65019 Report Prepared for Ministry of Public Health, Royal Thai Government Devolution of Health Centers and Hospital Autonomy in Thailand: A Rapid Assessment Report from HSRI and World Bank April 2009 Loraine Hawkins (World Bank) Jaruayporn Srisasalux (HSRI) Sutayut Osornprasop (World Bank) 1 DEVOLUTION OF HEALTH CENTERS AND HOSPITAL AUTONOMY IN THAILAND: A RAPID ASSESSMENT DRAFT 2: FOR COMMENT ACKNOWLEDGEMENTS The authors owe gratitude to all of the nearly 100 people we interviewed for this study: it draws upon their knowledge and wisdom as its main source. We would like to acknowledge in particular the great contributions of Dr Suwit Wibulpolprasert, Senior Advisor, Ministry of Public Health, Dr Viroj Tangcharoensathien of the International Health Policy Program, and Dr Pongpisut Jongudomsuk, Director of Health Systems Research Institute, without which we could not have carried out this assessment. 2 TABLE OF CONTENTS INTRODUCTION.................................................................................................... 5 PART I: DEVOLUTION OF HEALTH CENTERS ........................................... 5 A. DESCRIPTION OF DEVOLUTION PROCESS AND MODEL ............... 5 B. FINDINGS FROM FIELD VISITS ............................................................... 9 C. FINDINGS FROM DISCUSSIONS ABOUT DECENTRALIZATION WITH POLICY MAKERS AND RESEARCHERS ...................................... 13 D. DISCUSSION .............................................................................................. 15 PART II. HOSPITAL AUTONOMY UNDER THE AUTONOMOUS PUBLIC ORGANIZATION LAW....................................................................... 23 A. THE BAN PHAEO APO MODEL OF HOSPITAL AUTONOMY......... 23 B. SCALING UP THE APO MODEL ............................................................ 28 C. DISCUSSION ............................................................................................... 29 REFERENCES ....................................................................................................... 35 ANNEX 1: INTERVIEWS WITH POLICY RESEARCHERS AND CENTRAL GOVERNMENT MINISTRIES ...................................................... 37 Addendum: Policy Recommendations from ONDC Evaluation of Devolution of HCs.................................................................................................................. 47 ANNEX 2: INTERVIEWS AND VISITS IN PROVINCES: SUMMARY OF FINDINGS REGARDING DEVOLUTION ........................................................ 48 ANNEX 3: INTERVIEWS AND VISITS WITH HOSPITALS: SUMMARY OF FINDINGS REGARDING HOSPITAL AUTONOMY ............................... 65 ANNEX 4: LIST OF STAKEHOLDERS INTERVIEWED AND CONSULTED......................................................................................................... 73 ANNEX 5: TERMS OF REFERENCE FOR RAPID ASSESSMENT OF DEVOLUTION IN THE THAILAND HEALTH SECTOR REFORM .......... 76 ANNEX 6: SAMPLE MEMORANDUM ON TRANSFER OF HEALTH CENTER TO TAMBON ADMINISTRATIVE ORGANIZATION ................. 79 3 ABBREVIATION APO Autonomous Public Organization CMU Community Medical Unit CSMBS Civil Service Medical Benefits Scheme CUP Contracting Unit for Primary Care DHO District Health Office/Officer DLA Department of Local Administration, MOI EDL Essential Drug List GO Government Officer GPO Government Pharmaceutical Organization HC Health Center LAO Local Administrative Organization MOF Ministry of Finance MOI Ministry of Interior MOPH Ministry of Public Health NDC National Decentralization Committee NHSO National Health Security Office OCSC Office of the Civil Service Commission ONDC Office of the Secretariat of the National Decentralization Committee OP Outpatient OPD Outpatient Department PCMO Provincial Chief Medical Officer PCU Primary Care Unit PHO Provincial Health Office PTAC Pharmaceuticals Therapeutic Advisory Committee P&P Health Promotion and Disease Prevention P4P Pay for Performance SSS Social Security Scheme TAO Tambon Administrative Organization UC Universal Health Coverage Scheme VHV Village Health Volunteer 4 INTRODUCTION This rapid assessment was conducted for the Thailand Ministry of Public Health with the support of the World Bank in partnership with the Thailand International Health Policy Program and the Thailand Health Systems Research Institute. This work was done under the World Bank’s Country Development Partnership Agreement with the Government of Thailand. The Terms of Reference for the assessment are attached as Annex 6. The assessment team comprised: Ms Loraine Hawkins (World Bank, Health Systems Consultant), Dr Jaruayporn Sangchai (Health Systems Research Institute), and Dr Sutayut Osornprasop (World Bank, Human Development Program Specialist). PART I: DEVOLUTION OF HEALTH CENTERS A. DESCRIPTION OF DEVOLUTION PROCESS AND MODEL The Plans and Process for Decentralization to Local Administrative Organizations Act of 1999 called for ministries including the Ministry of Public Health (MOPH) to develop action plans for decentralization of functions, resources and staff to the elected Local Administrative Organizations (LAOs) by 2010. The Act also set a target for increasing the share of the central government budget that should be transferred to LAOs from 9% to 35% by 2006. In 2006, the Law was amended to remove the 2006 deadline, and set minimum share of national budget to be transferred of 25%, with a target of 35%. Devolution of health centers (HCs) to Tambon Administrative Organizations (TAOs) and municipalities was initiated in the second Action Plan for decentralization, prepared in 2006. Under the guidelines for devolution developed by the MOPH, devolution of HCs only occurs where the following criteria are met:  The TAO/municipality meets “readiness� criteria to manage the HC: the LAO must have received a good governance award, and demonstrated capacity for and commitment to health by establishing a Public Health Section and contributing funds to a Community Health Fund (an NHSO initiative to encourage local governments to lead and commit resources to disease prevention and health promotion activities, with NHSO co-financing);  At least 50 percent of HC staff support devolution of their HC and are willing to transfer to LAO employment, including the HC head; Organization and Responsibilities for Health Centers Prior to Devolution: Thailand’s health system has some features that lead to complexity and local variation in the scope of health responsibilities of HCs and the accountabilities and incentives of HCs before and after devolution. The creation of the Universal Coverage (UC) health financing scheme in 2002, managed by the National Health Security Office (NHSO), brought about a partial purchaser-provider split in the functions and budget of the MOPH. This also introduced dual accountability for the MOPH’s HCs. Most of the operating budget for HCs now comes from the NHSO (and to a lesser extent from the Civil Service Medical Benefits Scheme (CSMBS) and Social Security Scheme (SSS) for formal sector employees). The MOPH budget, supplemented by a top-slice from NHSO equivalent to 65% of the salaries budget, pays the salaries of all of the 5 government officers who work for it, including those who work in HCs. (Payment of civil servants from the budget is guaranteed under the constitution.) The NHSO’s payments to HCs cover for some personnel costs (such as contractual staff, overtime payments, a performance bonus scheme), as well as other operating costs: utilities, fuel, maintenance, supplies, and equipment, etc. Resource management decision-making for primary health care as well as disease prevention and health promotion is rather de-concentrated – in the MOPH to Provincial Health Offices (PHOs), and in the NHSO to district-level Contracting Unit for Primary Care (CUP) Board. The CUP Board is chaired by the Director of the hospital that services the District (which may be a district or provincial or regional MOPH hospital, or sometimes a hospital belonging to another Ministry or LAO). The HCs, DHO and PHO also have representatives on the CUP Board. For non-devolved HCs, the MOPH is responsible for employment and personnel management; most technical policy, regulation and supervision; and fiduciary supervision. The HC reports to the MOPH’s District Health Officer (DHO) on administrative matters, and the DHO or PHO approval is required for most personnel decisions in the HC, and most financial decisions involving MOPH funds. The NHSO allocates to the CUP Board a fixed per capita amount for curative primary health care for UC members in the District, and population-related allocations for disease prevention and health promotion (P&P). The CUP Boards have had considerable discretion to decide how to allocate these budgets for curative outpatient care (OP) and P&P among the HCs, the hospital’s own outpatient department (many patients go to hospital OPDs for primary care), and Primary Care Units (PCUs) which some hospitals have established to de-congest their own OPDs. UC patients register both with a HC and a hospital for their health care and are free to obtain primary care services from both (though some provinces require the user to visit the primary care unit first). As a result of patient choices, and local decision-making, the share of curative and P&P service provided by HC, CUP hospital OPD and PCUs respectively varies widely in different HCs, districts and provinces. The CUP Board contracts with HCs and other OP and P&P service providers. Under these agreements, the HC has a second line of accountability to the NHSO for use of funds, and activities covered by NHSO. CUP Boards also adopt a range of different approaches about the extent to which they provide resources to HCs as cash rather than in kind, in the extent to which they transfer risk to HCs for managing the drugs budget and other costs, and the extent to which they allow patient choice of HC (or PCU in some areas) rather than adopting mandatory catchment areas. As a result, there is variation around the country in the extent to which HCs face a “hard budget constraint�, the extent to which risks in the primary care budget are pooled, and the extent to which HCs are exposed to competition. The CUP Board supervises HCs, providing financial supervision over use of NHSO funds, and supervision of performance in delivery services. PHOs, DHOs and CUP hospitals vary in how they coordinate their respective supervision responsibilities for HCs. The CUP hospitals also provide technical support, advice and some training to HCs, to a varying extent. 6 Some non-devolved HCs (particularly those in more densely populated, better-off areas) have a considerable degree of managerial autonomy arising from the fact that they have multiple sources of revenue: MOPH, NHSO, CSMBS, SSS, user payments (mostly from out-of-area patients), and sometimes complementary resources from their LAO and donations. HCs operate their own bank accounts for cash they receive from NHSO, and other non-MOPH sources. They can retain any unspent funds in these accounts at the end of the year. They have considerable freedom to hire contractual staff, purchase supplies and equipment, and initiate new activities and services using these retained funds. Transfer of staff and assets: The devolution process transfers the HC physical assets to LAO ownership, and transfers willing MOPH government officers and contractual staff working in the HC to the employment of the LAO.1 Transfer of health responsibilities: A memorandum of transfer is signed by the provincial administration, the PHO and LAO chief executive officer (CEO) formally documenting the transfer of the “public health duties and responsibilities� of the HC to the LAO, and committing the LAO to “administer and manage the health center according to regulations, criteria, standards, and public health work methodologies set by MOPH� and the relevant PHO (see sample agreement in Annex 6). In practice, only some of the MOPH’s responsibilities are transferred to the LAO. The LAO takes over the MOPH’s responsibility for employment and personnel management, and fiduciary control. This transfer of responsibility is clearly specified in new regulations. The transfer to the LAO of responsibility for ensuring the HC functions effectively, and for improving the health of its population is not so clearly defined. The MOPH retains some responsibility for technical policy, technical supervision, technical training, and regulation of health professional work and coordination of public health matters. None of the NHSO/CUP Board responsibilities is transferred to the LAO. So the local variation and change over time that is already seen in the division of health responsibilities between the MOPH/PHO/DHO and the NHSO/CUP Board is carried over into the division of responsibilities between the LAO and the NHSO/CUP Board. Fiscal transfers: The devolution process transfers the MOPH’s budget allocation for HC salaries to the LAO, via the Ministry of Interior (MOI), as a specific (earmarked) grant. This budget allocation covers the basic salaries and benefits of government officers in the HC – typically two to three staff. Funds will continue to be transferred as a conditional grant to guarantee salary payment for ex-MOPH employees until they 1 HC staff who do not wish to transfer to LAO employment will not be allowed to continue working in the devolved HC in the long term. They can request transfer to another MOPH post and may continue to work for the devolved HC only temporarily.. Staff who transfer retain civil service pension and medical benefits, though any new HC staff hired by the LAO will not be eligible for these benefits. Those staff who do transfer to the LAO are eligible for annual bonuses of up to 3-5 months salary that LAOs are able to pay all their staff, within limits approved by the provincial administration – the provincial operations of the Ministry of the Interior (MOI), under the authority of the government- appointed provincial governor. The MOI is in the process of adopting regulations that will allow transferring HC staff to retain other benefits and allowances paid by MOPH. 7 retire. Any newly hired HC staff will be paid from the LAOs general revenues (which consist of a general grant, shared tax revenues and varying amounts of local revenues). In addition to salaries, the MOPH budget funds capital expenditure and staff training for HCs. These funds are not transferred when HCs are devolved. Future capital expenditure will be a responsibility of the LAO from their general revenues. In addition, the MOI budget includes discretionary specific grants for LAOs that are sometimes for public health activities. The MOI budget for 2009 includes a 22 million baht allocation for capital investment in the 28 devolved HCs, and an allocation for paying honoraria for village health volunteers in all LAOs. Although not formally stated in the Memorandum of Transfer, the PHOs interviewed appear generally to be committed to continuing to provide training for staff in devolved HCs. NHSO, CSMBS and SSS resources are not transferred to the TAO when the HC is devolved.2 These agencies continue to provide resources directly to the HC, as before. Changes in Decision Rights: LAOs – whether or not the HC is devolved - have some freedom to develop new health services using their own revenues, so long as they comply with MOPH regulations. Major developments, such as establishment of new hospitals, however, are subject to licensing approval by the MOPH. Additionally, the Office of the Auditor General discourages LAOs from spending on any curative care services that are deemed to duplicate the MOPH’s mandate. Following devolution, LAOs have rights and power to hire, promote, reward and discipline HC personnel. Local government officers enjoy protection of employment – firing permanent LAO staff is difficult. But LAOs may hire and fire contractual staff. LAOs also have financial decision-making authority over HCs in relation to their use of LAO revenues, including authority over procurement and capital investment within delegated limits set by regulation. LAOs have some latitude to determine the level of delegated authority they give to HC heads in relation to personnel and financial decisions. LAOs do not have authority over how devolved HCs use of NHSO, CSMBS, and SSS funds, which continue to be governed by regulations set by these organizations. The MOI sets personnel and financial policies applying to LAOs at national level, and more detailed regulations are adopted by the MOI’s provincial administration (under the authority of the provincial governor). For some decisions (such as creation of new permanent LAO posts, or more major investments), the LAO CEO requires approval of the provincial administration. Changes in Accountability, Supervision and Oversight: Devolved HCs retain their accountability to the MOPH for compliance with technical policies, standards and regulations. They retain an obligation to report public health data to the MOPH. The DHO and PHO continue to supervise and advise devolved HCs on technical matters. The MOPH’s Inspector Generals will continue to inspect devolved HCs, but their findings will be advisory only. It will be the responsibility of the LAO to act on the findings. 2 There is a partial exception. The NHSO provides co-financing of 37.5 baht per capita to TAOs that agree to establish community health funds, to finance local health promotion and disease prevention projects. This scheme operates for any TAO, regardless of whether the HC is devolved. 8 Devolved HCs retain the same accountability as before to the NHSO, via contractual agreements with the CUP Board. HCs now have a new third major line of accountability – to the LAO CEO – for overall performance, and for personnel and financial matters (in relation to the resources they receive from the LAO). Devolved HCs provide financial and personnel reports to the LAO, and provide copies of their public health/activity reports to the LAO, MOPH and NHSO. Some LAOs appear to have instituted somewhat independent supervision of the HC by their Public Health Section head. In others, LAOs have combined the roles of HC head and Public Health Section head. LAOs (like the MOPH) are subject to external audit by the Office of the Auditor General. The MOI’s provincial administrations play a role in oversight of LAOs, and hence of devolved HCs. The MOI has considerable powers to enforce and discipline LAOs that breach financial or personnel regulations, to investigate possible wrong-doing, and to adjudicate in case of personnel grievances or disputes. The provincial administration takes some responsibility for supporting LAOs with devolved HCs: LAOs can apply to the provincial administration for additional funding if needed, or for assistance in addressing personnel issues (staff shortages, staff transfer requests). The provincial administration (and the elected PAO) has a responsibility for coordination across LAOs within their territory. It is not clear that any one organization will take responsibility for ensuring that devolved HCs make proper, efficient and effective use of their resources from all the multiple sources of funding. Changes in Market Exposure or Hardness of Budget Constraint: The exposure of HCs to competition is determined by the policies of the CUP Board, CSMBS and SSS, rather than the LAO. That is, devolution does not change the exposure of HCs to market pressure (or lack of it). The MOI is putting in place regulations that will ensure HCs are able to continue to retain unspent balances of revenue they receive from NHSO, CSMBS, SSS and user fees. This will preserve the existing incentives HCs have to maximize these sources of revenue and the existing freedom they have to use this revenue. Because the LAOs currently have more discretionary “budget space� than the MOPH’s PHOs, devolved HCs have more scope to negotiate increases in budget allocation. Additionally, the MOI stands ready to receive requests from LAOs for additional funds if needed, and has an interest in supporting newly devolved services. Devolved HCs are thus likely to perceive a softer budget constraint. B. FINDINGS FROM FIELD VISITS The assessment team conducted interviews of Provincial Health Office (PHO), District Health Office (DHO), Provincial Governor’s Office and Department of Local Administration (DLA), Contracting Unit for Primary Care (CUP) hospital, Tambon Administrative Authority (TAO) and health center (HC) staff in five devolved health centers and five non-devolved health centers and a hospital-owned primary care unit (PCU). Tabulated summaries of the data gathered in field visits and interviews in 9 these provinces and facilities are provided in Annex 2. Annex 4 lists persons interviewed. In relation to the three stated objectives of this decentralization, the team’s findings were as follows: - Changes in Flexibility, Responsiveness and Participation Following Devolution of Health Centers to TAOs These criteria for assessment are derived from the stated objectives of decentralization of service delivery, based on The Plans and Process for Decentralization to Local Administrative Organizations Act of 1999, following the Thailand Constitution of 1997. Increased management flexibility: HC staff and TAO leaders in three of the devolved health centers (Naphu, Salabangpoo, and Pakpoon) have positive perceptions of improvement in management flexibility, in the sense that future decision making is expected to be faster and there should be greater scope for initiative. However, all noted transitional problems with finalization of regulations and some unresolved regulatory issues (such as licensing public health officers to provide curative medical care). Two of the devolved centers (DonKaew, Banprok) noted positive and negative changes, though both perceive net benefits from devolution, in terms of scope for innovation and a shorter chain of authority for most decision making. These centers have experienced delays in regulatory changes that led to delays in funds flows, though DonKaew was devolved only in October 2008, and so is at an early stage of transition. Banprok commented on delays and difficulty in gaining approval for a new nurse practitioner post from the Provincial Administration, and for the exam for selection and hiring. In part this seems to reflect the fact that these agencies are dealing with these procedures for the first time, and are not drawing upon the expertise of the MOPH. Increased responsiveness to the community and to patients: Three devolved health centers (Naphu, Salabangpoo, and Pakpoon) could point to a number ways in which service delivery had already improved and new services had been provided in response to the needs and preferences of the community. These include a stronger client service orientation, increased curative care services, and increase in promotion and prevention (P&P). However, in one TAO, some of the planned changes in service delivery do not appear to be evidence-based and may not be cost-effective. DonKaew TAO has asked the HC staff to increase outreach and initiate annual health checks for villagers within the same budget and is providing closer supervision. The Nakornping Hospital CUP reduced curative care resources for the DonKaew HC since decentralization, though it is not clear that this was because of decentralization. The transferred HC has very low utilization; the CUP hospital operates a competing PCU opposite this HC; the CUP board decided the HC should focus primarily on promotion and prevention (P&P), leaving curative care to the hospital and its PCUs. Banprok has introduced a new dental service, but has decreased outreach. Increased participation of the community: All five TAO CEOs and Councils are active in obtaining community input on health and health service delivery. Naphu, Salabangpoo and Pakpoon HCs have increased activities which involve community participation and increased activity of the VHVs. DonKaew’s TAO-CEO has used systematic community participation in identifying health and providing feedback on 10 health services, though it may be too early to assess whether the HC staff themselves will engage more actively with the community. Banprok HC staff are promoting their services and the benefits of devolution to the community, and report increased utilization by people outside their catchment area. In some provinces, a number of TAOs are already providing complementary funding and resources for non-devolved HCs and PCUs, over and above their contribution to the Community Health Funds co-financed by NHSO. One of the non-devolved HCs visited (Laem Chan in Phuket) was identified by the PHO as a case of good practice in partnership with an LAO (Wichit Municipality) without devolution. This HC demonstrates high levels of initiative; strong resource mobilization from LAO, PAO and community; development of new services in response to community wishes; and community participation. It has been able to achieve similar results to the devolved HCs through partnership with the municipality and community mobilization without devolution. HC staff perceive that this model may offer the HC more resources and more management freedom than devolution, because they are concerned that MOPH might reduce its support following devolution and concerned about the risk of increased political intervention if they are devolved. In DonKaew, partnership between the TAO and the Nakornping hospital was established some years before devolution. There is a hospital-run PCU (attached to Nakornping Hospital and adjacent to the DonKaew TAO office) which was the result of a TAO-hospital partnership in which the TAO and community raised substantial resources to build the PCU facility which was operated by the hospital staff. The TAO and community petitioned the Nakornping hospital to maintain this service when the hospital director changed and proposed to close the service. - Other Evidence of Results It is too early to assess the lasting effects of devolution on processes and outputs in health centers, and it is not yet possible to assess any effect on outcomes. Nonetheless, the rapid assessment gathered information on changes in major inputs, outputs and processes in the first year of devolution. It should be borne in mind that it is common for major organizational change to lead to some loss of productivity for a period of 6-18 months, before benefits of the change emerge. Efficiency and appropriateness of increased spending: Four out of five HCs have received increases in resources financed from the TAO budget. This includes civil works to upgrade the HC buildings. In three cases, the civil works have or will improve physical accessibility and will provide greater patient privacy. As noted above, in one case, there are some questions about the efficiency and effectiveness of proposed civil works expenditure. In the fifth case, the HC had very low utilization and low levels of outreach prior to devolution, and the TAO quite reasonably has asked for increased output (a tripling of outreach) within the same level of resources, before any increase in resources is provided by the TAO. This will increase efficiency. Output levels: Four out of five devolved HCs have increased either utilization or outreach or both. One HC has experienced transitional reduction in output (about 5% reduction in visits) because three out of five government officers have been transferred to other MOPH posts. This HC still has high utilization