89337 MOLDOVA HEALTH TRANSFORMATION PROGRAM (P144892) TECHNICAL ASSESSMENT REPORT (Last updated April 11, 2014) 1 Moldova Health Transformation Program Technical Assessment A. The Government’s program 1. The Government’s health reform program, which has intensified since the introduction of national health insurance in 2004, is comprehensive and technically sound. Its priority actions correspond to the key challenges in the health sector. In spite of the flux in the political situation over the last 10 years, the Government is committed to improving health for the population. The national health program’s objectives and key priorities remain constant despite the turn-over of senior management in the MOH. 2. The Health Care System Development Strategy 2008-2017 is the cornerstone of the Government program. This comprehensive strategy follows the WHO’s health system framework ( as defined in the World Health Report 2000) and is accompanied by an action plan and a monitoring framework. The government program has a strong focus on financial protection, efficiency and quality of health services. Health sector development in Moldova is guided by the 2007-2021 National Health Policy (the Policy). The Policy is further elaborated and operationalized by the 2008-2017 National Health System Development Strategy (the Strategy). Together, they form a robust reform agenda for the coming years with the following objectives: (a) continuous improvement of population health; (b) financial risk protection; (c) reducing inequalities in the use and distribution of health care services; (d) enhancing user satisfaction; and (e) restructuring the health system to improve performance and population health regardless of limited resources. 3. To achieve such objectives, the Strategy identifies four groups of priority activities (also known as “sections�): (i) Section 1: Improve the management/stewardship of the health system. This section includes interventions to: (a) increase the capacity of the MOH, MOH-affiliated institutions and local health authorities; (b) improve communications mechanisms; (c) strengthen multi-sectoral collaboration for better health; (d) increase the involvement of civil society, and (e) bring national health legislations to the level of European Union (EU) standards. (ii) Section 2: Improve the funding and payment mechanisms for health services. This section includes interventions to improve: (a) health funding; (b) mechanisms for payment and contracting of services, and (c) equity and transparency of resource allocation, as well as financial protection. (iii) Section 3: Organize and provide health care services to meet the people’s health care needs. This section includes interventions to (a) promote integration and continuity of health services; (b) develop priority sectors of the health system with large impacts on population health, and (c) improve quality of care and level of patients’ satisfaction. (iv) Section 4: Generate and ensure the necessary resources of for the health system. This section includes interventions to: (a) efficiently manage human resources for health through rational use of existing staff as well as other pre-service and in-service activities; (b) strengthen the technical and material base of the institutions and facilities, and (c) rationally manage the drugs provision. 2 4. A three-year Medium-Term Budgetary Framework (MTBF) is prepared on a rolling basis to finance the implementation of the Strategy. The MTBF has five programs: (i) “Health Policy and Management� which supports policy development, implementation management and monitoring through MOH, the National Health Insurance Company (CNAM) and other state institutions; (ii) “Priority Public Health Interventions" which focuses on disease surveillance and control for infections as well as for non-communicable diseases; (iii) “Personal Health Services� which concentrates on the delivery of a range of personal health care services rendered by family medicine, hospitals, specialized outpatient and community care providers; (iv) “Development of health system resources� which focuses on investments in the health sector and rational drug management, and (v) “Special medical programs� which focuses on special issues like forensic services, etc. 5. The MTBF’s five programs are further divided into 17 sub-programs. The main sources of financing for MTBF are the state budget (59 percent) and health insurance fund (40 percent). Total cost of MTBF for the next four years (2014-2017) has been estimated at 26,115 million Moldovan Lei (MDL) or US$1,964 million. B. Health Transformation Program Description 6. The proposed Program Development Objective (PDO) is to reduce key risks for non- communicable diseases, improve financial protection and enhance efficiency of health services in Moldova. 7. The proposed four-year (2014-2018) the Health Transformation Program will support a subset of the Strategy’s first three sections. The proposed Program-for-Results (PforR) covers a “time slice� of a portion of the GOM program. It is expected to contribute to the Government program and, therefore, health system as a whole, by disbursing funds against achievement of a subset of its key results. It correspond to 7 out of 19 sub-programs of the MTBF, namely: (i) Sub-program I: “Policies and management in health care� which provides recurrent funding to the MOH for the fulfilment of its stewardship functions for the health sector. No activities under this sub-program are excluded from the Program. (ii) Sub-program II: “Administration of mandatory health insurance fund� which provides recurrent funding to CNAM central and regional offices for the fulfilment of its management and administration of the health insurance fund. No activities under this sub-program are excluded from the Program. (iii) Sub-program V: “Primary Care Services� which provides recurrent funding for the delivery of primary care services. The Program is limited to two categories under this sub-program: (a) performance-based incentives for Family Medicine providers under CNAM contracts, and (b) CNAM reimbursement of the outpatient drug benefit package. (iv) Sub-program IX: “In-patient Care� which provides recurrent funding for the delivery of in-patient care. The Program is limited to one category under this sub-program, performance-based incentives to hospitals for in-patient services under CNAM contracts. 3 (v) Sub-program XIX: “National special health programs� which funds recurrent costs for special health programs. Under this category, the Program is limited to National Tobacco Control Programme activities only. 8. Total cost of the Program is US$153 million or 8 percent of the GOM 2014-2017 health program. In the boundaries of the Program, there is no other source of funding from other development partners. World Bank financing of the Program would be US$28 million or 18 percent of the Program’s cost. An incremental US$2 million in technical assistance is also needed to strengthen capacity for the Program and would be provided through a Technical Assistance Credit. Table 1: Program expenditure requirements according to MTBF and projections Planned (US$ million) PROGRAM 2014 2015 2016 2017 Policies and management in health care 0.71 0.80 0.87 0.91 Administration of national mandatory health insurance fund 5.06 4.80 5.37 5.64 Primary care (performance-based incentives for Family Medicine providers under CNAM contracts and CNAM reimbursement of the outpatient drug benefit package only) 17.88 19.53 20.86 22.10 In-patient care (performance-based incentives to hospitals for in-patient services under CNAM contracts only) 0.00 0.36 0.38 6.59 National special health programs (tobacco control activities only) 0.39 0.47 0.47 0.50 Total 24.04 25.96 27.95 35.73 Table 2: Program Financing Amount Source % of total (US$ million) Government 82.9 75% IDA (PforR) 28 25% Other Development Partners 0 0% Total Program Costs 110.9 100% IDA (IPF) for Incremental TA 2.7 Total Program Costs, including TA 113.7 4 C. Strategic Relevance and Technical Soundness of the Program C1. Strategic Relevance and Technical Soundness of the Program Regarding NCD Control 9. One of the foremost challenges facing Moldova is a growing burden of non-communicable diseases (NCD). NCDs account for the main causes of mortality and morbidity in the population, with the top causes of premature deaths being heart attacks, stroke, cancers and liver cirrhosis. Table 3: Top causes of life years lost (YLLs) in Moldova Source: Institute for Health Metrics and Evaluation 2013 10. Without effectively addressing NCD, Moldova will not be able to significantly improve health of its people and reduce NCD-related economic costs. Globally, it has been estimated that NCDs and their risk factors cost countries up to 6.7% of their GDP.1 For this reason, NCD control is prominent in the government Strategy as well as various sub-sector strategies and action plans, such as the National Public Health Strategy 2013-2020, the National Strategy to Prevent and Control Priority Non-communicable Diseases 2012-2020 and their corresponding action plans. In this context, it is highly relevant for the Program to focus on reducing NCD risks, especially smoking and hypertension which are two of the most important NCD risk factors. Smoking The situation 11. The burden of smoking is significant in the Republic of Moldova: about 50 percent of men and 7 percent of women smoke. Tobacco consumption is the second leading risk factor in men in terms of the cause of the disease burden in the Republic of Moldova, constituting 14.9 percent of the total DALYs. 