at SciVerse ScienceDirect

Social Science & Medicine 96 (2013) 223e231

Contents lists available

Social Science & Medicine

journal homepage: www.elsevier .com/locate/socscimed

Strengthening institutional and organizational capacity for socialhealth protection of the informal sector in lesser-developed countries:A study of policy barriers and opportunities in Cambodia

Peter Leslie Annear a,*, Shakil Ahmed a,1, Chhun Eang Ros b,2, Por Ir c,3

aNossal Institute for Global Health, The University of Melbourne, 4th Floor, Alan Gilbert Building, 161 Barry St., Carlton, 3010 VIC, AustraliabDepartment of Planning and Health Information, Ministry of Health, 151-153 Kampuchea Krom Ave, Phnom Penh, CambodiacNational Institute of Public Health, 2 Street 289, Khan Toul Kork, Phnom Penh, Cambodia

a r t i c l e i n f o

Article history:Available online 16 February 2013

Keywords:Universal coverageSocial health protectionInstitutional developmentCambodia

* Corresponding author. Tel.: þ61 3 9035 4065,fax: þ61 3 9347 6872.

E-mail addresses: pannear@unimelb.edu.auunimelb.edu.au (S. Ahmed), chhuneang@online.cyahoo.com (P. Ir).

1 Tel.: þ61 413 784 936 (mobile).2 Tel.: þ855 23 426 372, þ855 12 855 735 (mobile)3 Tel.: þ855 23 362 342, þ855 12 657 725 (mobile

0277-9536/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.socscimed.2013.02.015

a b s t r a c t

Reaching out to the poor and the informal sector is a major challenge for achieving universal coverage inlesser-developed countries. In Cambodia, extensive coverage by health equity funds for the poor hascreated the opportunity to consolidate various non-government health financing schemes under thegovernment’s proposed social health protection structure. This paper identifies the main policy andoperational challenges to strengthening existing arrangements for the poor and the informal sector, andconsiders policy options to address these barriers. Conducted in conjunction with the CambodianMinistry of Health in 2011e12, the study reviewed policy documents and collected qualitative datathrough 18 semi-structured key informant interviews with government, non-government and donorofficials. Data were analysed using the Organizational Assessment for Improving and StrengtheningHealth Financing conceptual framework. We found that a significant shortfall related to institutional,organisational and health financing issues resulted in fragmentation and constrained the implementa-tion of social health protection schemes, including health equity funds, community-based health in-surance, vouchers and others. Key documents proposed the establishment of a national structure for theunification of the informal-sector schemes but left unresolved issues related to structure, institutionalcapacity and the third-party status of the national agency. This study adds to the evidence base onappropriate and effective institutional and organizational arrangements for social health protection inthe informal sector in developing countries. Among the key lessons are: the need to expand the fiscalspace for health care; a commitment to equity; specific measures to protect the poor; building nationalcapacity for administration of universal coverage; and working within the specific national context.

� 2013 Elsevier Ltd. All rights reserved.

Introduction

All countries regardless of income level can set out on the pathto universal coverage (UC) through a mix of different pre-paymentand risk-pooling mechanisms, using a combination of tax-fundingand social health insurance (SHI) (Antunes & Saksena, 2009;Carrin & James, 2004; Chan, 2010; Evans & Etienne, 2010; OECD,

þ61 410 561 189 (mobile);

(P.L. Annear), shakila@om.kh (C.E. Ros), irpor@

.).

All rights reserved.

2009). While achieving UC in lesser-developed countries (LDCs)may be more difficult, many countries that recently moved toUC began the process when they were at an earlier stage of eco-nomic development (Mills, 2007; Tangcharoensathien, Prakongsai,Limwattananon, Patcharanarumol, & Jongudomsuk, 2007). There is,however, a gap in country-level evidence and understanding of theinstitutional and organizational arrangements that are needed inLDCs during the transition to UC (Carrin, Mathauer, Xu, & Evans,2008).

Generally, in LDCs the income tax base is narrow, even whereeconomic growth is strong, and revenue collection for social healthprotection (SHP) is constrained (Carrin, Waelkens, & Criel, 2005).The formal employment sector (government and private enter-prise) is small, populations are overwhelmingly rural, the informalsector (self-employed and subsistence) typically comprises morethan half the population and poverty is extensive. Nominally ‘free’

P.L. Annear et al. / Social Science & Medicine 96 (2013) 223e231224

government services may not reach the poor (who face additionalbarriers to accessing these services), and poorer households areoften unable to meet the costs of health insurance premiums(Tangcharoensathien et al., 2011).