and a very visible 11 patient/community service orientation. The HC is in the process of arranging back- up staffing from the CUP hospital. (In two cases, the staff were unwilling to transfer to work for the devolved HC; in one case, the TAO was not satisfied with the staff’s performance and initiated transfer of the staff back to MOPH from the devolved HC.) Equity: In two of the TAOs visited, only one of two HCs in the TAO has been devolved, because the majority of staff in the other HC voted against devolution. In one case, people in the catchment area of the non-devolved HC have complained about the fact that the non-devolved HC now receives less support than the devolved HC. In the second case, the TAO provided support to both HCs prior to devolution, and has maintained the same level of support to the non-devolved HC after devolution. However, it is not clear that the perceived inequity in this case arises from the decisions of the TAO. The devolved HCs are benefiting from substantial capital expenditure from the MOI budget for upgrading of building and equipment. Non-devolved HCs are receiving much smaller allocations of capital expenditure from NHSO and MOPH budgets. Accountability: The potential benefit of devolution is that direct local supervision and a shorter chain of accountability can improve service performance. This potential benefit has to be weighed against the potential disadvantages arising from the lower technical capacity of LAOs. Two of the TAOs visited had taken actions that pointed to increased accountability for “patient perceived quality�, and had initiated processes to provide a basis for increasing accountability for meeting community health need. One TAO provides feedback to HC staff based on patient complaints; in one case negative feedback on the performance of one staff did not lead to improvement in performance, and was handled by a request to transfer the staff member out of the HC to another MOPH position. One DHO commented that a positive result of devolution is that the TAO is better able to get HC staff to change their behavior than the DHO is, because of the very immediate accountability relationship. Another TAO has asked the HC staff to increase outreach from once a week to daily, in response to very low rates of HC utilization, and evidence of community need for more pro-active services (to address late presentation to health facilities by villagers when sick). This TAO uses supervision by the Public Health Section Head and village feedback meetings to ensure these services are provided. These two TAOs had also initiated surveys of community health status and health determinants, as a basis for establishing realistic local health plans that could be monitored by follow-up surveys. - Key “lessons learned� identified in interviews with TAOs, HCs and PHOs: a. Higher LAO bonuses and perceived higher prospects for promotion (to TAO Public Health Section head) make devolution attractive for substantial numbers of HC staff (estimated to be around 50% in some provinces). But substantial numbers of staff perceive higher promotion prospects in the MOPH or fear the loss of job mobility to other locations if the HC is devolved. PHOs and DHOs report that many HC staff are concerned about being closer to political decision-making under devolution. b. In four of the five devolved HCs, the HC staff, and TAO and PHO interviewees identified pre-existing good relationships between HC staff (particularly the HC head) and the TAO-CEO as a key success factor. In one case, however, interviewees noted that this close relationship could lead to 12 favoritism in promotion and budget allocation. This was one of the reasons why staff in the non-devolved HC in the same TAO were unwilling to transfer. In another case, some interviewees noted that the close relationship with the current TAO leadership creates risk of problems in the event of a change of TAO leadership. c. Changes to regulations and funds flows need to be resolved and complied with well in advance of future transfers, to ensure a smooth transition. d. HCs interviewed want clear and consistent policies on decentralization in the MOPH and MOI at both the strategic level (consistency on messages about the criteria for devolution, the timetable for implementation, whether devolution is voluntary or mandatory, whether devolved HCs can choose to return to MOPH…) and detailed level (for example, alignment of MOI regulations with the more generous MOPH regulations on overtime, and on retention of unspent funds). e. HC, TAO and some PHO and DHO personnel interviewed recommend that the MOPH/NHSO should ensure there is no change in the ways in which PHO, DHO and CUP hospital coordinate, provide support and resources and supervise HCs after transfer. Devolved HCs believe it is important for the MOPH and NHSO to continue to treat these HCs as part of the public health system. f. If devolution is to continue to be a voluntary choice for HC staff, HC staff recommend that the MOPH should provide clear, unbiased advice about benefits and risks, and prepare staff for scenarios in which the TAO-CEO, Council and other key personalities may change. C. FINDINGS FROM DISCUSSIONS ABOUT DECENTRALIZATION WITH POLICY MAKERS AND RESEARCHERS Interviews were conducted with policy makers, advisers and researchers from the health sector, NDC, MOI, MOF and Bureau of Budget, in addition to meetings with Deputy Governors, provincial administrations, PCMOs and other PHO managers involved in devolution, in a number of the provinces visited. Summaries of the points made in these interviews are provided in Annex 1. Annex 4 lists persons interviewed. There does not appear to be support at the highest levels for a forced implementation of the 2008 Action Plan for implementation of the Decentralization Law by a 2010 deadline, nor for exercising the “default option� in the Action Plan of transferring HCs to PAOs until TAOs are ready to manage them. None of the interviews found support for decentralization of hospitals, though some PAOs and municipalities have acquired or developed their own hospitals, and of course some municipalities have operated their own hospitals for many years. Further devolution of health services at this point appears to be optional, even for health centers. Among those who support some form of decentralization in principle, there is reasonable consensus that implementation in the health sector should be gradual and careful. Most health sector interviewees prefer models of decentralization that would devolve district, province or regional networks, rather than individual HCs. For many this has become a reason to oppose devolution of HCs. However, there is no evident 13 consensus about design of a network model, no systematic work is being done on design of this option, and this is acknowledged to be a much more radical reform. Positive opinions about potential benefits of devolution of HC to TAOs and municipalities were articulated by a number of non-health-sector interviewees, and by some health sector interviewees at provincial, district and HC level. The much shorter and more direct supervision and accountability relationship that devolution produces is seen as having the potential to increase HC staff productivity, pro-activity in addressing community health needs, and patient-service orientation. Although technical capacity is limited in TAOs, supporters of devolution believe this can and should continue to be provided by the PHOs and CUP hospitals. Supporters of devolution argue that the CUP Board and purchasing mechanism can provide the necessary coordination and integration of service delivery, as it increasingly does now. Based on the assessments of interviewees, continued implementation of the current guidelines for voluntary devolution of HCs may result in only 20 percent of HCs transferring to LAOs over time. The extent of implementation would be likely to vary across provinces. In some provinces visited, only a third of TAOs expressed interested in taking up HC devolution, whereas in others, 70-80% of TAOs are reportedly interested. Budget and capacity constraints in the large number of smaller TAOs will prevent a substantial share of TAOs from ever meeting readiness criteria, unless these TAOs are given additional funding and/or unless there is a further revision of the statutory limitation of 40% on the share of LAO budget that can be spent on personnel compensation. The unwillingness of a large share of HC staff to transfer to LAOs may be the binding constraint: interviews indicated that depending on the province, the proportion of HC staff supporting devolution ranges from 20- 50%. The most frequently cited reasons among both health sector and non-health sector interviews for unwillingness to transfer HCs to TAOs are: lack of career path and job mobility, uncertainty or lack of clarity about MOPH policy on devolution, uncertainty about future TAO funding for health, and risk of increased political influence as a result of increased proximity of the HC to political decision-making. The first two of these issues could be addressed by policy initiatives. Non-health sector interviewees express confidence in the mechanisms already established to protect merit-based hiring and promotion and deal with personnel grievance for LAO staff. The most likely scenario resulting from gradual HC devolution predicted by those interviewed is one in which the health system would change only at the margins, in an incremental way. The MOPH/NHSO will continue to regulate, coordinate, supervise, finance and support HCs and other primary health care provision as now. Gradual and cautious implementation of HC devolution will likely contribute to a gradual, modest increase in pluralism and heterogeneity in financing and ownership of health services. Many of those interviewed pointed out that this trend is nothing new. There is already substantial health care provider pluralism in public health service delivery in Thailand: the university hospitals, other Ministries’ health facilities, municipal health facilities, and private healthcare facilities participating in UC, SSS and CSMBS. There is already growing development of partnerships between LAOs and MOPH (at 14 the levels of PHOs, HCs and hospitals). So long as PAOs and TAOs continue to receive substantial fiscal transfers but have few mandatory transferred functions, their involvement in complementary financing of health services and in direct health service provision is likely to grow. Even if there were no further devolution of HCs, this would likely lead to increased variation in public health expenditure between different provinces, tambons and municipalities, because there appear to be marked differences in the electoral salience of health issues, and the interest of LAO leaders in health, in different parts of the country. D. DISCUSSION The assessment team found that the devolution of HCs in the sites visited is producing early positive results. Our over-riding impression is that this is a very limited, incremental change in the ownership and governance of HCs, which has the potential to produce benefits, and which carries relatively little risk in the short term. The major sources of financing and process for resource allocation remain with the NHSO. The major sources of technical support, training and supervision of the HCs remain divided between the DHO and PHO on the one hand and the CUP hospital, as now. The PHOs and CUP Boards are thus in a very strong position to prevent and manage any potential risks of devolution – to offset any risks arising from the limited capacity of the TAOs, to ensure coordination, and provide some incentives and sanctions for performance using the “purchasing� mechanism of the CUP Board. The PHOs and CUP Boards that adopted the mind-set that devolved HCs as part of the health system in the same way as before, continue to take responsibility for ensuring that health services are delivered continuously and appropriately in the devolved HCs. The devolved HCs and TAOs welcome this continuity in the role of the PHO and CUP Board. - Longer term risks and issues for devolution of health services: In some countries, some problems emerged and existing problems became worse only after some years following devolution. As the composition of staff in devolved health services shifted from being predominantly ex-MOH staff, to being predominantly staff who had never worked for the MOH, informal cooperation based on longstanding working relationships diminished. In some countries, LAOs progressively became more assertive about their mandate to act independently of national government on a range of policies (e.g. recent successful court action by Indonesian governors to challenge the centralized provisions of new social health insurance legislation). Decentralization also leads to changes in the required skill mix, business processes and organizational structures of the MOH. It commonly takes several iterations of organizational change over a period of many years before the MOH finds a firm footing for its leadership role in a decentralized health system. Common longer term problems identified in reviews of health decentralization include the following:  Deterioration in timeliness and completeness of reporting to MOPH and development of incompatible information systems;  Deterioration in compliance with technical guidance and protocols;  Decline in use of centralized or pooled medicines procurement and logistics services such as GPO, leading to higher costs of medicines procurement 15 because of loss of volume discounts, and also because of variable procurement capacity in local governments (though there is some evidence that local planning and budgeting for medicines and supplies is more efficient than centralized planning and budgeting and free delivery; in Thailand these functions are already de-concentrated to the CUP hospital);  Non-rational, uncoordinated investment in hospitals and other high-cost diagnostic and treatment capacity, sometimes leading to a “medical arms race� among rival LAOs;  Where accountability and responsibility under decentralized systems is ambiguous or overlapping between different levels of government, during periods of fiscal consolidation, there is a tendency for cost-shifting between different levels of government, and for central government to add unfunded or under-funded mandates to LAOs;  Gradual increase in inequity between high and low performing LAOs (though findings are mixed: some countries have achieved better allocation of resources to the rural periphery following decentralization because putting a greater share of public expenditure under the control of local authorities, and using non-discretionary, pro-poor grant allocation formulae for intergovernmental transfer overcame previous pro-urban bias in government expenditure). - Risk mitigation measures developed in decentralized health systems Mature decentralized health systems have evolved a range of mechanisms to achieve health system coordination and integration, economies of scale, and national public health priorities. They have also evolved mechanisms for dealing with externalities between different local governments, and for dealing with serious performance failures of sub-national governments. Newly decentralizing countries ideally need to design in these mechanisms as part of the devolution implementation plan. Additionally, newly decentralizing countries ideally need to build capacity not only in LAOs for their expanded responsibilities but also in communities to strengthen grass roots participation in holding democratic structures accountable. Various combinations of the following risk mitigation measures are found in different decentralized health systems:  The central government takes regulatory powers to deal with serious performance failure by local administrations – such as central appointment of a commissioner to manage the failing local government for a defined period.  The central government may take regulatory powers to coordinate health care capacity planning and investment – such as through development of a master planning or carte sanitaire, or through some form of certificate-of-need regulation. In some countries, major changes to health service delivery have been initiated by specific legislation that mandates local government to provide specific types of services to defined standards (e.g. specific legislation is adopted at national level from time to time on changes to service delivery, such as primary care development, new prevention programs or specific types of high priority secondary/tertiary services). As an alternative some countries use strategic purchasing and central capital investment grants as a means of creating incentives for rational coordination of major capacity development; 16  The central MOH needs regulatory powers and strong enforcement provisions in relation to major disease control issues and health emergencies.  Where there is universal coverage and a single or dominant funder/purchaser of health services, active and strategic use is made of the purchaser’s powers to coordinate service development and motivate quality improvement, standards compliance, and information provision.  Some countries provide a legal basis for joint-LAO service provision, and or for small LAOs to delegate responsibility for some functions to other LAOs. In some countries, the central Government has played a lead role in negotiating with LAOs to establish joint service provision structures or inter- LAO coordination arrangements.  In some countries, federations of LAOs play a role in coordinating policy and strategy and in consulting and negotiating with health professional associations, health care provider associations and unions of health workers.  Many countries have human resource policies for health care workers that make it easy for staff to move between LAOs and between LAOs and central government in devolved health systems or to move between autonomous public health care providers. These policies may include coordination of junior doctor/trainee doctor placement; portable or transferrable pension rights; mandatory open advertisement and recruitment for certain types posts; harmonized job classification, and grade structure for some types of staff; central subsidies for training and central initiatives to help fill “hardship� posts.  A central medicines and health goods procurement service to achieve economies of scale and volume discounts in medicines and supplies procurement.  Some central governments have instituted performance-linked specific grants to promote achievement of various national policy objectives.  Governments or donors supported NGOs or national institutes to develop local community capacity to be well-informed voters, advocates and monitors of governance and service delivery. - Risk Mitigation Measures Available Already In the Thai Health System As noted above, Thailand’s health system is already somewhat pluralistic, and already has a number of the mechanisms and levers available to the MOPH and NHSO that other countries use to achieve coordination, handle public health emergencies and protect patients from risk of institutional failure. The Memorandum of Understanding that LAO-CEOs sign when HCs are devolved commits the LAO to abide by MOPH regulations and guidance. The NHSO purchasing mechanism can continue to provide resources, create incentives and influence service delivery without change. The actual practice of some PHOs that have worked with the devolved HCs already demonstrates that the MOPH has the capacity and levers to manage the early transitional problems, and provides some assurance of capacity to manage future risks. The policy of many CUP Boards to supply unlimited “free� medicines and supplies to HCs provides strong incentives for LAOs and HCs to continue to source their drugs from the CUP hospital, which uses GPO’s centralized, pooled procurement services for essential medicines. However, the ONDC seems to be urging the NHSO to provide more of its support to devolved HCs “in cash� rather than “in kind�. This is 17 inadvisable. We found one case where the CUP Board limits the amount of medicines allocation to HCs – requiring them to use their own revenues to pay for any additional supplies needed. One CUP Board had begun to transfer the OPD allocation for medicines and supplies to the devolved HC as cash, rather than in kind. This CUP Board also has a policy of adding a levy of 15% on to the GPO prices for medicines supplied to HCs. It may be advisable for NHSO guidelines to discourage CUP Boards from doing this, in the interests of maintaining incentives for devolved HCs to use the hospital’s procurement and logistics management services. HCs are too small a risk pool for primary care drugs once they start providing care for chronic non- communicable disease patients. Additionally, because HCs vary in the extent to which they are providing curative care, CUP boards need to exercise flexibility in deciding how large a drugs allocation each HC receives. As well, because patients register with both HC and CUP hospital for primary care and can go to either place, there is scope for the devolved HC to refer to the CUP hospital any patients needing higher cost drugs – so there are incentives for cost shifting from any “hard budget� constraint imposed on the HC. The CUP hospital’s pharmacists are also better placed than the TAO to supervise inventory management at the HC to ensure efficient levels of stockholding and management of expiry dates to avoid waste, and to act as a check on leakage of products. Although concerns have been expressed by some about whether the MOPH has enough leverage over devolved HCs to manage public health emergencies, this concern is demonstrably misplaced. The MOU provides a formal basis for the MOPH to exercise the necessary authority. There is already actual experience of high levels of formalized and informal cooperation of even the for-profit private health sector in working with the MOPH in public health emergencies; LAOs have even greater formal duties and incentives to cooperate in these scenarios. The MOPH can use simulation exercises to ensure that coordination mechanisms for management of emergencies are well understood and effective, if it has not done so already. - Risk Mitigation Measures that May Warrant Further Attention At this stage, there is no well defined mechanism for dealing with cases of serious performance failure by LAOs in dealing with sector-specific responsibilities, such as health responsibilities. The MOI has considerable powers of supervision and sanctions for dealing with financial and administrative wrong-doing. Conceivably, an HC could be transferred back to MOPH management if other mechanisms are insufficient to ensure continuous, adequate service delivery. At this stage, there is no mechanism for coordinating planning of major new developments and investments in health sector capacity by LAOs. This does not seem to be perceived as a high priority concern at this stage, however, given that Thailand faces a need for increased expansion of facilities for curative care in most or all of the country. However, based on the stated intentions of some of the LAOs interviewed, the risk of inappropriate and inefficient scale and siting of new hospital developments by LAOs may arise quite soon. Already, some PHOs and CUP Boards are pursuing local strategies for developing and upgrading their primary health care network. The CUP Board can use their purchasing role to encourage LAOs to cooperate with these developments, but there may be cases where regulatory power is needed as well, to avoid inappropriate or duplicative development. 