Public Disclosure Authorized. Public Disclosure Authorized. Public Disclosure Authorized. Public Disclosure Authorized

Transcription

1 Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized 66014

2 July 2011 Lessons for Hospital Autonomy Implementation in Vietnam from International Experience Policy Note Issues identified from International Studies and a Public Hospital Survey in Vietnam Ministry of Health Department of Planning and Finance Health Strategy and Policy Institute The World Bank Vietnam Office

3 FOREWORD The Government of Vietnam sees hospital autonomy policy as important and consistent with current development trends in Vietnam. It is based on Government policies as laid out in Government Decree 10/2002/ND-CP on financial autonomy of revenue-generating public service entities; and to 2006, it is replaced by Decree 43/2006/ND-CP on professional, organizational, human resource management and financial autonomy of revenue-generating and state budget-financed public service entities. These policies apply to public service entities in all sectors, including the health sector and hospitals. This policy is an important element of public administration reform in Vietnam, helping service entities survive and develop under the socialist-oriented market mechanism. It aims to help hospitals in fulfilling assigned professional tasks by allowing them to restructure their organization and staffing. The government has also allowed public service entities to mobilize private capital and joint ventures to organize activities and services responding to social and people s needs. Joint ventures can promote dynamism, creativity and autonomy in the effective use of physical facilities, professional capacities and human resources and to expand the range of services, mobilize additional revenues from society, and improve service quality. Hospitals are given financial autonomy, including autonomy in using current budget and revenues for service related activities. They are allowed to develop their own internal spending regulations stipulating spending levels/ items that are appropriate with their own contexts, thus increasing the efficiency of financial resources. If a hospital performs well with additional revenues, staff can be rewarded with additional income on the basis of their performance. This is expected to encourage hospital staff to improve their productivity and work efficiency. In addition, various funds can be established, including welfare funds for the staff as well as for the expansion of business activities for re-investment. This study will show that since the implementation of Decrees 10/2002/NĐ-CP and Decree 43/2006/ NĐ-CP, a number of improvements have been demonstrated within hospitals with respect to physical facilities, service provision, medical techniques, service quality and staff incomes, thus creating stability and satisfaction among hospital workers. But it also describes the international evidence that implementation of hospital autonomy comes with a risk of unintended outcomes driven by powerful financial incentives from the market place to increase revenue. These include supply induced demand, cost escalation, inappropriate care. There are some indications that such risks may be emerging in Vietnam as well, although these would need further research. Fortunately, there is also international evidence about policies that can mitigate such risks, and these are also described in this report. We hope that this report will inspire further studies and encourage policymakers to think about continuous improvement of policies. FOREWORD iii

4 We welcome the study and strongly recommend it to policy makers, hospital managers, and researchers. We look forward to a fruitful collaboration with the World Bank on health system development in Vietnam. ACKNOWLEDGEMENTS Prof. Nguyen Thi Kim Tien Minister Ministry of Health, Vietnam This report has been prepared by the Vietnam Ministry of Health s (MOH) Department of Planning and Finance (DPF), in collaboration with the Health Strategy and Policy Institute (HSPI), World Bank (WB) and World Health Organization (WHO). The report is based on the work of Loraine Hawkins, a consultant supported by the World Bank, on the international experience relevant to Vietnam; a survey and analysis conducted by HSPI and supported by the Ministry of Health and WHO, on the implementation of Decree 43 in the public hospital system. The initial findings of the report were discussed at the workshop Hospital Autonomy: Distilling Lessons for Vietnam from International Experience on June 3, 2010 in Hanoi. Contributions by the following people is acknowledged: Nguyen Thi Kim Tien, Minister of Health; Pham Le Tuan, Director, DPF, MOH; Nguyen Hoang Long, Deputy Director, DPF, MOH; Le Quang Cuong, Director, HSPI; Tran Thi Mai Oanh, HSPI; Graham Harrison, Health Systems Advisor, WHO Vietnam; Nguyen Thi Kim Phuong, National Professional Officer, WHO Vietnam; Toomas Palu, Lead Health Specialist, WB in Vietnam; Dao Lan Huong, Health Specialist, WB in Vietnam; Jack Langenbrunner, Lead Economist, Health, WB; Gerard Martin La Forgia, Lead Health Specialist, WB; Nguyen Quynh Nga, Program Assistant, WB in Vietnam. The report also acknowledges the research team who conducted the public hospital survey, including Prof. Dr. Pham Le Tuan; Mr.Nguyen Nam Lien, MSc; Mr. Le Van Quan, MSc; Ms. Pham Minh Nga, MSc; Ms. Hoang Thi Bich Ngoc, MSc all from the Department of Planning and Finance, MOH; and Prof. Dr. Le Quang Cuong, Director; Dr. Tran Thi Mai Oanh, Vice Director; Dr. Tran Van Tien; Ms. Nguyễn Khanh Phương, MSc; Dr. Khương Anh Tuan; Dr. Duong Huy Luong; Dr. Hoang Thi Phuong; Mr. Trinh Ngoc Thanh, MSc; Vuong Lan Mai, MSc all from the Health Strategy and Policy Institute. iv ACKNOWLEDGEMENTS BACKGROUND ACKNOWLEDGEMENTS v

