C, Schmitz H, Tauchmann H, Ziebarth NR. How health plan enrollees value prices relative to supplemental benefits and service quality. SOEP-papers 714-2015; 2015.
Please cite this article in press as: Kifmann M. Competition policy for health care provision in Germany. Health Policy (2016), http://dx.doi.org/10.1016/j.healthpol.2016.11.014
G Model HEAP-3660; No. of Pages 7
ARTICLE IN PRESS M. Kifmann / Health Policy xxx (2016) xxx–xxx
[3] Schmitz H, Ziebarth N. Does price framing affect the consumer price sensitivity of health plan choice? Journal of Human Resources 2017 [forthcoming]. [4] Brosse J, Himmel K. Direct risk selection in the German health insurance system: a field experiment (unpublished manuscript). Universität Hamburg; 2016. [5] Bundesamt für Statistik. Kostenstruktur bei Arztpraxen 2011. Wiesbaden; 2013. [6] Roll K, Stargardt T, Schreyögg J. Effect of type of insurance and income on waiting time for outpatient care. The Geneva Papers on Risk and Insurance—Issues and Practice 2012;37:609–32. [7] Karlsson M, Kleibrink J, Ziebarth N. Exit, voice or loyalty? An investigation into mandated portability of front-loaded private health plans (unpublished paper). University of Duisburg-Essen; 2016. [8] Busse R, Blümel M. Germany: health system review. Health Systems in Transition 2014;16(2):1–296. [9] Bundesamt für Statistik. Grunddaten der Krankenhäuser. Wiesbaden; 2015. Krankenhausgesellschaft. Bestandsaufnahme zur [10] Deutsche Krankenhausplanung und Investitionsfinanzierung in den Bundesländern, Stand: August 2015. Berlin; 2015. [11] Lorson P, Wigger C, Telsemeyer L. Qualitätsberichte verfehlen ihre zielgruppe. Führen & Wirtschaften im Krankenhaus 2011;28:37–40. [12] Wübker A, Sauerland D, Wübker A. Does better information about hospital quality affect patients’ choice? Empirical findings from Germany. Jahrbücher für Nationalökonomie und Statistik 2010;230(4):467–90. [13] Pilny A, Mennicken R. Does hospital reputation influence the choice of hospital? Ruhr economic papers #516; 2014. [14] Cassel D, Ebsen I, Greß S, Jacobs K, Schulze S, Wasem J. Vertragswettbewerb in der GKV. Möglichkeiten und Grenzen vor und nach der Gesundheitsreform der Großen Koalition. Bonn/Berlin: Wissenschaftliches Institut der AOK; 2008. [15] Bundesversicherungsamt. Zulassung der strukturierten Behandlungsprogramme (Disease Management-Programme — DMP) das Bundesversicherungsamt; 2014. http://www. durch bundesversicherungsamt.de/weiteres/disease-managementprogramme/zulassung-disease-management-programme-dmp.html. [8 March 2016]. [16] Bundesversicherungsamt. Bericht des Bundesversicherungssamtes zur vergleichenden Evaluation von strukturierten Behandlungsprogrammen bei Diabetes mellitus Typ 2; 2009. http://www.bundesversicherungsamt.de/fileadmin/redaktion/ DMP-Veranstaltungen/Evaluationsergebnisse DM2 03-06 bericht. pdf. [8 March 2016]. [17] Fuchs S, Henschke C, Blümel M, Busse R. Disease management programs for type 2 diabetes in Germany — a systematic literature review evaluating effectiveness. Deutsches Ärzteblatt International 2014;111:453–63. [18] Achelrod D, Welte T, Schreyögg J, Stargardt T. Cost-effectiveness of the German disease management programme (DMP) for chronic obstructive pulmonary disease (COPD) — a large population-based cohort study. Health Policy 2016;120(9):1029–39.
