CONTRACTING PRIVATE HOSPITALS:EXPERIENCES FROM SOUTHEAST AND EAST ASIA

Chantal Herberholz1 and Siripen SupakankuntiCentre for Health Economics, Faculty of Economics, Chulalongkorn University

Abstract In resource-scarce settings governments have increasingly looked at ways of engaging the private sector in achieving national health system goals. This study is a comparative analysis of institutional contracting for hospital services in four Southeast and East Asian countries, namely Thailand, Singapore, the Philippines and South Korea, which were selected based on the degree of government involvement in these countries’ health sectors. Primary data were collected through in-depth-interviews and analyzed under a triangulation approach. In three out of four countries, institutional contracting is mainly used to increase or ensure access to hospital services, while the scale of private hospitals participation depends on contextual factors. Neither strategic provider selection mechanisms nor a preferred provider system is part of the institutional contracting models reviewed. In Thailand and the Philippines, simple contracting rather than performance-based contracting is conducted, with hardly any dialogue between the parties, indicating that the contracting tool has not been used to the fullest extent possible and suggesting that capacity development especially regarding contract management is needed. Although there is virtually no information available about the cost of contracting, the findings of this study suggest that the potential of institutional contracting arrangements should be explored further to improve health system outcomes and thereby support countries in their quest for universal health coverage.

Keywords: institutional contracting, hospitals, private sector

Key messages- Institutional contracting is used as a tool to engage the private hospital sector to achieve

health system goals - The scale of private hospital engagement depends on contextual factors- Capacity development, especially regarding contract management, is important - The potential of institutional contracting arrangements should be explored further as

these can support countries in their quest for universal health coverage

Acknowledgements This paper was prepared as one of the background papers for the course on “Engaging the private sector to achieve health systems goals – Private hospitals and the private sector as partners for universal health care coverage: Policies, strategies and tools”, developed by the Asia Network for Capacity Building in Health Systems Strengthening (ANHSS) in conjunction with the World Bank Institute and delivered in May 2013 at the School of Public Health and Primary Care, The Chinese University of Hong Kong. It has been accepted for oral presentation at the Health Systems Reform in Asia: Equity, Governance and Social Impact conference, which will take place in Singapore from December 13 to 16, 2013. The study was funded by the Faculty of Economics of Chulalongkorn University, Bangkok, Thailand. The authors thank the 11 experts from Thailand, Singapore, the Philippines and South Korea for sharing their views.

Authors’ contributions

1 Corresponding author. Faculty of Economics, Chulalongkorn University, Phyathai Road, Bangkok 10330. Telephone: (+66) 2-218-6238 ext. 4, Fax: (+66) 2-218-6212, Email: chantal.h@chula.ac.th.

Chantal Herberholz was responsible for conducting the research and wrote the first and final versions of the draft. Siripen Supakankunti provided comments on the draft. The in-depth interviews were jointly conducted, except the in-depth-interviews with experts from the Philippines, which were conducted by Chantal Herberholz.

Word CountAbstract: 212Body: 6,619

1. Introduction

Since the 2005 World Health Assembly resolution 58.33 for Sustainable Health Financing, Universal Coverage and Social Health Insurance, many countries around the world have shown significant progress in ensuring that people have access to needed health services without experiencing financial catastrophe.

Resource constraints within health sectors in Southeast and East Asian countries, like in many other countries, are amplified by the demographic and epidemiological transition, which increasingly results in inter alia greater demand for hospital services. Governments will not be able to cope with these challenges unless they consider alternative approaches such as for example engaging the private sector.

A sizeable number of studies deal with the role of the private sector in national health systems. While some focus on the characteristics of the private health sector or a segment thereof, others look at specific tools that exist to engage the private sector such as contracting, franchising and social marketing (Barnes, et al. 2009, World Bank 2003 among others).

A uniform typology of possible forms of private sector engagement in the health sector does not exist, but three broad areas based on the work by Taylor & Blair (2002), De Pinho Campos, et al. (2011), Harding & Montagu (2012) and Mills & Broomberg (1998), namely (i) Public Health and Social Programmes (i.e. programme-centred approaches), (ii) Services (i.e. service-centred approaches) and (iii) Health Facilities (i.e. facility-centred approaches), can be identified as indicated in Figure 1.

Figure 1 Typology of forms of private sector engagement to achieve health system goals

[insert Figure 1 here]

Contracting of services2 is one approach that can be employed to engage the private sector and in the case of hospital services it comprises two major types ( Vining & Globerman, (1999):

1. Institutional contracting, i.e. contracting among public payer/purchaser and

private for-profit and not-for-profit hospitals for clinical and clinical support services (ranging from comprehensive package of services to selected disease/condition-specific services)

2. Specific organizational contracting, i.e. contracting among public hospitals and private sector for clinical and clinical support services (disease/condition-specific services) and non-clinical services

While a private sector exists in all countries, the role of private hospitals in addressing key challenges in the healthcare sector differs across countries depending on the degree of government involvement in the respective countries. Figure 2 shows the extent of government involvement in healthcare sectors, focusing on its role as payer/purchaser and provider of hospital services. In most of the selected Southeast and

2 Harding & Preker (2003) state that contracting entails an ongoing exchange relationship and define contracting as “a purchasing mechanism used to acquire a specified service, of a defined quantity and quality, at an agreed-on price, from a specific provider for a specified period”, while Evans (2006) simply refers to contracting as “a tool that formalizes the relationships and obligations between the different actors in the health system, …”.

East Asian countries, the government is the dominant provider of hospital services albeit not necessarily the major payer/purchaser.

Figure 2 Government share in health expenditures and the provision of hospital beds

[insert Figure 2 here]

With varying degrees of government involvement in health sectors across countries, contracting models employed are likely to differ. To date, no comparative study to document and analyse contracting models for hospital services among public and private partners and the contextual factors that affect these exists. Comparative studies on contracting health services generally focus on primary health care services, public health services (e.g. tuberculosis prevention and control) and nutrition services in developing countries. (Harding & Loevinsohn, 2005, Loevinsohn, 2008, Liu, et al., 2008, Mills, 1998 among others), although several studies look at specific country experiences with contracting (Ashton, et al., 2004 among others).