5 CONTENTS A A. Background 5 B. Overview of international experience with hospital reform 5 and hospital autonomy A Framework for Understanding Hospital Organizational Reform 6 Governance and Supervision of Autonomous Hospitals 8 Reviews of International Evidence on the Impact of Hospital Reform 9 Characteristics of Successful Hospital Reforms 9 Less Successful Hospital Reform Experiences: Imbalanced Incentives in Public Hospitals 10 with Mixed Budget and Private Revenues Common Implementation Challenges: Factors That May Disable Reform 11 Enabling Conditions for Successful Hospital Reforms 12 Conclusion: Key Lessons from International Experience 12 Challenges in evaluating and review of hospital autonomy 15 Background C. Vietnam s initial experience with hospital autonomy: survey of 18 hospitals following 17 implementation of Decree 43 D. Policy options for managing risks from the unintended effects of hospital autonomy 19 Using multiple policy instruments to achieve balance among hospital policy objectives 19 Harnessing hospital managers as partners in achieving hospital policy objectives 20 Linking and integrating autonomous hospitals to improve the efficiency of the health system 21 Regulating the public-private interface in autonomous hospitals 21 Mobilizing and regulating investment in autonomous hospitals 23 Public financing of capital expenditure in public hospitals 23 Private financing of capital expenditure in public hospitals 24 D. Conclusions and Recommendations 25 References 26 vi CONTENTS BACKGROUND

6 The Government of Vietnam granted autonomy to public hospitals as part of a wider public administrative reform, with the aim of improving performance of these entities and reducing the burden on the Government budget. The 2002 Decree 10/2002/ ND-CP gave hospitals and revenue-generating entities in other sectors the rights of autonomy over how they used their non-budget revenues, subject to some regulations. Some hospitals implemented Decree 10 particularly urban hospitals with greater potential to attract private paying patients. In 2006, the Government promulgated Decree 43/2006/ND-CP to replace Decree 10. Decree 43 applied to all public hospitals. At the same time, the Government encouraged hospitals to implement social mobilization of capital to mobilize non-government finance for investment from private organizations and individuals, including current and former hospital staff. Social mobilization is a form of joint venture between public hospitals, investors and private companies to establish revenue-earning business units within public hospitals. These units are able to set their own fees for providing higher quality patient services. In cases when these fees are higher than prices that are reimbursed by Social Health Insurance, public hospitals can bill the patients for the difference. Staff investing in these services may also work in them and share profits with the joint venture partners and the hospital. Recognizing that public hospitals have specific public welfare characteristics that distinguish them from commercial revenue-generating enterprises, the Government has recently developed a new draft decree and is currently consulting and reviewing this draft. Hospital autonomy and social mobilization joint ventures have already resulted in significant changes in hospital financing, performance and management areas. However, as in most countries that have initiated this type of reform, there are some concerns with unexpected and unintended effects in the hospital sector - such as overuse of fee-generating services, increase in service prices, and increase in the health care expenditure burden on the people and the social insurance system (VSS). Recently, the Communist Party of Vietnam Politburo indicated that it is necessary to conduct an assessment on this issue. Taking all these facts into consideration, the Ministry of Health (MOH) commissioned the Health Strategy and Policy Institute (HSPI) in collaboration with the Department of Planning and Finance (DPF) to conduct an assessment on the implementation of Decree 43 with a special focus on the investment in medical equipment through "social mobilization" in public hospitals. The purpose of this report is to analyze emerging Vietnam experience in the light of international public hospital autonomy conceptual frameworks and experience. BACKGROUND 3

7 B Overview of international experience with hospital reform and hospital autonomy

8 Reforms of organization and management of public hospitals have occurred in every region of the world in the past 30 years. Reform has taken place in response to some common problems in public hospitals: inefficiency, waste, user dissatisfaction, a braindrain of qualified personnel to the private sector or through emigration, run-down assets and failure to serve the poor. In many countries, as in Vietnam, hospital reform introduced greater managerial and fiscal autonomy. Hospital autonomy was seen as a solution to the problem diagnosis about some (but not all) of the causes of poor hospital performance. Traditional public hospitals, operating as budgetary units of the MOH or local government, faced multiple constraints on performance: hospital staff were paid and their jobs were protected regardless of their productivity and quality of service; hospital staff salaries were often set based on rigid relativity to other civil service salaries (based on level of qualifications and years of experience). As private health systems developed, this resulted in a wide gap between public hospital salaries and private medical staff earnings, particularly in urban areas; hospitals were given budgets for specific inputs salaries, utilities, drugs, other operating costs, capital expenditure, etc.; (slow) procedures for MOH and MOF approval were necessary in many countries to re-allocate and execute budgets; if hospitals made efficiency gains, any unused budget was returned to the MOF/Treasury; any reduction of excessive staff positions resulted in lower hospital budgets, so that there was no incentive to make efficiency gains; capital expenditure from the budget was seen as a free good so every hospital had incentives to lobby for high levels of new capital expenditure, but little incentive to maintain the existing assets of the hospital, and make better use of existing assets. Studies of international experience with hospital autonomy have highlighted that autonomy alone does not lead hospitals to improve their performance towards all of the public policy objectives of higher quality, greater efficiency and access to hospital services for the poor. A critical mass of complementary reforms is needed to drive hospital performance towards all of these objectives, using multiple policy levers. Comprehensive hospital reforms usually include a number of the following elements: Management capacity and systems strengthening, including management information systems strengthening, was emphasised in countries with successful reforms. Financing and provider payment reform, often social health insurance or purchaserprovider separation was introduced in parallel to hospital reform; new payment methods such as case-based payment or global budget contracts were introduced. Quality improvement reforms: hospital licensing and accreditation systems were introduced. Hospital performance 7