12. Smoking among young people is of particular concern. In 2008, 13.4 percent of young respondents stated they consume tobacco (boys = 20.8 percent, girls = 7.1 percent). Among pupils who have ever smoked, 49.2 percent started smoking before the age of ten. 13. The share of smokers correlates with the socioeconomic status but the influence is different for men and women. A higher prevalence of smoking is recorded among men from poor households (60 percent) and with a lower level of education (54.1 percent), while in the case of women, the situation is reversed: smoking prevalence is higher among women from wealthier households and with a higher level of education. In terms of number of cigarettes smoked per day, 85 percent of men and 40 percent of women smoke ten or more cigarettes per day. 14. A significant proportion of smokers in Moldova are trying to quit, but they are not as successful in quitting as their counterparts in EU-15 countries (Figure 1). 1 Marc Suhrcke, Rachel A. Nugent, David Stuckler and Lorenzo Rocco. Chronic Disease: An Economic Perspective. London, 2006: Oxford Health Alliance. 5 Figure 1: Smokers are trying to quit, but not succeeding 15. The estimation of tobacco economic costs related to the productivity lost because of premature deaths, hospitalizations, and out-patient care of the smoking-related diseases amounted to about MDL 430 million in 2010, which is equal to the income collected to the State Budget from the sale of tobacco products. International experiences/best practices in tobacco control 16. Tobacco control is cost-effective and represents one of the “best buys� in NCD control as well as in health generally speaking as recommended by WHO, taking into account four key criteria: (i) health impact; (ii) cost-effectiveness; (iii) cost of implementation; and (iv) feasibility of scale-up, particularly in resource constrained settings such as in Moldova. Analysis shows that combined key tobacco control interventions provides the opportunity to save a total of 40 487 healthy life-years each year, at a cost of 1753 MDL per each healthy life-year gained. The results of the cost-effectiveness analysis regarding tobacco control interventions indicate that all the interventions under modelling are highly cost-effective: The quantum of resources required for their implementation is less than the Gross Domestic Product per capita, and the effects produced by implementing the interventions will be much higher compared to the requested resources. Soundness of the Program 17. Moldova ratified WHO Framework Convention on Tobacco Control (FCTC) in May 2007. To facilitate the implementation of FCTC, National Tobacco Control Programme (NTCP) was approved in 2012, and NTCP forms the basis for Program support in tobacco control. 6 18. NTCP is robust and encompasses all the key cost-effective, proven interventions recommended by FCTC on both demand and supply side to reduce tobacco consumption. These include provision of smoking cessation services, warning labels on cigarette packaging, and restriction on tobacco publicity and promotion. In particular, it encompasses tobacco tax increases which represent the most cost-effective measure to reduce smoking according to the global evidence base.2 NTCP also has a detailed, well sequenced four-year action plan with clear indicators and targets. For this reason, with regard to tobacco control, the Program is strategically relevant and technically sound. 19. Moreover, the population of Moldova is highly supportive of tobacco control interventions under NTCP, especially among women. For example, as per public opinion surveys, Moldova has higher percentage of people supportive of the ban of smoking in public places than in any country in Europe and Central Asia. More than half of the people are in favor of increasing tobacco taxation (Figure 2). This makes the Program more likely to succeed. Figure 2: Moldova’s population are highly supportive of tobacco control interventions Hypertension 20. Hypertension is the second leading health risk factor in Moldova, contributing to high rates of stroke and heart attacks. Screening of hypertension and managing it with anti-hypertensive drugs in the family setting is one of the “best-buys� in NCD control, by avoiding the most costly complications of untreated hypertension. 3 The Government program has been addressing this risk by (i) including hypertension screening and management in the basic benefit package, (ii) developing evidence-based 2 Gilbert A, Cornuz J (2003). Which are the most effective and cost-effective interventions for tobacco control? WHO Regional Office for Europe, Health Evidence Network report; 2003. 3 From Burden to “Best Buys�: Reducing the Economic Impact of Non-Communicable Diseases in Low- and Middle-Income Countries. WHO, 2011. 7 clinical practice guidelines and disseminating them to all primary care providers and (iii) reimbursing around 50% of the cost of out-patient hypertensive drugs. 21. However, less than one in three patients with hypertension in Moldova is taking medication daily, and consequently, only around 10 percent of people diagnosed with hypertension have their blood pressure under control4. It is likely that poor patient education combined with financial barriers lead to this low adherence rate, with 53 percent of respondents to a 2010 survey only taking antihypertensive medication “when needed� (rather than daily) and 5 percent only “when affordable�. 5 Figure 3: Hypertension is not under control Source: World Bank 2013 22. The Program plans to address this challenge through four channels: (a) First, public awareness campaigns will be conducted to educate the public on the importance of regular hypertension treatment for preventing death and disability. Campaigns will also promote lifestyle changes that can be used to manage or prevent hypertension, such as exercise and dietary change. As there are gaps in identification of hypertensive people, awareness campaigns will emphasize on the need for regular health checks. Specific campaigns will seek to improve treatment regularity among high risk groups, such as men, younger age groups and those without symptoms. 4 New Research: Irregular Treatment of Hypertension in Moldova. HITT-CIS, 2013. 5 Roberts, B, Stickley, A, Balabanova, D, Haerpfer, C, McKee, M (2012) “The persistence of irregular treatment of hypertension in the for mer Soviet Union� Journal of Epidemiology and Community Health. 8 (b) Second, to improve the quality of hypertension management, the Program will ensure that clinical guidelines (which have already been developed and reflect best practices in treatment) will be followed fully by providers. Health care providers will be made aware of the low rates of adherence of patients with hypertension, and receive training on counselling patients in a non-judgmental manner to help patients adhere better to treatment schedules. Performance-based incentives scheme for primary care physicians will be revised to include indicators related to hypertension management. Patients will be taught how to deal with missed doses, and what to do if drugs have adverse effects. (c) International studies show that reducing copayment levels would significantly increase the level of compliance with antihypertensive medication 6 . The Program will provide the incentive for the Government to (i) reinforce hypertension screening and management interventions by providers as per primary care protocols and mandates (ii) further reduce the co-payment level for anti-hypertensive drugs to achieve better results in hypertension control. (d) Such Program activities are technically sound and in line with international best practices in management of hypertension C2. Strategic Relevance and Technical Soundness of the Program Regarding Hospital Efficiency Hospital rationalization The situation 23. Hospital inefficiency is a major challenge in Moldova, with 73 hospitals serving a population of 3.6 million. The number of hospital beds for acute care is significantly higher than comparator countries (Figure 4). Hospital services account for more than 50 percent of total public expenditures on health, threatening sustainability of funding. In the past, there was considerable progress in hospital reform coinciding with the deepest economic and fiscal crisis of the country in 1997-98. The number of public hospitals was reduced from over 300 after independence in 1992 to 73 in 2012, and the number of hospital beds halved during the same period. Moreover, hospitals have been granted more managerial autonomy and there has been a gradual shift from a line-item based payment method to Diagnostic- Related Groups (DRGs) in all hospitals. However, recent progress has been slow and there remains much work to be done on the hospital agenda. 