18 At this stage, there are no well-developed LAO human resource policies that facilitate career path development and job mobility across different parts of the country for local government staff. It seems that addressing this issue may do more than perhaps any other intervention to increase the willingness of HC staff to transfer to LAO employment. Some aspects of this issue are multi-sectoral – for example pension transferability, or policies on open advertisement of LAO positions. But design of nation-wide policies for a local government health career service has sector-specific features that require MOPH’s expertise. Additionally, it is desirable to align and coordinate some elements of a local government health career service with MOPH human resources policies. Already, each province with devolved HCs is facing the need to develop its own job descriptions, hiring criteria and examinations for filling HC vacancies and new posts – although there would be clear benefits from coordinating this task nationally, and drawing upon the expertise and policies of the MOPH. If the number of devolved HCs increases substantially, the current problem of lack of local government career opportunities in health will ease. But if the pace of HC devolution continues as slowly as it has to date, it would seem to be necessary to allow individual staff more opportunity to move between central and local government service, while maintaining pension rights, and recognition of experience in both central and local government sectors. If there is a steady growth in the number of devolved HCs, and growth in LAO engagement in other primary care and hospital service provision, this growing pluralism will gradually increase the importance of the MOPH’s role in stewardship of organizations it does not directly own and control. One risk of such a process taking place very slowly and gradually is that it may delay the process of adjusting the MOPH’s organization and capacity to reflect this changing stewardship role. Gradualism means that there are not clear decision points at which new policy levers need to be created so that the Ministry may run the risk of progressive erosion of its leadership and influence. Currently LAOs receive substantial fiscal transfers, but have not yet received very major transfers of functions and staff. Until now, many LAO functions are “permissive� or “discretionary� functions, rather than mandatory obligations for public service delivery. As a result, LAOs have considerable freedom to provide complementary or supplementary resources for P&P or curative health care delivery if they wish, but are under no specific obligation to do so – they may choose other priorities. Our interviews with health sector and LAO informants reflect implicit expectation that LAOs will continue to have this level of discretionary resource available to them, so the health sector stands to gain from engaging with LAOs in order to tap into this resource. However, this favorable situation may not be sustainable in the face of pressures for fiscal consolidation arising from the financial crisis. It is possible that pressure will increase either to give LAOs more specific, mandatory accountability for some functions without full fiscal compensation, or alternatively to revisit the current expenditure and revenue sharing policies. Rising fiscal pressure is likely to increase the risks of cost-shifting or responsibility- shifting, so there may be a need for more clarification and specification of the LAO’s accountability for finances and service delivery. The ONDC review of HC devolution responds to this issue by recommending clearer national rules about what NHSO 19 funds should be transferred to devolved HCs. However, it does not seem feasible or desirable to specify and formalize these accountability relationships nationally, because of the local heterogeneity of how health services are organized, the mix of financing sources, and local variation in “who does what�, and where patients go for services. PHOs and CUP Boards operate with a high degree of delegated authority, for good reason. However, HCs devolution adds to the complexity and fragmentation of accountability for primary health care service delivery. Even before devolution, HCs have multiple sources of financing and complex accountability: – to MOPH, via the DHO and PHO, for personnel administration, most technical dimensions of service delivery, financial control and fiduciary audit; – to the CUP Board for service delivery to most patients, for some P&P for their community, and for a large share of their finances; – to SSS and CSMBS for other service delivery and a smaller share of their finances; – to the LAO for projects financed through the Community Health Fund. Devolution of the HC transfers to the LAO the MOPH’s responsibility for personnel administration and financial control. After devolution, the MOI also assumes some of the MOPH’s responsibility for fiduciary oversight of devolved HCs. The MOU signed when the HC is transferred gives the LAO responsibility for running the HC in compliance with MOPH regulations. But the LAO only receives a transfer of MOPH budget to cover the costs of salaries of an average of 3 government officers – i.e. the fiscal transfer is much less than the cost of complying with MOPH regulation. The financial burden of running the HC according to MOPH standards is a shared responsibility among the LAO and the CUP Board (and for some HCs, neighboring CUP Boards, SSS and CSMBS are important revenue sources). The MOPH also retains some responsibility for providing resources after devolution – for providing technical support and training. This complexity and interdependence makes it difficult to define what the LAO is accountable for – accountability is shared, and needs to be negotiated locally. This ambiguity about accountability in the relationship between the CUP Board and the LAO derives from the ambiguity that already exists in the relationship between the CUP Board and the MOPH, given that there is not a complete purchaser/provider split. The CUP Board could view the LAO as a co-funder and supervisor of the HC – because the LAO is taking over the MOPH’s funding role and parts of the administrative supervision roles of the DHO and MOPH management hierarchy. Under this view, the LAO and CUP Board would need to cooperate just as the MOPH and CUP Board do now and jointly hold the devolved HC accountable. The DHO and PHO would also continue to be part of this joint accountability arrangement – because they retain responsibility for technical supervision. This approach has the advantage of preserving existing roles and relationships to a substantial extent. This approach recognizes that the HC has some de facto autonomy. But formally, the HC is legally part of the LAO, and its performance could be substantially affected by LAO policies and performance over which it has little or no control. In future scenario’s in which LAO performance is at issue, the CUP Board may want to use its financial leverage to exact accountability from the LAO for ensuring service delivery to UC members. Additionally, the LAO will be the main source of future capital investment and other 20 development finance for upgrading and improving primary health care (PHC) in the community. If the MOPH nationally, or PHO and CUP Boards locally want to progress strategies for strengthening PHC, they will need to secure the financial support and cooperation of the LAO. The CUP Board could consider in future contracting with the LAO to provide specified services and specified funding for UC members – as it does now for Community Health Funds. These agreements could evolve into multi-year agreements that covered agreements over new developments in PHC, as well as payment for current service delivery. - The Criteria and Process for HC Devolution There are some calls to relax the criteria in the MOPH guidelines for HC devolution to enable more transfers to go ahead. Given the incremental nature of the change, the capacity of the MOPH and CUP Board to manage the risks involved, and the potential benefits illustrated by the experience of the first pilots, there is a prima facie case for reviewing the guidelines. The “LAO readiness� and good governance criteria are prudent as a means of managing risks of transition. There is empirical backing for these criteria: a Philippines study finds an inverse relationship between local government governance indicator scores and basic health indicators. A number of interviews identified the problem that small TAOs face in meeting the LAO readiness criteria because they have inadequate budget to establish a Public Health section without breaching the statutory limit on the share of budget spent on staff compensation. There seems to be a case for revising this percentage limit for LAOs that undertake health and education responsibilities. In both sectors, it is common for staff compensation to account for over half of the budget in middle income countries, and a higher share in upper income countries. In the case of HC transfer, the case for reviewing this limit is even stronger because only the salaries costs of the HC are transferred to the TAO. The rest of the HC’s operating costs are met by NHSO funds, and other sources. For the smallest TAOs, in the longer term, devolution may be difficult unless policies are developed to promote TAO mergers or provide a legislative basis for public service delivery organizations owned by more than one TAO. The “community willingness� criteria are means of providing additional local democratic legitimacy to the implementation of decentralization, which was described by a number of interviewees as a “top-down� initiative. Community members are also well placed to judge whether their LAO will give priority to health, and this is an opportunity for them to signal whether they have confidence in local governance and management capacity for health services. In one of the provinces visited, community members reportedly gave considerable weight to the view of the HC head before deciding whether to vote to support HC devolution. Under the “HC staff willingness� criteria, devolution only takes place if at least half of staff agree (including the HC head) to transfer to the LAO’s employment. Staff are not given the option of remaining on the staff of the MOPH but continuing to work in the HC in the long term. However, the staff who do not wish to transfer to LAO employment can request for transfer to another MOPH post. This step of the process has the effect of combining two distinct decisions. The first decision – whether or not devolution should take place - is a policy decision that should be made based on 21 public interest criteria - drawing upon analysis, evidence, wisdom and consultation with those affected. The second decision– on the staff member’s future employment status – is a personal choice that staff should feel free to make on the basis of “private interest� criteria - such as financial security, career aspirations, interests of family, etc. Because the devolution process uses this personal decision of the staff as one of the criteria that determines whether or not devolution takes place, it has the effect of combining both a public interest and private interest decision in a way that is leading to confusion about whether “public interest� or “private interest� criteria should apply. Some HC staff are deciding based on an unclear mixture of public interest concerns (such as whether or not devolution would mobilize more resources for the HC, or the risk of politicization or corruption), and private interests (e.g. their chances of being promoted to Public Health Section Head, their personal/familial/political affiliations to the TAO leadership or to the opposition political party). The personal concerns of staff can be addressed fully by the provisions in the MOPH guidelines that allow them choice of transfer to LAO employment, or transferring to another MOPH post elsewhere. This provides assurances of protection of staff employment rights and career opportunities, but one negative consequence of this policy is that some devolved HCs are left with an inadequate number of staff. Clearly individual staff’s private interests should not be the basis for making a public policy decision on whether or not to devolve HCs. The HSRI evaluation of the devolution pilots found that many HC staff are reluctant to vote to transfer to the LAO unless there is a clear policy on devolution from MOPH management. This appears to be in part because the current devolution process puts them in a situation where their personal choice may be the deciding factor that determines whether or not HC devolution will go ahead. It appears that many HC staff, understandably, are reluctant to do this if they are uncertain about the views of MOPH management. Consideration could be given to removing the “HC willingness� criteria as one of the factors that determines whether HC devolution will go ahead. Obviously, it would be difficult for a devolved HC to function accountability to the LAO unless the HC is an LAO employee, but there are alternative transition processes that could be devised to achieve this result. (For example, the ONDC Evaluation recommendations, included in Annex 1, propose a mechanism.) The field visits identified particular issues arose from the implementation of the “HC willingness� criteria in the case of TAOs with two HCs where staff in one HC voted to devolve, while staff in the other HC voted not to devolve. In these examples, personal or political factions, and rivalry over promotion prospects were evident. There were some early signs of higher risks in these contexts of visible inequity in resource allocation between devolved and non-devolved HCs in the same Tambon, and of poor communication between the HCs. Some MOPH staff interviewed in these areas identified potential risks of problems in the event of future change of local TAO CEO. Consideration could be given to revising the guidelines so that decisions on devolution are made for both/all the HCs in the LAO, not for selected HCs. This may also strengthen the message that devolution involves a transfer of responsibility for health objectives and health services for all of the citizens in the Tambon, without regard to political or personal affiliation. 22 PART II. HOSPITAL AUTONOMY UNDER THE AUTONOMOUS PUBLIC ORGANIZATION LAW Interviews were conducted with current and former managers of the Ban Phaeo Autonomous Public Organization (APO) Hospital, with the director of the Patong Hospital (which applied for APO status), and with members of the management team of the Phuket International Hospital. Discussions with policy makers, advisers and researchers from the MOPH, NDC, and Bureau of Budget also covered questions related to hospital autonomy. Summaries of the points made in these interviews are provided in Annex 1. Annex 4 lists persons interviewed. A. THE BAN PHAEO APO MODEL OF HOSPITAL AUTONOMY Ban Phaeo hospital was a 200 bed MOPH community hospital in 1999 at the time the decision was made to convert it into an APO – a form of government-owned, autonomous, non-profit organization, under a new law adopted in 1999. It is now a 300 bed hospital offering services at primary and secondary level, with some tertiary level services. It is seeking to upgrade its status to that of a general hospital. It is one of three public hospitals in Samut Sakhon province – a densely populated peri-urban province with a registered population of around 400,000 and an actual resident population of over 1 million. The model of autonomy implemented at Ban Phaeo was developed drawing upon preferred features from autonomous and corporatized hospitals in a range of countries, including Singapore, United Kingdom, Australia and New Zealand. The model represents a marked and generally consistent shift in the key dimensions for influencing organizational performance (described in Preker and Harding, 2003), from those appropriate for core government ministry functions to those of appropriate for an autonomous, non-profit service provider. Such changes create a strong set of incentives for improved performance and expansion of the business. The changes at Ban Phaeo hospital in these key dimensions that influence organizational performance are:  Management Decision Rights over organizational structure, organization of services, human resources, finances, logistics, and capital investment are shifted to the hospital board and its director, with the partial exception of decision rights over capital. Disposal of surplus land granted by the government or donors, and “equity injections� and borrowing rights for financing of major capital investment are not fully shifted to the hospital – these decisions require Cabinet approval. The hospital is also free to contract in or out and enter into partnerships with the private sector and non-health sectors.  Residual Claimant Status is fully with the hospital. There is not a clear regime for the event of financial failure or bankruptcy, though there is a precondition of demonstrating financial sustainability before APO status is granted. The Board is reported to be quite focused on ensuring that losses are avoided. The hospital retains the proceeds of most forms of efficiency gain, with the exception of efficiencies in management of granted and donated land and buildings.  Market Exposure is quite high because the revenues of the hospital are derived from fee-for-service (from CSMBS, which is the largest revenue source), case- 23 based payment systems of SSS and NHSO, and specific service contracts and project finance from NHSO, and other public health sector institutions. The location of the hospital in a densely populated area creates considerable potential to attract out-of-area patients, which increases its revenue from CSMBS and NHSO. Unlike MOPH hospitals, there is no salaries top-slice from the UC payments. The hospital is paid by UC at 100% of the case based payment and capitation rate. The MOPH budget does not guarantee payment of staff salaries, as is the case for MOPH health facilities. Staff remuneration comes from revenue for services delivered. The hospital manages the CUP for a defined catchment of UC patients, which gives it a somewhat protected market for this group of patients. Additionally, the neighboring CUP for the provincial general hospital has a policy of not charging out-of-area self referrals, which in theory could foster cost-shifting – though in practice, this does not seem to be a concern.  Accountability; accountability is to the purchaser (NHSO, CSMBS and SSS) for service delivery under the provider payment policies and regulations of these agencies. Accountability for performance, for service development, and for financial sustainability is to the Board (discussed in more detail below under Governance Structures).  Social functions (unfunded mandates and community participation): the hospital has a policy of providing exactly the same clinical care, including access to medicines, to UC patients as to other socially insured and private patients. Its main unfunded mandate is treatment for illegal migrants. The hospital does not refuse them treatment, on humanitarian grounds, but has progressively taken a tougher stance (e.g. risk of reporting to immigration authorities). The hospital has extensive engagement with its community, both in consultation over service development and hospital development, and in raising donations and accounting for their use. It undertakes some corporate social responsibility initiatives from its surplus. Preker and Harding (2003), drawing on evaluation of a range of country case studies, posit that reform is more likely to achieve benefits and minimize risks of unintended adverse effects if it makes changes to these five dimensions in a coordinated and consistent way. Reforms that make radical changes to some of these dimensions while neglecting others have been found to run risks of loss of financial control, loss of efficiency, or reduction in delivery of social obligations such as equitable access for the uninsured. Figure 1 maps the position of the Ban Phaeo model on these five dimensions, and illustrates the coherence and consistency of the organizational reform. 24 Figure 1: Critical Factors Influencing Organizational Behavior Ban Phaeo APO: Key Characteristics (Preker and Harding Framework) Ministry Autonomous Corporatized Private Directorate Non-Profit State Enterprise Enterprise Decision rights Market exposure Residual claimant Accountability Social functions Definitional notes: “Decision rights� refers to the extent to which management decision rights are moved from the MOH hierarchy to the hospital’s board and management team. “Market Exposure� refers to the extent to which the hospital is paid on the basis of outputs or services provided, and the extent to which patient choice or competition determines its revenues. “Residual Claimant� status refers to whether the hospital is able to retain surpluses earned through efficiency gains and revenue growth, and the extent to which it has a “hard budget constraint� (will not be bailed out if it runs deficits). “Accountability� refers to the extent to which accountability is transformed from vertical administrative accountability for compliance with the rules of the MOH and other public sector rules, towards accountability for organizational performance to a Board of Directors (and ultimately to stakeholder or shareholders). “Social functions� refers to the extent to which social obligations – such as providing health care for poor, uninsured patients - are explicitly recognized and paid for (e.g. under contracts) rather than treated as implicit and often “unfunded mandates�. It is useful to supplement the Preker Harding framework by looking at changes in the internal incentive environment for management and staff following autonomy. The incentive framework for management and staff adopted by the hospital’s Board appears to be strongly focused on aligning the incentives of staff with the objectives the board has for the hospital. The Director has a fixed four year term contract, renewable for only one term, and is subject to annual performance targets and review by the Board, which can decide on the level of Director’s bonus based on this assessment. Staff are no longer civil servants, and no longer participate in the civil service pension and medical benefits schemes.3 They are employed by the hospital 3 There was a transition process for four years during which staff could choose whether to retain civil service status while continuing to work in the hospital, or to shift to employment by the hospital under private law. Staff could also request to be transferred back to MOPH. By the end of the four-year period, all staff who chose civil service status had transferred out to other positions in the MOPH. 25 itself under private sector employment law, and their contracts can be terminated for poor performance more readily than is the case for civil servants. Doctors are paid a combination of salary, shared fees for service and performance rewards, and are subject to annual performance review. Medical staff are not permitted to work part time for other private sector hospitals or clinics, and in return are paid a substantially higher salary than the MOPH salary. They are permitted to earn additional fees for service in treating private patients after hours within the hospital. The hospital also hires on a part time basis specialists who are full time employees of other public hospitals. Prescribing is controlled by a hospital formulary, set by a Pharmaceutical and Therapeutic Advisory Committee, which is somewhat broader than the EDL and UC formulary (it includes more brand-name medicines). All patients, including UC patients, receive drugs based on this formulary. The hospital does not charge a profit margin on medicines and does not have any partnerships or profit sharing arrangements with private pharmacies. By contrast, conventional MOPH hospitals are constrained in the level of salaries and allowances they pay medical staff by national public sector salary regulations that keep remuneration substantially below the level of private sector doctor remuneration, and higher levels of private sector remuneration for other categories of workers. As a result, over 73% of MOPH doctors and 9.5% of professional nurses worked part time in the private sector in 2005 – up from 55.4% and 8.2% respectively in 2003. Although MOPH hospitals since 2005 have had considerably increased financial freedom (they can retain income from UC, SSS, CSMBS and user fees, and can offer private beds with enhanced quality of “hotel� services and enhanced choice of therapies, including drugs), they do not have the freedom Ban Phaeo has to use this revenue to increase staff remuneration. Linkages and Integration of the Hospital with the Public Health System: The hospital functions within the public health services network in the same way as other community hospitals – though it has now expanded its capacity and range of services to a higher level than is typical for community hospitals. It manages the CUP for outpatient services (largely curative primary care) and P&P for its District catchment area like any MOPH hospital. It has set up three PCUs/CMUs to provide PHC, which will supervise and support HCs in the catchment area as well as providing curative care. Two of the three PCUs also provide P&P. The hospital provides some P&P staff to work in HCs. Interviews with HCs in the catchment area found perceptions that the hospital is somewhat “less generous� than other CUPs that they are aware of – but this perception may be influenced by the fact that Ban Phaeo provide more support in kind and less in cash than most CUPs. Ban Phaeo’s staff – like devolved HC staff - are cut off from career mobility through the MOPH. The hospital plans career paths for its medical staff over a period of 9-10 years in the case of doctors, and provides scholarships in return for contracts to return to the hospital – but so far, it does this on its own, not as an integral part of MOPH hospital career paths. However, it has some linkages to the junior doctor training system. The hospital employs interns from public medical schools after graduation in the same way as other MOPH hospitals, and is seeking agreement of MOPH to also be assigned medical registrars who are providing public services in their first three years after graduation from a public university. 26 Governance Structures and External Accountability of the APO to its Owners and Other Government Stakeholders: The organization charts used by the Ban Phaeo hospital typically show it as subordinate to the MOPH. However, it may be more accurate to describe it as subordinate to the Cabinet, as the Cabinet holds some key decision rights (such as approval of Board membership and of capital finance or borrowing). There is no dedicated unit or agency in the MOPH nor any other part of Government responsible for independent monitoring and regulatory oversight of good corporate governance, nor of the financial performance of APOs or their performance in improving the value of the business. The Bureau of Budget reviews and provides advice to the Cabinet on any requests by APOs for capital finance. The criteria used by the Bureau are the same as for APOs in other sectors: low priority is given to APOs with substantial own source revenue (this was the main reason for rejecting Ban Phaeo’s application for capital finance to date); high priority is given to APOs responsible for investments that are part of a Government strategy or policy. The composition of the APO board includes ex officio the Permanent Secretary of the MOPH and PCMO, as a direct mechanism for enabling the MOPH to monitor and participate in decisions of the hospital, and the provincial governor who has a mandate to ensure local coordination of central and local government administration. The processes for appointing the other board members of the APO hospital involve a number of stakeholders and build in some checks and balances to seek to ensure both meritocratic selection and political accountability. The Royal Decree establishing the hospital as an APO specifies the composition of a search committee to identify candidates for the Board. The search committee is chaired by the provincial governor, and is appointed with the agreement of the provincial governor, PAO CEO, CEOs of the TAOs and municipalities making up the District, the District Officer and the PCMO. The Royal Decree specifies that the Board should include three community representatives and three experts in addition to the three ex officio members. The search committee is obliged to identify two candidates for the positions of Chair (whom may not be a government employee), and six community and six expert candidates for the three board posts for these categories. The Minister of Public Health selects candidates from the short lists proposed by the search committee, and submits the final board membership list to the Cabinet for approval. Interviewees who had reviewed lessons from experience with the Ban Phaeo Board concluded that the Board composition could be strengthened by inclusion of only professional Board members, on the basis of their skills as directors (including hospital management and service delivery skills). The hospital does not have LAO representation on the Board, and reportedly has had little take up from LAOs in response to initiatives to enlist their engagement in supporting the hospitals or working with it on joint initiatives. The Board meets for 3-4 hours every month and is described as an “activist� board. Its decisions are usually made by consensus, and rarely go to a vote. Where there is disagreement of 1-2 members, usually the proponent of a recommendation provides more information to address concerns raised until consensus is reached. MOPH’s role on the Board is characterized as more passive, and mostly focused on providing input and information on government and MOPH policy matters. The Board is reported to be focused on service delivery performance and new development of the hospital, but is not particularly focused on efficiency or cost containment – though it is concerned to ensure the hospital avoids losses. 