9 monitoring methods were strengthened. Health technology assessment and evidencebased clinical guidelines were developed. Modernizing infrastructure: some countries faced a need to rationalize outdated, excessive hospital capacity to increase efficiency. Fast growing countries faced a need to build new hospitals, and modernize their existing hospital capacity. Countries experimented with a range of mechanisms to encourage the private sector to invest in the public hospital sector, using a variety of models of public private partnerships. Organizational and governance reform: various models of hospital autonomy were implemented in different countries. Some countries converted hospitals into autonomous non-profit foundations or trusts. Some converted public hospitals into state-owned enterprises ( corporatization ). Some countries merged hospitals into autonomous networks, rather than giving autonomy to individual hospitals. Sometimes these networks were subordinate to a hospital authority. Some countries used public private partnerships to manage hospitals more efficiently; some contracted private management to run public hospitals another form of public private partnership; some leased public hospitals to private hospital organizations under contract to operate the public hospital. A Framework for Understanding Hospital Organizational Reform Hospital reform was expected to address the problems facing public hospitals by reforming some of the key drivers of performance: the incentives facing the hospitals managers and staff, the authority given to managers, the capacity of managers and staff to respond to these incentives, the intrinsic motivation or professionalism of staff, and the accountability mechanisms (Roberts, Hsiao, Berman and Reich 2001). Preker and Harding (2003) developed a framework to analyze how hospital reform changes the key drivers of hospital performance. The framework looks at five dimensions of hospital reform that affect how the hospital functions. Note that management autonomy is only one of the five dimensions: the other four dimensions are key contributing factors needed to make autonomous hospitals responsive to the public policy objectives of quality, efficiency and equity. The five dimensions are: Management autonomy: In autonomous or corporatized hospitals, decision-making authority is delegated from the Ministry of Health or local government to the hospital s own management or board; Financial incentives created by the provider payment mechanism: hospitals obtain their revenue from payment for services, and patients may have some choice of hospital exposing the hospital to competition. By contrast, traditional public hospitals obtain most or all of their revenue through budget allocations for input costs (salaries, other operating costs, capital expenditure); Autonomous or corporatized hospitals retain part or all of the financial surpluses they earn from reducing costs or increasing revenue; the reformed hospitals also face a hard budget constraint - the hospital is responsible for its own deficits and debts; Accountability of hospitals should change from following directions and seeking permission from the government hierarchy, to ex post accountability for performance to arms-length bodies including compliance with contracts, regulations, and accountability to its board, owners and the public; A summary schematic is provided in Figure 1. 1 Figure 1. Dimensions of Hospital Management Reform Social functions of hospitals (such as the costs of free or subsidized care for the poor or uninsured, compliance with ethical codes of conduct, meeting public health priorities, teaching and research) should be explicitly contracted and fully funded, rather than being an implicit expectation or unfunded mandate. 1 The terms used in Figure 1 for describing different levels of autonomy for hospitals are defined here: Budgetary Unit the hospital is a subordinate unit of the MOH or local government, financed from the budget, subject to the same rules and controls as government ministries; Autonomous Unit the hospital becomes a semi-autonomous or autonomous government-owned body; hospital directors are appointed by government; the hospital may have an external supervisory board; Corporatized Unit the hospital becomes a government-owned enterprise or corporation with a legal and financial structure and regulations that simulate those of private sector hospitals; but it remains in government ownership; corporatized hospitals have a board of external directors, usually appointed by the owner (government) or an appointments commission for state enterprises; Private Provider -- ownership of the hospital and decision-making transferred to private, non-state, legal entities or private shareholders. The private entity may be a non-profit foundation or trust. 8 9