7
[19] Sachverständigenrat zur Begutachtung der Entwicklung im Gesundheitswesen. Wettbewerb an der Schnittstelle zwischen ambulanter und stationärer Gesundheitsversorgung. Sondergutachten 2012; 2012. Bonn. http://www.svr-gesundheit.de/ fileadmin/user upload/Gutachten/2012/GA2012 Langfassung.pdf. [20] Hildebrandt Helmut, et al. Gesundes Kinzigtal Integrated Care: improving population health by a shared health gain approach and a shared savings contract. International Journal of Integrated Care 2010;10(2):154–60. [21] Deutscher Hausärzteverband. Hausarztverträge, nach KV-Regionen; 2016 geordnet https://www.hausaerzteverband.de/cms/Hausarztvertraege.988.0. html. [22] Deutscher Hausärzteverband. Pressemitteilung 21.01.2016: Patientensteuerung ist ein Erfolgsmodell; Hausärztliche https://www.hausaerzteverband.de/cms/uploads/media/ 2016 2016 01 21 PM DHAEV Hausaerztliche Patientensteuerung .pdf. [23] Deutscher Hausärzteverband. Einfaches Vergütungssystem. In: Sonderbeilage Der Hausarzt, Februar/März 2015; 2015 https://www.hausaerzteverband.de/cms/fileadmin/user upload/ redaktion/bundesverband/Hausarzt/SB HA 1503 16S HZV.pdf. [24] Weigeldt U, Tesic D. Hausarztzentrierte Versorgung nach § 73b SGB V: Vollversorgungsvertrag oder Add-on-Vertrag? Gesundheits- und Sozialpolitik 2015;69:37–45. [25] March S, Swart E, Robra B. Patientenzufriedenheit beim ambulanten/kurzzeitchirurgischen Operieren in einer Praxisklinik. Das Gesundheitswesen 2006;68:376–82. [26] Bayern KV. Qualitätsmaßnahme Ambulante Operationen — Ergebnisse zur Patientenzufriedenheit und zum postoperativen Patientenverlauf bei ambulanten Operationen; 2010. [27] Schulte H, Schulz C. Medizinische Versorgungszentren: Verbesserung der ambulanten Patientenversorgung versus Selektion und Exklusion von Patientengruppen. Baden-Baden: Nomos-Verlag; 2007. [28] Haas L. Präferenzmessung von Patienten in medizinischen Versorgungszentren (MVZ): eine Gegenüberstellung der Behandlungsqualität von MVZ und niedergelassenen Haus- und Fachärzten. Das Gesundheitswesen 2011;73:409–15. [29] Kassenärztliche Bundesvereinigung. Medizinische Versorgungszentren aktuell zum Stichtag; 2016. Berlin. [31 December 2014]. [30] Bundeskartellamt. Jahresbericht 2014. Bonn: Bundeskartellamt; 2015. [31] Bundeskartellamt. Bundeskartellamt gibt den Erwerb von 40 Kliniken und 13 medizinischen Versorgungszentren der RhönKlinikum AG durch Helios Kliniken (Fresenius) frei; 2014. [32] Preusker U, Müschenich und M, Preusker S. Darstellung und Typologie der Marktaustritte von Krankenhäusern Deutschland 2003-2013. Gutachten im Auftrag des GKV-Spitzenverbandes; 2014.
Please cite this article in press as: Kifmann M. Competition policy for health care provision in Germany. Health Policy (2016), http://dx.doi.org/10.1016/j.healthpol.2016.11.014
ARTICLE IN PRESS
HEAP-3660; No. of Pages 7
Health Policy xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Health Policy journal homepage: www.elsevier.com/locate/healthpol
Competition policy for health care provision in Germany Mathias Kifmann Universität Hamburg, Esplanade 36, 20354 Hamburg, Germany
a r t i c l e
i n f o
Article history: Received 15 June 2016 Received in revised form 24 October 2016 Accepted 16 November 2016 Keywords: Hospitals General practitioners Competition policy Germany Regulated competition Corporatism
a b s t r a c t Since the 1990s, Germany has introduced a number of competitive elements into its public health care system. Sickness funds were given some freedom to sign selective contracts with providers. Competition between ambulatory care providers and hospitals was introduced for certain diseases and services. As competition has become more intense, the importance of competition law has increased. This paper reviews these areas of competition policy. The problems of introducing competition into a corporatist system are discussed. Based on the scientific evidence on the effects of competition, key lessons and implications for future policy are formulated. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Traditionally, Germany’s social health insurance system is governed jointly by associations. The national association of sickness funds, the national and regional associations of physicians and the German Hospital Federation negotiate contract conditions and payment schemes for ambulatory and hospital care. In the 1990s, however, competitive elements were introduced in the system. The first major reform concerned sickness funds. Since 1996, almost all individuals insured in the social health insurance system can choose freely among sickness funds. This has fundamentally transformed the funds from administrative entities into service-oriented institutions. Initially, sickness funds had few possibilities to influence the provision of health care services as these were predominantly determined by collective negotiations and legal requirements. Inspired by managed care in the US, sickness funds were given some freedom to sign selective contracts with providers in the late 1990s. Today sickness funds offer Disease Management Programs, Integrated Care
E-mail address: [email protected]
Contracts and Gatekeeping Contracts. The vast majority of health care services, however, continue to be managed by the corporatist system. A prominent feature of the German health care system is the strong separation of the outpatient and inpatient sector. Hospitals are mainly restricted to inpatient care. Outpatient specialist care is provided by physicians with specialist training. In the last decade, some competition between the sectors was introduced. Hospitals were allowed to offer ambulatory care for specific diseases and for certain specialized services. Hospitals also opened medical treatment centers for patients seeking outpatient care. Both, hospitals and ambulatory specialists, offer ambulatory surgeries. As competition in German health care has become more intense, the importance of competition law has increased. Selective contracting by sickness funds is subject to competition law. The competition authority is also responsible for mergers of sickness funds and of hospitals. Between 2004 and 2014, 182 hospital mergers were approved and 7 prohibited. This report provides information on these areas of competition policy in Germany. In Section 2, a short overview of the German health care system is given. In Section 3, com-
http://dx.doi.org/10.1016/j.healthpol.2016.11.014 0168-8510/© 2016 Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Kifmann M. Competition policy for health care provision in Germany. Health Policy (2016), http://dx.doi.org/10.1016/j.healthpol.2016.11.014