Consequently, universal coverage schemes that focus primarilyon the formal sector through compulsory insurance, without tar-geted schemes for the poor, risk creating a health financing systemthat is more inequitable (Gwatkin & Ergo, 2011; WHO, 2008).Allotey, Verghis, Alvarez-Castillo, and Reidpath (2012) and Allotey,Yasin et al. (2012) argue that to be truly universal, coverage must beextended to all population groups regardless of individual eco-nomic, social or geographic position. However, expanding the riskpool for voluntary and community-based insurance schemes in theinformal sector faces significant cultural and economic constraints(Chankova, Sulzbach, & Diop, 2008;Mills, 2007; Nguyen & Knowles,2010; Pauly, Blavin, & Meghan, 2008; Poletti et al., 2007; Thorntonet al., 2010). Generally, the poor and the informal sector have low,irregular, non-taxed incomes, are dispersed geographically, andmay be difficult to identify as beneficiaries (Canagarajah &Setharaman, 2001).

What then are the institutional, organisational and healthfinancing arrangements required to provide social health protec-tion for the poor and the informal sector? And to what extent areLDC governments ready to meet these challenges? In practice,broad and inclusive approaches are needed along with targetedschemes for hard to reach population groups, including measuressuch as direct subsidies (health equity funds, vouchers, cashtransfers), voluntary affiliation to formal-sector national insuranceschemes, public subsidies to enrol the poor or informal workers, ormandatory participation (Ahmed & Khan, 2011; Ekman, Nguyen,Ha, & Axelson, 2008; Gottret & Schieber, 2006; Ir, Heang, & Sim,2010; Ir, Horemans, Souk, & Van Damme, 2010; Mathauer,Schmidt, & Wenyaa, 2008; Morris, Flores, Olinto, & Medina, 2004).

One consequence of implementing targeted schemes within amultiplicity of SHP arrangements is fragmentation of administra-tion and funding (Carrin et al., 2008; Li, Yu, Butler, Yiengprugsawan,& Yu, 2011; McIntyre et al., 2008; WHO, 2005). Creating uniformand effective systems therefore depends on effective planning andmaintaining a balance of power among health-system stakeholders(El-Idrissi, Miloud, & Belgacem, 2008). The need for governmentintervention in addition to social-health or other insurance marketarrangements has been recognised (Allotey, Yasin et al., 2012; Chua& Chee, 2012; Mathauer, Cavagnero, Vivas, & Carrin, 2010). Thechallenge is to develop regulatory and financing mechanisms byengaging with the political process, build government adminis-trative and technical capacity, include all stakeholders with cross-sectoral coordination and recognise patient views of quality andaccessibility (Allotey, Verghis et al., 2012).

Social health protection in Cambodia

In Cambodia, out-of-pocket spending is almost two-thirds oftotal health expenditure; more than one-quarter of the populationlive below the poverty line and are, by definition, unable to pay forhealth care, and debt for health care is known to be amajor cause ofimpoverishment (Van Damme, van Leemput, Ir, Hardeman, &Meessen, 2004). Removing financial barriers at the point of careand developing SHP mechanisms is a key Ministry of Health (MOH)objective (MOH, 2008a). There are currently a number of differentSHP schemes for the poor and the informal sector, including user-fee exemptions, Health Equity Funds (HEFs), a Government Sub-sidy Scheme (SUBO), Community-Based Health Insurance (CBHI),targeted vouchers and conditional cash transfers (CCT) (Annear,Bigdeli, & Jacobs, 2011; Ir, Heang et al., 2010; Ir, Horemans et al.,2010; Jacobs et al., 2008; Jacobs, Price, & Sam, 2007; Saneth et al.,

2010). The main features of these schemes are summarised inTable 1.