10 When international experience of hospital reform is analyzed using this framework, it has been found that two of these five dimensions provider payment incentives and surplus retention - are particularly important in giving hospitals incentives for efficiency. If provider payment mechanisms use consumer choice and competition, this can also drive improvement in Figure 2. Organizational Dimensions and Incentives for Performance Management autonomy aspects of quality that are visible to consumers such as polite service, clean and modern facilities, new equipment. Two of these dimensions - reforms to accountability and social functions - are particularly important in giving hospitals incentives for clinical quality, rational use of health resources and for equity. These incentives are illustrated in Figure 2. major capital investment and also for appointing an external board of the hospital (the governance body) if it has a board. The board usually appoints the hospital director. For autonomous public hospitals, often the Ministry of Health carries out the responsibilities of the owner in conjunction with the Ministry of Finance and Cabinet. Governance functions for autonomous hospitals require specialized, high level expertise and good hospital performance data and data analysis. These functions require new capacity and skills in Figure 3. Governance of Autonomous Hospitals the MOH. Some countries have created a special semi-autonomous hospital authority to carry out the governance functions of the government with respect to public hospitals, including Hong Kong SAR, and a number of provinces and cities in China. These specialized authorities have highlevel expert capacity to monitor the performance of hospitals, and oversee the appointment and development of hospital management. Figure 3 summarizes the main elements of good governance for autonomous hospitals. Provider payment incentives, choice Surplus retention Efficiency, visible quality Objectives Clear Consistent Performance Targets Oversight Clear role & duties Vai trò và nhiệm vụ rõ ràng Professional, technical capacity Accountability Social functions Equity, clinical quality, rational use of services HOSPITAL Source: Adapted from World Bank, Preker, et al., 2003 Governance and Supervision of Autonomous Hospitals The supervision and control functions carried out by the owner of the hospital are sometimes termed governance. In a traditional public hospital, the owner usually the national or local government - exercises direct control over the hospital manager, and has the authority to approve many operational management decisions. The government as owner (represented by the Ministry of Finance or Treasury) also takes any nongovernment revenue or surpluses earned by the hospital. When hospitals become autonomous, the owner steps back from involvement in daily operational decisions, and instead focuses on policy and strategy setting objectives for the hospital, agreeing on strategic plans, and putting in place effective checks and balances and supervision. The owner becomes more focused on performance assessed ex post and less focused on prior control through permissions and approvals. The owner of the autonomous hospital is usually responsible for approving and financing Reviews of International Evidence on the Impact of Hospital Reform Some cross-country reviews and syntheses of the evidence on the impact of hospital autonomy have been carried out. These reviews have attempted to find correlations between key characteristics of reform design and implementation and the results External Accountability Ex -post, performanceoriented, arms-length Published financial & annual reports of reform using the type of framework described above. Most of the countries studied had high levels of health spending and universal or near universal health insurance coverage. The findings are often specific to particular country contexts and cannot always be readily transferred that needs to be taken into account when interpreting the lessons learned. (Figueras, Jakubowski and 10 11

11 Robinson 2005; Preker and Langenbrunner 2005; Harvard School of Public Health 1996; Preker and Harding 2003; La Forgia and Couttolenc 2008; McKee and Healey 2001). Characteristics of Successful Hospital Reforms There is evidence from hospital reform in upper and middle income countries that hospital reform has led to improved efficiency when the organizational reforms were accompanied by the following policies: a credible budget constraint for the hospital (for example, either a global budget contract with some performancebased-payment, or contracts that controlled both prices and volume of services); merit-based hiring and promotion of managers based on their managerial qualifications and management track record; hospital management training; good information systems for clinical and financial management and reporting; complementary reform to strengthen accountability of managers for the performance of the organization (often through creation of a board of external directors or trustees or a hospital authority to supervise the manager), particularly in relation to the objectives of equity and costeffectiveness; and, increased managerial authority to shape workers incentives (e.g. through freedom to hire, promote, set tasks and hours of work, decide on performance rewards and sanctions). Less Successful Hospital Reform Experiences: Imbalanced Incentives in Public Hospitals with Mixed Budget and Private Revenues Hospital autonomy can reduce equity, reduce the less visible dimensions of clinical quality, and contribute to irrational excessive medical intervention in profitable areas of treatment unless there is balanced, complementary reform and institutional capacity building to strengthen accountability for equity and cost-effectiveness, and financing and provider payment reform to create incentives for hospitals to carry out their social functions and contain costs. Many low income and low-middle income countries have had less positive results from hospital autonomy. Many countries introduced partial, limited reform focused only on fiscal autonomy - enabling hospitals to increase their non-budget revenues by giving hospitals freedom to charge fees and retain fee income, to sell medicines at a profit, and to sell higher-quality services to private patients. Many countries gave hospitals fiscal and managerial autonomy over use of their private revenues without reforming how budget funds were managed. For example, studies of Indonesia s Swadana Hospital reform found that this limited approach to hospital reform focused only on fiscal autonomy - reduced equity, and did not reduce the need for budget finance for hospital care. The first phase of China s hospital reforms also took this approach and led to similar problems (see China Case Study in Box 2). Figure 4: Hybrid public-private hospital rules for budget funds and private revenue Management Authority Provider payment incentives Surplus retention incentives Accountability Social functions Traditional MOH hospital Note: Budget revenue settings indicated in yellow; private revenue settings indicated in red. There are few successful examples of hospital autonomization in countries that introduced this type of partial public hospital autonomy. In this type of partial hospital autonomy, public hospitals often operate under two different sets of rules traditional budget unit rules for the subsidies they continue to receive from the budget, and another more autonomous set of rules for the non-budget payments they receive from patients and insurance funds. The hospital functions as a hybrid of a traditional public hospital and a private hospital. These hybrid organizational settings are illustrated in Figure 4. Countries who adopted this flawed approach to reform encountered problems with profit-driven excessive intervention rates for paying patients. Autonomous hospital Corporatized hospital Private hospital Unregulated, unaccountable autonomous hospitals engaged in wasteful, duplicative investment in profitable high-technology services (e.g. Lebanon, Armenia). When hospitals operate under the hybrid mix of rules illustrated in Figure 4, as budget revenue falls and private revenue grows, the budget subsidy becomes ineffective as a mechanism to ensure care for the poor, or other social functions of the hospital. In many countries, budget revenue falls to the level where it covers little more than low basic salaries for staff. The budget allocation and salaries are not affected by the output of the hospital or other dimensions of hospital performance, and so do not create incentives for the hospital to improve its performance