G Model HEAP-3660; No. of Pages 7
ARTICLE IN PRESS M. Kifmann / Health Policy xxx (2016) xxx–xxx
2
petition policies are reviewed. Section 4 concludes with key lessons and implications for future policy. The focus is on competition between health care providers. Section 2 also provides summary information about health insurance competition. 2. Institutional set-up 2.1. Basic characteristics In 2014, Germany spent 11.2% of its GDP on health care [1]. Most health care expenditure (66%) was financed by “Statutory Health Insurance” (SHI, Gesetzliche Krankenversicherung), the public health insurance program in which sickness funds provide health insurance. A share of 9% was paid for by the private health insurance system which insures public servants, self-employed persons and individuals who opt out of SHI. The remaining share was financed by private households and non-profit organizations (13%), by other social insurance programs or from public budgets (7%) and by employers (4%). In the SHI system, currently 118 sickness funds insure about 87% of the population. Mergers have decreased the number of funds considerably. In 1990, there were still 1147 funds. Sickness funds are quasi-public corporations and non-for-profit institutions. A major characteristic of SHI in Germany is self-governance. Payers and providers are organized in “corporatist bodies” with mandatory membership and special legal status. In particular, this comprises the national association of sickness funds, the national and regional associations of SHI-physicians and the German Hospital Federation. These associations negotiate contract conditions and payment schemes for ambulatory and hospital care. The regulatory framework is provided by the Social Code Book V (Sozialgesetzbuch). The federal and state (Länder) authorities supervise the self-governance of the associations. The most important decision-making body is the Federal Joint Committee (Gemeinsamer Bundesausschuss) which consists of ten representatives of the associations and three neutral members. Five patient representatives can put topics on the agenda, but have no voting rights. To finance SHI, a contribution rate of 14.6% is levied upon labor income up to an income ceiling of D 4237.50 in 2016. This leads to a maximum levy of about D 619 per month. The contributions are transferred to a Central Reallocation Pool which is administrated by the Federal Insurance Authority (Bundesversicherungsamt). For each member, sickness funds receive risk-adjusted capitation payments based on age, sex, disability status and morbidity. In addition, sickness funds can charge an additional contribution rate on the same income base. It currently ranges from 0.3 to 1.7%. Information on prices and supplemental benefits, such as alternative health care or integrated care, is readily available (see, e.g., https://www.gesetzlichekrankenkassen.de/). However, there is no official publication of performance indicators. As the benefit package of sickness funds is to a large extent specified by law, competition between sickness funds focuses on price. A study by Bünnings et al. [2] with
data from 2007 to 2010 finds that price differences are the main factor for the choice between sickness funds. Service quality and supplemental benefits only seem to play a minor role. From 2009 to 2014, premium differences between sickness funds were expressed in absolute Euro values as opposed to percentage point differences in contribution rates. Schmitz and Ziebarth [3] show that this policy change strongly increased the willingness to switch funds. Demand elasticities increased by a factor of four. There is an ongoing debate about the risk adjustment scheme. Among the issues discussed are the number of diseases considered and the manipulation of diagnostic information. In a field experiment, Brosse and Himmel [4] find evidence for risk selection by sickness funds against elderly and those with chronic diseases. In contrast to sickness funds, private health insurers charge premiums related to risk. In 2015, about 11% of the population were covered by private health insurance. Of the 42 members of the association of private health insurers, 24 are for-profit companies and 18 are mutual insurance societies. The privately insured consist of active and retired civil servants, self-employed and employees who have opted out of the sickness fund system. Opting out requires that income exceeds a threshold (in 2016, a yearly income of D 56,250). Most physicians treat both SHI-patients and private patients in the same practice. Private patients pay according to a different fee schedule for outpatient services which is considerably more generous than payment for SHI-patients and not subject to volume restrictions. In hospitals, private patients often pay for additional services, e.g. for treatment by the head doctor. Payments by private patients are an additional and important source of income for both physicians and hospitals. According to a study by the Federal Statistical Office, private patients generated 28% of physicians’ income compared to 69% for SHIpatients [5]. Roll et al. [6] find that private patients obtain appointments earlier and remain shorter in the waiting room. Private health insurers compete on price and on the level of coverage, e.g., on the size of the deductible. There is also competition with sickness funds for high earners who have the possibility to opt out of the sickness fund system. Private health insurers typically offer lower premiums for young, single and healthy individuals. However, social health insurance is attractive for those with pre-existing medical conditions. In contrast to private health insurers, sickness funds also insure children and a non-working spouse without additional contributions. Private health insurers save part of the premium income in early years to dampen premium increases in old age. Until 2009, these old-age provisions were not transferable. This strongly restricted competition within the private health insurance system. On the other hand, this regulation avoided that individuals who remain healthy opt out of contracts, leaving insurers with an adverse selection of individuals. Since 2009, a share of old-age provision has become portable. Karlsson et al. [7] examine the data of a private insurer and find only a small effect of portability on