This study seeks to develop a framework that allows a comparative analysis of institutional contracting3 for hospital services. It aims at examining hospital contracting models for clinical and clinical support services employed in four Southeast and East Asian countries, i.e. Thailand, Singapore, the Republic of Korea and the Republic of the Philippines4, via a review of databases, academic and grey literature and in-depth interviews (IDIs). Contracting arrangements are mainly compared in terms of key contract design and process features, while the perceived impact of these contracting models in achieving stated goals is assessed in terms of access, equity, quality, efficiency and sustainability.

The policy implications of this study are relevant for decision-makers in countries with similar health systems where governments are interested in contracting with private hospitals.

2. Methodology

Primary data for the four countries were collected through IDIs with key experts. A total of 11 experts from academia (4), private hospitals (2), department of health (1), purchasing organisations (3) and a consulting firm (1) were selected purposively and invited to participate. IDIs were conducted by the researchers in March and April 2013 in person (Thailand (3), Singapore (3) and Republic of Korea (3)) and on the phone (Republic of the Philippines (2)). Experts from academia were chosen based on their research interests and publications in this field. IDIs were taped to facilitate a qualitative content analysis. Guideline questions for the IDIs are presented in Appendix 1. To validate the primary data and deepen the discussion triangulation was used. Secondary data were collected through a review of databases, academic and grey literature.

3 Specific organizational contracting is very common and mostly done for non-clinical services (e.g. catering, cleaning, laundry, parking and security) and clinical support services (e.g. laboratory tests and diagnostic procedures), although it has also been used for clinical services (see for example Herberholz & Supakankunti, 2011). Contracts for non-clinical and clinical support services are typically negotiated on a bilateral basis by the public hospital and the private company for each service and contain organization specific content that makes a meaningful comparison across a representative number of cases extremely difficult. In fact, specific organizational contracting for non-clinical and clinical support services had been used in all countries prior to the introduction of institutional contracting, indicating that a contracting learning curve exists.4 One country was chosen from each of the quadrants in Figure 2 to reflect the different environments in which public and private sectors interact.

The framework to compare contracting arrangements for hospital services among public and private partners across countries (shown in Figure 3) draws on England & HLSP Institute (2008), Liu, et al. (2007), Mudenda, et al. (2008) and de Roodenbeke (2004) and analyses contracting models within the context of the principal-agent theory. A principal-agent relationship occurs whenever a principal (the contractor in Figure 3) engages an agent (the contractee in Figure 3) to perform certain tasks on her behalf. Conflicts in the relationship between the two parties arise since the agent may have incentives to make decisions that differ from those desired by the principal, given that the agent typically has more information than the principal. Therefore, the principal has to ensure that the agent is acting in the principal’s best interest by choosing appropriate design and process features for the contracting arrangement, taking contextual factors into account, so that the objectives of principal and agent are more aligned. Key design and process features include information about the (i) nature of the contract, (ii) provider payment method, (iii) incentives, (iv) key performance indicators, (v) review and adjustment mechanism and (vi) dispute settlement, while contextual factors refer to for example healthcare system characteristics. Information about the nature of the contract captures information about the involved parties and the way providers are selected, “contractability”5 of the contracted services, type of contract and contract duration. Provider payment refers to the mechanisms used to pay and implicitly or explicitly incentivise providers, the key variables being price, quantity and payment methods. The key performance indicators (KPI) specify the performance requirements of the contracting arrangement to achieve stated objectives such as increased access to services, more equity in the provision of services, higher quality of care, higher efficiency in service delivery and sustainability of the service provision as well as their measurement. KPIs within the framework of a balanced scorecard, which has been applied to hospitals, packages of health services and health systems, for example relate to the following four areas, financial situation, patients, internal business processes and innovation. (Peters, et al., 2007, Ten Asbroeck, et al., 2004 among others) Review and adjustment mechanisms look at ways in which future contingencies and re-allocations of risk may be addressed given that internal or external factors such as for example technology, demographic and disease patterns are likely to change as time passes by. The dispute settlement arrangements determine if any disputes are to be solved in or out of courts, the latter through for example negotiations, mediation or arbitration depending on the type of the contract.

Figure 3 Conceptual framework

[insert Figure 3 here]

Transaction cost economics then explains (i) the choice between internal production and contracting and (ii) in case of the latter that the contracting arrangement that minimises agency costs, given contextual factors, is chosen. Transaction costs consist of production costs and governance (bargaining costs and opportunism costs) (Vining & Globerman, 1999). Donato (2010) distinguishes between two extreme contracting types, namely classical-type contracting, based on rigid and complete contracts in which the identities of the parties hardly matter, and relational (i.e. incomplete and flexible) contracting models in which the identities of the parties are crucial. The higher asset specificity, the higher uncertainties and the more contestable markets are, the greater the transaction costs and the more likely the occurrence of more incomplete contracts. (Donato, 2010; Bose, et al., 2008)

5 According to Liu, et al. (2007) “contractability” refers to the degree to which the quantity and quality of services can be measured and monitored as well as the level of contestability in the market.

3. Country context

Thailand’s Universal Coverage Scheme (UCS) was introduced in 2001 at a per capita GDP of 1,808 (current) USD (World Bank, n.d.) and as of 2009 the entire Thai population was covered either by the UCS, the Social Security Scheme (SSS; formal sector employees and dependents, some voluntary informal sector members) or the Civil Servant Medical Benefit Scheme (CSMBS; civil servants and dependents) as shown in Table 1. While the UCS and the CSMBS are funded through general taxes, the SSS collects social insurance contributions in form of a payroll tax. Provider payment methods differ across the three schemes. While the UCS and the SSS both use closed-end provider payment mechanisms, the CSMBS is a public reimbursement scheme, using fee for service for outpatient services and rates based on DRG for inpatient services. All three schemes offer a comprehensive benefit package although some differences exist. Co-payments are generally not required, except in a few cases such as for example for drugs that are not listed in the essential drug list. 75 per cent of hospitals in Thailand are public, reflecting that Thailand has a publicly-dominated, yet mixed, healthcare system, while the remaining 25 per cent are mostly for profit private hospitals, mainly located in urban areas.