12 Bureaucratic controls make it difficult for managers to spend budget revenue efficiently. The incentives facing the public hospital in these cases are dominated by their private revenues. Private revenues create more powerful incentives because they are linked to service delivery, and they are easier to spend: the public hospital has a high degree of autonomy over private revenue, and the ability to retain profits on chargeable services. So the hospital behaves like an unregulated private hospital in many respects. There are some examples of good practice in mitigating the problems of partial public sector autonomy with a mix of public and private organizational rules for controlling budget and non-budget revenues. Turkey, for example, introduced tighter control, accounting and reporting for non-budget revenues of hospitals. It regulated prices. Turkey also regulated how revenues and surpluses were used to pay staff performance bonuses. Rather than paying staff bonuses based on the individual performance in earning fee-for-service income for the hospital, bonuses were redistributed, under a scheme that applied to all staff and paid them fairly on the basis of their contribution to the hospitals objectives including its non-revenue earning functions and social objectives. Many countries introduced hospital autonomy with too little attention to putting in place new mechanisms for supervision and accountability, and before health information systems were good enough to generate data for effective accountability. Where external monitoring of autonomous healthcare providers is weak, some case studies have found evidence of abuse of resources by autonomous providers (Harding and Alvarado, 2005). In smaller rural hospitals where internal financial capacity may be weak, health care providers may be unable to cope with the increased accounting and reporting requirements associated with financial autonomy. In some countries where there was a lack of credible hard budget constraints on autonomous providers, hospital autonomy was followed by weaker cost containment. Autonomous hospitals in some countries ran unsustainable operating deficits until more stringent mechanisms for supervision were put in place, combined with stronger coordination between the social insurance agency or purchaser on the one hand, and the financial supervision of the hospital (e.g. New Zealand, UK). In some countries with fee-for-service provider payment systems or open-ended case-basedpayment systems, autonomous hospitals rapidly expanded output, leading to unsustainable cost pressure on the social health insurance system (e.g. Hungary, Slovakia). In Indonesia and China, allowing hospitals to operate joint ventures with for-profit pharmacies was found to result in irrational and inappropriate prescribing and uncontrolled expenditure on medicines. Some countries introduced autonomy before they had put in place effective mechanisms for motivating public hospitals to treat poor and uninsured patients. In the worst cases (e.g. Armenia) hospitals found themselves in a lowutilization, low-revenue trap because high user charges and informal fees deterred patients from using hospitals. Common Implementation Challenges: Factors That May Disable Reform Many countries found themselves unable to implement hospital autonomy reforms as planned in the design stage. Some common implementation challenges arose because of the concerns of stakeholders about reform leading decision-makers to abandon controversial elements of reform or negotiate changes to policy. Hospital clinicians in many countries felt alienated by or excluded from the reforms; in some countries there was a perception that management energy was too focused on structural and organizational change, leading to a relative loss of focus on clinical quality and clinical healthcare outcomes (e.g. UK, New Zealand). Other implementation challenges arose because of a lack of capacity in hospitals for the new and more complex challenges of management under autonomy, and a lack of capacity in MOH and other central government agencies for new supervision and regulation tasks. Weak and unreliable hospital information systems for management and accountability acted as a constraint on effective implementation in many countries. Successful reforms invested substantially in management training, health information systems development, and development of institutions for supervision and regulation of autonomous hospitals. A common cause of problems in the implementation of hospital reforms has been the failure to increase hospital management authority over personnel. In many countries, it has not been possible to transfer public hospital staff from civil service status to employment by the hospital under more flexible labour rules. Fear of loss of civil service job security, pension security and career mobility has led to blocking of reform by staff and unions in a number of countries (e.g. Pakistan). In some countries, the MOH leadership has also been reluctant to transfer staff to autonomous hospitals because reduction in the size of the MOH affects civil service seniority and influence of senior management in the MOH (e.g. Thailand). Some countries were able to negotiate an acceptable package of conditions for the transfer of staff to autonomous hospitals with staff unions. The UK, for example, adopted policies to protect the employment rights of transferring staff, and provided a package of protection and support for any staff made redundant during re-organization. The UK also kept its national pay agreements and created a single, portable pension scheme for all medical staff in autonomous hospitals. This addressed staff concerns about loss of career mobility. Enabling Conditions for Successful Hospital Reforms Conversely, some aspects of the political or government context and institutional capacity were found to be important enablers of successful reform: consistent and coordinated approach to the complex, interacting elements of hospital reform, sustained over time. Coordination between the different institutions involved in hospital-related policies becomes more challenging but more important when hospitals are autonomous; coordination is needed among the Ministry of Health, social health insurance agency (or purchaser organizations), and other Ministries or agencies responsible for hospital planning and capital investment in hospitals; strong government leadership, accompanied by effective communication, to explain and defend essential elements of the reform policies that may be controversial; 14 15