6 Copayment Level and Compliance with Antihypertensive Medication: Analysis and Policy Implications for Managed Care. Tara et el. (Am J Manage Care. 2006; 12:678-683). 9 Figure 4: Hospital beds for acute care per 100,000 population in selected countries 600 500 400 300 200 100 0 Source: European Health for All database. 24. Firstly, there is still significant overcapacity and duplication of services. Many of these hospitals are in poor physical condition, with dilapidated infrastructure and obsolete equipment. The safety of some laboratory, imaging and anesthetic equipment is of particular concern. This overcapacity negatively impacts on quality of care, as some rural hospitals have such small patient volumes that there are low rates of interventions that require a minimum to maintain skills (for example, Caesarean sections and many surgical operations). Plans to convert inefficient hospitals into long-term/rehabilitation care or outpatient service centers, however, have met with strong opposition by local governments and hospital managers. A detailed National Hospital Master Plan developed in 2009 is yet to be implemented and there is currently no hospital networking. Decisions on the implementation of a hospital regionalization plan are ongoing. 25. Overcapacity is particularly acute in the capital Chisinau, with over 50 percent of the total number of hospitals and 56 percent of beds for a population of less than 700,000. Specialist institutions are concentrated in Chisinau and many of these are organized for single conditions as per the Semashko model (the so called monoprofile hospitals that include trauma, cancer, cardiology, TB, neurology and neurosurgery and dermato-venerology). These monoprofile hospitals are not well suited to modern hospital care based on multidisplinary teams managing patients with multiple conditions. It also leads to significant waste with duplication of support services including imaging, laboratory, and management. The monoprofile hospitals co-exist with municipal hospitals and hospitals affiliated to line ministries, again leading to overlap and duplication that is compounded by weak referral protocols and patient education with the consequence that many patients are treated at a higher level of care than necessary. 26. Secondly, there is an overreliance on hospital care, with underuse of cost-effective, preventive services that would prevent hospital admissions. Figure 5 displays the proportion of the adult population in several ECA countries and the EU-15 that have received an X-ray over the last year compared to the proportion of over 65-year olds that have received a flu vaccine. Although the hospital admission rate for acute care is near the EU-15 average at 17 per 100 population, there is evidence that patients are being treated in hospital for conditions that are managed in primary or ambulatory care elsewhere in Europe. There is also significant variation in admission rates, indicating that some hospitals may be more efficient 10 than others (Figure 6 illustrates the variation in admission rates for hospitals in Chisinau). Average length of stay has fallen, but recently this trend has stalled. There are opportunities to increase the use of day treatment and to use home care and social care to improve the discharge of patients once their treatment is complete. Figure 5: (a) Proportion of the 15+ population that have received an X-ray in the last 12 months and (b) Proportion of the 65+ population that have received a flu vaccine in the last 12 months in selected European countries 11 Figure 6: Hospital admission rates per 100,000 population for hospitals in Chisinau Source: CNAM: unstandardized data. 27. Lastly, the hospital payment system is not being utilized to its full potential to achieve improved efficiency and quality. A DRG system was developed and piloted in nine hospitals in 2012, with subsequent rollout nationwide. However, DRGs are currently being used as a method of recording activities combined with historical budget allocations, rather than as a basis for payments to hospitals. Unit prices have been allowed to remain different for different facilities and are not yet based on actual country data. Increasing the proportion of hospital payments based on DRGs would allow strategic purchasing of hospital services by CNAM to support improved efficiency and quality. International experience/best practice 28. International evidence related to hospital reforms identifies a number of key measures that have been used successfully to achieve an efficient hospital sector7,8,9, including: • Retention of central provision in the capital city or, if appropriate, regional centers where it is necessary to ensure a critical mass of services for reasons of quality, safety, workforce constraints or economies of scale. • Increasing the size of population served by smaller general hospitals to create regional hospitals to ensure high quality and cost-effective care. • Decentralization of some specialist services where this is appropriate and affordable. 7 Improving the hospital system in the Republic of Moldova. World Health Organisation, 2011. 8 Hospitals in a changing Europe. European Observatory on Health Systems and Policies, 2002. 9 Governing public hospitals. Reform strategies and the movement towards institutional autonomy. European Observatory on Health Systems and Policies, 2011. 12 • Creation of multispecialty centers of excellence to bring mono-profile hospitals together. • Strengthening of supporting infrastructure and services. • Orientation of the health care system towards primary and community care and reduce the overall proportion of hospital expenditures. • Strengthening governance mechanisms including autonomy of management structures. • Use of prospective payment mechanisms to incentivize improvements in quality and efficiency of hospital services with shorter lengths of stay, increased day case work and other more efficient practices. 29. While hospital overcapacity and inefficiency is a common problem in the region due to the Semashko legacy, other countries have managed to overcome resistance from multiple stakeholders through innovative approaches in order to achieve ambitious hospital reforms. For example, the hospital sector in Estonia after independence was very similar to the current situation in Moldova. The first response in Estonia was to close many small hospitals followed by the development of a detailed hospital master plan based on performance targets from Sweden. This succeeded in reducing acute hospitals from 68 to 13 and halving the number of acute beds. Factors identified as integral to the success of these reforms included (i) extensive use of mergers; (ii) a coordinated development plan for hospitals in the capital city; (iii) oversight of the change process by a wide group of stakeholders10-11. 30. In Croatia, the average length of hospital stay was 9 days in 2011, compared to 4.5 days in Norway and Sweden, 5.5 in France and the Netherlands, 6.5 in Spain and the United Kingdom. In order to achieve a shorter average hospital stay and so rationalize expensive hospital care, Croatia has started upon an ambitious agenda of hospital reforms. To accelerate these reforms, the MoH recentralized management of hospitals to implement common management structures between groups of hospitals that will facilitate future hospital rationalization and reduction in duplication of services12. 31. Finally, it is agreed that most hospital reform programs tend to fail because of the approach taken to change management rather than the technical design. Unless the leadership, at all levels of the system, convey a clear vision of how the system will look in the future and how different actors need to adapt and change, there is a high risk of failure1. Soundness of the Program 32. In order to incentivize further action in this strategically relevant area, the Program focuses on a set of results and interventions related to: (i) further reduction in acute care beds and hospital admissions for acute care and (ii) strengthening the DRG provider payment mechanism in hospitals. The targets for these indicators have been based on international best practice. 33. The Program pertaining to (i) is based on extensive background technical work carried out with World Bank support, including the National Hospital Master Plan, a Regionalization Strategy, and reports on decentralization of chemotherapy and radiotherapy services. These strategies are based on sound prin- ciples which have been assessed to be in line with those used in large-scale, successful strategies in other countries3. Priority actions arising from this technical work and within the timeframe of the Program include (i) regionalization of hospitals (ii) common management for public hospitals in Chisinau and (iii) 10 Improving the hospital system in the Republic of Moldova. World Health Organisation, 2011. 11 Governance of hospitals – experience from Estonia in past decades. Presentation by Urmas Siigur, Second National Health Forum, 26 November 2013. 