Please cite this article in press as: Kifmann M. Competition policy for health care provision in Germany. Health Policy (2016), http://dx.doi.org/10.1016/j.healthpol.2016.11.014
G Model HEAP-3660; No. of Pages 7
ARTICLE IN PRESS M. Kifmann / Health Policy xxx (2016) xxx–xxx
switching behavior. However, internal switching between contracts offered by this insurer increased substantially. 2.2. Physicians in the social health insurance system About 98% of all ambulatory care physicians are affiliated with the SHI system. They are obligatory members of one of the 17 regional associations of SHI-physicians (Kassenärztliche Vereinigungen). The associations receive morbidity-adjusted capitations for each SHI-member from the sickness funds. In return, the associations guarantee the health care provision for SHI-members (Sicherstellungsauftrag). Total revenue is distributed by the associations to their members based on a complex catalog which specifies fees for cases as well as for specific services. Physicians face target budgets. If they exceed their budgets, the reimbursement for further cases is strongly reduced. Physicians are obliged to take part in external quality assurance measures and to implement internal quality management. Each year, the physician associations randomly select physicians and check whether specific medical services have been provided according to the quality standards. The Federal Joint Committee can define penalties if physicians do not meet their quality assurance obligations (see Ref. [8,pp. 81–82]). Most SHI-affiliated physicians work in private practices. In 2012, 46% were GPs (Hausärzte) while 54% were specialists (Fachärzte). Specialists who are employed by hospitals are not SHI-affiliated. In contrast to many other health care systems, patients can access specialists without first seeing a GP. This, however, applies only for ambulatory specialists and not for specialists in hospitals unless patients suffer from specific diseases (see Section 3.2). Patients also have the possibility to join Gatekeeping Contracts which require them to see a GP initially before seeing a specialist (see Section 3.1). 2.3. The hospital sector In 2014, the hospital sector was composed of a mix of public hospitals (29.7%), private non-profit hospitals (35.2%) and for-profit hospitals (35.1%). The total number was 1980 hospitals, down from 2166 ten years before. This number includes both single and multi-specialty hospitals. In particular, 273 hospitals only treat patients with psychiatric or neurological diseases. Between 2004 and 2014, take-overs and mergers increased the number of private for-profit hospital from 555 to 695 [9]. A considerable number of private for-profit hospitals are owned by chains. The two biggest are Helios with 111 and Asklepios with 102 hospitals. Hospitals are paid per case according to the German DRG-system. Relative weights of DRGs are determined at the national level by the German DRG Institute InEK (Institut für das Entgeltsystem im Krankenhaus) based on the cost data of selected hospitals. Multiplication with the statespecific case rate yields the per-case payment for a DRG. The state-specific case rates are negotiated by sickness funds and hospitals at the state level each year. In 2015, they ranged from D 3191 to D 3396. Furthermore, sickness funds and hospitals negotiate the kind and quantities of services with each hospital. German states also pay for
3
capital investments in hospitals. However, these payments have fallen considerably in the last decade. Between 1991 and 2014 they have almost halved in real terms [10]. A number of measures are implemented to secure hospital quality. Based on data reported by hospitals, federal or state quality-assurance institutions enter into a “structured dialogue” with a hospital if there are indications for low quality. The hospital must either explain the findings with special circumstances or show that the quality problem has been taken care of. Furthermore, hospitals are required to publish quality reports. These are lengthy documents which are not easily accessible for patients and are criticized for providing little valuable information [11]. There is also no official platform which allows patients to compare hospitals. However, some sickness funds and other institutions provide online guidance on hospital choice. Hospitals can participate in voluntary quality inspections and certification procedures ([8,p. 81]). Minimum service volumes are defined for several projectable services, e.g., transplants. The introduction of a pay-for-performance scheme is planned for the next years. Two studies have examined how active patients are in choosing different hospital services. Wübker et al. [12] find that voluntary publication of quality data attracts more patients. Furthermore, hospitals which publish data increase their market share if their quality is above average, while hospitals with lower than average quality lose market shares. Pilny and Mennicken [13] find that a significant share of patients undergoing a coronary artery bypass graft are willing to accept additional travel time to be treated in a hospital with better reputation.
3. Review of competition policies Two elements of competition are very present in German SHI. First, sickness funds compete for SHI-members. Second, SHI-members have free choice between SHIaffiliated physicians (general practitioners as well as ambulatory specialists) and between hospitals. This creates competition for patients in areas with a high density of suppliers. In addition, the introduction of DRGs has intensified competition in the hospital sector. However, the German SHI system is mainly run by collective negotiations in which the scope of services and payments are determined. This runs counter to the idea of competition. A further obstacle for competition is that hospitals are mostly confined to inpatient care [14]. Since 2002, a number of policies have been introduced which potentially increase competition between providers. One approach was to allow and promote selective contracting by sickness funds with selected providers, in particular with ambulatory care providers. Some of these contracts provide an add-on to the services agreed in the collective negotiations, others substitute for these services. They will be reviewed in Section 3.1. In recent years, competition between ambulatory care providers and hospitals has also increased to some extent. This is the focus of Section 3.2. Finally, Section 3.3 reviews the application of competition law in the German health care sector.