27 Social Accountability to the Community Served by the Hospital: In addition to community representation on the Board, Ban Phaeo raises donations from the community and consults and reports to the community on how donated resources are used. As well, the hospital’s Board has chosen to adopt some other forms of community participation, though these are not mandated by law or Royal Decree. It sends representatives to meetings of community leaders at District, Tambon and village level to provide information about the hospital, consult and seek support for service development and receive feedback. The Board also commissions the (independent) Thailand Rating Information System to conduct annual patient satisfaction surveys. These practices in Ban Phaeo are not unique to APOs, and may not be linked to APO status: a survey of 209 MOPH hospitals found that many have some form of community participation in hospital consultative committees, primarily focused on fund-raising for the hospital, though Ban Phaeo is the only one with formal community representation in governance. A number of other MOPH community hospitals have committees and processes for community participation in development plans, for coordinating between the hospital and the community, for management of infrastructure and non-medical activity, and for community support to health service implementation. Performance and Evaluation of Processes and Results: The Ban Phaeo hospital APO has been highly successful in increasing the outputs, range of services, and turnover of the hospital, as has been documented in a series of before-after evaluation studies. Some early studies noted some transitional issues in developing management capacity and systems, but found generally appropriate development of capacity to manage autonomously. A study of quality found no adverse effect on clinical practices or outcomes of care in the three clinical areas studied, though record keeping deteriorated. Patient satisfaction rates rose after autonomy, then flattened and decreased slightly in the last 2-3 years. However, satisfaction remains high at 86%. Declining satisfaction is perceived to be due to increased utilization – giving rise to increased waiting. But it is the only APO hospital, and was a self-selected candidate for APO status. The initiative and final decision to grant the hospital APO status involved strong advocacy by the hospital’s management and many doctors at the hospital, for reasons that may be associated with successful performance under any status. It is viewed by some of those interviewed as a unique or atypical case. However, its former and current managers believe the model is replicable in other larger community, general and regional hospitals with a diversity of revenue sources. In smaller community hospitals serving small or sparse populations and reliant almost entirely on UC and MOPH finance, the positive dynamic achieved in Ban Phaeo that led to a virtuous cycle of revenue growth and expansion would be difficult to achieve, and such hospitals would best be given autonomy as part of a larger network of hospitals. B. SCALING UP THE APO MODEL A survey of 209 MOPH community hospitals found 25 percent were interested in pursuing autonomous status. During 2006, a formal invitation to hospital to express interest in autonomous status attracted around 45 expressions of interest, although only one of these (Patong Hospital) pursued this process to the stage of submitting a 28 formal proposal for decision. In this process, a semi-autonomous “Service Delivery Unit� status within the MOPH was also on the agenda. In the context of deliberations on implementation of the Decentralization Law, the NDC and Commission on Public Sector Reform have been opposed to further creation of APO hospitals. They do not regard APO status as a form of decentralization. Under current law, it is not possible to transfer APO’s to LAO ownership. There is no inconsistency in principle, however. Transferring a well functioning autonomous hospital should be less of a financial risk and managerial concern for a PAO and the MOPH than transferring a conventionally managed ex-MOPH hospital. More recently, the Commission on Public Sector Reform and the Cabinet have halted any further creation of APOs. This decision was taken in response to cases in which some of the many non-health-sector APOs established in the 2002-2005 period have performed poorly and run into financial difficulty. Boards of these APOs outside the health sector in some instances have been criticized for awarding themselves high remuneration, relative to the duties performed. It appears that the Royal Decree establishing Ban Phaeo APO hospital and the draft Royal Decree for Patong APO hospital incorporated many important features, drawing upon lessons from hospital autonomy in other countries – including appropriate focus on social objectives, criteria and processes for selecting a professional Board, and role of the Board – that were not adopted by other sectors that created APOs. C. DISCUSSION If at some point in the future, there is renewed interest in scaling up hospital autonomy to include significant numbers of hospitals, there are some further elements of the policy and institutional framework that would need to be developed. Supervision of APO Boards - As the experience of APOs in other sectors illustrates, the APO model cannot rely on an assumption that boards of such organizations will be competent, and motivated to act in the interest of the organization as a whole, nor the wider public interest. Upper income countries with large numbers of APO-type organizations have established arms-length monitoring, oversight and regulation units for APOs as a safeguard against Board failure. In countries such as the UK, where all public hospitals are organized as an APO-type organization, the health sector has long had its own specialist monitoring and oversight functions. Monitor, the UK regulatory agency established to regulate National Health Service (NHS) Foundation Trusts is one relevant example to consider. The Ban Phaeo model has similar autonomy in many respects to NHS Foundation Trusts (Ban Phaeo has more diverse revenue sources, greater human resource autonomy and less autonomy over land and buildings than Foundation Trusts). Systems for Managing Capital Finance If Thailand over time were to adopt an APO model for most MOPH hospitals, with tens to hundreds of health sector APOs, it would be necessary also to establish more systematic policies and dedicated capacity to review capital investment and borrowing proposals. Private capital markets will view loans to APOs as implicitly government 29 guaranteed, even if there is no explicit guarantee. APO borrowing will be viewed as a component of government debt, under broad definitions. Accordingly, it will be necessary to carry out the same kind of economic and financial appraisal of major investment proposals as are appropriate for conventional public sector investment. To provide some increase in capital autonomy within prudent limits, consistent with fiscal policy, the UK gives NHS Foundation Trusts freedom to borrow up to a modest limit, defined and supervised by Monitor. Career Paths and Human Resource Mobility As with decentralized systems, health systems with predominantly or entirely autonomous public providers usually develop human resources policies that facilitate movement of staff between autonomous hospitals. More “liberal� regimes (e.g. Estonia, New Zealand 1993-2000) for achieving this usually involve:  portable pension rights,  open advertisement of all posts in certain occupational and grade categories in a common health sector journal/website;  a common occupational classification system, and a broad-banded common grade structure;  co-ordinated policies for posting and rotation of junior doctors and doctors in specialization training. More “regulated� regimes (e.g. UK NHS, Australian states) may also have a nationally negotiated scale of pay and allowances for the main occupational groups, with individual hospitals having freedom over hiring, placement of new hires in the scale, promotion, and discipline. These regimes may give hospitals some freedom to pay bonuses and pilot various reforms (such as P4P). Networks In scaling up APO policy, consideration would need to be given to autonomizing networks, rather than individual facilities. Within the public sector, networks offer the advantage that they de-concentrate decision-making about how to adjust the organization and configuration of health service delivery and internalize this task within a single organization (assuming the necessary managerial competence, authority, motivation and incentives exists for making optimal changes). This gives the network more freedom to respond to some of the trends and drivers of change in health systems – such as the increasing concentration and specialization of hospital services combined with the shift out of hospital to community settings of a larger range of curative care services. (Richard Scott, 1994) Some countries have adopted a geographic catchment area approach to establishment of autonomous networks. Others (notably the State of Victoria, in Australia, in the 1990’s) attempted to design networks in such a way as to permit inter-network competition. Some interviewees advocate that it would be more efficient to give autonomy to district or provincial networks, rather than individual hospitals. There is not a strong evidence base for the commonly expressed preference for bringing the network of facilities into a single legal entity. The presumption that coordination and communication between organizations is inferior to coordination and communication within organizations may not be true, in case of large, complex multi-site organizations with delegated/de-concentrated management. There is some evidence of diseconomies of scale in very large hospitals, and some studies postulate that these may arise from the additional costs of coordination and communication in 30 large campuses or multi-site facilities. There is some theory and evidence, however that networks of organizations with differentiated and heterogeneous nodes (a characteristic of the Thai public sector health facilities network) require local, de- concentrated and personal coordination (as distinct from formal coordination based on rules, processes, and impersonal information exchange). (Alter and Hage 1993; Wadman et al. 2009) The CUP hospital and CUP Board already provides this type of coordination, to varying degrees, at district level – even though the management hierarchy for HCs is to the DHO, rather than the CUP hospital. The PCMO and PHO already plays this role at provincial level, and their coordination role already encompasses private and local government health providers that participate in UC in a number of cases. This provides a natural opportunity for research to assess whether network coordination functions are more or less effective across organizational boundaries or within the MOPH in the Thailand context. Some see stronger potential for improving HC performance, developing primary care and de-congesting the hospital outpatient departments if HCs are under the managerial control of the CUP hospital. For example, with APO status, Ban Phaeo would be very strongly placed to improve staffing and motivation of staff in HCs, after a transition period. There is little evidence on the effects of integrating primary care under the management of autonomous hospitals. Most countries that have autonomized their hospitals have either private provision of primary care with public finance, or decentralized provision of primary care at a lower level of government from the hospitals. (One exception is Sweden, where public primary care centers are managed by the same level of government as most public hospitals. See Wadman et al. (2009) for a discussion of how coordination between primary and secondary care functions in the Swedish system.) Additionally, strong advocacy movements for primary health care or family medicine in a number of countries have opposed proposals for any form of merger, out of concern that primary care will have less power over negotiation for the hospital’s discretionary resources (such as capital investment) and receive lower priority for management attention than higher profile specialist services. Some countries (e.g. parts of Australia, New Zealand, parts of Canada) have broadened the role of their public hospitals to become “area health boards� with responsibility for ensuring provision of primary care and P&P in a defined geographic catchment. In areas where (dominant) private sector provision of primary care is absent (e.g. in sparsely populated rural areas, some deprived urban locations), these boards sometimes provide primary care and a range of community P&P services directly, though this has become less common. Increasingly, boards have contracted NGO providers or used incentives to attract or partner with private providers. Community Participation Finally, there is potential to develop the role of community participation in hospital APOs, though there is limited evidence about the benefits of citizen participation in hospital governance among OECD countries. The UK and New Zealand have mechanisms for electing community representatives to the boards of Foundation Trusts and District Health Boards, respectively. These mechanisms have encountered problems with conflict of interest in the Board, and have not produced the desired shift away from centralized Ministerial accountability to local participatory accountability for local health service delivery. Under models similar to the APO model, the Board members identify closely with the hospital and their duties require 31 them to be loyal to it and defend it. The complexity of hospital management makes community Boards relatively weak in influencing the performance of more expert, more powerful managers and senior clinicians in the hospital. The case studies of successful models of community participation on hospital boards or in hospital supervision are found in countries where public sector hospital performance is very low, and community boards are able to play a role in detecting and deterring very evident problems such as absenteeism, stock outs and losses of medicines and supplies. The Thailand context is not directly comparable to either of these types of example. It may be more effective to separate the role of citizen and patient feedback from the role of the governance Board. Some countries (including high-income countries such as the Netherlands, and some developing countries where there is a much lower level of trust in public institutions) have experimented with having a second community board to supervise the hospital. Successful cases of hospital community oversight or supervision boards commonly enlist “altruistic, expert elites�, with sufficient power to challenge hospital performance, rather than ordinary patients and citizens that are not able to influence the hospital. These powerful community boards can act as a channel for patient and citizen complaints. However, rather than relying on direct community participation, many high-income country Governments establish independent expert commissions with a mandate to inspect hospitals both routinely and proactively, and in response to patient complaints, and design these institutions with substantial lay representation so as to avoid “professional capture�. In the shorter term, it would be possible to build incrementally on the Ban Phaeo model of community participation in governance, and evaluate the effects of these changes. Community participation processes could be expanded to encompass accountability for use of public as well as donated funds. This might entail community consultation on service strategy, more in-depth patient and community surveys and focus group feedback on service delivery, and community participation in reviews of annual financial and service performance. Future Reform of MOPH Hospitals: Issues and Questions for Further Exploration Although the Ban Phaeo model has shown success, there is not yet a clear consensus that this model of autonomy is appropriate for most MOPH hospitals in Thailand. In 2005, all MOPH hospitals were granted increased autonomy in a number of dimensions. They now enjoy freedom to retain revenues from UC, CSMBS, SSO and user fees (from out-of-area patients and private patients) and have considerable freedom about how they allocate these revenues. They are free to establish private- paying beds offering a higher standard of services, and their staff can earn additional income by part-time work in these private units. They are able to raise donations from the community and have considerable freedom over how they spend donated funds. They have some capacity to earn additional revenues by, for example, entering into joint ventures with the private sector, including retail pharmacy. They are free to contract out some services. They can use their additional income to increase staff salaries through bonuses of up to 25 percent, and are free to hire contractual staff. 32 It is outside the scope of this study to assess the current organizational and governance framework for MOPH’s directly managed hospitals. However, the assessment of the APO model touched on some advantages that the APO model still has over the MOPH directly managed hospitals even after the increase in autonomy these hospitals received in 2005. It points to some major areas in which these hospitals face constraints that stand in the way of optimizing hospital performance and development:  Human resource management constraints: Hospital directors still face many constraints on their ability to hire, fire, promote and reward staff.  Constraints on major hospital investment: There is an acknowledged need to increase hospital capacity in many parts of the country, but public sector capital finance for hospital development is very constrained.  Absence of independent governance or supervision: Governance and supervision for MOPH hospitals is internal to the Ministry and lines of accountability for provincial and regional hospitals are centralized. In the absence of a consensus about options for reform of public hospital governance, such as the APO model, decentralization or community participation, it may be useful to study these and other constraints on MOPH hospital performance more systematically. It could be useful, for example, to explore whether the growth in the high share of MOPH doctors who rely on part-time private earnings to supplement their income is continuing, and whether this is having adverse effects on staff productivity and service quality in public hospitals. It could be useful to study the effects of the 2005 freedoms introduced into MOPH hospitals on their performance in delivering both publicly financed and privately financed services. It could also be useful to review existing mechanisms for supervision and oversight and assess whether there is demand from citizens, patients and professions for greater independence and transparency in hospital supervision and oversight. Exploration of specific, tailored institutional reform options for addressing the identified constraints – in human resource management, capital finance, and incentives for performance - could be based upon these types of studies. The international evidence base for adopting one form of hospital governance over another is difficult to interpret and apply to a specific country context, and so does not generate unequivocal recommendations for the future reform of Thailand’s public hospitals. The nature of both the positive drivers of hospital performance and the dysfunctionalities of public hospitals in any specific country context interact in complex ways with the design of the health system, the system of public administration and finance, other institutions (such as the power and ethos of professional associations and trade unions for different cadre’s of health workers), the relationship between the public and private health sectors, and the relationship between patients, citizens and communities and the hospital. Since the wave of piloting and scaling up of various models of hospital autonomy in the 1990’s in many OECD countries, there has been a second wave of reform of hospital governance in the last five to ten years. This new wave of reform is quite diverse in the governance models different countries are adopting. Each country is forging solutions based on the specific nature of their problem diagnosis for their country’s public hospital system – there is no “blue print�. So for example, the UK’s Foundation Trust reforms in the past five years have increased autonomy more 33 decisively - cutting more of the ties with the UK’s Department of Health and Treasury that progressively reversed the changes brought about by the first wave of autonomization in the early 1990’s. The UK has also introduced community and staff participation in hospital boards, to respond to public and political perceptions of an over-centralized and unresponsive NHS bureaucracy. By contrast, the French Hospital 2007 Plan appears to strengthen the role of doctors in hospital governance and management, and reduce the role traditionally played by trade unions representing other staff groups. There is no focus on community representation or participation in the French Plan, and a centrally determined model for organizational reform of all public hospitals has been adopted. (See Eeckloo et al. 2007) However, there are some common themes in the more recent wave of reforms, which may be relevant in further development of hospital policy in Thailand. Clinical and safety-related dimensions of governance receive greater emphasis than in the reforms of the 1990’s. Methods and institutional arrangements for performance assessment and monitoring are much more elaborated. As a result, accountability and oversight of public hospitals has become increasingly multi-faceted. 34 REFERENCES Alter, C. and J. Hage, Organizations Working Together, Newbury Park, SAGE (1993) Ban Phaeo Hospital (Autonomous Public Organization), Annual Report 2007 Chulalongkorn University Faculty of Commerce and Accountancy, Ban Phaeo Management Development for a Sustainable Hospital, Health Services Research Institute, Bangkok (2003) Deerek Phatamasiriwat, et al., A joint learning program together with Tambon Administrative Organizations to enhance good governance and public participation: phase II, Health System Research Institute, Bangkok (2005) Eeckloo, Kristof and Luc Delesie, “Where is the Pilot?: The Changing Shapes of Governance in the European Hospital Sector� in Journal of the Royal Society for the Promotion of Health, March 2007, Vol. 127 (2) Jiruth Sriratanaban, Sukalaya Khongsawad, Porranee Laoitti: Measuring the Impact of the Autonomization of a public hospital on health Services; the case of Ban Phaeo hospital (Section 3), Health System Research Institute, Bangkok (2005) Khunpa Wachanasara, The study of civil society and its impact: Ban Phaeo district, Samutsakhon province, Health Services Research Institute, Bangkok (2005) Kittiphan Saentaveesuk, Danai Phahuyut, Daraporn Detphonmat, Patterns of the Relationship between the Provincial Administration and the Local Administration with Good Governance, Health Services Research Institute, Bangkok (2005) Dr. Komatra Chuengsatiansup, Deliberative Action: Civil Society and Health Systems Reform in Thailand, National Health Commission Office (2005) Ministry of Public Health, Ten Important issues about the devolution of Health Centers (HC) to local government in Thailand and Guidelines for health decentralization, Ministry of Public Health, Thailand (2007) Office of the National Decentralization Commission, Evaluation of Devolution of Health Centers, Office of the National Decentralization Commission (2009) Preker, Dr. Alexander S. and April Harding (Eds) Innovations in Health Service Delivery, The World Bank, Washington DC (2003) Scott, Dr. Richard W, “The Organization of Medical Care Services: Toward an Integrated Theoretical Model� in Medical Care Review 50th Anniversary Edition (1994) Surasak Taneephanitsakul, Yupha Aontuam, Wiphan Phachuapmuam, Vinus Udomphasertphan, A study of a model and evaluation of an autonomous district hospital on administration and management with in the hospital : a case study of Ban Phaeo hospital, Health System Research Institute, Bangkok (2003) 35 Wadman, Sarah, Martin Strandberg-Larsen, and Karsten Vrangbaek, “Coordination Between Primary and Secondary Healthcare in Denmark and Sweden�, in International Journal of Integrated Care, 12 March 2009 Dr. Suwit Wibulpolprasert (Ed.), Thailand Health Profile 2005-2007, Ministry of Public Health, Bangkok (2008) Dr Tipicha Posayanonda (Ed), The Movement of Thailand’s Health System Reform: A Paradigm Shift, National Health Commission Office, Bangkok (2008) Westcott, Clayton, World Bank Support for Public Financial Management: Conceptual Roots and Evidence of Impact; Independent Evaluations Group Working Paper; World Bank, Washington DC (2008) Dr Wiput Phoolcharoen, Quantum Leap: The Reform of Thailand’s Health System, Health Systems Research Institute, Bangkok (2004) Preeda Taerarak, Nipapan Suksiri, Ramphai Kaewwichian, and Kirana Taerarak, Review of Decentralization in Health during 1999-2007, Health Systems Research Institute, Bangkok (2008) World Bank, Thailand: Country Development Partnership Governance and Public Sector Reform: Progress Assessment and Implementation Completion Report, World Bank, Washington DC (2006) Jadet Thammathataree et al., Autonomization Process in Ban Phaeo Hospital, Health Systems Research Institute, Bangkok (2001) Somsak Chunharus and Supatra Srivanichakorn, Human Resources for Health in Thailand: Technical Report, Health Systems Research Institute, Bangkok (1999) Supasit Pannarunothai, Macro-economic Indices for Measuring Equity in Health Finance and Delivery 1986-1989, Health Systems Research Institute, Bangkok (2001) Siriwut Tiptaradol, Sub-district Administrative Organizations’ Role Toward Public Health and Environmental Development in Nongkhai Province (in 1996/97). Health Systems Research Institute, Bangkok (1997) Kowit Pueongngam, The Tasks of Tambon Administration Organizations (TAOs) and the Strength of Tambon Civil Society (Sub-district Community) in Dealing with Tambon Health Problems (in 1998-1999), Health Systems Research Institute, Bangkok (2001 36 ANNEX 1 ANNEX 1: INTERVIEWS WITH POLICY RESEARCHERS AND CENTRAL GOVERNMENT MINISTRIES B. Background The Plans and Process for Decentralization to Local Administrative Organizations Act of 1999, following the Thailand Constitution of 1997, called for all ministries including the Ministry of Public Health (MOPH) to develop action plans for decentralization of functions, resources and staff to LAOs by 2010. The Act also set a target for increasing the share of the central Government budget that should be transferred to LAOs from 9% to 35% by 2006. Following the coup and change of government in 2006, some pertinent provisions of the Decentralization Act of 1999 were amended in 2006, specifically:  The amended law revises the provisions regarding the share of government expenditure that should be transferred to LAOs. The current provisions are for a minimum of 25 percent to be transferred, with a target of 35 percent.  