12 Croatia Program-for-Results Project Appraisal Document for the Improving Quality and Efficiency of Health Services Project. 13 establishment of a university hospital. However, it has been noted that the first three actions require incorporation into the new national health strategyrationalization strategy, which is currently under development by MoH. Cognizant that international experience has shown that political commitment is vital to successful hospital reforms1-3, the Program encompasses the Parliamentary approval of this strategy. 34. The Program related to (ii) updating of DRGs using country data and their use for hospital payment brings Moldova’s provider payment mechanisms in line with international best practice and is therefore technically sound. 13 This area will require some technical assistance in order to guide the updating process for the Moldovan context and initiation as a basis for hospital payment. Strategic Relevance and Technical Soundness of the Program Related to Performance-based Incentive Scheme The situation 35. While performance-based incentives have been introduced in primary care in Moldova, the current design undermines the potential benefits to efficiency and quality of care. Moreover, the operation of CNAM as a strategic purchaser of efficient and high quality care is hindered by the lack of performance based indicators for hospital payments. 36. In primary care, services have been purchased on a per capita basis since 2004. Initially, primary care services were funded prospectively according to an unweighted capitation estimate, using the resident local population as the denominator. From 2009, the capitation formula has been risk-adjusted for age and based on the number of patients registered at a given practice. From 2013, up to 15 percent of primary care salaries have been paid by CNAM as retrospective “bonus� payments based on a formula that includes 22 indicators of performance. These indicators include provision of TB care, cervical smears, care of women in the first trimester of pregnancy and children in the first year of life. While the achievement of broad support for performance-based payments represents an important milestone, the existing formula rewards volume of care in a few selected areas rather than holistic quality improvement, in particular the need for strong primary prevention for NCDs. Clinical pathways designed to improve quality of care have been introduced in primary care, yet have not been incorporated into the performance-related scheme. Furthermore, additional factors that are highly correlated with patient needs, for example deprivation and rurality, should be taken into account to enable more equitable budget allocation to primary carepractices. 37. Although plans for hospital performance payments were drawn up in 2008, they have never been implemented. Currently, payments to hospitals are based on historical budgets alone, with no recognition of performance related to key quality or patient experience indicators. There is evidence of wide variations in efficiency and quality of care at hospital level. Figure 1 displays the average number of surgeries performed annually per surgeon by rayon, with the number ranging from an average of 32 per surgeon (all specialties combined) in Briceni to 90 in Floreşti. If best performers are taken as a normative target for what is achievable within current health system constraints, performance based payments have the potential to shift efficiency and quality of the worst performers closer to that of the better performers. 13 Diagnosis Related Groups in Europe. European Observatory on Health Systems and Policies, 2011. 14 Figure 7: Average number of surgeries performed in 2011 per surgeon by rayon 100 90 80 70 60 50 40 30 20 10 0 Rayon Source: Sanigest, 2012. 38. The Roadmap 2012–2014 for the Republic of Moldova has among its main objectives the provision of quality services to the entire population. The document emphasizes the need to improve the payment mechanism at the primary care level by taking into account demographic, geographic and performance aspects; to implement systems of quality management in all health facilities; and to implement information systems for the evaluation of health facilities based on quality and performance indicators. International experience/best practice 39. Globally, there is a growing body of evidence in the effectiveness of performance-based incentives in improving results delivered by health service providers in both primary careand hospital settings. A large systematic review assessed 128 studies on pay for performance between 1990 and 2009,14 with encouraging results for the effect of such schemes when the primary purpose was to support minimal quality standards. Evidence to support an improvement in equity and the cost effectiveness of such schemes was also identified, with only a few studies finding negative effects such as gaming or neglect of non-incentivized areas. If pay for performance is intended to boost performance of all providers, however, its capability to do so is confirmed for only a number of specific targets, for example diabetic care. In addition, less evidence was found for positive effects on coordination, continuity and patient-centeredness. 40. In primary care, there can be rapid achievement of quality targets associated with pay for performance. A study assessing the impact of a nationwide pay-for-performance contract for family medicine practitioners in England and Wales found high levels of achievement after one year with respect to 146 quality indicators covering clinical care for 10 chronic diseases, organization of care, and patient 14 Van Herck P, De Smedt D, Annemans L, et al. Systematic review: Effects, design choices, and context of pay-for-performance in health care. BMC Health Services Research 2010, 10:247. 15 experience.15 Another study of a similar national scheme in Scotland found improved NCD and risk factor management with pay for performance schemes and reduced referrals to secondary care.16 41. There is also growing evidence in the effect of pay for performance at hospital level. A large study looked at changes in quality indicators over two years at 613 US hospitals that voluntarily reported publicly information on quality of care and of which one third also participated in a pilot pay for performance scheme 17. The study found that, compared to the control hospitals, pay for performance hospitals showed greater improvement in all composite measures of quality, including measures of care for heart failure, acute myocardial infarction, and pneumonia. Baseline performance was inversely associated with improvement, ie the worst performing hospitals at the start of the period made the most improvement. Overall, pay for performance was associated with improvements ranging from 2.6 to 4.1 percent over the two-year period (see Figure 8). 15 Doran T, Fullwood C, Gravelle H , et al. Pay-for-Performance Programs in Family Practices in the United Kingdom. New England Journal of Medicine 2006; 355:375-384. 16 Simpson CR, Hannaford PC, Ritchie LD. Impact of the pay-for-performance contract and the management of hypertension in Scottish primary care: A 6-year population-based repeated cross-sectional study. British Journal of General Practice 2011; DOI: 10.3399/bjgp11X583407. 17 Lindenauer PK, Remus D, Roman S, et al. Public Reporting and Pay for Performance in Hospital Quality Improvement. New England Journal of Medicine 2007; 356:486-496. 16 Figure 8: Improvement in performance indicators among hospitals engaged in both pay for performance and public reporting and those engaged only in public reporting In an analysis matched for hospital characteristics, pay for performance was associated with improvements in performance indicators ranging from 4.1 to 5.2 percent over two years, including those in four key areas: acute myocardial infarction (Panel A), heart failure (Panel B),pneumonia (Panel C), and a composite of 10 measures (Panel D). The performance rate is the percentage of patients who were given the specified care for the condition. Q denotes quarter. Source : Lindenauer PK, Remus D, Roman S, et al. Public Reporting and Pay for Performance in Hospital Quality Improvement. New England Journal of Medicine 2007; 356:486-496 42. Two comprehensive reviews have noted that the effects of pay for performance schemes vary according to design choices and contextual factors18. In particular, the evidence is not as extensive for low and middle income incomes as for high income countries. The following general recommendations were made for future programs drawing on the available evidence19: (1) select and define performance targets 18 Witter S, Fretheim A, Kessy FL, Lindahl AK. Paying for performance to improve the delivery of health interventions in low- and middle-income countries . Cochrane Database of Systematic Reviews 2012, Issue 2. Art. No.: CD007899. 19 Van Herck P, De Smedt D, Annemans L, et al. Systematic review: Effects, design choices, and context of pay-for-performance in health care. BMC Health Services Research 2010, 10:247. 