Table 1 Healthcare systems in Thailand, Singapore, Philippines and South Korea

[insert Table 1 here]

Singapore’s healthcare system is also a mixed system. It is mainly administered through the Central Provident Fund (CPF), which was established by the British in 1955 and consists of three accounts, two for retirement and one for healthcare (the so-called MediSave account). (Reisman, 2009) MediSave, a national medical savings account scheme, established in 1984, is one of Singapore’s three “Ms”, together with MediShield and MediFund as shown in Table 1. Employed and self-employed individuals (the latter with income above a certain threshold) must have a MediSave account, in which they make income-dependent contributions that can be withdrawn subject to certain limits to pay for healthcare received at public and private hospitals. MediShield, on the other hand, is an opt-out medical insurance aimed at helping insureds pay for catastrophic healthcare expenses and is operated by the CPF. The insurance premia for MediShield can be paid with MediSave savings. In 1993, an endowment fund, MediFund, was established to help those who cannot afford to pay for healthcare despite MediSave and MediShield. In light of Singapore’s demographic transition, ElderShield, a severe disability insurance run by three private insurance companies, was set up in 2002. ElderShield premia can also be paid using MediSave savings. Individuals who seek higher insurance coverage may purchase supplementary insurance under MediShield and ElderShield, reflecting that the Singaporean system heavily relies on individual responsibility. This is confirmed by the fact that co-payments are typically required at the point of service. In addition, co-insurance and deductibles to avoid moral hazard are required under MediShield. Around 80 per cent of hospital services are provided by 15 (corporatized) public hospitals, since private hospitals are generally smaller than public hospitals. (WHO WPRO n.d.). Public hospitals receive subsidies based on DRG-costed service delivery and the type of ward. (Reisman, 2009) Public hospitals have five ward types, namely A, B1, B2+, B2 and C, that mainly differ in terms of the comfort they offer. Beds in C wards receive subsidies of up to 80 per cent and in B2 wards of up to 65 per cent (Reisman 2009), while beds in B1 wards are less subsidised (up to 20 per cent of cost) and beds in A class wards are not subsidised (Ministry of Health Singapore 2012, Lim 2004). A means-test for B2 and C wards to determine the co-payment was introduced in 2009 to avoid increased utilisation of these subsidised wards by the better-off.

In contrast to the healthcare systems in Thailand and Singapore, the Philippine system is based on social health insurance. The Philippine Health Insurance Corporation (PhilHealth) was established in 1995, when per capita GDP stood at 1,070 (current) USD compared to 2,140 (current) USD in 2010, and mandated to achieve universal health coverage in the Philippines. As of 2012, 84 per cent of the population were enrolled in PhilHealth and 72 per cent covered (see Table 1). As a social health insurer, PhilHealth collects contributions from its members. There are four groups of members, namely (i) the formal sector (who contributes through payroll taxes; 41 per cent of total registered members), (ii) the informal sector (individual/family- or group-based members who make payments through a network of collecting agents; 18 per cent of total registered members), (iii) overseas workers and lifetime members (12 per cent of total registered members) and (iv) sponsored members (identified through a proxy means test and paid for by national and local government; 29 per cent of total registered members). (PhilHealth, 2012) PhilHealth, which functions as a single payer, decides the benefits package and provider payment mechanisms. The benefit package has been expanded continuously and now also includes a so-called Z-benefits package for catastrophic high cost care. (PhilHealth, 2012) Balance billing, under which patients are charged the difference between what PhilHealth pays and the cost of care, is common, but a no balance billing scheme started in 2011 for sponsored members and selected treatments. (PhilHealth, 2012) The private sector plays a very important role in the Philippine health care sector as 73 per cent of hospitals are private for profit and not for profit hospitals.

South Korea, which employs a Bismarck-type social security system, achieved universal health coverage in 1989 at a GDP per capita (in current USD) of 5,438. Since 2000, the National Health Insurance Service (NHIS) has functioned as a single public health insurer providing a comprehensive benefit package for all employed and self-employed Korean citizens. Not all services, however, are included in the benefit package and a provider-driven expansion of non-covered services has occurred over the past years given low fee schedules. (Kang, et al., 2012) As of 2011, the NHI covered 96.8 per cent of the population, 65.3 per cent of the total population were categorized as employed insureds and 31.5 per cent as self-employed insureds. 3.2 per cent of the total population were covered by the Medical Aid programme. The NHIS is also a single payer, in charge of collecting contributions (through payroll taxes). To deal with moral hazard type behaviour, co-payments, subject to an income-dependent ceiling, are imposed on insureds. (NHIS, 2013) Only 17 per cent of hospitals in South Korea are public, while the remaining 83 per cent are non-profit private hospitals. The NHIS itself runs one 745-bed model hospital. (NHIS, 2013)

What emerges from the above description of healthcare systems in Thailand, Singapore, the Philippines and South Korea is that each of the four countries has chosen a quite different way to progress towards universal health coverage. Among the four countries, only the Philippines have not achieved universal health coverage yet, but significant progress has been made since the establishment of PhilHealth. Irrespective of the chosen path, in all countries private providers have contributed to achieving the goal of universal health coverage, albeit to much lesser extent in Thailand and Singapore compared with the Philippines and South Korea. Thailand as a publicly-dominated system had the lowest out of pocket expenditures in 2010, only 14 per cent of total health expenditures compared with 54 per cent in Singapore and the Philippines and 31 per cent in South Korea. Yet, Thailand’s total health expenditures were only 3.9 per cent of GDP, whereas in South Korea the ratio stood at 6.9 per cent in 2010, reflecting Thailand’s cost-containment efforts. Average life expectancy was highest and the infant mortality rate lowest, however, in Singapore.