13 a capable, trained cadre of hospital managers; strong, expert institutions to supervise, regulate and audit autonomous hospitals; health information systems and hospital management information systems in place that generate timely, reliable data for measuring hospital performance, and for providing the basis for more sophisticated systems to pay hospitals; and participation and support from hospital clinical staff in the reform process and in leadership of reformed hospitals. Conclusion: Key Lessons from International Experience Detailed design of organizational reform needs to be adapted to the local context. These are complex reforms, requiring policy consistency, stability, strong government institutions with good implementation capacity, and government credibility. If the reforms take an incomplete or inconsistent approach - implementing some, but not all, of the five key dimensions of organizational reform listed in Figure 1 above (management autonomy, provider payment incentives, surplus retention, accountability and social obligations), disappointing or unintended results are likely to occur. Organizational reform requires close coordination between policy design and implementation. Most countries have had to go through several rounds of review and revision to address all of the dimensions of reform and to adjust reforms to changing circumstances and priorities; willingness to review the impact of reform and adjust policy should be seen as a strength, not an admission of failure (see China Case Study in Box 1, page 14). There have been few successes with hospital organizational reform in countries that have lacked levers for creating firm budget constraints for their hospitals. This does not necessarily mean a rigid limit on annual expenditure but it does require control of cost-plus pricing, a move away from fee-for-service payments, and some limits or levers of influence over the volume of services provided. There have been very few successes with hospital organizational reform in countries or local governments with weak capacity for governance and stewardship (regulation, performance monitoring, and institutions of accountability for public service providers), and in hospitals with limited management capacity and unreliable health information systems. Management systems and management skills strengthening are a prerequisite for successful hospital autonomy. The more successful implementation strategies have involved: Identifying the pre-conditions for reform success including management systems and capacity, contracting capacity and supervision capacity; Assessing hospitals (and local health administrations in decentralized countries) to assess whether these pre-conditions are in place; Piloting in hospitals/localities where these conditions are in place; and Building these conditions where they are not yet in place before rolling out reform. Challenges in evaluating and review of hospital autonomy High quality evaluation evidence of the impact of hospital organizational reforms is rather sparse. Most reviews are case studies or comparisons of performance before and after reform. The best quality studies are from the UK, but the context of reform in the UK was very different from that facing low and middle income countries and Box 1. China Case Study: Adjustment of Hospital Reform Based on Review of Experience China introduced partial autonomy in its public hospitals after 1980, with the goal of reducing the Government financial burden and expanding and upgrading hospital infrastructure. Most public hospitals still have a government-ministry model of governance and management, e.g. there is continuous direct intervention via the government hierarchy, even though hospitals are autonomous, Central personnel controls - staffing structure and grades are still imposed by government. Typically, most public hospitals are a form of hybrid public-private hospital (similar to Figure 3 above). Hospitals are paid for on a fee-for-service basis by social insurance funds and patients. There are few viable private hospitals, and a lack of real competition among hospitals. Across China s many provinces and cities, a variety of different models of hospital reform have been carried out. Alongside hospital reform there has been development of social health insurance, which has generated new mechanisms for financing and paying hospitals. The Government has also implemented changes to mechanisms for setting and regulating prices, including drug prices. Based on studies and lessons from earlier experiences, China has been seeking to identify what works best and move to more unified models of public hospital management. First Phase of Hospital Reform: Social mobilization of capital & financing from private capital markets: a common form is project cooperation a sub-hospital within a hospital, operated by investors, or by a joint venture that leases hospital space and supplies equipment, with profit-sharing between investors and the hospital; Hospitals allowed to charge higher prices for higher quality services; Hospitals allowed to pay incentives to staff and retain surpluses to develop facilities; Regulated fee schedule sets prices below cost for basic services, while drug sales and high tech services earned profits; Hospitals continued to receive budget subsidy to cover basic salaries, but over time, this became a small share of hospital revenue (as low as 10 percent). Policy Achievements Identified in Reviews Increased number of hospitals and volume of hospital care; Multiple experiments with alternative models of governance and management, including: trustee-models of hospitals, corporatized hospitals, contracting, leasing, shareholding, outsourcing of support functions; Development of new provider payment methods, mirroring international models. Unintended Effects of Policies Identified in Reviews Hospitals pursued economic benefits and expanded infrastructure and equipment in a chaotic way: the growth and distribution of hospitals has been inequitable: strong growth of higher level hospitals, at the expense of primary care and outpatient care; 16 17