17 on the basis of baseline room for improvement, (2) make use of process and (intermediary) outcome indicators as target measures, (3) involve stakeholders and communicate information about the programs thoroughly and directly, (4) implement a uniform design across payers, (5) focus on both quality improvement and achievement, and (6) distribute incentives to the individual and/or team level. Soundness of the Program 43. This Program will support the revision of the performance-based incentive scheme in primary care, including (i) performance indicators and (ii) verification methods. The Program builds on existing reforms in primary care with broad support that require modification to bring them in line with international best practice. Strengthening the performance-based incentive scheme in primary careis likely to contribute to improve overall sector efficiency by encouraging more resolution of care in primary care and avoiding hospital admissions. Clinical pathways, already introduced in primary care, will be used as part of the criteria for performance-based financing to promote quality and efficiency of care. Targeted quality indicators are also likely to improve NCD management and their risk factors, thus supporting other areas under the scope of the Program. The government’s plan to introduce verification methods to ascertain the achievement of results and deter potential cheating by providers is in line with common practices of performance-based incentive schemes elsewhere. The Program is therefore strategically relevant and technically sound with regard to primary care. 44. Drawing on the evidence for the effectiveness of performance-based incentives in secondary care, the Program will also support the introduction of pay for performance mechanisms in hospitals. Here, the Program is supporting the implementation of previously developed work and the strengthening of CNAM as a strategic purchaser, again in line with the international evidence base. In this context, the introduction of the performance-based incentive scheme for hospitals supported by the Program is expected to enhance hospital efficiency and quality, and is therefore strategically relevant and technically sound. D. Institutional Arrangements and Capacity 45. The two most critical stakeholders involved in implementing the proposed Program are the MOH and CNAM, with responsibility for the Program scope of activities rather equally divided between MOH and CNAM. In the context of the proposed Program, the MOH and CNAM are the primary beneficiaries that will be responsible for using the funds available through the proposed Program to implement the healthcare reforms according to priorities and directions defined in the National Healthcare Development Strategy 2007–2021. Their commitment to the Program activities is high as the scope of activities is based on nationally identified priorities and/or ongoing reforms supported under the previous World Bank project. 46. The MOH is well placed and empowered to carry out its stewardship functions for the health sector. Its ability to work across sectors and collaborate with various stakeholders in health has been steadily increasing. The MOH therefore is able to design and implement national-level interventions related to the Program. However, three is still room to further strengthen MOH’s capacity in some strategic areas, including (i) implementation of hospital rationalization measures (ii) communication with stakeholders to galvanize their support for key health reforms, and (iii) monitoring and evaluation of results. One of the key challenges in MOH institutional capacity is high staff turnover. For instance, the Policy Analysis and Monitoring unit in the Ministry of Health is currently missing two thirds of its staff. Continued high turnover may affect institutional memory and the achievement of the Program’s results. 18 47. NCPH, as a subordinated agency of the MOH, has increased its mandate in recent years beyond the Soviet-era sanitary-epidemiological functions to include the development and implementation of population-level health promotion and disease prevention activities (including NCD control). Nevertheless, NCPH still needs to strengthen its analytical capacity as well as the ability to conduct well- coordinated public information campaigns. 48. CNAM is the sole purchaser of public healthcare services responsible for revenue raising through mandatory health insurance and contracting providers in line with a defined benefits package. CNAM plays a key role in the achievement of PforR results related to healthcare efficiency and financial protection of population, although as an independent agency directly subordinated to the Government it must coordinate extensively with MOH. In particular, CNAM is responsible for review and deepening of the antihypertensive drugs benefits (in consultation with the MOH) and ensuring its adequate funding. By further strengthening of the DRG payment mechanism to hospitals, CNAM will ensure transparency and efficiency in financing healthcare providers, as well as provide better evidence on the efficiency and quality of healthcare treatment. Although making hospitals accountable for quality of care is a broad function that involves many stakeholders (the MOH, accreditation agency), CNAM plays a major role by using its contracting leverage to stimulate hospitals and primary care providers to produce the desired results. 49. CNAM has significantly built up its capacity since the introduction of social health insurance in 2004. By and large, CNAM has been able to fulfill its mandates of contracting services with providers. The on-going performance-based incentive program in primary care is a good example of CNAM capacity in service purchasing. It also provides a good foundation for expansion of this approach to the hospital sector. However, CNAM still needs to further build its capacity to (i) strengthen provider payment mechanisms, (ii) regularly analyze utilization data, identify cost outliers and address such cases, and (iii) negotiate with and contract providers as well as hold them accountable for results in a more effective manner. Currently CNAM is going through a restructuring in order to strengthen its core functions of strategic purchasing of health services, while ensuring effectiveness, efficiency and cost control of its internal organization. 50. The MOH and CNAM are committed to drawing on technical assistance to build capacity in key areas identified above in order to strengthen institutional capability to implement the Program and proposed Program Action Plan. These capacity building measures are placed into the context of the regular program of training and capacity building in each institution 51. An adequate governance structure and coordination arrangements are in place to implement the Program. Coordination mechanisms for the MOH and CNAM to collaborate with each other as well as to coordinate with country stakeholders and external development partners are well established. At an executive level, CNAM’s Administrative Council includes a representative from the Ministry of Health along with other relevant stakeholders. A Steering Committee will be established to provide project oversight and guidance. The Committee will meet at least twice a year. 52. Overall, the MOH has moderate to high capacity to undertake the activities included in the Program, but will benefit from capacity building measures in the areas of communications and monitoring and evaluation. NCPH has high capacity for its related activities under the Program, however will need technical assistance to carry out special surveys to monitor DLI 1 and 2 (see below). CNAM has moderate to high capacity for its related activities under the Program, however will require technical assistance for certain technical studies related to Program activities (see Annex 8). 19 E. Assessing the Program Expenditure Framework 53. As evidenced by budget allocations, the GOM is highly committed to health. During the last five years public expenditures on health amounted to no less than 5 percent of GDP and at least 12 percent of total government expenditures. This trend is expected to continue in the medium term during the Program’s implementation. However, further increases in the health sector’s share in government budget are not likely. The 2013 government health budget was 1,415 MDL per capita (77 MDL more than in 2012). The largest part of expenditure was for individual health services (80.5 percent), followed by public health (8 percent) and health system resources investment (7.4 percent). MTBF forms the cornerstone of the expenditure framework and its quality has been gradually increasing. 54. As noted in the 2011 assessment based on Public Expenditure and Financial Accountability (PEFA) methodology, Moldova scores relatively well on budget credibility, comprehensiveness and classification, treasury operations (including budget, payroll and expenditure controls), in-year reporting and public access to government budget and financial information. With the World Bank support (under the BOOST initiative20), Moldova became one of the few countries in the world to publish budget execution data at the item level for all institutions on the internet. The BOOST database, available on the MOF website, is a useful analytical tool to monitor the efficiency and transparency of the use of public funds. In line with the new Classification of Programs approved by the MOF, starting in 2014, the Sector Strategy of Expenditures in Health will cover 17 subprograms, by priority intervention areas. 