4. Findings

This section presents the findings from the analysis of institutional hospital contracting arrangements for clinical and clinical support services in Thailand, Singapore, the Republic of Korea and the Republic of the Philippines.

Nature of the ContractIn Thailand, both the SSO and the NHSO contract with private hospitals. As of

2009, the SSS worked with 2,563 contracted hospitals, 152 of which were public, 98 private and 2,313 network hospitals. (SSO, 2010) The 98 private hospitals are main contractors, which in turn may contract with providers that provide lower levels of care (sub-contractors) or higher levels of care (supra-contractors). Before entering into a contract, the SSO screens private hospitals based on a set of criteria such as number of beds and number of medical personnel. A rather broad, standard contract for the provision of general services is used for all private hospitals, irrespective of size. The contract duration is one year. The contract is written by the SSO and non-negotiable. The SSO does not require public hospitals to sign contracts, although public hospitals receive a copy of the standard SSO contract for their information. Similarly, although to a lesser extent, the NHSO contracts with private hospitals (49 private hospitals compared with 844 MoPH hospitals and 75 other public hospitals). Private hospitals can choose to join the UCS, provided they are Thai HA accredited (although HA accreditation is not explicitly required in the contract), whereas public hospitals cannot opt out. (Hughes, et al., n.d.) The NHSO also signs rather broad, one-year contracts for general services with private providers. Private providers may either function as Contracting Units for Primary Care (CUPs; a CUP typically comprises so-called primary care units (PCUs; one for 10,000 to 15,000 registered beneficiaries) and the contracted hospital), Contracting Units for Secondary Care (CUSs) or Contracting Units for Tertiary Care (CUTs). In addition, the NHSO also contracts private hospitals for specific diseases such as for example cataract surgery or disease management of chronic diseases. Contracts are generally not negotiated with private hospitals, but imposed by the NHSO. Moreover, as also reported in Sriratanaban (2011), the NHSO announced policy and other changes at rather short notice, without giving sufficient time for (budgetary) preparations.

Institutional contracting is also used by PhilHealth. In the Philippines, however, institutional contracting has been tied to a voluntary accreditation process that all public and private providers, should they wish to get reimbursed by PhilHealth for services covered by the benefits package, have to follow. As of 2012, 1,670 hospitals were accredited. (PhilHealth, 2012) To remain eligible for PhilHealth reimbursement, accreditation must be renewed annually. Up to 2012, hospital accreditation was vested in PhilHealth, but an administrative order was announced early 2013 that requires the establishment of a “Hospital Accreditation Commission which shall be the National Accrediting Body for Hospitals utilizing the PhilHealth Benchbook Standards to improve the Quality of Services of Hospitals in the Philippines pursuant to the strategic thrust of “Kalusugang Pankalahatan” (Universal Health Care)” (Office of the Secretary, Departmentof Health, Republic of the Philippines, 2013), which was set up in 2013. The separation of payer and accreditor is aimed at reducing conflicts of interest. In the future, contracts will probably simply replace the current accreditation warranties. (Bultman, et al., 2008) Some hospitals have decided to opt out and simply declare to patients that they do not accept PhilHealth payments. The terms and conditions of this soft contracting model like for example the rates paid for services are imposed by PhilHealth on the providers.

Since the integration reform, which was completed in 2000, institutional contracting has also been used in South Korea for covered services. Provider participation is mandatory and NHIS patients cannot be denied treatment. Institutional contracting in South Korea is, however, not tied to hospital accreditation6. The quality of care is mainly ensured through (i) compulsory (large hospitals) and voluntary quality evaluations and subsequent public announcement of rankings (Chun, et al., 2009) and (ii) Health Insurance Review & Assessment Service (HIRA) evaluations of the

6 The Korea Institute for Healthcare Accreditation was only established in November 2010. (Chang & Lee, 2012)

reasonableness of medical care benefits. One contract is negotiated annually with one representative of the Korean Hospital Association, who is nominated by the chairperson of the association. This contract then applies to all public and private (non-profit) hospitals. Negotiations start three months prior to contract expiration and it typically takes around fifteen days to negotiate a contract. All public and private hospitals belong to the Korean Hospital Association, but medium-sized and small hospitals have their own respective sub-associations under the Korean Hospital Association. The NHIS collects contributions, negotiates fee schedules with providers, monitors volume changes and pays providers, while the HIRA decides the scope of services (i.e. manages benefits), reviews medical fees, evaluates medical claims and assesses the quality of providers.

Compared to Thailand, the Philippines and South Korea, the situation is quite different in Singapore, where public hospitals are corporatized and JCI accredited. Institutional contracting with private hospitals does not exist.

“The Singaporean government is very pro-active and ahead of the curve. … Through corporatization, the public sector has become as competitive as the private sector and public hospitals pay competitive salaries. There is no [institutional] contracting with private hospitals; it is up to the individual to decide where to go. The government must get the pricing of medical services right as private hospitals follow public hospitals. Otherwise the system will be distorted.” Academic expert, National University of Singapore

In Thailand, the Philippines and South Korea, provider selection appears to be rather underdeveloped if not non-existent. In Thailand for example all public hospitals are required to participate in the UCS, reflecting their monopoly situation in many rural areas, while in South Korea it is mandatory for all public and private hospitals to treat NHIS patients. Besides, only Thailand’s NHSO explicitly attempts to strengthen the referral system and achieve an integrated delivery system through its contracting arrangements with the CUPs. In general, quality improvements may be achieved through a more selective contracting approach based on for example epidemiological profile of the area (unless of course there is limited provider competition), especially in larger cities with a sufficient number of hospitals to select from (Bultman, et al., 2008). In none of the three countries is a formal, competitive selection processes with clear selection criteria used. This in turn would require a continuing dialogue between the contracting parties. At present, however, providers can hardly influence contract content in Thailand and the Philippines, where contracts are simply imposed on providers.