14 Rapid growth of hospital expenditure, exceeds growth rate in financing sources for social insurance and budget finance; Self-financing incentives did not promote efficiency: hospital efficiency has declined; Social mobilization of capital weakened management control and led to non-standard operations forbidden by government; Price setting & insurance did not protect the social responsibility of hospitals to provide quality services for the poor; Irrational over-provision of drugs and high-tech services and private facilities for more affluent patients; Conflict and mistrust between hospitals and patients, as costs to patients continuously increase. Continuing Implementation Challenges Inconsistent approach to reforms; Lack of evaluation of reforms and study of transferability of alternative models; Slow pace of development of health information systems; Slow pace of development of an adequate cadre of well-trained hospital managers. New phase of adjustment to hospital reform: : Constructing a Socialistic Harmonious Society China s new direction is based on strengthening the welfare function of public hospitals and health services. The new direction will be piloted in 16 hospitals, beginning in The main elements of the new direction are: Strengthening the role of public revenues; Strengthening government input in planning and supervision; Improving management and quality of services; Promoting efficiency; Reducing patient expenses; Separating ownership and management, and strengthening governance bodies; Separating non-profit and for-profit activities of hospitals and other health service providers; Reforming the payment method and price level to eliminate incentives for irrational over-provision of some services. Compensation & incentives policies for staff improved to motivate staff in line with the new directions. transition economies. However, some crosscountry syntheses of case study evidence have been carried out, and attempts have been made to look for correlations between key characteristics of reform design and implementation and the results of reform using the type of frameworks discussed above. The Harvard School of Public Health published a series of case studies of hospital autonomy experiences in developing countries and developed guidelines based on analysis of these cases (Chawla et al. from Harvard School of Public Health, 1996). Preker and Harding (2003) has a set of case studies of upper and middle income and transition countries and includes an evaluation based on a synthesis of evidence from these case studies. The European Health Reform Observatory in 2001 published a set of studies from transition economies and EU member states. These and other studies are listed in the References section of this paper. There are many reasons why evaluation of hospital autonomy is difficult methodologically: Reform models are diverse within and across countries, making it hard to characterize the reforms. It is difficult to control for the external environment typically other reforms and public finance changes occurred in parallel with hospital reform, often implemented simultaneously. Few countries implemented reform in a way that permitted evaluation with a control group. No countries adopted a randomized approach to hospital autonomy as a policy intervention. It is hard to control for differences across hospitals when measuring performance, particularly in countries where there is no reliable data for adjusting for case-mix. In many countries, medical records, coding and costing data before reform were incomplete and unreliable, but improved with reform. It is difficult to deal with systematic changes in data completeness. It is hard to distinguish effects of reform design from effects of implementation. It is hard to measure all the dimensions of hospital performance anyway, and there are trade-offs between different dimensions of performance (e.g. between cost-containment and incentives for responsiveness). It is particularly difficult to measure clinical quality and rational and appropriate use of services, except in countries with established standards and protocols for care. As a result of these difficulties, most countries set realistic expectations for evaluation, and have been prepared to use qualitative evaluation methods (e.g. from expert stakeholder interviews) and to rely on limited, but best available, data for policy review and identifying lessons for improving policy design and implementation. The methodology used in Preker and Harding (2003) whose findings were discussed in Section II is practicable in many countries. Cross-country case studies were used to derive testable hypotheses and construct a plausible counterfactual, using the framework for analysing hospital reform described in Section II. The evaluation methodology then identified indicators of four dimensions of hospital performance: efficiency, quality, equity and allocative efficiency (which captures the rational use of resources and services). The study looked at performance indicators of inputs, processes and final results in these four domains. (See Table 1 below for examples. A fuller list of indicators is given in Preker and Harding, Chapter 3.) The analysis then reviewed seven country case studies, to identify evidence of whether hospitals demonstrated improvement after reform by looking at the differences before and after reform in as many of these performance 18 19