55. The Program’s estimated cost is reasonable and represents around 18 percent of the estimated cost of the total government expenditures for the health sector. The majority of Program’s cost is recurrent. F. Program’s Results Framework and Monitoring: 56. In order to monitor progress toward achieving the PDOs, the Program’s Results Framework details four PDO-level Results Indicators and tenIntermediate Results Indicators, all of which were agreed with the GoM to monitor progress of the Program. In particular, the intermediate results DLIs have been chosen to address the bottlenecks along the results chain that require incentivizing, with both ambition (“stretch�) and feasibility (“realism�) taken into account. Four PDO indicators and all Intermediate Results Indicators were selected as Disbursement-Linked Indicators (DLI) because of their ability to measure and incentivize key intended changes in the health sector. Surveys 56. Two PDO-level indicators will be monitored through two rounds of a special survey. These pertain to the monitoring the prevalence of adult smoking (DLI 1)) and the percentage of adults with hypertension under control (DLI 2). 57. The STEP-wise approach to chronic disease risk factor surveillance is a survey designed to be an entry point for low and middle income countries to begin NCD surveillance activities while strengthening country monitoring capacity. The survey has three “steps� for monitoring risk factors such as blood pressure and cholesterol levels: questionnaire, physical measurement and biochemical (biomarker) measurement. 20 BOOST is not an acronym. It is the name of a new data tool developed by the World Bank to help enhance public sector performance 20 58. The first STEPS survey was conducted in November 2013 by NCPH with the technical support of WHO. This survey included the questionnaire and physical measurement elements, but not biochemical measurements. Evaluation of this survey is still ongoing, but initial reports from WHO indicate that data collection was undertaken successfully and data quality meets the required standards. 59. This “streamlined� survey will be repeated in Year 2 and Year 4 of the Program. The survey will be commissioned by MOH and coordinated by NCPH. The data collectors trained for the 2013 survey are members of the NCPH staff, and thus a pool of specifically trained field staff will be available for the two survey rounds. Routine monitoring systems 57. Existing administrative information systems in MOH and CNAM have full capacity to monitor the remaining Program indicators. The remaining two PDO-level indicators, reduction in the number of annual hospital admissions for acute care (DLI 3) and reduction in the number of acute hospital beds (DLI 4), are collated and reported by the MOH’s Department of Monitoring and Evaluation on an annual basis. 58. The Results Framework includes several indicators that can be more efficiently monitored through electronic reporting, in order to incentivize the institutionalization of these systems in CNAM and MOH and build upon the activities of the previous World Bank Project. 59. The remaining intermediate indicators (mainly Yes/No) rely on self-reported data from CNAM and MOH. Independent entities will be contracted to follow agreed verification protocols to confirm achievement of submitted results. Given its stewardship role, the MoH will ultimately be responsible for monitoring progress on these indicators and for ensuring timely collection and reporting of monitoring data and provision of necessary verification documents to the World Bank and MoF. 60. As part of monitoring progress towards the PDO, the Bank team will conduct regular implementation support missions based on the detailed Implementation Support Plan , whose focus would be on timely implementation of the agreed Program Action Plan , provision of necessary technical support, conduct of fiduciary reviews, and verification of results, where appropriate. G. Economic Analysis 61. The proposed PDO is to contribute to reducing key risks for NCDs and improving efficiency of health services in Moldova. Therefore, the development impact considered in the economic analysis of the Program is the improvement in the population health status as well as the efficiency gains deriving from the interventions. The analysis is inclusive of the benefits of the IPF component, since it mainly consists of technical assistance to achieve the main benefits of the PforR component. 62. Costs of the Program: The costs of the Program are considered the entire costs of the Government supported through the PforR operation of the Bank. (i.e. the five sub-programs from the 2008-2017 NHSDS chosen for support through the PforR). These costs amount to about US$132.20 million from 2014 to 2017, US$30.8 million (28.0 million PforR and 2.7 million IPF) of which are Bank funds. Since the majority of Program costs is expected to be recurrent, the cost-benefit analysis is based on the assumption that the entire yearly Program costs (inflation-adjusted) will accrue as recurrent costs beyond the end of the NHSDS implementation phase in 2017 until 2030 (the time period considered in the cost-benefit analysis). This assures that the proposed NPV and IRR estimations are conservative. 21 63. Benefits of the Program: The benefits from a higher share of patients controlling their high blood pressure will be measured through Disability Adjusted Life Years (DALYs) averted, which represent the sum of years of potential life lost due to premature mortality and the years of productive life lost due to disability. The revision of the outpatient drug benefit package in order to increase the reimbursement rate of generic anti-hypertensive drugs to 70 percent will not be considered separately. Rather, it is assumed that this intervention will help to achieve the target of a higher control rate of hypertension among the population. The benefits from this intervention are therefore already incorporated through the estimated reduction in DALYs caused by hypertension. 64. The beneficial effect of more efficient primary healthcare services (through a performance-based incentive scheme) as well as the improved efficiency and quality of care in hospitals (through a performance-based incentive scheme in hospitals) will be estimated using the impact on population health status measured in terms of the DALYs averted. 65. The benefits from the reduction in smoking prevalence among young people and the ensuing reduction in DALYs related to smoking will not be accounted for. The effects of tobacco consumption are cumulative and take place over time. Hence the current burden of disease reflects the consumption of tobacco in previous decades and not in current ones.21 Changes in tobacco-related mortality follow the changes in prevalence and consumption only a significant period of time (17-20 years) after tobacco consumption has been restrained. There are (substantial) long-term benefits from policy interventions aimed at reducing tobacco consumption, but they will not accrue in the period of time (2014-2030) considered in this analysis. The achievement of greater efficiency in the hospital sector is linked to two of the key results of the Program, namely the reduction of the number of acute hospital beds from 17,586 to 15,000 per 100,000 persons and the reduction of annual hospital admissions from 17.6 to 15.6 per 100 persons by the end of the Program in 2017. The expected savings related to the reductions in acute care beds and annual admissions constitute the first source of monetary benefits considered in this analysis. The update of the DRG payment mechanism in hospitals can be expected to contain the future growth of public expenditures for acute hospital care (e.g. growth through medical technology improvements). However, this cost containment is not expressively being modelled in the performed cost-benefit analysis to allow for a more conservative estimation of the overall benefits from the hospital reform interventions. 66. The additional benefits deriving from reduced out-of-pocket expenditure among the poorest 40 percent of the population mainly consist in better equity outcomes of the health system. The possibly beneficial impact on population health outcomes due to less self-rationing and underutilization (in particular of cost-effective preventive treatment of NCDs) among the poor is considerable, but is not part of the considered benefits, since it is hard to monetize. Furthermore, the establishment of autonomous hospital networks is not expressively considered as a benefit either, but rather considered to support the overall reforms initiated by the Program. Therefore the estimated profitability of the Program (i.e. NPV and IRR) is likely to be conservative. 