Provider paymentIn Thailand, the SSS is entirely based on capitation rates that are determined by

the SSO, whereas the UCS employs capitation rates for outpatient and health promotion and disease prevention services, with capitation rates estimated by the NHSO, and weighed DRGs under a global budget for inpatient services (NHSO determined amount per adjusted relative weight). The NHSO makes capitation payments to the CUPs, which are responsible for referring patients subject to predetermined rates. If patients are referred to inpatient departments, payments based on DRG are made.

In the Philippines, on the other hand, providers have traditionally been paid by PhilHealth on a fee for service (FFS) basis, but case rates were introduced for common medical and surgical procedures in 2011 to contain cost and apply to both public and private hospitals. Already in 2012, 60 per cent of total claims were paid using case rates. (PhilHealth, 2012)

Hospitals in South Korea are also paid by FFS, with the fee schedule in FFS determined by quantity and prices of provided services (Health Insurance Review &Assessment Service, 2013a). The service prices are calculated by multiplying a Resource-Based Relative Value Score or RBRVS, which is announced by the Ministry of Health and Welfare and revised every five years, with the unit price per medical procedure, which is determined annually by the NHIS and providers. (NHIS, 2013) Annual unit price increases depend on annual volume increases. The unit price is the same for all hospitals,

with the exception of large hospitals. Since 2002, DRG have been used for seven diagnoses to contain costs. A new DRG system, which is a combination of FFS and DRG, was eventually introduced in 2009. In total, however, only around 2 per cent of payments are made based on DRG. Per diem payments are used to pay long-term care hospitals. (Health Insurance Review & Assessment Service, 2013a)

Thailand’s SSO and UCS both emphasise cost-containment through closed-end provider payments, thus avoiding the cost escalation inherent in the FFS payment mechanism, while PhilHealth is transitioning from FFS to a case rate system. FFS is still the main payment mechanism used in South Korea as the introduction of DRGs has faced opposition from the medical society, raising questions of financial sustainability. (HealthInsurance Review & Assessment Service, n.d.) With the exception of South Korea, providers can hardly exert any influence over prices and payments rates.

IncentivesOn the basis of the outcome of the screening process by the SSO, a patient ceiling

for the contracted private hospital is determined, ranging from 25,000 registered patients to 130,000 registered patients at present. Contracted private hospitals with Thai HA receive a top up from the SSO on the uniform capitation rate that increases as the hospital achieves higher stages of Thai HA. Similarly, the NHSO may reward hospitals with hospital accreditation and make performance based payments according to certain performance indicators such as meeting cervical smear test target numbers. (Sriratanaban, 2011) These payments, however, vary across time and region. (Sriratanaban, 2011) Besides, a higher than average payment rate is offered by the NHSO for specific diseases under disease-specific contracting arrangements to attract private providers. (NHSO, 2009)

There is one payment standard and PhilHealth does not use an incentive pay structure.

“The incentive is still the volume as we enrol more Filippinos under the universal health care. We have actually pumped in about 25 million people into the health insurance. So now, even private hospitals that used to cater 10 per cent of their hospital beds for what we call indigents or charity patients ... These wards can now be reserved for the poor who already have health insurance. These hospitals will have an incentive to treat the poor as they will get something from PhilHealth by billing them the reimbursement rates.” Executive, Department of Health of the Republic of the Philippines

Yet, as of 2011, the share of social insurance in total health expenditures was only 8.9 per cent (National Statistical Coordination Board of the Republic of the Philippines, n.d.) In addition, PhilHealth’s does not fund re-admissions and complications, i.e. these do not constitute a new care episode, which is expected to incentivise providers to provide high quality care.

Since 2007, the NHIS has paid more to high-quality hospital and less to low-quality hospitals as assessed by HIRA under its Value Incentive Programme. The number of institutions and items will be expanded as the Value Incentive Programme has resulted in quality improvements. (Health Insurance Review & Assessment Service,2013b) Additional payments, which are publicly announced, are made by NHIS for services provided at higher level facilities (20 per cent for hospitals with more than 30 beds, 25 per cent for hospitals with more than 100 beds and 30 per cent for tertiary hospitals).

Pecuniary incentives aimed at improving the quality of care are used by Thailand’s SSO and, to some extent, NHSO as well as the NHIS in South Korea. None of the institutional contracting arrangements, however, uses a preferred provider system through which additional incentives, such as for example advisory services and enhanced communication channels, could be offered to hospitals that perform well according to pre-determined criteria, such as for example number of fraudulent claim cases relative to

best practice hospital or adherence to clinical practice guidelines, as additional incentives. Pecuniary disincentives are only used by the NHIS, which increase financial risk for the providers and therefore require careful monitoring.

Key performance indicatorsMain SSO contractors are required to submit (i) reports containing claim and

treatment information, (ii) financial statements and (iii) profit and loss accounts according to a pre-determined schedule to the SSO. While these criteria touch upon the areas of financial situation, patients, internal business processes and innovation, they are not viewed as KPIs as no explicit targets against which performance is measured are set. No other KPIs than those used for performance based pay are employed by the NHSO.

Although PhilHealth accredits providers, there is basically no provider performance review apart from monthly mandatory hospital reports (Bultman, et al., 2008) and no KPIs are employed.

The NHIS uses KPI related to the quality of services. The quality assessment is done by HIRA for each medical service based on the reasonableness of the medical care provided, not based on a set of facility-based criteria. Initially, HIRA’s quality assessment focused on high frequency or high expense diseases, but the assessment areas have continuously been expanded. (Health Insurance Review & Assessment Service, 2013b)

Although “possibly the single biggest advantage of contracting is that it allows purchasers and contractors to focus on results” (Loevinsohn, 2008: 26), none of the contracting arrangements reviewed comprehensively utilizes key performance indicators and corresponding benchmarks against which to measure the performance of hospitals in terms of for example financial situation, patients, internal business processes and innovation. Hence, it is difficult to formulate conrete action plans to achieve the objectives of the contracting arrangement.