Please cite this article in press as: Kifmann M. Competition policy for health care provision in Germany. Health Policy (2016), http://dx.doi.org/10.1016/j.healthpol.2016.11.014
G Model HEAP-3660; No. of Pages 7
ARTICLE IN PRESS M. Kifmann / Health Policy xxx (2016) xxx–xxx
4
3.1. Selective contracting Initially, sickness funds barely had possibilities to differentiate regarding the scope of services provision which was highly regulated at the level of self-governance. Competition between the funds was mainly limited to price competition. This motivated policymakers to allow selective contracting. It was hoped that quality competition between sickness funds would intensify and that health care providers would compete to contract with sickness funds beyond the collectively negotiated payments. Overall, the objective was to achieve a better coordination and cooperation in health care, resulting in higher quality, higher efficiency and the introduction of innovations [14]. In 2002, sickness funds were given the option to introduce Disease-Management Programs (DMPs) for chronic diseases (asthma, breast cancer, COPD, diabetes type I and II, ischemic heart disease). The aim was to organize care for the chronically ill according to evidence-based guidelines and to coordinate services among different providers, mostly at the ambulatory care level. Sickness funds recruit patients with participation being voluntary. Providers are reimbursed for the additional services they provide for the DMP. The design of reimbursement can be freely chosen and is determined in negotiations. Fee-for-service, capitations and pay-for-performance elements are used. However, little information is available on the combination of these elements as contracts are usually not disclosed. Sickness funds had strong financial incentives to set up DMPs because the risk-adjustment scheme considered membership in a DMP as an adjustor until 2008, leading to considerably higher risk-adjustment payments for DMP patients. Today, sickness funds receive an administratively set capitation for each member inscribed in a DMP. Furthermore, payments from the risk-adjustment scheme tend to be higher as DMP indications are used as risk adjustors. In 2012, more than 10,000 programs existed with 7.16 million participants [15]. Sickness funds are obliged to externally evaluate the DMPs and publish the results. However, a limitation is that most evaluations do not use a control group receiving routine care. In a review of evaluations until 2008, the Federal Insurance Authority concluded that most evaluations of DMPs for diabetes type II report positive effects on patient outcomes [16]. Considering only evaluations of DMPs for diabetes type II with control groups, Fuchs et al. [17] find that these DMPs have positive effects with respect to mortality and survival time. There are also improvements in process parameters, indicating a positive impact on the health of patients with diabetes. No clear effect emerges for other outcomes and economic parameters. In a controlled observational study using administrative data on a DMP for COPD, Achelrod et al. [18] find significant improvements in mortality, morbidity and process quality. Costs were found to be significantly higher. Another important form of selective contracts are Integrated Care Contracts (ICC, Integrierte Versorgungsverträge). These aim at overcoming intersectoral barriers to care and use managed care concepts from the US, e.g., case management and coordinated patient
pathways. ICCs can integrate different providers at the same system level (horizontal integration, e.g., within the ambulatory care sector) or integrate providers across sectors (vertical integration, e.g., ambulatory care and inpatient care). Indication-based contracts focus on case management for a certain indication, e.g. hip replacements or stroke patients. Population-based contracts aim at providing care for a certain population in a broad field of indications. Sickness funds negotiate ICCs with single providers or a network of providers. In addition to physicians and hospitals, rehabilitative care providers and other health professionals can participate. The design of reimbursement is determined in negotiations between the contracting parties and can involve any scheme like fee-for-service, service-based or capitation-based lumpsums [14]. The participation of patients in ICC is voluntary. With the enrolment in a program, patients are in principle committed to the contracted health care providers. In practice, however, sickness funds do not penalize patients for seeking care from different providers. Initiatives to foster Integrated Care Contracts already started in the 1990s. However, several barriers made such contracts unattractive for sickness funds, in particular the requirement that a regional physician association needed to be a contractual partner and that spending should not be increased. These barriers were removed in 2004. Furthermore, between 2004 and 2008, one percent of the resources for ambulatory physicians were earmarked for ICCs. By 2008, about 6400 contracts were in place. This number remained roughly constant until 2011 but the number of participants increased from 1.66m to 1.92m [19]. In contrast to DMPs, services provided within ICCs can substitute for services provided in the collectively negotiated contract. A correction of the transfers from sickness funds to the regional physician associations becomes necessary in order to avoid that sickness funds pay twice. This has been a controversial topic as physician associations resisted the loss of revenue. Only in 2009, the parties agreed on a procedure to correct the transfers from sickness funds. The majority of ICCs are indication-based. Among the population-based contracts, the project “Gesundes Kinzigtal” which provides integrated care for a valley in the Black Forest has received particular attention. A health management company, two large sickness funds in cooperation with a physicians’ network, hospitals and several other providers have rearranged the organization of health care, incentivized by sharing of financial gains. Reaching half of the population of the region, they improved patient pathways and fostered cooperation and communication between providers. Evaluations point to financial gains for all partners and improved health outcomes for the population [20]. Overall, little is known about the cost-effectiveness of ICCs as there is no obligation to evaluate the contracts. The fact that over 6000 contracts are in place could be interpreted as a sign of cost-effectiveness because otherwise the contractual parties would not continue these contracts. This view, however, neglects that selective contracts may also be useful instruments to influence the coding of patients to generate higher payments from the risk-adjustment scheme.