Further revenue transfer is to be based on functions transferred. C. Summary of Interviews with Health Policy Makers and Researchers - The early phase of implementation of decentralization Following adoption of the Plans and Protocols for Decentralization Act of 1999, there was discussion of options for health sector decentralization that were consistent with parallel development of health reform policy. Discussion and negotiation took place between the (MOPH) and National Decentralization Committee (NDC) over these options, and was reflected in the first Action Plan for Decentralization of 2001. Devolution of some promotion and prevention functions to LAOs was agreed. But the MOPH argued, successfully, that it would be undesirable to split curative health services across different levels of government. This policy debate drew upon negative perceptions of the experience of “multi-level� models of decentralization in the Philippines and Indonesia. During this period, a devolution model based on “area health boards� was developed that would oversee the whole network of MOPH health facilities and coordinate all health services. The whole network could then be devolved. There was debate among health reform circles about the appropriate scale and level for devolved networks. The model was not without controversy within the MOPH. Some advocated a regional health board model – comprising all of the health facilities that fall within the catchment area of a regional tertiary hospital, encompassing 5-6 provinces and around 5 million population – on the grounds that this structure minimizes inter-board patient flows and internalizes the full referral chain within a single organization. Some advocate provincial level networks – since province-based networks would best map to the responsibilities of directly elected PAOs. Others advocate a district model of area health boards that would be closer to the communities served by health facilities and in rural areas would have a strong focus on primary care. The district health boards would manage the health budget for 37 populations of around half a million or less – the catchment area of a provincial hospital CUP – and build in participation. But there is no LAO at regional or district level so the design of either regional or district area health boards would have required new legislation to create a new legal structure. Additionally, the district model would have required design of linkages to higher levels of the referral system. Such boards could include representatives of groups of LAOs or more direct forms of community representation. Some work was done in the past on drafting such a law. The full details of this model were not fully defined – for example it was not clear whether the area health board would develop as a funder/purchaser – with increasing autonomy of hospitals or provider networks - or as an integrated provider network. Pilot development of provincial health boards took place in 2002 in 10 provinces. These boards were not legal entities. The pilot model introduced a multi-stakeholder committee within a de-concentrated health administration, but did not involve transfer of staff and facilities to Provincial Administrative Authorities (PAOs). Momentum in implementing decentralization stopped following the change of Government in 2001 and the adoption of the National Health Security Act 2002, which introduced the Universal Coverage (UC) scheme. The establishment of the National Health Security Office (NHSO) and the transfer of much of the MOPH health services budget to the NHSO became the Government’s priority. It also ushered in a partial form of purchaser-provider split, which changed the health budget allocation system in ways that are unique to the sector. This would have necessitated re-design of the area health board model, but in fact, there was no subsequent effort to develop this model. Additionally, the priority given to the very major task of introducing the UC scheme reduced the MOPH’s focus on decentralization. - The recent phase of implementation of decentralization Renewed impetus to implement the 1999 Decentralization Act began following the coup of 2006, under the new Constitution and amended Decentralization Law that followed. In the MOPH’s input to the second Action Plan for decentralization that was prepared in 2006-07, the MOPH came up with the idea of transferring individual HCs to TAOs and municipalities, as a means of at least starting a process of devolution. This might be described as a “minimalist� model for voluntary devolution of individual health centers to Tambon Administrative Organizations (TAOs) and municipalities. Under very detailed guidelines developed by the MOPH, devolution only occurs where the following criteria are met:  TAO/municipality “readiness�, based on good governance awards, and LAO commitment to health in the form of establishment of a Public Health Section and contribution to a Community Health Fund (co- financed by NHSO);  HC willingness: at least 50 percent of HC staff willing to transfer, including the HC head;  Community support of at least 50 percent. The cautiousness of this approach was in part a response to the Education Sector’s experience with attempting to push for mandatory devolution of schools, which encountered strong protest action by teachers. It is also a response to perceptions of weak LAO capacity and poor governance in many LAOs. 38 Many of the health policy experts and MOPH personnel interviewed continue to oppose multi-level models for decentralization that would fragment primary, secondary and tertiary health care across different levels of government. This viewpoint appears to be a reason many people cite as a concern about the current model of devolution of individual health centers. However, others argue that the funding mechanism through the CUP Board can continue to provide the basis for integration and coordination of both devolved and non-devolved health centers, PCUs and CMUs – noting that there are already well established precedents for CUP Boards to include primary care and outpatient facilities owned by LAOs and private providers. There are also precedents for municipal hospitals to manage the CUP (as in Nakhon Si Thammarat municipality). One of the policy makers interviewed advocates a provincial area health board model for decentralization, on two grounds: this keeps the primary and secondary levels of the health facilities network together, and also maps onto the political/administrative structure of the PAO and provincial administration. The reasons why the MOPH did not return to the area health board or network model of decentralization are not entirely clear, but interviewees cited the following factors:  PAOs or provincial or regional area health boards would be too large and too remote from communities to provide an effective level for effective “grass roots� participation; there seems to be more optimism about development of community participation and mobilization at TAO level;  While devolution of district networks of health facilities might offer theoretical advantages, there is no directly elected LAO at district level; voluntary collaboration among TAOs is difficult; and it is no longer seen as feasible to adopt new law necessary to underpin creation of new district boards;  Hospital autonomy with community participation (along the lines of the Ban Phaeo APO) is seen as preferable to transfer of hospitals or health facilities networks to PAOs or provincial boards, because PAOs lack health sector capacity and perceptions of PAO governance are negative;  The funding/purchasing mechanism of the district Contracting Unit for Primary Care (CUP) can be used as the integration mechanism for devolved HCs;  Given the widespread caution if not opposition to decentralization among many in the senior levels of the MOPH bureaucracy, voluntary devolution of HCs represents the “least harmful� mode of complying with the decentralization provisions of the Constitution and decentralization laws. The most recent Action Plan for implementation of Decentralization calls for HCs to be transferred to TAOs and municipalities. As a “default option� the Plan calls for any HCs not transferred by end of 2010 to be transferred to PAOs as a transition measure, until TAOs are ready. All health sector interviewees (and a number of non- health sector interviewees) do not regard this “default option� as credible or sensible. This would require two transition processes, rather than one. PAOs do not have health sector knowledge or capacity for managing HCs or managing their transfer to TAOs and municipalities, nor are they close to the individual communities served by HCs. Additionally, this default option could risk disruption to the vital linkages between HCs and CUP boards and CUP hospitals. 39 Few of the health sector informants interviewed expressed opposition to any form of decentralization – the selection of interviewees was not random or representative. But the fact that there are both variations in the level of understanding of decentralization and substantial opposition among MOPH staff is evident in the December 2007 MOPH document: Ten Important Issues on the Devolution of Rural Health Centers to Local Government in Thailand and the Guidelines for Health Decentralization. - Summary of Comments in Stakeholder Workshop on April 7, 2009 Thailand already has some of the laws and institutions that can be used to mitigate risks of further devolution and autonomy of health care providers. The Health Act could easily be modified to allow Certificate-of-Need regulation if this becomes needed. This and other legislation provide a basis for mandating public health programs and management of emergencies. Additionally, the health professional associations and professional ethics already are able to play a role in promoting compliance with appropriate clinical standards and protocols. The DHO Association chair, however, questioned whether use of regulation would be effective as a means of dealing with information requirements and emergency response, perceiving that law enforcement is generally weak. The Thai health system is already quite pluralistic and has long had LAO hospitals and PCUs operating in some areas as part of the health system (though the largest example, the BMA, is not regarded as a good example), and has already decentralized a range of public and environmental health functions. Career path concerns and job mobility to transfer from LAO to LAO (as well as from LAO to central government) is a major and unresolved issue for LAO staff – not only in the devolved HCs, but more generally. Provincial administration committees deal with a high volume of personnel issues including transfer requests. These are difficult to handle in the absence of a standardized or harmonized human resource management system across LAOs, given the wide range of LAO sizes and responsibilities. This issue seems to call for a systemic policy response from the NDC, MOI and other relevant authorities. Concerns were raised about the risk of changes in policy and strategy with changes in LAO leadership. The TAO response to this issue is that it is possible for TAO policies and regulations to be entrenched by adopting good policy and regulatory procedures – for public consultation, TAO CEO and Council decision, DHO and/or governor co-signing, and publication. Future TAOs would have to repeat the full process before changing policy. Concerns were raised about accountability, and whether devolution is diluting or fragmenting accountability. Currently, the laws permit LAOs to carry out health responsibilities but do no give them clear duties and responsibilities, nor protect them from future addition of unfunded mandates. (Though the MOU signed when HCs are devolved commits the TAO to manage the health center according to current and future regulations, criteria, standards, and public health work methodologies set by the MOPH and PHO.) Several factors were mentioned that underpin the concerns about accountability: the HCs already draw revenue from multiple sources and already have more than one line of accountability (to the MOPH via the DHO, to the CUP Board, 40 to CSMBS and SSS). Adding the TAO into this group adds to the existing complexity of accountability, and may add to the existing anomalies in Thailand’s purchaser-provider split (the separation and protection of the public sector salaries component of the budget for health services). Additionally, the split of primary care and P&P responsibilities between the HC and the rest of the network covered by the CUP is negotiated by each CUP board and varies from location to location. As a result, it is not clear to whom the HC is accountable for overall performance in relation to variables such as efficiency or patient satisfaction, nor is it clear who has overall responsibility for the health of the enrolled population and for provision of primary health services. Exactly what the TAO is accountable for in relation to primary care provision and the devolved HCs is somewhat ambiguous and will vary from place to place depending on the local CUP board. TAO and devolved HC representatives in the workshop believe there are gains in local accountability for HCs and for health of the community following devolution. There is closer supervision and direct accountability of HC staff to the LAO that employs them and answers to the local community. TAO representatives see problems with network models for devolution – the immediacy of supervision and clear accountability of HC staff to the LAO that employs them would be lost in a network model with only indirect and shared accountability. Additionally, some of the TAOs with devolved HCs have instituted systematic health status, risk factors and health needs assessment through census or survey of the population, and plan to use this data to set realistic local health plans and targets, and to monitor progress against these. D. Summary of Interviews with Other Agencies Engaged with Decentralization Policy (NDC, MOI, MOF, Bureau of Budget) - Views on Experience with Health Sector Decentralization to Date Some of the agencies tasked with supporting implementation of decentralization that in recent years perceive that the MOPH has not given priority to providing health sector input into the development of policies and plans through the NDC and the Health Sub-Committee. Some non-health sector interviews also take the view that the criteria in the MOPH’s guidelines for devolution and the interpretation of the guidelines are unduly restrictive, and recommend that LAOs should be accepted so long as they have a public health section and adequate budget allocation for health. The ONDC has been monitoring HC devolution pilot sites and has produced an evaluation. The recommendations are attached as an Addendum to Annex 1. Many of its findings relate to the multiple factors underlying the reluctance of HC staff to transfer to LAOs, which has been the main impediment to voluntary devolution. The evaluators conclude that clearer policy intentions on the part of MOPH may address the hesitancy of many staff who prefer to “wait and see� what happens with the first pilots, before making a decision. Additionally, the report recommends that there should be a coordinated process of transfer of larger numbers of HCs in a way that allows exchange of HC staff who are willing to work for LAOs with those who are not. The report recommends harmonizing staff compensation and benefits provisions between the LAO/MOI and MOPH regulation to whichever provisions are more generous or beneficial to staff. It also recommends that the Bureau of Budget should allocate funds for LAO HC staff training, because some PHOs have reduced their 41 support for training of devolved HC staff. The evaluation does not attempt to judge whether these increases in the costs of HC services are likely to be offset by commensurate increases in HC output or outcomes. The report calls for further joint work by the MOPH, ONDC, MOI and Office of the Civil Service Commission (OCSC) on some human resources policies that are important for ensuring appropriate technical meritocratic criteria in hiring and promotion, and for fostering career path development for staff. It also calls for clarification of the future role of DHOs in supervising devolved HCs. In practice the pace of transfer of functions, assets and staff in other sectors has also been very gradual and limited in extent. The OCSC initially made it mandatory for staff to transfer when functions transferred, but in response to protest action by civil servants later changed its stance so that staff transfer is now based on “willingness�. Only around 4,000 central government staff across all sectors have transferred to LAOs, and most of these are from the Rural Assessment Department which was destined to be abolished at central Government level. In other words, only those staff that faced an uncertain future in the central government have been willing to transfer in significant numbers. Pensions and benefits differences on transfer to LAOs have been a factor in the unwillingness of staff to transfer. But the Central Provident Fund rejected a request from the NDC to take in LAO officers. Although a number of central government staff (especially at district level) have voluntarily resigned to take up jobs in LAOs outside of the devolution/ transfer process, mobility of LAO staff back in to central government career paths is very difficult. In the education sector, there is now a large enough pool of devolved teachers to permit some transfer between LAOs. But it is recognized that the health sector has distinctive and complex career paths and personnel systems, and that there is a need for highly technical advice and supervision/regulation of the work of devolved professionals, to a greater extent than in other sectors. Despite extensive discussion between the MOPH and the NDC and ONDC about human resource development and career path mobility, no concrete policy and administrative mechanisms for addressing these issues have been agreed. Interviewees from outside the health sector expressed more confidence than health sector interviewees in the mechanisms that have been put in place to promote merit- based hiring and promotion by LAOs and protect LAO staff from politicization. National, central (DLA) and provincial committees oversee policies, regulations and actual decisions on personnel management, appointment and promotion. Three levels of The Merit Protection Committee is supposed to play this role. The Permanent Secretary position in the TAO is a permanent post. There has been a lot of training of local political leaders. Some of the relevant provisions in the decentralization plan have not been implemented yet. The Plan calls for province-level education boards and health boards that will monitor devolved services and have authority to oversee technical matters, including technical and professional aspects of personnel policies and processes in their respective sectors. Only a few provinces have established education boards; none have health boards. 42 - Fiscal Decentralization, Fiscal Transfers and Health At this stage, the Government is already transferring around 25% of its budget to LAOs, and any further transfers are likely to be limited to transfers of the central budget allocations for specific health functions, staff and assets that are transferred. The NDC sees the NHSO as central funder/purchaser of health services, and sees its funding pool as outside the scope of fiscal decentralization. It is only the MOPH budget allocation to HCs that is decentralized when HCs are transferred to LAOs. This budget predominantly finances core salaries of government officers and little more than the salaries budget is devolved when HCs transfer to LAOs. 4 PHOs also hold small budgets of MOPH funds for non-salary recurrent costs such as training. Supervision of HCs continue to be managed by the PHO and are not being devolved. The MOPH budget occasionally finances some capital investment. But in the case of the devolved HCs, the MOI has budgeted for substantial capital investment per HC – so there has been no need for MOPH’s more limited capital investment budget to be transferred. The amount of fiscal decentralization from the MOPH budget that accompanies the transfer of HCs cover only part of the costs of the transferred functions – it does not even cover the full staff compensation costs for transferred staff. The share of HC costs covered by these transfers varies from HC to HC. The reason for this is that the MOPH and NHSO budgets are interdependent, and the processes for determining the allocation of these two funding sources to HCs are highly de-concentrated. PHOs, provincial health security budget committees, and CUP boards have considerable flexibility. Part of the budget for compensation for HC staff is derived from NHSO funds which remains centralized after devolution. HCs and other health facilities use NHSO and other HC revenue sources (such as user fees, CSMBS and SSS payments) to pay for hiring contractual staff, OT, hazard allowances, some training allowances – though the extent of this varies by province and by CUP board. The ONDC does not envisage that NHSO’s budget for curative care – currently managed through the CUP – should be transferred to LAOs, and are content for this aspect of the CUP to continue to operate as before, and for the CUP hospital to continue to provide in-kind or cash support to HCs. The ONDC has the impression that most of the work of HCs is P&P, though this team found that 50-80% of HC utilization is for curative care, and that most TAOs aspire to increase the role of the HC and TAO in curative care. Based on this perception, the ONDC advocates for the P&P part of the NHSO budget to be allocated directly to LAOs (or to the devolved 4 Sixty five percent of the MOPH’s salaries budget for staff in healthcare facilities is sourced from a top-slice from the allocation to the NHSO, reflecting the approximate share of healthcare users covered by the UC scheme. The non-salaries recurrent budget and the budget for minor capital expenditures for MOPH health facilities now comes from the NHSO, plus payments from the other social health insurance schemes, some user fees, donations and LAO complementary funding. This creates some anomalies (for example, there is no top-slice of funds from CSMBS and SSS to pay for their “share� of the salaries budget). It also has the result that there is not a clear funder/purchaser provider split in Thailand’s public health system. In relation to devolution of health facilities, the fact that the MOPH portion of the budget for HCs is to be devolved, while the NHSO portion of the budget is to remain central, entrenches the split of responsibility for salaries versus non-salaries recurrent budgets. It adds to the complexity and anomalies in the system, because LAOs’ own revenues can be used to complement the NHSO’s funding of non-salary recurrent costs and capital expenditure. 43 HC). The ONDC evaluation report recommends that NHSO develop a clear national guideline regarding the amount that should be transferred to LAOs for devolved HCs. However, this team found wide local variation in how CUP boards allocate funds to HCs for both curative and P&P services. This in part is because CUP boards are responding to the wide variation across HCs in the level and mix of services they provide and in the other sources of revenue HCs receive. Ironically, the ONDC recommendation would amount to centralization of decision-making over HC re- allocation, by comparison with the degree of local adaptation and responsiveness that is achieved through existing de-concentration. Changes to the NHSO’s guidelines for P&P “express demand� should see a larger share of this budget going as cash to HCs (devolved and non-devolved) but this team found that there is still wide variation as of 2009 in how different CUPs are allocating the P&P budget to HCs, with the exception of finance for community health funds which follows standardized guidelines. Non-devolved HCs currently receive their MOPH budgets via the PHO in two lines – for recurrent and investment costs – and return unspent balances to the Treasury. Both devolved and non-devolved HCs retain unspent balances of NHSO, CSMBS, SSS and user payment revenues. Devolved HCs receive their devolved budget via the MOI and TAO as a single line, and the TAO can retain unspent balances. This devolved budget is managed as an integral part of the TAO budget. LAOs are free to allocate additional amounts from their other revenues. The devolved budget allocation to pay the salaries of (named) government officers who transferred to the devolved HCs will continue to flow to them via MOI to the LAOs an earmarked, specific grant until these staff retire. But for any new staff hired by the LAO, there is no earmarked funding. These costs will be met from the general grant and other revenues the LAO receives. Uncertainty about the future financial position of the LAO and its future allocation to health is one of the concerns MOPH staff raise as a reason to oppose transfer to LAOs. However, to date fiscal decentralization has proceeded faster than functional decentralization, so many LAOs have enough budget space to increase spending on health above the levels the MOPH was able to allocate. Many LAOs have potential to increase their local tax collection, as well as benefiting from future growth of the shared taxes. Additionally, there is a redistributive component to the general grants to LAOs.5 But uncertainty exists both about whether the currently loose fiscal position of LAOs will be allowed to persist in the face of the more constrained fiscal situation facing Thailand at the national level, and about whether health expenditure will remain a high local priority after future LAO elections. As well, specific grants from central government to LAOs are subject to discretionary, politically influenced allocation, and so are perceived as being at risk for LAOs affiliated with opposition political parties. Interviews with provincial Deputy Governors, DLAs and LAO chief executives found that the local electoral salience of health and health services differs quite widely in different provinces, municipalities and tambons. 5 On average, LAOs receive about 9% of their revenue from local taxes and revenue; 50-55% from shared national taxes, and 38-40% from central government grants. About 80% of the latter are general grants, allocated on a combination of functional criteria (e.g. per capita allocations for pupils in devolved schools) and need-based criteria (e.g. population, area, income of the LAO). About half of the general grant is allocated on the basis of these needs-based criteria. 