Targeted on the poor, Health Equity Funds are the country’smajor social protection mechanism (Annear, 2008; Annear, Bigdeli,Ros, & Jacobs, 2008). Cambodia was the first country to introduceHEFs, beginning in 2000, as a third-party payer for services atpublic health facilities. Until now, each HEF has been managed atdistrict level by a local agent (commonly a local NGO; known as theHEF operator) and supervised at the national level by an interna-tional NGO (known as the HEF implementer). HEFs are financedmainly by donor and government counterpart funds through thenational MOH-donor Health Sector Support Project (HSSP) (CARD,2010). The aim is soon to achieve national coverage (MOH,2008b). By 2012, HEFs and SUBOs together provided protectionfor the poor in 56 of a total 77 health operational districts (ODs),covering at least 78% of the population living below the povertyline, or 3.2 million people (Flores, Ir, Men, O’Donnell, & vanDoorslaer, 2011; Health Messenger, 2011; Men, Ir, Annear, & Sour,2011; MOH, 2011; Ros, 2011). Using a national proxy means testnow administered by the Ministry of Planning (MOP), the benefi-ciary pre-identification system has been shown to be accurate atthe time of survey, though changes in socio-economic status overtime and the difficulty in reaching migrant workers and thosewhose work requires mobility introduces an element of inclusionand exclusion error (Ir, 2009; Ir, Heang et al., 2010; Ir, Kristof,Hardeman, Horemans, & Van Damme, 2008). The evidence thatHEFs, CBHI and other demand-side schemes increase financial ac-cess to services for the poor and the near poor, reduce health costsand minimize impoverishment to households and reduce out ofpocket health expenditures is well documented elsewhere (Annear,2010; Flores et al., 2011; Hardeman et al., 2004; Ir, Horemans et al.,2010; Noirhomme et al., 2007).

The current operation of these schemes by government,different donors and various international and local NGOs hasresulted in a fragmented administrative systemwith high overheadcosts (Crossland & Conway, 2002; Jacobs et al., 2007; MOH, 2009),high monitoring and evaluation costs and complex reporting re-quirements. Fragmentation has complicated the stewardshipfunctions of the MOH, stretches the capacity of the MOH to coor-dinate and support the different arrangements and undermines thefinancial sustainability of the schemes (Jacobs & Price, 2006; Jacobset al., 2007).

These longer-term developments in the Cambodia health sectorhave laid the foundation for the current process aimed at defrag-mentation. These circumstances are discussed further below, butinclude increased government funding in the health sector, furtheralignment of donors with national priorities, an ongoing concernamong policy makers with health equity, and in particular theexpansion of HEFs to national coverage (Martinez et al., 2011; Menet al., 2011; MOH, 2011).

Methods

This article investigates the main policy and operational barriersto, and opportunities for, creating a national SHP agency for thepoor and the informal sector. While the paper analyses policychallenges rather than the process of making policy, we draw onthe approach developed by Walt and Gilson and the broadertradition of policy analysis (Gilson, 2012; Gilson & Raphaely, 2008;Walt & Gilson, 1994; Walt et al., 2008). We focus on the Cambodiancontext and look particularly at the approach taken by key policymaking actors, including the views of scholars, politicians, bu-reaucrats and other prominent figures (Kingdon, 1984). The re-searchers themselves have an advantaged position as long-termobservers of health policy development in Cambodia. Our vantage

Table 1Overview of social health protection schemes in Cambodia.

Scheme Implementer/Operator Target group Benefit/Service Coverage/Location

Government service provision:Tax funding via

Government budgetMEF/MOH/PHD/OD/RH/HC All population sectors Recurrent budget, drugs and

material suppliesNationwide public healthfacilities

Compulsory, formal-sector schemes:Social Health Insurance NSSF/MOLVT Formal sector workers

(est. 1.2 million workersout of total 7 millionworkforce)

Currently only occupational risk;user fees, transport, disabilitybenefit, funeral expenses andsurvivor benefit; proposedmedical coverage not yet begun

In 3 national hospitals and12 referral hospitals in 7provinces, covering approx.40% of formal sector workers;a pilot for formal sector workersis being tested in Phnom Penh

Social Health Insurance NCSSF/MOSVY Civil servants Still to be defined A pilot for formal sector workersis being tested in Phnom Penh

Occupational riskinsurance

NSSF/MOLVT Formal private sectorworkers

Medical treatment, temporary/permanent disable, funeralexpenses and survivor benefit

Covered 2429 enterprises (88%of enterprises) with 591,955workers

Maternity benefits NSSF/MOLVTNCSSF/MOSVY

Pregnant women formalsector workers and civilservants (spouses)

3 month maternity leave with50% salary for workers. For civilservants, 3 month maternity leavewith full salary and cash incentiveof USD150 per newborn

National coverage

Demand-side schemes for the informal sector and the poor:Global health initiatives

and vertical programsNational programs Patients with TB, malaria,

AIDS, and children forvaccination,

Medicines and vaccines Public facilities nationwide

User feea exemptions MOH/health facilities Poor patients User fee schedule Public facilities nationwideHealth Equity Funds NGOs for the formal HEFs The eligible poor (those

under the national povertyline; est. 4 m pop.)