Review and adjustment mechanismAs the SSO contract expires after one year, providers have to apply for a renewal

and, upon successful screening and acceptance, sign a new one-year contract. The SSO’s decision to renew contracts is mainly based on an assessment of (i) the quality of patient services (based on information from the SSO call/complaints centre), (ii) adherence to general building requirements and safety standards and (iii) appropriateness of filed claims and supporting documentation.

“Contracts are signed and revised annually. … Minor changes to the contract can be made by the SSO, while major revisions require the approval of the Attorney-General. So far, no major revisions have ever been made. Minor revisions were not made this year and the year before.” Contracting Executive, SSO

NHSO contracts are renewed automatically, unless either party gives notice of termination 90 days prior to the end of the contract. Contract monitoring and management by the NHSO has remained rather weak (Hughes, et al., n.d.), although the NHSO also operates a call/complaints centre to monitor patient satisfaction.

As PhilHealth’s soft contracting is tied to the accreditation of providers, no additional review and adjustment mechanisms exist. To remain eligible for PhilHealth reimbursement, accreditation must simply be renewed annually.

The contract between the NHIS and the representative of the Korean Hospital Association is also renewed annually. The unit price is re-negotiated annually between the NHIS and providers, whereas the RBRVS is revised by HIRA every five years. Should the parties not be able to reach an agreement on the unit price, then the Health Insurance Policy Deliberation Committee (HIPDC) under the MOHW makes the decision.

The institutional contracting arrangements in all three countries require annual renewal of the underlying contracts. Given the relatively short contract lengths, most risks associated with internal or external changes are therefore shifted to the providers.

Longer contract periods, however, may not be possible due to annual budget requirements. In addition, some of the contracts employed are rather broad, rendering periodic adjustments basically redundant. Non-renewal of contracts in case of weak performance is quite uncommon, probably due to the possible interruption of patient care and negative media coverage as suggested in Ashton, et al. (2004).

Dispute SettlementDisputes in all three institutional contracting models are ultimately subject to the

national legal system, i.e. disputes are to be settled through the judicial process and appropriate alternative dispute resolution measures. Some contracts, like the one used by the SSO, encourage the parties to settle disputes through negotiation. In Thailand and South Korea, however, hospitals often simply accept the decisions made by the government agencies.

Perceived impact of contracting arrangementPatients are perceived to benefit most from the contracting model employed by

the SSO as they may choose from a larger number of facilities near home, especially in urban areas. The main disadvantage of this contracting model is viewed to be the uniform capitation rate, which may induce providers to engage in inappropriate cost saving with potentially negative health impacts. Yet, no evidence supporting this claim exists. Limwattananon, et al. (2010), however, report practice variation across the SSO, UCS and the CSMBS. Similarly, the main advantage of contracting between the NHSO and private hospitals is also perceived to be the possibility of offering patients more choice near home, particularly in urban areas. The main threat seems to be the lack of interest from hospitals due to the low closed-end payment methods coupled with an essentially open-end benefit package. While the number of contracted private hospitals decreased from 88 in 2003 to 49 in 2010, the number of private clinics, however, has almost doubled, from 89 in 2004 to 169 in 2010. (Hughes, et al., n.d.)

The inclusion of private hospitals into the national health insurance system, which has resulted in improved access, given low bed-to-population ratios across the country, equity and efficiency, is perceived to be the main benefit of PhilHealth’s soft contracting approach. The main challenges on the other hand are the relationship between PhilHealth and providers as well as the infrastructure gap, especially in disadvantaged areas. There seems to be a feeling of mistrust between the two parties due to long waiting times for payments, inefficient claims processing and insufficient communication between PhilHealth and providers. The infrastructure gap on the other hand implies that the tool of provider selection could only, if at all, be used in certain cities, as in many areas there is only one hospital or no hospital at all.

The main advantage of the South Korean contracting model is perceived to be the mutual agreement of parties. There are negotiations and the terms and conditions are not mandated by government. It is important to have a continuing dialogue and reach a consensus with providers. On the other hand, the current negotiations seemed to be based on reaching a strategic consensus rather than on scientific evidence. In addition, the one-size-fits all approach may be unfair for some hospitals. In the future, the NHIS may have to contract with each hospital or groups of hospitals, depending on transaction costs, rather than with one hospital only. Most importantly, the current contracting system is seen as unsustainable due to FFS payments and resulting cost pressures. The adverse incentives FFS payments create for providers are well documented in the literature (Langenbrunner, et al., 2009 among others) and capitation and/or a global budget will eventually have to be introduced. Another main challenge is viewed to be the management of the relationship between NHIS, HIRA and the MOHW.

“The relationship between NHIS, HIRA and the MOHW is very, very complicated. … According to the law, NHIS is the insurer. … HIRA acts like an insurer - we have three

insurers. … HIRA looks like an insurer. They do a lot of duties of an insurer. They decide the scope of services and they decide the RBRVS and things like that. … HIRA used to be the NHIS. When we integrated the funds, the provider said we have to have a separate independent assessment review institute. HIRA got separated from us, but they got to do what they used to do … That’s why it looks complicated.” Contracting Executive, NHIS

5. ConclusionThis study examined institutional hospital contracting arrangements in Thailand

Singapore, Republic of the Philippines and the Republic of Korea. Primary data were collected through IDIs with key experts and supplemented with secondary data obtained from a review of available statistics, academic and grey literature.

While the healthcare systems in the four countries are strikingly different, some general observations can be made. First, institutional contracting with private hospitals is used in Thailand, the Philippines and South Korea, but not in Singapore, given the strength of its corporatized hospitals. Institutional contracting can be viewed as a mechanism to complement hospital services provided by the government and improve health service delivery. Indeed, the inclusion of private hospitals in the network of contracted providers is viewed as the main advantage of the contracting arrangements in Thailand and the Philippines. The role of private hospitals in institutional contracting, however, has remained much smaller in Thailand compared with the Philippines and South Korea, which is not surprising given that the former tends towards the Beveridge model, while the latter leans towards the Bismarckian model.