44 There have been transition problems in getting the new funds flows operating for the devolved HCs. TAOs needed to prepare budgets for the HCs as a basis for funds to flow, and many lacked capacity to do this. There is additional complexity where some HC staff declined to transfer to the LAO but continue to work in the devolved HC, and draw their salary from the MOPH budget. There has also been controversy over aligning some of the MOPH/NHSO regulations for HCs – which allowed HCs quite a high degree of financial autonomy in relation to funds received from NHSO and other sources, including ability to operate their own bank accounts and ability to retain unspent NHSO funds and user fee revenue – with MOI/TAO regulations. A number of LAOs already allocate parts of their budgets for what might be called complementary funding for health programs and projects and also allocate funds to non-devolved health facilities. This is leading to rather ad hoc variation in health budget allocation in different areas. Some PHOs now obtain more non-salaries budget from PAO allocations than they receive from the MOPH budget, but other PHOs receive none. Some PAOs and municipalities have established or acquired hospitals and PCUs or health centers. Some TAOs and municipalities provide complementary funding to MOPH’s non-devolved HCs in their territory, commonly for capital upgrading and equipment, but in some cases HCs receive recurrent budgets from LAOs that are used to pay for additional drugs and supplies and for hiring of additional contractual staff (e.g. contracted doctor or dentist to provide clinics). However, the Office of the Auditor General discourages LAO spending on MOPH health facilities care as duplicative and wasteful. There is a new regulation that will allow TAOs to support non-devolved schools. To date there is no policy on non- devolved HCs, though this team found an example of a municipality finding a way around the OAG’s objections, in order to maintain recurrent funding support for a non-devolved HC. - Views on Future Development and Implementation of Decentralization Policies At least some members of the NDC as well as senior MOPH officials interpret the amended Decentralization Act and second decentralization Action Plan as making it optional to devolve health services further. There seems to be a general consensus that transfer of MOPH hospitals to PAOs is difficult – and there is little interest from PAOs. Even the devolution of HCs is viewed as optional – in the sense that any transfer should be based on willingness of the MOPH and HC staff, as well as TAO/municipal readiness – in line with the MOPH’s guidelines. Some even take the view that devolution should be reversible – if, for example, evaluation found net negative results from transfer of HCs, they could be transferred back; and individual HCs where staff wished to transfer back could be considered case-by-case. However, there appear to be different perceptions among policy advisers in the Office of the NDC, who interpret the law and Action Plan as entailing an obligation for the MOPH to increase the pace and extent of HC devolution. The Office will request the Prime Minister as Chair of the NDC to write to the Minister of Public Health to ask for compliance with the current Decentralization Action Plan. There is recognition that the very constrained fiscal and administrative capacity of small TAOs is a barrier to devolution and to efficiency in devolved services. Only around 800 of the 6,500 Tambons have over 10,000 population; average population 45 size is around 4,000. Only the larger TAOs that have built up staff of around 100 officers have a public health division. The 40% limit on share of budget that can be spent on staff compensation prevents smaller TAOs from establishing a separate public health section. It is also recognized that voluntary associations or syndicates of LAOs are not workable. A new draft law governing LAOs will contain provisions to create a framework for dealing with this problem, including options that would allow multiple LAOs to participate jointly in common infrastructure development and service provision, outsourcing, jointly owned enterprises, among other options. However, specific sub-laws under this law would need to be enacted for specific options, such as LAO-owned APOs. 46 ADDENDUM Addendum: Policy Recommendations from ONDC Evaluation of Devolution of HCs 1. MOPH needs to set up clear goals on how many health centers will be devolved in the next phase, and turn it into operational plan for the NDC to consider. This way, LAOs can plan for the devolution. In addition, MOPH should give the opportunity for all health personnel in the country to express interests in devolution, so that there could be exchanges of those who volunteer to go to devolved health centers, and those who do not. If all health center staff (5 personnel) in a health center support devolution, MOPH should then proceed to devolution. 2. For health centers where some or all of their staff do not support devolution, NDC is to support budget to hire health personnel to work in those health centers. 3. To motivate health personnel to support devolution, Central Committee of Local government officers should exempt certain criteria and qualifications, e.g. exempt certain qualification of Head of Health Section of LAO so that Health Center Chief will be able to take that post. 4. MOPH should reconsider criteria and conditions in evaluating the readiness of LAOs. MOPH may consider only revenues and readiness of LAO personnel as criteria. 5. NHSO should set clear policy direction and guidelines on national health security system, following the devolution, including amount of NHSO budget which will be allocated to LAOs that will have devolved health centers. 6. For clarity and to build confidence among the staff of the devolved health centers, Department of Local Administration should (i) set regulations to allow health center staff to receive the same level of overtime (OT) as that provided by MOPH; (ii) accelerate issuance of MOI regulations on grants for LAOs (in the meantime MOI provide clear guidelines on LAO issuing receipts to patients who can reimburse their medical fees); (iii) solve problem of scholarship students who volunteer to go to LAO and make it clear whether they can be hired as local government officers or contractual staff. 7. Issues concerning promotion and salary increase of staff in devolved health centers should be studied by sub-committee on devolved personnel, power and duties, with Secretary-General of Office of Civil Service Commission as chairperson, and subject to further consideration by NDC. 8. Bureau of Budget should support budget on staff training both before and after devolution, so that the staff will be able to develop their skills and deliver high quality services to the public. 9. Staff in devolved health centers should be able to maintain their membership with Central Provident Fund. Sub-committee on devolved personnel, power and duties should consider this. 10. Office of Civil Service Commission and Office of Civil Service System Improvement should analyze structure and staff scale of health centers following the devolution to consider e.g. (i) how to deal with health personnel who do not want to join devolved health centers, and how to proceed forward; (ii) should DHOs continue to monitor and visit devolved HCs. 47 ANNEX 2 ANNEX 2: INTERVIEWS AND VISITS IN PROVINCES: SUMMARY OF FINDINGS REGARDING DEVOLUTION Udon Thani Summary of Key Facts and Findings re Devolved HC Naphu HC Transferred to TAO November 2007 TAO Population 12,500; 18 km from province capital ; not poor Unique features TAO has 2 HCs; 1 devolved, 1 non devolved; HCs in 2 different CUPs. UC, SSS registered at HC 5243 UC, 600 SSS, 432 CSMBS Utilization - 2007 24,628 - 2008 23,002 Outreach – 2007 40 patients per month - 2008 200 patients per month Staffing before transfer 4 GO (2 recent); 4 contractual Staffing after transfer 4 GO; 6 contractual; plan to recruit dentist, pharmacist Current service delivery About 50% curative care; 50% P&P Changes in service Increased outreach; upgraded facilities for staff office, delivery; evidence of upgraded facilities for Thai traditional medicine and responsiveness massage; upgraded parking and signage; outdoor youth exercise facility; new building for patient’s families, with beds for observation of patients and overnight stay of VHVs attending training. Community health fund used for bicycle subsidy; road safety training; infrastructure improvement to reduce water pooling to prevent dengue; market sanitation improvement. Not very clearly linked to priorities or concerns expressed of citizens. Change in revenue and Increased resources from TAO budget of 5-6 M baht per other resources from year for more contractual staff, bonuses of 5 months TAO, MOI salary, training, and new P&P initiatives; 3M baht for capital investment from MOI Change in management Faster decisions and closer communication with TAO- flexibility CEO compared to DHO; increased flexibility to hire personnel (linked to increased budget); NHSO OPD 60 baht/card now provided in cash to HC so free to procure own drugs (15% cheaper to buy from GPO directly than from UT hospital which added mark-up) Change in UC provider HC continues to receive cash from NHSO for “fixed payment costs� and maintenance and P&P express as before; 60 baht per card OP allocation for drugs and supplies now provided in cash, previously in-kind (and if HC exceeded this amount in past, had to use user fee revenue or borrow from other HCs if lacked funds). Changes in incentives Closer supervision (next to TAO office); VHVs on TAO Council; higher salaries for staff with some ability to 48 link bonus to performance; lack of incentives for efficiency and technical effectiveness; planning better matched to community health needs (e.g. cervical and mammography screening viewed as national priority but low priority for this TAO) Change in reporting and All reports now copied to TAO. PHO no longer receives information flows routine financial and personnel information from devolved HC; HC continues to report health data and NHSO reports online, linked to PHO, NHSO, TAO. Change in referral and No change; little personal contact with hospital doctors other linkages to Udon for advice or referral (apart from use of formal referral Thani hospital and other system for patients HC has seen before); coordinates HCs with Ban Luang HC as before – lend each other resources and supplies; expect large amounts to be repaid; provide back up. Change in supervision and No change except TAO now provides scholarships for training training; PHO/DHO and hospital continue to invite HC for training 3x/year; DHO continues monthly supervision visits; PHO continues annual supervision Change in participation of Increased linkage of HC to participation mechanisms of citizens and patients TAO; CEO convenes meetings with village leaders which discuss health; village elders come to CEO to raise issues; Councillors discuss and approve health projects. Continue to conduct monthly meetings of community leaders and monthly meeting of VHVs. Governance issues Limited internal checks and balances if Naphu HC head is also Acting Head of Public Health Section; non- devolved HC in the TAO perceives risk of discrimination against her in promotion, due to political allegiance of family members Equity issues Perceived inequity because non-devolved HC receives less support from TAO than devolved HC HSRI qualitative score of +++ transfer process Reasons for willingness to Close personal relationship between TAO-CEO, HC transfer Head; HC Head Acting Head of Public Health Section and hopes to be promoted to this post; bonus payment. Views of stakeholders on decentralization policy and process View of PHO/DHO PHO and DHO have maintained supervision, training and support for devolved and non-devolved HCs. View of TAO-CEO CEO gives P&P and curative health care a high profile in local politics, but plans do not appear to be linked to health needs assessment or evidence of effectiveness and efficiency; wants devolved HC to become a health promoting hospital; Views of DLA and Small TAOs in province not ready for devolution Deputy Governor because of budget limitations – though expects future TAO budget increase. Inter-TAO cooperation difficult. Attractions of devolution: faster procurement, bonuses and scholarships for staff. 49 View of Community Naphu is not in the CUP of the Community Hospital Hospital located in its District (2 CUPs cover 2 HCs this TAO). Other findings re implementation of UC and Decentralization Act as they affect PHOs, DHOs, and PAO developments PHO top slices 12.6% of NHSO allocation to the province to fund a range of “input� costs for PHO, DHO, HCs (e.g. “fixed� operating costs and maintenance, hazard pay, overtime, contractual staff) and some province-wide P&P; these funds are administered through PHO. Devolved HC continues to receive these funds as before. PAO development priorities include projects health promotion, disease prevention, mental health promotion, exercise promotion, healthy families, organic agriculture. Does not provide support directly to PHO. 50 Chiang Mai Summary of Key Facts and Findings re Devolved HC Don Kaew HC Transferred to TAO end of 2008 TAO Population 14,500 (plus 10,000 unregistered population, many students and government employees) Unique features TAO has a PCU built by TAO from community fund raising, and operated by Nakornping hospital; it is adjacent to the TAO office and over the road from the HC; patient choice of registration with HC or PCU means that PCU competes with the HC for patients; both facilities have relatively low utilization per staff member; PCU is staffed by 4 nurses, 1 assistant nurse and 1 contractual, and has visiting doctor clinic 2x per week; doctor and Thai massage account for about half of PCU utilization UC, SSS registered at HC 2286 UC, 1692 CSMBS, 1760 SSS Utilization - 2008 300 per month; no change yet Outreach – 2008 Half day per week - 2009 5 half days per week Staffing before transfer 5 GOs, 2 contractuals Staffing after transfer No change Current service delivery About 50% P&P Before devolution had doctor clinic half day per week, staffed from CUP hospital Planned changes in CUP Board wants HC to focus on P&P and PCU to service delivery, and focus mostly on curative care, so hospital has cancelled evidence of weekly doctor clinic in HC; TAO-CEO wants increased responsiveness outreach, introduction of health checks for villagers (in response to evidence of late presentation and diagnosis), health survey to be conducted as basis for realistic health plan and targets which would be monitored; longer term would like to develop community hospital in the TAO (around 50 beds, with doctors, dentist, pharmacy) Change in revenue, other CEO sees no need for additional TAO budget for HC yet resources - expects increased efficiency from existing resources, given past low productivity Change in management Closer supervision, faster decision and lower threshold flexibility for delegated authority for procurement expected – but so far, are still working out changes in regulations and funds flows Change in UC provider CUP Board has increased P&P allocation to HC since payment devolution and reduced OPD allocation – introduced policy of allocating on basis of 60 baht per actual users for drugs and supplies from hospital, rather than 60 baht per UC card holders registered with HC; patients can 51 now choose to register with PCU or HC; HC receives cash back from OPD savings but has to absorb overspends from own funds; P&P allocated on project basis Changes in incentives Increased service demanded and increased supervision by TAO; UC payment changes could motivate increased staff to seek to increase utilization; TAO bonus of up to 3 months salary will be adjusted based on HC performance review. Change in reporting and Same as UT information flows Change in linkages to No change in referral relationships, which are described hospital (Nakornping as “smooth�; but HC has little personal contact with General Hospital manages hospital; visiting doctor clinic provided by the hospital the CUP) has ceased Change in supervision and DHO has reduced supervision from 1x per month to 1x training per 2-3 months; PHO has adopted policy of only funding devolved HCs’ training if budget left over after funding non-devolved HCs and seeks increased TAO contribution; but in practice, this DHO continues to invite HC staff for training; TAO’s Public Health Section will also supervise the HC and participate jointly in some of the HC’s outreach Change in participation of TAO-CEO has 10 meetings with village heads per year, citizens and patients and holds meeting in each village once per year Other governance issues This TAO is regarded as a good practice example – receives many study tours from other countries HSRI qualitative score of Not covered – part of second phase of pilots transfer process Reasons for willingness to Bonuses and improved prospects for promotion in TAO transfer compared to MOPH (less competition for TAO Public Health Section Head than for DHO jobs) Views of stakeholders on decentralization policy and process View of PHO/DHO Reservations about HC transfer; prefer provincial network model. DHO unwilling to meet. HCs that volunteer to transfer are those with close personal relationship to the TAO-CEO – some disadvantages (e.g. risk of favoritism). Receives complaints from devolved HC staff about political pressure on procurement. Negative perceptions of HC staff – view that require close and continuous supervision or their productivity will be low and illegal clinical practice will occur. Have been disputes with TAO-CEOs over “who pays for what� – e.g. training leave and allowances. View of TAO-CEO A champion of devolution, but emphasized importance of HC readiness. E.g. thinks PHO should not have transferred Thapha HC with only one GO on staff. Views PHO and HC staff as very passive – little outreach, wait for instructions, wait for patients. View of Deputy Governor Strong advocacy for further, faster HC devolution, and 52 for upgrading devolved HCs to “hospitals� (meaning facilities with a doctor) in future. Governor’s office involved in seeking to resolve staffing problems in devolved HCs. Reservations on hospital devolution based on perceived disappointing results from Chiang Mai municipality’s hospital and PCUs. View of CUP Hospital Did not meet Nakornping Hospital (a general hospital, upgrading to regional level) except for PCU staff in Don Kaew. Other findings re implementation of Decentralization Act as they affect PHOs, DHOs, and PAO developments PHO now derives more non-salary budget funds from PAOs than from MOPH 60 million baht in 2008; 10 million baht in 2009; PHO has close relationship with governor and PAO and is very engaged in the province-wide development planning process as a result. These funds have a strong health promotion/disease prevention/quality of life focus. PAOs described as “floating like a balloon� by Deputy Governor: significant budgets but very few mandatory, tangible functions. 53 Phuket Summary of Key Facts and findings for “good practice� example of non- devolved HC working jointly with Wichit Municipality Laem Chan HC Municipality Population 40,000 plus 20,000 unregistered population Unique features Province has more unregistered (about 400,000) than registered population (about 300,000) in addition to high numbers of illegal migrants and tourists. Many seasonal workers with no address in Phuket cannot shift their UC registration. PHO/NHSO does not operate a catchment area system for HCs or hospitals (attempts led to citizen protest). UC patients have free choice to register with any of 21 HCs, or 2 hospital-run PCUs. – plus free choice of hospital. There are 2 HCs in Wichit municipality. Laem Chan was identified by the PHO as an example of best practice in collaboration between an HC head and an LAO. UC, SSS registered at HC 9,406 total users; 5,450 UC users; 2,070 SSS users; 663 CSMBS users Utilization - 2008 24,742 Outreach – 2008 3 afternoons a week Staffing 4 GOs; 4 contractuals; doctor 5 days per week paid for by municipality Current service delivery Open 7 days; on call after hours; over 80% curative; dental services; doctor clinics 5 days a week; equipped gym room for exercise classes; some lab equipment; emergency section set up for infusions, COPD management (nebulizer, oxygen). Revenue: sources of HC receives revenue and resources from 8-9 sources. In revenue and resources addition to MOPH, UC, CSMBS, SSS, out-of-area patients; the municipality spends about 10% of its budget on health and provided substantial support to HCs for over 10 years - built/acquired 2 additional buildings for the HC, and provides recurrent revenue for drugs and supplies, staff overtime. PAO occasionally provides support. Significant donations from the community; religious donations. Significant revenue from sale of medicines and supplies to out of area patients. Management flexibility HC is has substantial own revenue and has made significant surpluses in last 5 years, giving quite a lot of de facto financial flexibility, flexibility to hire contractuals, pay overtime for 7 day service; can recommend staff for training and scholarships. UC provider payment 100 baht allocation per card holder of drugs and supplies (larger allocation, longer list due to doctor clinics); savings returned to hospital; overspends charged to P&P budget; free choice of HC/PCU and hospital registration all over Phuket; P&P allocated on project basis. 54 Municipality has a community health fund. Linkages to Wachira Referral is smooth. Little regular contact, because can General Hospital (CUP get expert opinions from the doctor in the HC manager) Supervision and training Annual CUP hospital supervision; annual PHO supervision; DHO supervises 2-3 times a year Participation and Health is one of the main election platform issues for the responsiveness municipal mayor and Council; citizen’s issues of concern include: dengue, HIV/AIDS, traffic injuries, cardiovascular disease, diabetes. Municipality canvasses user satisfaction with health services: high for Laem Chan; low for Wachira General Hospital HC staff also have substantial community engagement of their own, through clubs for elderly, exercise clubs, ANC/PNC education, home visiting, raising donations…. Linkages to PAO and Municipality has 9 health staff. Its staff do school and Municipality village outreach jointly with HCs. Provides resources to HCs. Works with them in Community Health Fund. HC head coordinates with them regularly. (More frequent contact than with MOPH.) Occasional support from PAO – e.g. some medical equipment. Reasons for HC Risk of losing MOPH support if transfer. Thinks may unwillingness to transfer get more resources by being able to tap both LAO and to TAO MOPH rather than transferring. Recent change of political allegiance by PAO and municipality teams. Other findings re implementation of Decentralization Act as they affect PHOs, DHOs, and LAO developments Views of PHO PHO has good relationships with PAO and governor’s office. Most of their discretionary budget funds come from the PAO, rather than MOPH. Received 20 M baht from PAO budget in the past, 5 M baht in 2009 for non- salary operating costs. Because of this, PHO is able to pass through all of the CUP funds to the CUP Boards without any top-slicing for the province. PHO encourages HCs to collaborate with LAOs and seek to mobilize resources from them, but it finally depends on HC head’s negotiating ability. Two HCs have negotiated LAOs to finance doctor services. Many HCs not good at negotiating. Leads to inequity in level of resourcing across different HCs. Only 1 of 18 TAOs/municipalities interested in HC transfer, but this HC’s staff unwilling to transfer because wanted to progress in career within MOPH. Patong Hospital APO proposal would incorporate 1 HC into the hospital and upgrade it to a CMU. PAO bought a bankrupt private hospital and wants to operate it as a hospital but cannot afford the high up front cost to get it to license standards. Currently operating it as a home health care service for chronic disease patients – similar to services of HC. PAO not interested in decentralization of MOPH 55 hospitals. Views of Municipality Municipality wants to continue to provide high level of (Mayor, PS, Council support for health and would like to have a health member) promoting hospital. Municipal regulations prevent them transferring the buildings they built/bought to the MOPH for the HC so they operate them in a partnership. OAG recently objected to municipality spending on curative care – argues it duplicates role of MOPH. Municipality can find a way around this using their Community Health Fund. 56 Nakhon Si Thammarat Summary of Key Facts and Findings re 2 Devolved HCs in Pakpoon TAO TAO Population 40,000 Unique features TAO is donut shaped, with a municipality of 4,500 people in the center which has no health facilities, but has ambulance and one public health officer. Two HCs in TAO, both devolved. 3 CUPs – 1 is a for a military hospital which does not provide resources to HCs but acts as catchment hospital for OPD for 1 village under each of the 2 HCs. Province will pilot Minister’s policy of free patient choice of hospital. HCs receive support from Thai Health Foundation, Walailak University and MOPH for FAPS project to map, analyze and monitor community health status and health determinants with community participation. Changes in service TAO CEO identified that people want more delivery, and evidence of compassionate, considerate service, longer hours of responsiveness in both service, better transport to facilitate access, and has HCs instituted changes based on this. TAO financed renovation of HCs. Emergency van for TAO for transport to hospital; increased VHV activity; recruitment of 200 youth VHVs; increased home care and social support for elderly and disabled people (each HC staff is assigned particular people to care for); FAPS mapping of community health status and determinants of health; village competition for “leader for change� and disease control initiatives; knowledge center for VHVs and visitors from other areas for training/study; longer term would like to have doctor clinics 5 days a week but difficult to attract doctor. Change in revenue, other TAO budget provided about 400,000 baht to each HC resources in both HCs for upgrading (building more private, accessible treatment rooms downstairs) and TAO budget is financing increased numbers of contractuals, emergency van, and community initiatives, including 8M baht in health and social services for elderly, 380 baht per month to 423 VHVs for visiting elderly. 13% of TAO budget spent on health since devolution (10 M baht, compared to 2 M baht before transfer of HCs). Change in management Quicker decisions. TAO can set own rules. Has given flexibility in both HCs HC heads delegated authority on signing up to 50,000 baht. Increased supervision and audit reduces the wrong sort of flexibility for poor performance. TAO able to cope with delays in changing regulations by finding ways around them – alternative ways of channeling resources. Changes in incentives TAO-CEO uses citizen and VHV feedback on staff’s services. Led to transfer back of to MOPH of 1 HC staff who was perceived to have poor service-orientation to 57 patients; closer supervision; TAO bonus for staff of up to 3 months salary; NHSO performance bonus unchanged; Civil Service Commission bonus will cease in 2010 for devolved HCs. Change in reporting and No change, except that TAO receives all reports, and information flows personnel reports no longer come to DHO/PHO, but HCs still provide an annual report on HC resources which covers staffing, facilities, supplies, vehicles, etc DHO no longer has authority to demand delayed reports – but reporting delays already an issue before devolution. Change in supervision and No change. Twice a year joint DHO/CUP hospital team training supervision and performance assessment for NHSO HC staff bonuses. Twice a year PHO visits. Twice a year PHO CFO does internal audit of HCs, report checked by Governor’s Bureau of Internal Control. DHO includes devolved HC staff in training and planning, and they want to participate. Change in participation of FAPS: volunteers trained to survey all households and citizens and patients enter health status and risk factor data in GIS and analytical software, as basis for identifying local health needs. TAO provides funds for projects initiated by the people – village competition. TAO getting HCs to involve VHVs and people more in their P&P work. TAO-CEO using citizen feedback on service delivery and using this to discipline staff if necessary. Other governance issues One HC head is acting as TAO Public Health Section Head, and HC staff cover the Public Health Section’s health and environment responsibilities, with assistance from VHVs. Checks and balances are achieved by TAO internal audit function to check services and evaluate user satisfaction, plus citizen feedback. FAPS will provide basis for systematic planning, monitoring and accountability in future. Pakpoon HC Transferred to TAO November 2007 UC, SSS registered at HC 13,000 UC; SSS &CSMBS 9,000 Utilization - 2007 Approx. 