Please cite this article in press as: Kifmann M. Competition policy for health care provision in Germany. Health Policy (2016), http://dx.doi.org/10.1016/j.healthpol.2016.11.014
G Model HEAP-3660; No. of Pages 7
ARTICLE IN PRESS M. Kifmann / Health Policy xxx (2016) xxx–xxx
Gatekeeping Contracts (Hausarztzentrierte Versorgung) are another example for selective contracting. In 2004, sickness funds were given the option of signing such contracts with GPs. The idea was to strengthen the position of GPs in the health care system and to improve the coordination of health care received from different providers, potentially saving costs. Initially, sickness funds were very reluctant to conclude such contracts because they faced additional expenditure for GPs while it was not clear whether these programs would lead to cost reduction in other areas. In addition, there was uncertainty about the correction of their payments to the physician associations. The law was changed in 2007 obliging sickness funds to conclude such contracts with a contracting partner representing more than 50% of GPs in an area. This gave the members of the German Association of Family Physicians (Deutscher Hausärzteverband) a very strong position as a contracting partner. However, sickness funds can also conclude “add-on” contracts in cooperation with a regional physician association, embedding Gatekeeping Contracts within the traditional corporatist system. Contracts with members of the German Association of Family Physicians, by contrast, replace parts of the services provided within the corporatist system. As in the case of ICCs, a correction of the transfers from sickness funds to the regional physician associations becomes necessary. Currently, 91 such contracts are active [21]. Overall, almost 4m SHI-members are inscribed and over 16,500 GPs participate [22]. Reimbursement of the contracts with members of the Deutsche Hausärzteverband consists of a mix of lumpsum payments, surcharges and fees for specific services. An exemplary contract pays a GP a lump-sum of D 60 Euro annually for each patient who is inscribed in the contract. If the patient visits the patient, a lump-sum of D 40 is paid per quarter with a maximum of D 120 per year. For patients with chronic diseases, the GP receives a further D 20 per quarter. This is complemented by a number of surcharges and fees which can be related to the physician’s performance [23]. Only few Gatekeeping Contracts have been evaluated. These evaluations point to improvements in a number of dimensions, e.g., there are fewer avoidable hospital cases and the number of dangerous drug combinations has been reduced. Patients also show a higher level of satisfaction. Yet, there are also surprising results, such as more contacts with specialists and higher pharmaceutical expenditure [24]. The cost effects are not clear. While two evaluations of contracts with the Deutsche Hausärzteverband point to cost reductions, the experience with add-on contracts points to cost increases, at least in the short run. 3.2. Competition between ambulatory care providers and hospitals The German health care system is characterized by a strong separation between outpatient and inpatient care. Sickness funds negotiate separately with ambulatory physicians and hospitals. Services are reimbursed in different ways (see Section 2). Also different rules apply for the use of new technologies. In the outpatient sector, these
5
need to be approved by the Federal Joint Committee before they can be used. In the inpatient sector, hospitals can use them as long as the Federal Joint Committee does not explicitly exclude them. Reimbursement is on top of DRG payments after approval from the German DRG Institute ([8,p. 73]). Traditionally, hospitals are mainly restricted to inpatient care. Specialists in the ambulatory sector strongly oppose that hospitals offer outpatient specialist services. Still, there are areas where hospitals can offer such services and where competition has increased in the last decade. Since 2004, hospitals can offer ambulatory care for patients with certain rare diseases and special forms of disease progression, as well as highly specialized services. In particular, this covers the treatment of cancer patients. First, it was up to the sickness funds to conclude such contracts. After 2007, the states decided whether a hospital can offer specialized services. Typically, larger hospitals offer these services. Today, the Federal Joint Committee defines the diseases and the requirements. Specialized physicians and hospitals can apply to provide these services. For the treatment of cancer patients, ambulatory physicians and hospitals are required to enter into cooperation agreements. In 2014, payments to hospitals for these services amounted to D 174.2m, equivalent to 0.25% of total expenditure on hospital care. An indication for strong competition is that 32.7% of hospitals reported conflicts with ambulatory physicians and regional physician associations in a questionnaire study ([19,p. 256]). Another area of competition are ambulatory surgeries. Both, specialists in the ambulatory sector and hospitals, can offer such surgeries. Since 2006, such surgeries have become particularly attractive for the first group because budgetary restrictions were lifted. In 2014, spending amounted to D 1.8bn for ambulatory surgeries by specialists and D 0.6bn for ambulatory surgeries by hospitals. Studies show high levels of satisfaction for patients who were operated upon by ambulatory specialists [25,26]. However, there are no studies examining how competition has affected outcomes. One conjecture is that the ambulatory surgeries by specialists have substituted for more expensive surgeries in hospitals. However, it has been criticized that the lifting of budgetary restrictions has created incentives for ambulatory specialists to induce demand ([19,p. 290]). In 2004, a new form of outpatient treatment was introduced. Medical treatment centers (Medizinische Versorgungszentren) are headed by a physician and employ physicians and other health care professionals. The idea was to foster cooperation across medical disciplines. Two studies find that patients judge quality to be higher in medical treatment centers compared to standard outpatient care [27,28]. However, both studies only elicited the preferences of patients who obtained treatment in these centers. Since hospitals are allowed to run these centers, medical treatment centers can intensify competition between ambulatory care providers and hospitals. Strategically, opening a center can be attractive for hospitals. It can be used to prepare patients for inpatient stay and provide follow-up treatment. Furthermore, it can be a means to channel-in patients. In 2014, 2073 medical treatment cen-