Second, none of the contracting arrangements has a mechanism in place to select hospitals in a strategic way and there is no system of preferred providers, both of which could allow the contractor to exert more leverage if applied in certain geographic areas.

Third, performance-based contracting as defined by Loevinsohn (2008) has remained rather underdeveloped in all three models, although to varying extent. The NHIS for example specifies explicit KPIs, aimed at improving the quality of care, to facilitate monitoring and evaluation. No KPIs are, however, formulated to achieve improvements in equity, efficiency and sustainability.

Fourth, while the NHIS negotiates with providers annually, institutional contracting follows a top-down approach in Thailand and the Philippines. Yet, to be able to further engage private providers, a continuing dialogue between the parties must be created. Ashton, et al. (2004) among others state that a good relationship between the contracting parties is important, and that especially small providers benefit from face-to-face interaction. These findings indicate that capacity development especially regarding contract management is important.

Fifth, the results confirm that provider payment mechanisms are essential for the success of contracting. Yet, Thailand’s closed-end payments seem to have prevented several private hospitals from joining the SSS or the UCS. On the other hand, the FFS payments are seen as the main drivers of cost escalation in the Philippines and South Korea. A final observation is that there seems to be a lack of rigorous evaluations of institutional contracting initiatives in Southeast and East Asian countries or results of such evaluations have not been disseminated. There is virtually no information available of the cost of contracting.

Nevertheless, the findings of this study suggest that the potential of institutional contracting arrangements should be explored further to improve health system outcomes and thereby support countries in their quest for universal health coverage.

BibliographyAshton, T., Cumming, J. & McLean, J., 2004. Contracting for health services in a public health system: the New Zealand experience. Health Policy, Volume 69, p. 21–31.

Barnes, J. et al., 2009. Kenya private health sector assessment, Bethesda, MD: Private Sector Partnerships-One project, Abt Associates Inc.

Bultman, J. et al., 2008. Philippines: Developing capacity for contracting of providers by PhilHealth (Financed by the ADB and GTZ), Manila: Asian Development Bank.

Chang, H.-S. & Lee, S.-H., 2012. Establishment, present condition, and developmental direction of the new Korean healthcare accreditation system. Journal of Korean Medical Science, Volume 27, pp. S61-69.

Chun, C. B., Kim, S. Y., Lee, J. Y. & Lee, S. Y., 2009. Republic of Korea: Health system review. Health Systems in Transition, 11(7), p. 1–184.

De Pinho Campos, K., Norman, C. D. & Jadad, A. R., 2011. Product development public-private partnerships for public health: A systematic review using qualitative data. Social Science & Medicine, Volume 73, pp. 986-994.

de Roodenbeke, E., 2004. Purchasing inpatient and outpatient care through hospitals, Washington: World Bank.

Donato, R., 2010. Extending transaction cost economics: Towards a synthesised approach for analysing contracting in health care markets with experience from the Australian private sector. Social Science & Medicine, Volume 71, pp. 1989-1996.

England , R. & HLSP Institute, 2008. Provider purchasing and contracting mechanisms, s.l.: Rockefeller Foundation.

Evans, D., 2006. Bulletin of the World Health Organization interview with Dr David Evans: Contracting and health. [Online] Available at: http://www.who.int/bulletin/volumes/84/11/who_news.pdf[Accessed 3 October 2013].

Harding, A. & Loevinsohn, B., 2005. Buying results? Contracting for health service delivery in developing countries. Lancet, Volume 366, p. 676–81.

Harding, A. & Montagu, D., 2012. A zebra or a painted horse? Are hospital PPPs just infrastructure partnerships with different stripes, or are they a separate species?. World Hospital and Health Services, 48(2), pp. 15-19.

Harding, A. & Preker, A. S., 2003. Private participation in health services, Washington: World Bank.

Health Insurance Review & Assessment Service, 2013a. Introduction of HIRA. [Online] Available at: http://www.hira.or.kr/eng/news/01/__icsFiles/afieldfile/2013/10/15/brochure_1.Introduction.pdf[Accessed 1 October 2013].

Health Insurance Review & Assessment Service, 2013b. Comprehensive quality report of National Health Insurance 2012, Seoul: Health Insurance Review & Assessment Service.

Health Insurance Review & Assessment Service, n.d. Payment system. [Online] Available at: www.hira.or.kr[Accessed 10 October 2013].

Herberholz, C. & Supakankunti, S., 2011. Thai contracting case. Bangkok: ANHSS training course on Strategies for private sector engagement and PPPs in health (May 30 – June 4, 2011) .

Hughes, D. et al., n.d. Assessment of a decade of Thai UCS implementation. [Online] Available at: www.hsri.or.th/sites/default/files/browse/tor3.pdf[Accessed 24 August 2012].

Kang, M. S., Jang, H. S., Lee, M. & Park, E.-C., 2012. Sustainability of Korean National Health Insurance. Journal of Korean Medical Science, Volume 27, pp. S21-24.

Langenbrunner, J. C., Cashin, C. & O’Dougherty, S. eds., 2009. How-to manuals: Designing and implementing health care provider payment systems. Washington D.C.: The World Bank.

Limwattananon, C. et al., 2010. Different provider payment methods: variation in clinical practice, lessons from Thailand. [Online] Available at: http://ihppthaigov.net/publication/attachresearch/259/chapter1.pdf[Accessed 10 September 2013].

Liu, X., Hotchkiss, D.R. & Bose, S., 2008. The effectiveness of contracting-out primary health care services in developing countries: a review of the evidence. Health Policy and Planning, Volume 23, pp. 1-13.

Liu, X., Hotchkiss, D. R. & Bose, S., 2007. The impact of contracting-out on health system performance: A conceptual framework. Health Policy , Volume 82, p. 200–211.

Loevinsohn, B., 2008. Performance-based contracting for health services in developing countries : A toolkit, Washington: World Bank.

Mills, A., 1998. To contract or not to contract? Issues for low and middle income countries. Health Policy and Planning, 13(1), pp. 32-40.