6,000 - 2008 Approx. 7,500 Outreach 2 half days per week – no change since devolution Staffing before transfer 3 GOs, 3 contractuals Staffing after transfer 3 GOs, 8 contractuals Current service delivery 80% curative; 20% P&P Change in UC provider No change. Maharat CUP provides 15,000 baht per payment month for fixed costs, capital expenditure of 20,000 baht per year (varies each year), 30 baht per card holder for P&P Express Demand and various other categories of P&P totaling 100,000 baht per year, and unlimited drugs, supplies and small tools/equipment. The CUP also receives the SSS registration payments and passes on 15% to the HCs. 58 Change in linkages to No change. Referral is smooth. Maharat sends a nurse CUP hospital (Maharat practitioner and sometimes a doctor for 2 half day clinics Regional Hospital) per week; and a visiting dentist with mobile equipment once a month. Maharat also provides back up staff if they have a staff shortage (e.g. while staff on extended leave or training). Maharat pays for these staff. Patients can self refer to Maharat without HC referral except for high cost cases. HSRI qualitative score of +++ transfer process Reasons for willingness to Close relationship with TAO-CEO and VHVs on transfer Council; all staff settled in this community and don’t want to move; promotion prospects better (to Public Health Section head in TAO) compared to fighting with more senior people for promotion within MOPH Salabangpu UC, SSS registered at HC 17,000 UC; 4,000 SSS and CSMBS Utilization - 2007 8,500 (1200 a month including out of area and - 2008 municipality patients) 8,500 (but transitional dip in output due to staff losses) Outreach Two half days a week; some reduction in HC staff outreach since devolution because of staff losses, and more work done by organizing VHVs Staffing before transfer 5 GOs, 2 contractuals Staffing after transfer 2 GOs, 4 contractuals (Plus HC head is also acting Public Health Section Head and TAO expects HC staff to cover environment responsibilities, with volunteer support, in addition to health.) Current service delivery 75% curative; 25% P&P. Nurse practitioners from Walalai University provide weekly clinic for chronic disease patients (DM, COPD, hypertension) Change in UC provider No change. Tha Sala CUP provides more in cash and payment less in kind than Maharat CUP. P&P 35 baht per card per year for fixed costs and P&P express. Unlimited drugs and consumables but not equipment. Change in linkages to No change. Tha Sala hospital provides visiting dentist CUP hospital (Tha Sala once a week. Director helping HC by providing a nurse Community Hospital) and practitioner to help offset the loss of staff. HC will pay other hospitals 600 baht per day over time for the nurse HSRI qualitative score of ++ transfer process Reasons for willingness/ Two GOs transferred. Already worked closely with unwillingness to transfer TAO before devolution – good relationship; TAO-CEO committed to health. Now have greater TAO ownership and can “do anything to serve the people� through it. Two GOs did not want to transfer for, for reasons not closely related to devolution (both unhappy in their job already – e.g. one was from another area and different culture). A third GO transferred back to MOPH soon after devolution at the initiative of the TAO-CEO 59 because of complaints from patients about uncompassionate treatment, and unwillingness to of staff member to change working practices. HC is now short of staff. Views of stakeholders on decentralization policy and process View of PHO/DHO Has adopted the approach that the PHO and CUP Boards should continue to act as the coordinating, integrating and supervising mechanism for devolved HCs. About 50% of HC staff are willing to transfer to LAOs, but for others lack of job mobility and higher level career path options is a disincentive. This is a wider problem for all LAO staff – Governor’s Administration Committee has to deal with many personnel transfer cases and personnel issues – it needs a systemic solution. Small TAOs don’t have enough budget to meet minimum capacity requirements – mergers difficult but needed longer term. It may take a 10 years transition to complete decentralization. Sees more problems with idea of decentralizing hospitals – breaking the community – general – regional referral and integration linkages more difficult than breaking the HC-hospital link, because in the latter case, the CUP Board and purchasing/funding mechanism can provide the integration. However, there is a positive experience with NST municipality establishing a hospital and PCUs which now acts as a CUP hospital. It provides PHC, minor injuries, diagnostics, and low-level hospital services (convalescence, etc – no deliveries). It helps decongest Maharat. The PAO doesn’t provide support to this PHO. PHO has a vision of upgrading HCs to “ideal PCUs� over time – with nurse practitioner and eventually doctor services. Maharat hospital OPD dealing with an impossible burden of primary care cases. Decentralization of HCs need not be an obstacle to this development. Much depends on whether HC head understand health system and cares/is motivated to provide better services. Many are not. View of TAO-CEO TAO-CEO participated in NDC health sub-committee. He sees health as less of an election issue than education, but sees it as important for development of the community all the same (internal motivation). Gives priority to P&P but also wants to upgrade curative care. Wants to improve the compassion and care of services. Strategy focuses on community empowerment and mobilization – focus on youth and elderly. Complains of delays in changes to regulations, and differences in MOI and MOPH policies regarding decentralization, and detailed policies. View of Deputy Governor About 30% of TAOs and municipalities want devolution of HCs. Believes more are ready, but not interested – 60 health not a priority for them. Anticipates more and more will devolve over the next 10 years, based on the same three criteria applied in the pilots. PAO in this province not interested in health; has no health section, though in principle he thinks PAOs are capable of managing hospitals. He does not think PAOs should be given management of HCs – too far from the people. Believes mergers of smaller TAOs would be difficult and would cause divisions in the province, but believes coordination among TAOs is already happening. Aware of problem of expiry of licenses for HC staff to practice curative care in Nov. 2009 and proposes TAO-CEOs should be authorized to renew the licenses, given that hospitals, DHOs and PHOs will continue as before. Believes the mechanisms available through the Governor’s office for dealing with staff grievances are adequate to address the risks of politicization or favoritism among staff transferred to TAOs. View of CUP Hospitals Have maintained their existing support, and helped with transitional staffing problems. 61 Samut Songkram Summary of Key Facts and Findings re Devolved HC Ban Prok HC Transferred to TAO end of 2007 TAO Population 8560 Unique features Two HCs in the TAO; one devolved, one non-devolved. UC, SSS registered at HC 3,162 users; 1,487 UC users; 950 SSS users; 632 CSMBS users Utilization - 2007 6,610 - 2008 6,980 noted increased utilization from out of area Outreach Has reduced since devolution; stopped outreach during election to avoid risk of perceived political campaigning; have also had less time due to time taken in hosting visits and giving presentations about the experience of the devolution pilot. Staffing before transfer 5 GOs, no contractuals Staffing after transfer No change in full time staff. Visiting dentist hired. Current service delivery About 50% curative; 50% P&P Changes in service A dental clinic was equipped and dentist hired in delivery, and evidence of response to what people in the community wanted; plan responsiveness to hire nurse practitioner; would like to hire doctor to provide evening clinics Change in revenue, other MOI is financing upgrade of the HC building – filling in resources ground floor area with more accessible, private treatment areas. TAO provides 150,000 baht per year; UC provides 109,160 baht per year. Capital is allocated by provincial NHSO board. Change in management Some frustration due to delay in changing regulations flexibility and find MOI personnel procedures difficult (obtaining Governor’s Office approval for a new nurse post and developing exam for appointment hiring). But TAO is able to respond more rapidly to initiatives and budget requests. Change in UC provider No change. CUP provides 10,000 baht per month for payment fixed costs; unlimited drugs, supplies and equipment; will introduce activity based payment for P&P Express this year. Within each CUP area, UC members are free to choose which HC or hospital CMU/PCU they register with. Catchment areas for hospitals are treated flexibly for people who live near borders or on main roads. (In practice the Samut Songkram General Hospital doesn’t charge out of area patients from the province.) TAO has a community health fund. Changes in incentives Bonus of 3-5 months salary Change in reporting and Same as others. information flows Change in linkages to No change in referral relationships or support. Hospital hospital (Samut Songkram sends a team of doctor, pharmacist and nurse once a 62 General Hospital manages month to a zone of HCs (4 in this zone); sharing of this the CUP) resource is coordinated by one of the zonal HCs. Change in supervision and No change. DHO and CUP Board conduct joint training supervision twice a year. Bonus committee assesses performance once a year for NHSO bonuses. Province is implementing HCA. Training provided as before. Change in participation of HC active in communication. TAO has processes for citizens and patients obtaining people’s concerns and wishes for health services. Other governance issues PS is currently acting as TAO Public Health Section Head; regular informal communication but no formal supervision or performance assessment yet carried out by TAO. Relationships with non-devolved HC not so good. HSRI qualitative score of ++ transfer process Reasons for willingness to Close personal relationships between HC staff, TAO- transfer CEO and TAO PS. Dissatisfaction with MOPH. Views of stakeholders on decentralization policy and process View of PHO/DHO Has adopted the principle of treating devolved HC as a part of the health system, as before. Envisages there will be a mix of devolved and non-devolved HCs for some time, though eventually think all HCs should be devolved to TAOs/municipalities because they are closest to communities. Recommends allowing 3-5 years preparation and planning to sort out changes in regulations. Sees a need for one clear model that will be implemented nationwide (alternative models are still under discussion). On licensing issue – still need MOI to issue a regulation making clear that devolved HC staff are still licensed by and under the supervision of the PCMO. PHO has a primary care development strategy. Gaps in the network filled by new hospital PCUs. Hospital CMUs/PCUs support a zone of surrounding HCs in parts of the province (though Ban Prok is not in one of these zones). Preferred model in theory would be to devolve whole CUP network, but TAOs and networks of TAOs not ready. PAO has a small budget and lacks health capacity. Finances only small project requests from PHO. View of TAO-CEO Positive about transfer of one HC, but views the other non-devolved HC in the TAO as passive, poorly performing. TAO Council member from catchment of non-devolved HC says community there would like to see the BC devolved, but HC staff are not willing. View of DLA About 70% of TAOs want to take over HCs but only 2 in province have good governance awards. He thinks the criteria should be relaxed to allow more transfers. Municipalities are not interested in HCs – have their own health sections that do P&P including 63 immunization, and their citizens usually use hospital OPD. Thinks “politicization� is not likely to be a problem – although PHO perceives this as a common reason why HC staff do not want to transfer. View of CUP Hospital Has adopted philosophy of continuing to treat devolved HC as part of the network/zone system, as before. Concern about future decentralization because some TAO leaders give low priority to health, and because willingness to transfer so far is based on good personal relationships in BanProk, and may change if the opposition win future elections and personnel change. 64 ANNEX 3 ANNEX 3: INTERVIEWS AND VISITS WITH HOSPITALS: SUMMARY OF FINDINGS REGARDING HOSPITAL AUTONOMY 1. Hospital Autonomy under the APO Law– the Ban Phaeo Hospital Case The model of autonomy implemented at Ban Phaeo represents a marked and generally consistent shift in the key dimensions for organizational reform described in Preker and Harding (2002), which creates a strong set of incentives for improved performance and expansion of the business: a. Management Decision Rights over organization of services, structure, human resources, finances, logistics, capital investment, and partnerships with private and other sectors are fully shifted to the hospital board and director, with the partial exception of decision rights over disposal of surplus land granted by the government or donors, and “equity injections� and borrowing rights (which require Cabinet approval).  service organization: the hospital is free to open new services (e.g. it has added some additional specialties, developed mobile eye surgery services delivered in other areas, has established a second tier of services for private patients with higher “hotel� standards, and acquired another private health facility); it has not so far faced decisions on closure of services; MOPH approval is required for upgrade of hospital status (to general hospital, for example) but not for specific service developments; the hospital is free to contract out some services (they contract out security and established a co- operative to provide catering) and enter into partnerships with the private sector (such as framework agreements with private hospitals for referral, and for some diagnostic laboratory work such as reading PAP smears).  human resources: the Board is free to set the remuneration of the Director; the management is free to propose its own salary scales and other features of the remuneration package, which are submitted to the Board for approval (though decisions about appointment, promotion and placement on the salary scale rest with management, not the Board, as is appropriate); staff are employed by the hospital under private employment law (they are not civil servants), are subject to annual performance review, and their contracts can be terminated for poor performance by the hospital; the Board has set a policy that only 40% of expenditure can be spent on staff compensation – but this decision rests with the Board – there is no regulatory constraint;  finances: the hospital has full freedom and responsibility for budgeting, and retains all revenues; for private services it sets its own prices; for services for UC, SSS and CSMBS it is paid the same prices/capitation rates as private hospitals participating in these schemes; the hospital has established its own internal audit function which reports directly to the board; the hospital is subject to external audit by the Office of the Auditor General – i.e. it is only subject to ex post audit, not to any prior audit; it adopts full accrual accounting; its annual reports and audited financial statements are published;  logistics: the hospital does all its own procurement of drugs and supplies following government procurement rules, but with its own thresholds for Director or Board approval, and for competitive tender; it sets its own hospital formulary which includes some more expensive drugs including some brand 65 name drugs which are not included in the MOPH’s EDL; it does not source all its supplies from GPO (which is less than 3% of its supplies) both because GPO does not supply all the brand name or more expensive products on its formulary but also because some GPO prices are above market; it does all its own procurement of drugs and supplies;  capital investment and capital finance: the hospital has authority to determine capital investments and has freedom over financing them from retained revenue and donations; they can apply to the Bureau of Budget for capital finance; they can request Cabinet approval to borrow but have so far not done so. The hospital is not free to sell land or buildings it was granted on its establishment from the MOPH, nor to dispose of donated land and buildings; it is free to buy land, buildings and other businesses and dispose of these. b. Residual Claimant Status is fully with the hospital. There is not a clear regime for the event of financial failure or bankruptcy, though there is a requirement to conduct assessment of financial sustainability before APO status is granted, and the Board is highly focused on avoiding losses. c. Market Exposure is quite high because the revenues of the hospital are derived from fee-for-service, cased based payment systems and specific service contracts and project finance. Unlike MOPH hospitals, there is no salaries top-slice from the UC scheme payments to guarantee payment of staff salaries: staff remuneration comes from the revenue for service delivery. The hospital manages the CUP for a defined catchment of UC patients, which gives it a somewhat protected market for this group of patients. Additionally, the neighboring CUP for the provincial general hospital has a policy of not charging out-of-area self referrals, which in theory could foster cost-shifting – though in practice, this does not seem to be a concern. d. Accountability and Governance Structures; accountability is to purchasers (NHSO, CSMBS and SSS) for service delivery under the laws and regulations of these agencies. Accountability for performance, for service development, and for financial sustainability is to the Board. The Board is active. Interviewees who had reviewed lessons from experience with the Ban Phaeo Board concluded that the Board composition could be strengthened by inclusion of only professional Board members, on the basis of their skills as directors (including hospital management and service delivery skills). The hospital has selected community representatives on the Board but does not have LAO representation on the Board, and reportedly has had little take up from LAOs in response to attempts by the hospital to enlist their engagement in supporting the hospitals or working with it on joint initiatives. The hospital participates in the Hospital Accreditation program operated by the HA Agency subordinate to the MOPH, and is subject to inspection by the Inspector General, PHO, NHSO and other insurers as before, but the recommendations of the MOPH’s officers are now advisory rather than mandatory. The PHO supervises Ban Phaeo, but less than other hospitals (in part because it is regarded as a good hospital). It no longer receives financial and personnel reports from the hospital (though they have access to this information from Ban Phaeo’s published annual report). e. Social functions (unfunded mandates and community participation): the hospital has a policy of providing exactly the same clinical care, including access to medicines, to UC patients as to other SHI and private patients. Its main unfunded mandate is treatment for illegal migrants. The hospital does not refuse them 66 treatment, on humanitarian grounds, but has increasingly taken a hard-line stance with them (risk of reporting to immigration authorities). The hospital has extensive engagement with its community, both in consultation over service development and hospital development, and in raising donations and accounting for their use. It undertakes some corporate social responsibility initiatives from its surplus, and earmarks 4% of own source revenue for P&P initiatives. The internal incentive environment for management and staff adopted by the hospital’s Board appears to be strongly focused on aligning the incentives of staff with the objectives the Board has for the hospital. The Director has a fixed four-year term contract, renewable for only one term, and is subject to annual performance targets and review by the Board, which can decide on the level of Director’s bonus based on this assessment. Hospital staff are no longer civil servants, and no longer participate in the civil service pension and medical benefits schemes.6 They are employed by the hospital itself under private sector employment law, and their contracts can be terminated for poor performance more readily than is the case for civil servants. Doctors are paid a combination of salary, shared fees for service and performance rewards, and are subject to annual performance review. Medical staff are not permitted to work part time for other private sector hospitals or clinics, and in return are paid a substantially higher salary than the MOPH salary. They are permitted to earn additional fees for service in treating private patients after hours within the hospital. The hospital also hires part time specialists who are full time employees of other public hospitals. Prescribing is controlled by a hospital formulary, set by a Pharmaceuticals Therapeutic Advisory Committee (PTAC), which is somewhat broader than the EDL and UC formulary (includes more brand name medicines). All patients, including UC patients, receive drugs based on this formulary. The hospital does not charge a profit margin on medicines and does not have any partnerships or profit-sharing arrangements with private pharmacies. By contrast, conventional MOPH hospitals are constrained in the level of salaries and allowances they pay medical staff by national public sector salary regulations that keep salaries substantially below the level of private sector doctor remuneration, and higher levels of private sector remuneration for other categories of workers. As a result, over 73% of MOPH doctors and 9.5% of professional nurses worked part time in the private sector in 2005 – up from 55.4% and 8.2% respectively in 2003. Although MOPH hospitals since 2005 have had considerably increased financial freedom (they can retain income from UC, SSS, CSMBS and user fees, and can offer private beds with enhanced quality of “hotel� services), they do not have the freedom Ban Phaeo has to use this revenue to increase staff remuneration. Linkages and Integration of the Hospital with the Public Health System: The hospital functions within the public health services network in the same way as other community hospitals – though it has now expanded its capacity and range of services to a higher level than is typical for community hospitals. It manages the CUP for outpatient services (largely curative primary care) and P&P for its District catchment 6 There was a transition process for four years during which staff could choose whether to retain civil service status while continuing to work in the hospital, or to shift to employment by the hospital under private law. By the end of the four-year period, all staff who chose civil service status had transferred out to other positions in the MOPH. The majority of staff requesting transfer were in more junior grades and non-medical positions. 67 area like any MOPH hospital. It has set up three PCUs/CMUs to provide primary healthcare, which will supervise and support HCs in the catchment area as well as providing curative care. Two of the three PCUs also provide P&P. It provides P&P staff from the hospital to work in HCs. Interviews with HCs in the catchment area found perceptions that the hospital is somewhat “more stingy� than other CUPs that they are aware of – but this may relate to the fact that Ban Phaeo provide more support in kind and less in cash than most CUPs. Ban Phaeo’s staff – like devolved HC staff - are cut off from career mobility through the MOPH. The hospital plans career paths for its medical staff over a period of 9-10 years in the case of doctors, and provides scholarships in return for contracts to return to the hospital – but so far, it does this on its own, not as an integral part of the MOPH hospital career paths. It also has some linkages to the junior doctor training system. The hospital employs interns from public medical schools in the same way as other MOPH hospitals, and is seeking agreement of MOPH to also be assigned medical registrars who are providing public services in their first three years after graduation from a public university. Governance Structures and External Accountability of the APO to its Owners and Government Stakeholders: The organization charts used by the Ban Phaeo hospital typically show it as subordinate to the MOPH. However, it may be more accurate to describe it as subordinate to the Cabinet, as the Cabinet holds some key decision rights (such as approval of Board membership and of capital finance or borrowing). There is no dedicated unit or agency in the MOPH nor any other part of government responsible for independent monitoring and regulatory oversight of the performance of APOs in relation to financial sustainability, value of the business or good corporate governance. The Bureau of Budget reviews and provides advice to the Cabinet on any requests by APOs for capital finance. The criteria used by the Bureau are the same as for APOs in other sectors: low priority is given to APOs with substantial own source revenue (this was the main reason for rejecting Ban Phaeo’s application for capital finance to date); high priority is given to APOs responsible for investments that are part of a Government strategy or policy. The composition of the APO board includes representation of the MOPH by the Permanent Secretary, as a direct mechanism for enabling the MOPH to monitor and participate in decisions of the hospital. The processes for appointing the board members of the APO hospital involve a number of stakeholders and build in some checks and balances to seek to ensure both meritocratic selection and political accountability. The Royal Decree establishing the hospital as an APO specifies the composition of a search committee to identify candidates for the Board. The search committee is chaired by the provincial governor, and is appointed with the agreement of the provincial governor, PAO CEO, CEOs of the LAOs making up the District, the District Officer and the PCMO. The Royal Decree specifies that the Board should include three community representatives, three experts, and three ex officio members (the Permanent Secretary of the MOPH, the PCMO of the province, and the provincial governor). The search committee is obliged to identify two candidates for the positions of Chair (who may not be a Government Officer), and six community and six expert candidates for the three board posts for these categories. The Minister of Public Health selects candidates from the short lists proposed by the search committee, and submits the final board membership list to the Cabinet for approval. 