Please cite this article in press as: Kifmann M. Competition policy for health care provision in Germany. Health Policy (2016), http://dx.doi.org/10.1016/j.healthpol.2016.11.014
G Model HEAP-3660; No. of Pages 7
ARTICLE IN PRESS M. Kifmann / Health Policy xxx (2016) xxx–xxx
6
ters were registered, 38.4% of which were run by hospitals [29]. 3.3. Application of competition law in the German health care sector Collective contracts between sickness funds and associations of health care providers are the backbone of the German public health care system. The Social Code Book obliges the different associations to conclude such contracts. For this reason, collective contracts are exempted from German competition law. However, the Social Code Book states explicitly that competition law with respect to “abuse control” applies to selective contracting by sickness funds. Furthermore, drug discount contracts between sickness funds and pharmaceutical companies are subject to competition law. The competition authority is also responsible for mergers of sickness funds and of hospitals. With respect to selective contracts, only one case is documented in which the competition authority (Bundeskartellamt) intervened. In 2013, it started a case against a selective contract between a sickness funds and an association of ophthalmologists in Brandenburg which banned ophthalmologists located in other states from joining the association and therefore the contract. As a result, the contract had to be opened for other ophthalmologists. Hospital mergers are subject to German Antitrust law. Mergers should be prohibited if the merged firm obtains more than 40% market share or if the merger leads to significant market concentration (joint dominance with 3 or less firms with 50%, or of 5 or less firms with 66% market share). If hospitals intend to merge, the competition authority defines the market according to comparable services from the perspective of patients, e.g., the market for acute care or rehabilitative care. Only hospitals which can be reached by patients within a certain time are included. Between 2004 and 2014, the competition authority approved 182 mergers and prohibited 7 mergers [30]. Mergers were prohibited to prevent significant market power in the relevant regional market. The biggest merger approved yet was the take-over of 40 hospitals and 13 medical treatment centers of Rhön-Klinikum AG by Helios Kliniken, both private hospital chains. Initially, a take-over of 43 hospitals and 15 medical treatment centers was planned. The number was reduced after the competition authority raised concerns about four hospital markets [31]. Little is known about the economic effects of hospital mergers. In a report for the association of sickness funds, Preusker et al. [32] find that hospital mergers played only a minor role for hospital closures. 4. Key lessons and policy implications In the last decades, Germany has introduced a number of competitive elements into its corporatist public health insurance system. Selective contracting between sickness funds and providers was regarded as a way to improve quality and efficiency by encouraging better coordination and cooperation in health care. However, it took some time before sickness funds took initiative. An important lesson
is that contracting only got started when either further financial incentives were given, as in the case of DMPs and Integrated Care Contracts, or when sickness funds were forced to conclude contracts as in the case of Gatekeeping Contracts. The likely explanation for this phenomenon is that sickness funds are reluctant to provide special care options because these can attract high-risk individuals. This, in turn, may result in a competitive disadvantage for a sickness fund. This problem was particularly relevant until 2008 when the risk-adjustment scheme did not yet use morbidity-related data. Even with improved risk adjustment, however, it is not clear whether the problem is resolved. The scheme is restricted to 80 diseases and only compensates average cost. The incentives to set up selective contracts which improve quality at higher cost remain low. However, selective contracting can also be a means to benefit from risk adjustment if sickness funds can influence coding in their favor. The experience of Germany also demonstrates that introducing competition into a corporatist system produces frictions. The corporatist stakeholders resist change. A controversial topic has been how payments for the collective systems should be reduced if services are provided via a selective contract. Some competition between hospitals and outpatient care providers has also been introduced in recent years. Hospitals can provide selected services for outpatient services, physicians can perform ambulatory surgeries. Whether competition has improved care for certain groups of patients and reduced costs still needs to be corroborated with empirical data on outcomes and costs. Overall, it is difficult to judge to what extent the competitive reforms have achieved their objectives since only few studies evaluate the effects on both outcomes and costs with good data. In a corporatist system in which each group strongly guards its interest, such data is hard to obtain. For selective contracting, there are either no evaluation requirements as in the case of Integrated Care Contracts or few methodological requirements as in the case of Disease Management Programs. Nevertheless, the few high-quality studies indicate improvements in the quality of care. The effects on costs are not yet clear. The evidence rather points to cost increases, at least in the short run.
Acknowledgments This research is supported by the Health Foundation, an independent charity committed to bringing about better health and health care for people in the UK. The views expressed in this paper and in the case studies are those of the authors and do not necessarily reflect the views of the funders.