Mills, A. & Broomberg, J., 1998. Experiences of contracting health services: an Overview of the literature, London: London School of Hygiene and Tropical Medicine.

Ministry of Health Lao PDR, 2010. National strategy for malaria control and pre-elimination 2011-2015, Vientiane: Ministry of Health Lao PDR.

Ministry of Health Singapore, n.d. Costs and financing - Schemes & subsidies. [Online] Available at: http://www.moh.gov.sg/content/moh_web/home/costs_and_financing/schemes_subsidies/medisave.html[Accessed 4 October 2013].

Ministry of Public Health Thailand, 2009. Health policy in Thailand, Bangkok: Ministry of Public Health, Bureau of Policy and Strategy.

Ministry of Public Health, 2011. Thailand health profile report 2008-2010. [Online] Available at: http://www.moph.go.th/ops/thp/index.php?option=com_content&task=view&id=176&Itemid=2[Accessed 21 August 2012].

MoPH, 2011. Thailand Health Profile Report 2008-2010. Bangkok: Ministry of Public Health.

Mudenda, D. et al., 2008. Provider purchasing and contracting for health services: The case of Zambia, s.l.: Rockefeller Foundation.

National Health Security Office (NHSO), 2009. NHSO Annual Report 2008, Bangkok: National Health Security Office (NHSO).

National Statistical Coordination Board of the Republic of the Philippines, n.d. Philippine National Health Accounts. [Online] Available at: http://www.nscb.gov.ph/stats/pnha/[Accessed 18 October 2013].

NHIS, 2013. National health insurance system of Korea 2013 - Pursuing the best in the world, Seoul: Health Insurance Policy Research Institute, NHIS.

NHSO, 2009. Annual Report 2008, Bangkok: NHSO.

Office of the Secretary, Department of Health, Republic of the Philippines, 2013. Administrative order no. 2013-0002 Establishment of the Hospital Accreditation Commission as the national accrediting body for hospitals in the Philippines, Manila: Office of the Secretary, Department of Health, Republic of the Philippines.

Peters, D. H. et al., 2007. A balanced scorecard for health services in Afghanistan. Bulletin of the World Health Organization, Volume 85, pp. 146-151.

PhilHealth, 2012. Philippine Health Insurance Corporation 2012 Annual Report. [Online] Available at: http://www.philhealth.gov.ph/about_us/annual_report/[Accessed 4 October 2013].

Reisman, D., 2009. Social policy in an ageing society: Age and health in Singapore. Cheltenham: Edward Elgar.

Rokx, C. et al., n.d. Health financing in Indonesia - A reform roadmap, Washington: World Bank.

Sriratanaban, J., 2009. Health care purchasing and provider responses in Thailand. Bangkok: WHO training course on Strategic purchasing for social protection (Nov. 30 - Dec 4, 2009).

Sriratanaban, J., 2011. Impact of universal coverage policy implementation on public hospitals, and their responses that affect medical services. [Online] Available at: www.hsri.or.th/sites/default/files/browse/tor5-1-2.pdf[Accessed 1 September 2012].

SSO, 2010. Annual Report 2009, Bangkok: SSO.

Taylor, R. & Blair, S., 2002. Public hospitals: Options for reform through public-private partnerships. [Online] Available at: http://www.ihf-fih.org/en/Resources/Contracting-and-Public-Private-Partnerships/Public-hospitals-Options-for-reform-through-Public-Private-Partnerships[Accessed 3 October 2013].

Ten Asbroeck, A. H. A. et al., 2004. Developing a national performance indicator framework for the Dutch health system. International Journal for Quality in Health Care, 16(Supplement 1), p. i65–i71.

Thammatacharee, J., 2010. Provider payment of UC scheme. [Online] Available at: http://jointlearningnetwork.org/sites/jlnstage.affinitybridge.com/files/Thailand_provider_management_presentation_0.pdf[Accessed 4 October 2013].

Vining, A. R. & Globerman, S., 1999. Contracting-out health care services: a conceptual framework. Health Policy , Volume 46, pp. 77-96.

WHO WPRO, 2011 Revision. Country profiles. [Online] Available at: http://www.wpro.who.int/countries/en/[Accessed 12 January 2013].

World Bank, 2003. Private sector assessment for health, nutrition and population (HNP) in Bangladesh, Washington: World Bank.

World Bank, n.d. World Development Indicators. [Online] Available at: http://data.worldbank.org[Accessed 15 January 2013].

World Health Organization, 2010. The world health report: health systems financing: the path to universal coverage. [Online] Available at: http://www.who.int/whr/2010/en/[Accessed 21 May 2013].

WPRO, 2011. Singapore. [Online] Available at: http://www.wpro.who.int/countries/sgp/30SINpro2011_finaldraft.pdf[Accessed 24 December 2012].

Appendix 1 Guideline questions for IDIs

1. Who are the partners?2. Which party initiated the contracting arrangement? Why? How has the contracting

arrangement changed since inception?3. What are the sources of funding?4. How are providers selected? Is there competition?5. What is the design of the contracts?

a. When did the contracting arrangement start? Has it been renewed? How?b. What is the nature of the contract? (objectives; scope, volume and quality of

services; price, etc.)c. What is the extent of the population covered?d. How are providers paid? e. What incentives are providers given to perform well?f. What are the KPIs and how are providers monitored and evaluated?

6. Are uniform contracts used for all providers or do variations exist? In what way? 7. Do you think the contracts used are too complex or overly formal and detailed? Why?8. How are disputes resolved? Is there any recourse mechanism in case of non-

performance?9. Can the terms and conditions of the contract be reviewed and adjusted/renegotiated? 10. Who handles contract performance management?11. Are good, cooperative relations between the parties important? Why? Can strong

relational ties perhaps be considered a strategic objective that adds value to contracts?12. How large are related transaction costs? (in % of total contract value)13. What do you perceive to be the most important achievements/challenges of this

contracting arrangement? (In terms of access, equity, quality, efficiency and sustainability.)