15 indicators as were available. Looking at final results indicators, for example, the study found evidence of efficiency improvement after hospital autonomy in six out of seven countries, and of quality improvement in four of the countries, but equity was protected or improved in only three countries and rational resource and service use was maintained or improved in only one of the seven countries. More than half of the countries experienced deterioration in equity and rational use of services after hospital autonomy. Although causation cannot be inferred from this type of study, qualitative information may shed light on causal mechanisms. It may be useful to identify typical cases and elaborate qualitative case studies of what changes took place after reform and what drove these changes, based on in-depth anonymous interviews with MOH/PPC 2 officials responsible for hospital governance, hospital managers, staff, and well-informed patients. These qualitative methods can bring out lessons and recommendations for adjustment and refinement of hospital autonomy policies and implementation. Table 1. Examples of a Balanced Set of Hospital Performance Indicators Indicators Indicators of Rational Resource Use (allocative efficiency) Input Monetary and physical inputs per patient and per bed Availability rates of functional essential equipment Physical input mix (share of resources allocated to personnel, drugs, other operating costs, capital) Input price ratios (e.g. prices of drugs and consumables, salaries as ratio of national or international benchmarks) Process Average length of stay Bed occupancy or bed turnover rate; equipment utilization rate Compliance with mechanisms for priority-setting and evidence-based practice Compliance with evidencebased hospital drug formulary Intermediate or Final Outcome Unit cost per patient admitted, adjusted for case-mix Financial balance (revenue minus expenses) per patient or per bed day Incentives in place for provider to improve quality of care and provide social functions Number of patients bypassing primary care Quality Indicators Number and mix of qualified medical staff Availability of functional medical equipment and essential drugs Percentage of treatment according to defined protocol; compliance with safety and quality checklists Availability of complete patient medical records Adverse outcome rates adjusted by case-mix/severity (mortality, hospital infection, re-admission, repeat surgery) Patient satisfaction Public expenditure per patient by socioeconomic category or insurance status Utilization rate of hospital care or specific services by patient socioeconomic category Health outcome for conditions amenable to treatment by patient socioeconomic category Equity Indicators Share of patients covered by fee-waiver, social insurance or low-cost beds Waiting time for nonurgent surgery by patient socioeconomic category or insurance status Mean out of pocket expenditure per visit or admission by patient socioeconomic category Ratio of out-of-pocket health expenditure to food expenditure by patient socioeconomic category Poverty gap before and after out-of-pocket health expenditure by socioeconomic category 2 PPC Provincial Peoples Committee 20 21

16 C Vietnam s initial experience with hospital autonomy: survey of 18 hospitals following implementation of Decree 43 3 BACKGROUND 23

17 3 This section of the report presents the Ministry of Health and HSPI own analysis of the results of the 18-hospital public hospital survey without international or World Bank input. The Vietnam MOH s HSPI and DPF conducted a survey of a purposive sample of 18 hospitals including a sample of fully autonomous hospitals generating enough non-budget revenue to cover their operational costs and investments, partially autonomous central and provincial hospitals with varying levels of non-budget revenue, district hospitals and a specialized mental hospital with limited scope to generate non-budget revenue. The hospitals spanned some cities and six provinces with a range of geography and poverty rates. The survey attempted to identify how performance changed in response to the level of autonomy granted under Decree 43, using available performance data before and after implementation of the Decree. (Baseline data are from 2005 and post-reform data are from 2008.) The survey finds indicative evidence that the effects of hospital autonomy in Vietnam are similar to China and other fast-growing transition countries where private out-of-pocket spending is the largest source of health financing. Positive changes since implementation of Decree 43 include: Significant growth in total hospital revenues, including Government budget, health insurance payments, user charges. Comparing between 2005 and 2008, the hospital revenues increased by 1.8 times in fully autonomous hospitals, 3 times in centrally managed hospitals, 2.9 times in provincial level hospitals, and 2.5 times in district level hospitals. Most growth came from increase of social health insurance payments, as the share of Government budget for recurrent expenditures decreased by 2.7 times in fully autonomous hospitals, 2.5 times in centrally managed hospitals, and 1.3 times in both provincial and district level hospitals; and, the share of revenues from direct user fees declined by 1.2 times in fully autonomous hospitals, 1.3 times in centrally managed and provincial level hospitals and 1.16 times in district level hospitals, although the revenues from both the latter sources still increased in absolute terms; Increased capital investment in hospitals, particularly in medical equipment; Expansion of the range of healthcare services and an increase in hospital utilization between 2005 and 2008of 25% in fully autonomous hospitals, 17% in centrally managed hospitals, 14% in provincial level hospitals, and 16% in district level hospitals, comparing. During the same time period the number of hospital consultations and admissions increased by times, and times, respectively. Number of operations in provincial and district level hospitals increased by times; Substantial growth in incomes of public hospital medical staff. The average additional income per one hospital staff is higher than monthly salary in most hospitals, but varies by hospitals: 2.3 times of monthly salary in fully autonomous hospitals, 1.3 times in centrally managed hospitals, 1.4 times in provincial level hospitals, and 0.5 times in district level hospitals. The additional income for staff in 2008 is higher than that in 2005: it increased by 1.2 times in fully autonomous hospitals, 1.7 times in centrally managed hospitals, 3 times in provincial level hospitals, and 3.5 times in district level hospitals. VIETNAM S INITIAL EXPERIENCE WITH HOSPITAL AUTONOMY: SURVEY OF 18 HOSPITALS FOLLOWING IMPLEMENTATION OF DECREE 43 25