References [1] Bundesamt für Statistik. Gesundheitsausgaben. Wiesbaden; 2016. [2] Bünnings C, Schmitz H, Tauchmann H, Ziebarth NR. How health plan enrollees value prices relative to supplemental benefits and service quality. SOEP-papers 714-2015; 2015.
Please cite this article in press as: Kifmann M. Competition policy for health care provision in Germany. Health Policy (2016), http://dx.doi.org/10.1016/j.healthpol.2016.11.014
G Model HEAP-3660; No. of Pages 7
ARTICLE IN PRESS M. Kifmann / Health Policy xxx (2016) xxx–xxx
[3] Schmitz H, Ziebarth N. Does price framing affect the consumer price sensitivity of health plan choice? Journal of Human Resources 2017 [forthcoming]. [4] Brosse J, Himmel K. Direct risk selection in the German health insurance system: a field experiment (unpublished manuscript). Universität Hamburg; 2016. [5] Bundesamt für Statistik. Kostenstruktur bei Arztpraxen 2011. Wiesbaden; 2013. [6] Roll K, Stargardt T, Schreyögg J. Effect of type of insurance and income on waiting time for outpatient care. The Geneva Papers on Risk and Insurance—Issues and Practice 2012;37:609–32. [7] Karlsson M, Kleibrink J, Ziebarth N. Exit, voice or loyalty? An investigation into mandated portability of front-loaded private health plans (unpublished paper). University of Duisburg-Essen; 2016. [8] Busse R, Blümel M. Germany: health system review. Health Systems in Transition 2014;16(2):1–296. [9] Bundesamt für Statistik. Grunddaten der Krankenhäuser. Wiesbaden; 2015. Krankenhausgesellschaft. Bestandsaufnahme zur [10] Deutsche Krankenhausplanung und Investitionsfinanzierung in den Bundesländern, Stand: August 2015. Berlin; 2015. [11] Lorson P, Wigger C, Telsemeyer L. Qualitätsberichte verfehlen ihre zielgruppe. Führen & Wirtschaften im Krankenhaus 2011;28:37–40. [12] Wübker A, Sauerland D, Wübker A. Does better information about hospital quality affect patients’ choice? Empirical findings from Germany. Jahrbücher für Nationalökonomie und Statistik 2010;230(4):467–90. [13] Pilny A, Mennicken R. Does hospital reputation influence the choice of hospital? Ruhr economic papers #516; 2014. [14] Cassel D, Ebsen I, Greß S, Jacobs K, Schulze S, Wasem J. Vertragswettbewerb in der GKV. Möglichkeiten und Grenzen vor und nach der Gesundheitsreform der Großen Koalition. Bonn/Berlin: Wissenschaftliches Institut der AOK; 2008. [15] Bundesversicherungsamt. Zulassung der strukturierten Behandlungsprogramme (Disease Management-Programme — DMP) das Bundesversicherungsamt; 2014. http://www. durch bundesversicherungsamt.de/weiteres/disease-managementprogramme/zulassung-disease-management-programme-dmp.html. [8 March 2016]. [16] Bundesversicherungsamt. Bericht des Bundesversicherungssamtes zur vergleichenden Evaluation von strukturierten Behandlungsprogrammen bei Diabetes mellitus Typ 2; 2009. http://www.bundesversicherungsamt.de/fileadmin/redaktion/ DMP-Veranstaltungen/Evaluationsergebnisse DM2 03-06 bericht. pdf. [8 March 2016]. [17] Fuchs S, Henschke C, Blümel M, Busse R. Disease management programs for type 2 diabetes in Germany — a systematic literature review evaluating effectiveness. Deutsches Ärzteblatt International 2014;111:453–63. [18] Achelrod D, Welte T, Schreyögg J, Stargardt T. Cost-effectiveness of the German disease management programme (DMP) for chronic obstructive pulmonary disease (COPD) — a large population-based cohort study. Health Policy 2016;120(9):1029–39.
7
[19] Sachverständigenrat zur Begutachtung der Entwicklung im Gesundheitswesen. Wettbewerb an der Schnittstelle zwischen ambulanter und stationärer Gesundheitsversorgung. Sondergutachten 2012; 2012. Bonn. http://www.svr-gesundheit.de/ fileadmin/user upload/Gutachten/2012/GA2012 Langfassung.pdf. [20] Hildebrandt Helmut, et al. Gesundes Kinzigtal Integrated Care: improving population health by a shared health gain approach and a shared savings contract. International Journal of Integrated Care 2010;10(2):154–60. [21] Deutscher Hausärzteverband. Hausarztverträge, nach KV-Regionen; 2016 geordnet https://www.hausaerzteverband.de/cms/Hausarztvertraege.988.0. html. [22] Deutscher Hausärzteverband. Pressemitteilung 21.01.2016: Patientensteuerung ist ein Erfolgsmodell; Hausärztliche https://www.hausaerzteverband.de/cms/uploads/media/ 2016 2016 01 21 PM DHAEV Hausaerztliche Patientensteuerung .pdf. [23] Deutscher Hausärzteverband. Einfaches Vergütungssystem. In: Sonderbeilage Der Hausarzt, Februar/März 2015; 2015 https://www.hausaerzteverband.de/cms/fileadmin/user upload/ redaktion/bundesverband/Hausarzt/SB HA 1503 16S HZV.pdf. [24] Weigeldt U, Tesic D. Hausarztzentrierte Versorgung nach § 73b SGB V: Vollversorgungsvertrag oder Add-on-Vertrag? Gesundheits- und Sozialpolitik 2015;69:37–45. [25] March S, Swart E, Robra B. Patientenzufriedenheit beim ambulanten/kurzzeitchirurgischen Operieren in einer Praxisklinik. Das Gesundheitswesen 2006;68:376–82. [26] Bayern KV. Qualitätsmaßnahme Ambulante Operationen — Ergebnisse zur Patientenzufriedenheit und zum postoperativen Patientenverlauf bei ambulanten Operationen; 2010. [27] Schulte H, Schulz C. Medizinische Versorgungszentren: Verbesserung der ambulanten Patientenversorgung versus Selektion und Exklusion von Patientengruppen. Baden-Baden: Nomos-Verlag; 2007. [28] Haas L. Präferenzmessung von Patienten in medizinischen Versorgungszentren (MVZ): eine Gegenüberstellung der Behandlungsqualität von MVZ und niedergelassenen Haus- und Fachärzten. Das Gesundheitswesen 2011;73:409–15. [29] Kassenärztliche Bundesvereinigung. Medizinische Versorgungszentren aktuell zum Stichtag; 2016. Berlin. [31 December 2014]. [30] Bundeskartellamt. Jahresbericht 2014. Bonn: Bundeskartellamt; 2015. [31] Bundeskartellamt. Bundeskartellamt gibt den Erwerb von 40 Kliniken und 13 medizinischen Versorgungszentren der RhönKlinikum AG durch Helios Kliniken (Fresenius) frei; 2014. [32] Preusker U, Müschenich und M, Preusker S. Darstellung und Typologie der Marktaustritte von Krankenhäusern Deutschland 2003-2013. Gutachten im Auftrag des GKV-Spitzenverbandes; 2014.
Please cite this article in press as: Kifmann M. Competition policy for health care provision in Germany. Health Policy (2016), http://dx.doi.org/10.1016/j.healthpol.2016.11.014