67. The assumptions used in the cost-benefits analysis are listed below: • Basic discount rate. Financial costs and financial savings are discounted at 8 percent to account for inflation (5 percent)22 and the time value of money (TVM) (3 percent). A higher discount rate of 11 percent (reflecting a 6 percent discount rate accounting for the TVM) is also applied to 21 WHO: The global burden of disease: 2004 update, 2008. 22 The IMF World Economic Outlook, October 2013 estimates an average inflation rate of 4.79 percent for the 2013-2018 period. 22 verify the sensitivity 23 of the results to this assumption. A higher inflation estimate would decrease the present value of the financial costs (increasing the estimated benefit as measured by the NPV) more than the present value of the benefits 24 and is therefore not considered in a sensitivity analysis. • Period of time considered. The cost-benefits of the interventions are calculated over the 2014- 2030 period in order to account for the long-term benefits in terms of reduced DALYs from the interventions aiming at: (i) a better control of conditions of hypertension and (ii) a higher efficiency and quality of care at both the primary and secondary level. • Population covered. In general it is assumed that all interventions will be implemented nationwide. The interventions targeting a better control of hypertension only affect a subgroup of the population. 25 Other than that, the interventions will affect health results for the entire population, (around 3.6 million people in 2012) or the efficiency level of all facilities. Population growth up to the year 2030 is based on the WB HNP Statistics. • Expected disbursements of investments. When discounting the financial costs of the Program, it is assumed that the funds provided by both the Government and the Bank are disbursed according to the estimated disbursement schedule (see Program Financing Data). • Benefits of reforms beyond 2030. Although the benefits from the Program will likely persist beyond Program completion in 2030, benefits beyond the year 2030 are not accounted for in this analysis due to the increasing uncertainty about the counterfactual scenario without the Program. This approach contributes to the overall conservativeness of the estimated benefits. • Due to the inter-correlation of the supported interventions and the PforR nature of the project (making disbursements for a given intervention less related to the actual cost of implementing this particular intervention), the NPV and IRR are only calculated once for the whole package of interventions. Expected Benefits from Hospital Reforms 68. The calculation of the benefits from the reduction in the number of acute care beds and the reduction in annual admissions is based on the following additional assumptions: • The annual cost per bed equals hospital costs divided by total number of beds.26 Costs for the hospital sector are assumed to grow by 5 percent (in order to account for future expected inflation) plus 1 percent (in order to account for price increases due to improvements in medical technology). • Fixed costs account for 75 percent of total hospital sector costs and variable costs account for the remaining 25 percent.27 23 The estimated benefits are sensitive to a higher TVM, since it decreases the present value of the more distant benefits from averted DALYs. The related reforms are long-term oriented, and it takes some time until benefits materialize. 24 The benefits of the interventions to reduce the smoking prevalence among young people are not considered. 25 World Bank. 2008. “Romania Human Development Program Knowledge Note - Health Sector.� World Bank, Washington, D.C. The evidence in table A4.1 reflects this estimate. 26 Roberts RR, Frutos PW, Ciavarella GG, Gussow LM, Mensah EK, Kampe LM, Straus HE, Joseph G, Rydman RJ JAMA 282(7) 630. Distribution of variable vs. fixed costs of hospital care and see: http://www.post- gazette.com/business/Biz-opinion/2013/10/06/Regional-Insights-How-hospitals-are-driving-up-the-cost-of-health- care/stories/201310060112. 27 Marc Suhrcke, Rachel A. Nugent, David Stuckler and Lorenzo Rocco. Chronic Disease: An Economic Perspective. London, 2006: Oxford Health Alliance. 23 • The reduction in acute care beds leads to a proportionate reduction in the overall costs of hospital costs, whereas the reduction in annual hospital admissions only leads to a proportionate reduction in the variable costs of the hospital sector. • The counterfactual in the absence of the Program is that the number of acute hospital beds and the admission rates do not change until 2030. Table 4. Benefits from the Reduction in the Number of Acute Hospital Beds and in Annual Hospital Admissions (USD ’000s) 2014-2030 2014-2030 Discount Factor Benefits from Acute Bed Reductions Benefits from Annual Admissions Reductions (in US$) 3% 249,596 USD 82,788 6% 199,957 USD 68,402 Expected Benefits from More Efficient Primary Healthcare Services, Improved Efficiency and Quality of Care in Hospitals, and a Better Control of Hypertension among the Population 69. The benefits deriving from the above-listed interventions are estimated using the impact on the population health status measured in terms of DALYs from NCDs (The proposed PDO of the PforR is to reduce risks for non-communicable diseases. Although the interventions aimed at increasing the efficiency and quality of care at healthcare institutions could also contribute to a reduction in the burden of disease due to injuries or communicable diseases, these potential benefits are not taken into account). 70. The additional assumptions made in the economic analysis of these interventions are: • Reduction in DALYs: DALYs, which represent the sum of years of potential life lost due to premature mortality and the years of productive life lost due to disability, have a built-in age weighting and discount rate of 3 percent. The reduction in DALYs from the integrated interventions on healthcare efficiency and quality supported by the Program are set at 0.3 percent across all NCDs and expected to become effective in 2020 after the completion of the Program. The adverse health effects of hypertension are estimated to account for 20 percent of the global burden of disease.28 Assuming that this proportion is representative of the situation in Moldova as well, an increase of the rate of cases of hypertension being under control from 10 to 15 percent over the course of the Program would reduce the burden of disease and hence DALYs in Moldova by 1 percent. It is assumed that the rate of hypertension cases under control increases according to the specified DLI target rates. To allow for a delayed evolution of the beneficial impact, it is assumed that the stepwise reduction in DALYs due to hypertension sets in with a lag of three years. • Counterfactual Scenario for DALYs: The baseline DALYs were calculated for the various conditions from WHO estimates for the Eastern and Central Asia region, adjusted for the population size of the project (around 3.6 million people) and the age structure of Moldova (from the WB HNP Statistics). These include the forward projections of DALYs averted (that is, healthy life years gained) from 2013 to 2030. • Valuation of DALYs used a very simple rule. Each DALY saved is valued at yearly per capita income (using a starting value of about US$2387 for 2014). The Disease Control Priorities Project and Copenhagen Consensus guidelines mention three times per capita income as a still 28 See: D. Jamison, P. Jha, and D. Bloom, “Copenhagen Consensus 2008 Challenge Paper: Diseases,� 2008; http://www.givewell.org/files/DWDA%2 02009/Stop%20TB/Copenhagen%20Consensus%20Paper-Diseases.pdf. 24 conservative estimate for the value of each DALY averted.29 Studies of valuation of life in the United States even utilize much higher values for a year of life that would produce more extreme results. • Discount Rates for DALYs: The monetary value of future stream of health benefits (i.e., annual DALYs saved) is discounted at 3 percent (a higher rate of 6 percent is used for the sensitivity analysis), per guidelines from WHO and the Disease Control Priorities Project.30 • GDP Growth: An annual growth rate of 4.5 percent in real per capita GDP is used, being slightly more conservative than the estimates provided by the IMF. 31 The reduced growth rate of 3 percent is used for the sensitivity analysis. Figure 9. Total DALYS averted by year compared to Europe and Central Asia counterfactual projections, baseline scenario 9000 8000 7000 6000 DALYs averted 5000 4000 3000 2000 1000 0 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Year Table 5. Benefits from DALYs averted (USD ’000s), baseline scenario Discount 2014-2017(US$) 2018-2030 (US$) Factor (%) Benefits from DALYs Benefits from DALYs averted averted 3 3,366 272,889 6 4,009 202,460 Expected benefits from the interventions 71. Table 4 presents the NPV and the estimated IRR of the considered interventions. The sum of costs and benefits (i.e. the NPV of the interventions) is largely positive and the estimated IRR ranges 29 The IMF World Economic Outlook, October 2013 estimates a growth rate of 4.8 percent for the 2014-2018 period. 