68 The Board meets for 3-4 hours every month and is described as an activist board. Its decisions are usually made by consensus. Decisions rarely go to the vote. Where there is disagreement of 1-2 members, usually the proponent of a recommendation provides more information to address concerns raised until consensus is reached. MOPH’s role on the Board is characterized as more passive, and mostly focused on providing input and information on government and MOPH policy matters. The board is focused on service delivery performance and new development of the hospital, but is not particularly focused on efficiency or cost containment – though it is concerned to ensure the hospital avoids losses. Social Accountability to the Community Served by the Hospital: In addition to community representation on the Board, Ban Phaeo raises donations from the community and consults and reports to the community on how donated resources are used. As well, the hospital’s Board has chosen to adopt some other forms of community participation, though these are not mandated by law or Royal Decree. It sends representatives to meetings of community leaders at District, Tambon and village level to provide information about the hospital, consult and seek support for service developments and receive feedback. The Board also commissions the (independent) Thailand Rating Information System to conduct annual patient satisfaction surveys. Satisfaction rates rose after autonomy, then flattened and decreased slightly in the last 2-3 years. However, satisfaction remains high at 86%. Declining satisfaction is perceived to be due to increased utilization – giving rise to increased waiting. 2. Patong Hospital – Candidate for APO Status, Rejected by Public Sector Development Commission Patong Hospital is one of around 45 MOPH hospitals that expressed interest in autonomous status in response to an initiative of the Minister of Public health in 2006 to begin to scale up the Ban Phaeo model. At this time, hospitals were also considering the option of becoming a semi-autonomous service delivery unit (SDU) within the MOPH, which is not a separate legal entity, and has a MOPH-appointed Government Officer as Director. Patong Hospital is the only one that pursued this initiative through the feasibility study stage, and finally to the stage of preparing a draft Royal Decree and seeking the necessary approval of the Public Sector Development Commission to become and APO. Its proposal was rejected, reportedly on the grounds that APO status is not seen by the NDC or the Public Sector Development Commission as a form of decentralization, and it is not possible under existing law to transfer APO hospitals to LAO ownership. The context at that time was one in which the NDC and Public Sector Development Commission were seeking to push the MOPH to make faster progress on health sector devolution, so that there was a potential concern that creating APOs could block devolution. Like Ban Phaeo, Patong Hospital is a community hospital that has grown (from 10 beds in 1996 to an official number of 60 beds now, but actually 90 beds, including some private beds) and has attracted very substantial donations from the community, country and abroad. About a third of its patients are foreign tourists. The Patong Hospital proposal differed slightly from the Ban Phaeo hospital model. The composition of the Board would have been modified to include two members who are representatives of hospital staff (in place of one of the community representatives and 69 one of the experts, and cannot be staff of Patong Hospital) – to ensure expertise in hospital management and service delivery is available to the Board. Additionally, the Patong Hospital APO would have included one of the HCs in its District – a HC that required substantial new investment (following damage in the Tsunami) and performance improvement. 3. Siriroj Phuket International Hospital The purpose of interviewing management of a private for-profit hospital was to identify similarities and differences in the key factors driving organizational performance by comparison with the Ban Phaeo APO model. This is a privately owned for-profit hospital, founded by a local doctor, whose family remains the majority shareholders, though the hospital is pursuing public listing. It is a 150-bed secondary hospital, which does not participate in UC. It offers a “single tier� of private service – unlike private hospitals that participate in UC. Key differences from the Ban Phaeo model:  Full autonomy over decisions on land, buildings and capital finance;  Licensed by MOPH, but only inspected annually or less for a brief “check- list� form of inspection – licensing focuses on start up and major facilities changes only;  SSS inspects the hospital once a year and does a “walk around� inspection;  Private health insurance is a significant source of revenue and the major insurers are a significant driver of quality, standardization, including assessing of medical necessity/appropriateness; the five major insurers carry out inspections of medical records, financial statements, and facilities, individually and sometimes as a group;  The hospital is ISO9001 accredited and is inspected twice a year to maintain ISO rating; it is HA to Stage I);  The board is the main driver of efficiency and cost control;  Although it is a privately owned company, its financial reports are published;  The hospital has a customer focus and a philosophy of also serving the community, but as a for-profit, “corporate social responsibility activities� need to have a financial case; as a for-profit the hospital does not raise any donations – though it has received recognition for its pro bono response during the Tsunami, for example, during which it handled about 600 cases and about 10% of the major surgery);  The hospital carries out more extensive patient opinion and feedback research than Ban Phaeo to find out what actual and potential patients want – including surveys and focus groups;  There are some linkages to the public health system – mostly focused on priority public health risks, disaster management and emergencies: the hospital reports notifiable diseases, and for some important higher-risk circumstances it reports daily (e.g. over Thai new year); there is a provincial master plan for natural disaster management and all private as well as public hospitals have their own plans; the hospital’s ambulance is part of an organized emergency call out system in Phuket coordinated by the PHO; the PCMO visits irregularly (around every 2-3 years); the hospital has referral relationships with the public hospitals in the province, with private hospitals outside of the province; the hospital sometimes agrees for their doctors to provide some part- 70 time back up to public hospitals in the province to deal with short-term staff shortages.  Career paths for staff – aside from the first 3 years of mandatory public service for medical graduates of public universities, medical staff mostly have a purely private sector career path; nursing staff are mostly recruited with some prior experience, generally from non-profit private hospitals, rather than public hospitals. 4. Perceptions of Hospital Autonomy Reform and Prospects for Scaling Up The Ban Phaeo case is viewed as successful, by most of those interviewed who are aware of it, and the before-and-after evaluations back up this perception. However, some take the view that APO status is only replicable in specific circumstances - Ban Phaeo Hospital and Patong Hospital are both able to earn substantial revenue from non-UC sources, and are in communities with growing demand, and reasonably dense populations (permitting inter-hospital competition). However, the managements of these hospitals take the view that what they have achieved would be replicable in any larger community, general and regional hospitals with some diversity of revenue sources. In smaller community hospitals serving small or sparse populations and reliant almost entirely on UC and MOPH finance, the positive dynamic achieved in Ban Phaeo that led to a virtuous cycle of revenue growth and expansion would be difficult to achieve, and such hospitals would best be given autonomy as part of a larger network of hospitals. Some interviewees advocate that it would be more efficient to give autonomy to district or provincial networks, rather than individual hospitals. In the case of Ban Phaeo, both the hospital management and Samut Sakhon PCMO believe that the hospital would be far better able to manage the HCs if they were part of the hospital, and directly accountable and controllable by the hospital. Incorporation of an under-performing HC was built into the proposal for Patong Hospital for this reason. Among the reasons why no further hospitals have been autonomized, in spite of expressions of interest from over 40 hospitals, is the fact that the NDC and Commission on Public Sector Development declined to approve APO status for Patong Hospital, on the grounds that this model is not a form of decentralization. However, there is currently no push from the NDC for hospital decentralization. Autonomization it is not inconsistent with decentralization, in principle. A well- functioning autonomous hospital should be less of a financial risk and managerial concern for a PAO than a conventionally managed ex-MOPH hospital. However, a new law would be required to authorize the transfer of APOs to LAO ownership. Interviewees identified some other reasons why scale up has not occurred:  The APO model has not been well explained to the public by the Public Sector Development Commission; many people misunderstand it as a form of privatization, that could lead to profiteering and foreign take-over;  Although it is believed that a majority of doctors would support the model in many hospitals, other staff will oppose it (as is usual with public sector pay compression, opposition often comes from lower grade staff and non-medical staff for whom private sector pay and benefits is often less attractive than civil service status); individual hospital directors are reluctant to manage this kind of staff opposition without external back-up and support; 71  Career path mobility concerns – similar to those identified in relation to devolution of HCs – are a concern for many staff with a model of autonomy that treats individual hospitals as “islands� rather than as part of a common career service;  Lack of support for the model by the MOPH senior management;  Concerns from other sectors about risks of the very “light handed regulation� of the APO model – some Boards have awarded themselves high pay for low output, in the absence of any guidelines or regulatory oversight of these decisions; some APOs have run into financial difficulty, in the absence of any financial supervision and bankruptcy regime; recently this has led the Public Sector Development Commission and Cabinet to halt any further creation of APOs;  Concern that the model so far developed needs modification before scale-up would be feasible – e.g. if hundreds of APOs are created in the health sector, Permanent Secretary representation on the Board would need to be re-thought, and dedicated systems and capacity would be needed to supervise and participate in the governance of APOs;  Some express concern that APO hospitals, like devolved HCs, might weaken the MOPH’s ability to ensure public health reporting and management of public health emergencies and disasters. However, the private hospital example given above suggests that this is unfounded – private hospitals have both professional and reputational motivation to work closely with the MOPH on these issues – and the MOPH/Cabinet has greater leverage over APOs and devolved health facilities than it has over the private sector. 72 ANNEX 4 ANNEX 4: LIST OF STAKEHOLDERS INTERVIEWED AND CONSULTED 18 MARCH – 7 APRIL 2009 Bangkok 1. Dr. Supakit Sirilak, Director, Bureau of Planning and Strategy, MOPH 2. Dr. Viroj Tangcharoensathien, Director, International Health Policy Program (IHPP) 3. Dr. Pongpisut Jongudomsuk, Director, Health System Research Institute (HSRI) 4. Dr. Mongkol Na Songkhla, Former Minister of Public Health 5. Dr. Suwit Wibulpolprasert, Senior Advisor, MOPH 6. Dr. Amphon Jindawatthana, Secretary-General, National Health Commission Office 7. Dr. Winai Sawasdivorn, Secretary-General National Health Security Office (NHSO) 8. Mr. Napong Sirikantayakul, Director, Bureau of Public Sector Receipt and Disbursement Administration, Department of Comptroller-General, MOF 9. Ms. Supanee Lertchaiyalit, Director/Senior Economist, Division of Local Fiscal Policy, Fiscal Policy Office, MOF 10. Ms. Chumsri Pojanapreecha, Advisor, Bureau of Budget 11. Mr. Kampon Jitgaroon, Advisor, Bureau of Budget 12. Mr. Narong Chueaboonchuay, Policy Officer , Office of the Decentralization to Local Government Organization Committee, Office of Permanent Secretary Government House 13. Mrs.Supranee Janrattanawong, Director of Decentralization Policy and Plan, Office of the Decentralization to Local Government Organization Committee, Office of Permanent Secretary Government House 14. Mr. Teerapat Kutchamat, Director of Division of Public Health and Social Welfare, Department of Local Administration, Ministry of Interior 15. Mr. Prawing Nujam, Member of NHSO Board, Rep. of LAO. 16. Mrs.Yoavaman Suasangthong, Director, Office of Supporting Public Health Decentralization, bureau of health policy and strategy 17. Mr.Noppadol Kaewsupat, Member committee of NDC 18. Assoc.Prof.Wutisan Tunchai, Member committee of NDC 19. Mr.Somsak Tathaisong, representative of Dr.Somreong Yangkathok, Nakornratchasima PAO 20. Ms.Samruay Yothavichit, Policy analyst, Udonthani PHO 21. Mr.Prajak Thongngam, Ubonratchathani DHO, Ubonratchathani province 22. Maureen Birmingham, Resident Representative of WHO 23. Patrick Brenny, Country Coordinator of UNAIDS 24. Tongta Khiewpaisal, Program Manager of UNDP 25. Toomas Palu, Lead Health Specialist, World Bank 26. Dr.Phusit Prakongsai, Policy researcher, IHPP 27. Dr.Supol Limwattananon, Senior researcher, IHPP Udonthani Province 28. Dr.Samit Prasuntanakarn, Deputy PCMO, Udonthani Provincial Chief Medical Office 73 29. Mrs.Rungtip Aekkapong, Director, Division of Health Insurance, Udonthani Provincial Chief Medical Office 30. Mr.Teeradej Wongratch , Vice Governor, Udonthani Provinceal Hall 31. Mr.Udom Oonjun, Head of General Administration (Representative of Thongthin Changwat) 32. Mr. Amnuay Intarathirach, Chief Executive of the TAO, Naphu TAO 33. Mr.Prasong Chaichana, Head of Naphu Health center 34. Mr.Rungrueng Punnaratch, Pen District Chief Health Officer 35. Dr.Kriengsak Aekkapong, Director of Pen Community Hospital 36. Mrs.Phongsai Sodavichit, Head of BanLuong Health Center Chiangmai Province 37. Dr.Surasingha Vitsarutrattana, Deputy PCMO, Chiangmai Provincial Chief Medical Office 38. Ms. Rungtawan Hutamai, Head of Public health strategy development 39. Ms. Jutamas Wongkam, Public health strategy development officer 40. Mr.Chumporn Saengmanee, Vice Governor, Chiangmai Provincial Hall 41. Mr.Noppadol NaChiangMai, Chief Executive of the TAO, DonKaew Tumbon Administration Organization(TAO) 42. Mr.Suriyan Paesri, Head of Donkaew Health center 43. Ms.Ubon Yawai, DonKaew Permanent Secretary TAO 44. Dr.Wichai Jaikaew, Deputy Director , PCU Nakornpink General Hospital 45. Mrs.Wasana Taecharoen, PCU Nakornpink General Hospital officer Phuket Province 46. Dr.Pongsawat Ratanasang, PCMO, Phuket Provincial Health Office 47. Dr.Prapornsri Narinrak, Head of Public health strategy development , Phuket Provincial Health Office 48. Dr.Suwanna Lorlowhakarn, Public health strategy development officer, Phuket Provincial Health Office 49. Mr.Kreeta Saetan, Mayor of Vichit Municipality, Phuket Province 50. Mr.Nimit Aekvanit, Head council of Vichit Municipality, Phuket Province 51. Mrs.Sirikul Limnukul, Head of Public Health Division, Vichit Municipality, Phuket Province 52. Mrs.Yupa Sukwattanavijit, Head of Laemchun, Vichit Municipality, Phuket Province 53. Mr.Kampanat Limmanee, Public Health officer, representative of Muang District Health Officer 54. Dr.Thaweesak Netwong, Director, Patong Hospital 55. Dr. Toranis Tantipiriyakij, Medical Director, Siriroj Phuket International Hospital 56. Mr.Ekapol Tharasiriroj, Deputy Managing Director, Siriroj Phuket International Hospital Nakornsrithammarat Province 57. Dr.Nopporn Chuenkhin, Provincial Chief Medical officer, Nakorn Sri Thammarat Provincial Health office 58. Mrs.Weena Thitipraserth, Deputy PCMO, Nakorn Sri Thammarat Provincial Health office 59. Mr.Sumran Pitakuldilog, Public health analyst officer 60. Mr.Prasong Chaowaphasi, Muang District Health Officer 74 61. Dr.Kithi Rattanasombat, Director of Thasala District Hospital 62. Dr.Somchai Nimwattanakul, Director of Maharaj Nakornsithammarat Hospital 63. Mrs.Marasri Kuanhin, Head of Public health strategy development 64. Mrs.Oraphan Khlongsrichay, Public health analyst officer 65. Mrs.Sarapee Sriporn, Head of Pakphun Health center 66. Mr.Prapas Jarayprapas, Head of Bansalabangpu Health center 67. Mr.Thanawut Thawornbrahm, Chief Executive of Pakphun TAO 68. Mr.Bunyawat Cheechang, Vice governor, Nakornsithammarat Province Samutsakhon Province 69. Dr.Chairat Wechpanich, Provincial Chief Medical officer, Samutsakorn Provincial Health office 70. Dr.Kasem Supawannakit, Deputy PCMO, Samutsakorn Provincial Health office 71. Mr.Mittraprap Anusatnun, Head of public health division, Luksam TAO, Banphaeo district, Samutsakorn Province 72. Mr.Paoen Srinivet, Chief executive of Luksam TAO, 73. Mr.Tawipak Hunchoroen, Permanent Secretary of Luksam TAO 74. Ms.Tanyatip Yingyuod, Banphaeo District Health Officer 75. Ms.Wannee Tongsawad, Head of Luksam Health Center 76. Mr.Narinat Putchakarn, Head of Tunginsee Health Center 77. Ms.Ladda Amornlukpreecha, Public health officer, Tunginsee Health Center 78. Mrs.Krisana Klovutisatein, Public health analyst officer, Samutsakorn Provincial Health office 79. Ms.Jaree Sripharat, Deputy director (Quality improvement), Banphaeo Hospital 80. Mrs.Kanokwan Sengkhampha, Assistant director, Banphaeo Hospital 81. Dr.Witit Arttavetkul, Director of the Government Pharmaceutical Organization (Former Director of Banphaeo Hospital) Samutsongkram Province 82. Dr.Veerachai Peetawan, Provincial Chief Medical officer, Samutsongkram Provincial Health office 83. Ms.Ketsuda Losachitanon, Head of Public health strategy development, Samutsongkram Provincial Health office 84. Dr.Somporn Natiruttakorn, Public health officer, Samutsongkram Provincial Health office 85. Mr.Kittipong Thonglua, Muang District Health Officer 86. Mr.Pumit Leenawong, Local Administration Officer, Samutsongkram DLA 87. Ms.Anchalee Keowwan, Head of Health Promotion division, Somdejpraputtalerdla Hospital, Samutsongkram province 88. Mrs.Chaweewan panpradit, Head of Community curative care sector, Somdejpraputtalerdla Hospital, Samutsongkram province 89. Mrs.Tussanee Jantarasutti, Nurse, Somdejpraputtalerdla Hospital, Samutsongkram province 90. Mr.Chana Intarachot, Chief Executive of Banprok TAO 91. Mr.Manoj Trairattanayon, Permanent Secretary of Banprok TAO 92. Ms.Vilailuk Tiwakorakoj, Dentist, PHO 93. Ms.Porntip Mingmontien, Head of Banprok HC 94. Mr.Chatchai Chantarattanachok, Banprok HC staff 95. Mr.Veerapong Hengboonme, Kaewfa HC staff 75 ANNEX 5 ANNEX 5: TERMS OF REFERENCE FOR RAPID ASSESSMENT OF DEVOLUTION IN THE THAILAND HEALTH SECTOR REFORM Background. Decentralization is emerging as one of the important issues at the current stage of health sector reform in Thailand. The 1997 constitution (as well as revised in 2007) and consequent Act on Operationalization of Decentralization in 1999, mandated all sectors to develop decentralization plans of their functions, facilities and personnel by 2010. The same law also mandates that 35% of Government revenues be retained and spent at the local government level. The legislation assigned six key groups of functions to local administrations: providing for essential infrastructure; improvement of quality of life (including health and education); social and community management; planning and local investment; and, tourism. The Ministry of Public Health initially developed plans for decentralization of responsibility of health care to municipalities and sub-districts – tambons with local Area Health Boards covering several local administrations to undertake key roles to coordinate to be decentralized health functions but these reforms stalled shortly after instigation. But more recently, the MOH has launched 28 pilots of devolving primary health care facilities providing a range of preventive and curative care to local administrations. The 1997 economic crisis also launched reform ideas for public institutions and policies in social sectors supported by the ADB social sector reform loan. Autonomization of public hospitals was part of the social reform package. The 1999 Act of Public Organizations provided an enabling environment for reform. The main objective of the reform was better responsiveness of health care providers to local communities and to allow more flexibility in management instead of rigid central bureaucratic and financial controls. However, only one 120 bed community hospital (Ban Phaeo Hospital in Samut Sakhon Province) has become autonomous, further roll-out was stalled because of political concerns about “privatization,� implementation difficulties in particular about changing the way public finances were managed, and resistance by civil servants whose employment conditions were to change. On parallel track, Thailand introduced in 2002 Universal Health Insurance Coverage Scheme (UC) extending insurance coverage to more than 95% of population. This significantly altered the way how health services were financed by introducing purchasing care for the insured by National Health Security Office (NHSO) and application of population based block grants for district level primary health care and DRG based global budgets for inpatient care. The UC complemented the Civil Service Medical Benefits scheme and Social Health Insurance for formal sector employees that already acted as purchasers of care for their beneficiaries. Overall, the health financing landscape has changed significantly requiring corresponding changes in the management of public health services. In addition, the National Health Security Act 2002 also requires the decentralization of UC budget management to the local administration and NHSO responds this by 76 transferring budget for community care (37.5 Baht or approx 1 US$ per capita) to sub- district administrative organization (SAO). The SAO has to co-fund with this transferred UC budget to set up a community fund managed by a committee comprised of all local stakeholders. There were 888 SAOs participating in this initiative in 2006 and the number has been increasing to almost 4,000 in 2009. As of 2009, decentralization agenda as called by constitution and laws is very much alive and health sector would need to act. Also, public hospital autonomy reforms are gaining again increasing attention given the changes in health sector financing and political climate. In order to facilitate policy discussion among stakeholders about options for further actions in decentralization and autonomous management of hospitals, a review of the enabling policy environment, experience with decentralization and autonomization pilots, relevant international experience and lessons learned is called for. Objective of the assignment is to conduct rapid assessment of current status of decentralization in the Thai health sector, identification of issues; developing of options for addressing decentralization in the health sector. Scope of Work. The consultant is expected to undertake the following: o desk review of applicable policy documents and reports; o conduct interviews with the key stakeholders for decentralization policies overall and in the health sector; o conduct field visits to and review experience of at least: two primary health care facilities devolved to local administrations (including visit their related local administrations); the one autonomous public hospital; one un-reformed local administration and public hospitals; as well as one private hospital to gain better understanding on market exposure, cost structures and governance arrangements; time and resources allowing, more locations and health providers can be visited; o on decentralization analysis, ensure that different dimensions of health and health care covered, including public health surveillance and safety, disease prevention and health promotion, health care at primary and secondary level, investments and maintenance. However, in depth analysis is expected to focus on devolution of health care provision including its financing) o on autonomy - applying consistent methodology for analysis, e.g. Harding-Preker dimensions of hospital autonomy: decision rights, residual claimant, social functions and exposure to markets; o pay particular attention to governance and capacity issues, including engagement of local communities and applying purchasing contracts as accountability instruments; o pay particular attention to incentives of health personnel before and after devolution and autonomization o discuss and analyze the implications on parallel health sector reforms (e.g. UC) or cross-sectoral reforms on the health sector decentralization and hospital autonomy agenda; 77 o Identify international best practices applicable to Thai context as well as case studies highlighting the risks of application of decentralization policies in the health sector. 78 ANNEX 6 ANNEX 6: SAMPLE MEMORANDUM ON TRANSFER OF HEALTH CENTER TO TAMBON ADMINISTRATIVE ORGANIZATION Memorandum on the Transfer of Public Health Duties and Responsibilities of Ban Pa Ngo Health Center to Don Kaew TAO in accordance with the Operational Plan and Steps on Decentralization This memorandum is the evidence that Chiang Mai Provincial Health Office (PHO), as the handover organization, represented by PCMO Dr. Wattana Kanjanakamol, has successfully transferred public health duties and responsibilities of Ban Pa Ngo Health Center (Don Tambon Kaew, Mae Rim District, Chiang Mai), in care of Chiang Mai PHO, Office of Permanent Secretary of Public Health, MOPH, to Don Kaew Tambon Administrative Organization, Mae Rim District, Chiang Mai province, as the recipient organization, represented by TAO CEO Mr. Nopadol Na Chiang Mai. This devolution is in accordance with Plan and Steps in Decentralization Act, A.D. 1999. Don Kaew TAO is to administer and manage the health center according to regulations, criteria, standards, and public health work methodologies set by MOPH and Chiang Mai PHO, or those which are set by MOPH/ PHO following the devolution of the health center as necessary or required by health situation, will also apply to the devolved Health Center. The details are attached after this memorandum. Signed______________________ Chair of the Transfer Process (Mr. Choomporn Saengmanee) Deputy Governor of Thailand Chair of the Sub-Committee on Promotion of Devolution of Health Center, Chiang Mai Province Signed_____________________ Handover Organization (Mr. Wattana Kanjanamol) Chiang Mai PCMO Signed_____________________ Recipient Organization (Mr. Nopadol Na Chiang Mai) Don Kaew TAO CEO Signed____________________ Witness (Mr. Charoenrit Sasguansat) Local Administration Officer Signed____________________ Witness (Mr. Choocherd Puanpinta) District Health Officer, Mae Rim District Signed_____________________Witness 79 (Mr. Suriyan Praesee) Chief of Ban Pa Ngo Health Center 80