30 Net of inflation. 31 Net of inflation. 25 between 6.13and 9.10 percent, depending on the GDP growth rate used, which clearly shows the positive development impact of the considered Program interventions. Table 6. Net Present Value and Internal Rate of Return of the Program (USD ’000s) for different GDP growth scenarios TVM Baseline GDP growth Low GDP growth Discount Factor NPV IRR32 NPV IRR33 3% 70,005 USD 31,467 USD 9.10% 6.13 % 6% 31,189 USD 3,446 USD The rationale for public sector involvement 72. The Program does not involve the introduction of new programs that could be alternatively implemented by the private sector. Instead, it aims to strengthen existing government programs (NCD prevention and management, performance-based contracting with service providers, rationalization of hospitals in Chisinau, etc.) and ongoing reforms in order to alleviate binding constraints to public sector performance. The case for public provision and financing of interventions under the Program is therefore justified. H. Implementation Support 73. This Implementation Support Plan is in line with the Program-for-Results operational guidelines. The Borrower is in charge of the implementation of all Program activities in support of achievement of the agreed DLIs, as well as of elimination of inefficiencies/bottlenecks identified in the social, environment and fiduciary assessments. The Bank will tailor implementation support to ensure the following: • Provide technical advice to the implementation of HTP Program Action Plan, the achievement of DLIs and elimination of other social, fiduciary or governance-related bottlenecks relevant to the Program; • Review program implementation progress, verify achievement of DLIs, review program progress reports, audit reports and such other relevant information; • Monitor health system performance with particular emphasis on the program result areas and monitoring compliance with legal agreements, keep records of risks and propose remedy actions to improve program performance, if and as needed; • Provide support in resolving any operational issues pertaining to the project, including review of grievance redress mechanisms 74. In particular, the Task Team will be working with the Government and other key stakeholders in the following main areas: (a) Monitoring and evaluation: the Team will provide technical support for building capacity at MoH and CNAM for monitoring Program results (DLIs) and verifying the achievement of DLIs in line with the agreed protocols. For the PDO-level indicators monitored through the STEPS survey , the Task Team will ensure that adequate internal quality control 32 Net of inflation. 33 Net of inflation. 26 mechanisms are in place for accurate and trustworthy results reporting. The Team will co- operate with WHO to ensure standard methodology for surveys is applied, or to seek any ad hoc capacity strengthening support, if needed. For the DLIs monitored through a survey conducted at healthcare facilities, an independent entity will be contracted to perform a random audit in order to validate the submitted results. In case the DLIs are achieved partially, the Bank’s team will review the submitted documentation, request any other supporting information to verify the achievement of the reported DLIs (from MoH, CNAM or other counterparts) and will then communicate in written to MoH/CNAM and MoF on the achievement of particular DLI informing about the level of financial proceeds for disbursement or partial disbursement of each particular DLI. (b) Environment and social: in relation to environmental issues, the team will focus on the medical (and pharmaceutical) waste management capacity building (regulatory and institutional), although the Program does not support any specific activity that could have a direct link to it. Regarding social, the task team will focus on the following main areas: (i) change management and mitigation of resistance from health system players (ii) social inclusion and equity in access to healthcare services (iii) Social Accountability, Transparency and Responsiveness in the healthcare. Given that cca 40% of DLIs will have potential impact on employees and the way the healthcare services are rendered, the team will also pay special attention to the effect of specific DLIs implementation on the medical (and non-medical) staff within the healthcare system. (c) Fraud and corruption: the Task team will monitor implementation of fraud and corruption mitigation measures and will stay abreast of any developments that could potentially create conflict of interest situations thus providing guidance in resolving any emerging issues or early prevention of fraud and corruption practices. (d) Procurement: the project is of hybrid type which entails two-fold task for the procurement. First, the procurement filter is being applied at the assessment stage to identify country system risks related to implementation of project-related activities by the Borrower in a transparent and accountable manner; secondly, the SIL TA component will apply World Bank procurement rules and procedures to allow procurement of project-specific services that are key for achievement of project results and that otherwise would be difficult to contract out. In particular, the Bank team will ensure that terms of references are non- restrictive, well-defined and are in line with the existing needs. The team will use the project to transfer knowledge to MoH and CNAM on the procurement and contract management and will help to install adequate contract administration and monitoring system, evidence of contract performance in terms of time, quality and cost, enforcement of contractual remedies in cases of contract’s non-compliance. (e) Financial management: the Bank’s team will focus its support around (i) strengthening of internal financial control mechanisms, particularly the internal audit departments of MoH and CNAM (ii) implementation of Program external audit (ii) ensuring regular flow of funds to MoH and CNAM for implementation of project-related activities. (f) Inter-institutional cooperation and coordination: The Bank team will, wherever necessary, ensure that there is a good inter-institutional cooperation between MoH, CNAM and all the other relevant project counterparts (Mayor’s health department, Medical University, LPAs, etc).In order to mitigate the risk of resistance to reforms and raise awareness about some specific healthcare challenges, a well-targeted, proactive information 27 campaign will be held in the first six months of effectiveness to inform medical personnel and patients with the reforms in the sector . 75. Furthermore, the Task Team will continue to stay engaged in the policy dialogue on health reform agenda with the Ministry of Health and CNAM, as well as with other development partners, particularly WHO. The major project risks concern the capacity of the Ministry of Health to absorb the assistance quickly and efficiently and the willingness of the hospitals staff to accept new working practices based on performance. However, by the time when the proposed PforR operation becomes effective, the World Bank will complete its ongoing SIL “Health Services and Social Assistance project� (HSSAP). Thus, the small team of consultants currently working under HSSAP can continue providing their assistance thus ensuring continuity to the implementation support and knowledge transfer. 76. Key of the Bank’s implementation support team members (technical, fiduciary, M&E and social systems), are either based in the Country Office or in the Region (mostly Kiev, Ukraine). This will help to ensure timely, efficient, and effective implementation support to the MoH and CNAM. Table 7: Main Focus of Implementation Support Time Focus Skills Needed Resource Estimate Partner Role First twelve Capacity strengthening Technical, M&E, Two full missions, WHO (local office) months in implementation of fiduciary first one to launch and other partners MoH and CNAM for the project and will also be invited DLIs monitoring and second one to to join the missions’ verification of results supervise and key meetings provide implementation support 12-48 months Timely implementation Technical, social, Regular mission WHO and other of program action plan M&E, fiduciary every six months. partners will be and elimination of invited to participate bottlenecks for in policy dialogue achievement of Program results Mid-Term Revisiting the Program Technical, Beginning of 2017 WHO and other Review design and making fiduciary, social and partners will be necessary adjustments environmental, invited to attend key based on the M&E meetings implementation progress and its likelihood to achieve the PDO within the project period Table 8: Task Team Skills Mix Requirements for Implementation Support Number of Staff Weeks Number of Trips Skills Needed Comments (per year) (per year) Task Team Leader 10 2-3 HQ based Technical specialist 36 2 International M&E consultant 6 1 Country-based Financial Management 4 1 Country-based Procurement 4 1 Country-based Social 3 1 National 28 29