User fees, food, transportLimited funeral expenses andother basic items

In 56 Referral hospitals and361 health centres, coveringapprox. 78% of the targetgroup or 3.2 m people

Government Subsidyschemes (SUBO)

MOH/PHD/OD The eligible poor (thoseunder the nationalpoverty line)

User fees In 6 National Hospitals, 10referral hospitals and 89health centres

Community-BasedHealth Insurance

Mainly NGOs Mainly informal sectorpeople living abovepoverty line

User feesLimited food and transport,funeral cost

18 schemes in 16 hospitalsand 164 health centres,covering 237,000 personsor approx. 1.6% of thepopulation

Vouchers MOH/NGOs Poor pregnant women User fees and transport for 3ANCs, delivery and PNC athealth centres

In 5 ODs with 5 hospitalsand 78 health centres

a User fees do not provide for full cost recovery but are set according to affordability by the majority of people, in principle with the involvement and support of thecommunity and local authorities before approval by the MOH. User-fee exemptions are poorly administered and largely ineffective. Source: updated from Saneth et al. (2010).

P.L. Annear et al. / Social Science & Medicine 96 (2013) 223e231 225

point is in problem identification and policy formulation, the firsttwo of four recognised stages of policy development (including alsoimplementation and evaluation).

The findings are based on a review of policy documents and onqualitative data collected through key informant interviews carriedout in Phnom Penh, Cambodia, during 2011e12. The identificationand selection of documents was based on the extensive experienceof the key authors and on previous reviews (see Annear, 2010). Wesearched the published and grey literature by themes related tosocial health protection for supplementary and current titles. Weincluded all relevant reports and publications from other minis-tries, development partners and NGOs, and selected all documentsthat specifically proposed institutional and organizational ar-rangements for social health protection. These documents weresourced directly from the MOH, development partners and NGOs,verified during the interview process, and are listed in Table 2below. Following accepted principles of qualitative analysis, thedocuments were reviewed against a prepared list of topics todetermine the level and nature of support for establishing a na-tional agency, to understand the most common constraints and toidentify possible options for policy development.

We conducted key informant interviews to examine the viewsand values of policy makers and technical advisors about the issues

raised in policy documents and outstanding questions (Antunes,Wanert, Bigdel, & Ros, 2009; Brenzel et al., 2009). Respondentswere identified during preliminary investigations and purposivelyselected as the most important national and international technicalstaff working with the government at national level on schemeimplementation and/or policy development covering the institu-tional and organisational arrangements for social health protection.We included all key technical staff involved centrally in policymaking and available for interview; there were no exclusions forany reason.We did not interview respondents involved primarily inpolitical decision-making because the discussion had not yetmoved from the technical level into that arena.

A total of 18 key informants comprising 10 national policymakers, four representatives of development partner organisationsand four program managers were selected, including: (i) nationalpolicy makers at the MOH, the Ministry of Economics and Finance,the Council for Agriculture and Rural Development and the Councilfor Administrative Reform; (ii) technical advisors working withdevelopment partners; and (iii) scheme implementers and opera-tors. Technical experts, civil society, academicians and politiciansall have a significant influence on health policy and each seeks topreserve their constituency interests (Seddoh et al., 2012). Benefi-ciaries in this study were represented by the HEF and CBHI

Table 2Major policy documents and reviews.

Document title Authors Published by Source

Official documents:Health Strategic Plan 2008e2015 MOH, 2008b Ministry of Health, Royal Government

of Cambodiahttp://www.moh.gov.kh/files/dphi/HSP0815En.pdf

Strategic Framework forHealth Financing 2008e2015

MOH, 2008a Ministry of Health: Department ofPlanning and Health Information

http://www.moh.gov.kh

Implementation of HealthEquity Funds (and GovernmentSubsidy Scheme) Guideline

MOH, 2009 Ministry of Health, RoyalGovernment of Cambodia

http://www.moh.gov.kh/files/hssp2/hefim/HEF%20Implementation%20Gudelines-Eng.pdf

Draft Master Plan on Social HealthProtection

Kingdom ofCambodia, 2009

Prepared by the Ministry of Health http://www.medicam-cambodia.org/publication/details.asp?publication_id¼19&pub_language¼English

National Social Protection Strategyfor the Poor and Vulnerable

CARD, 2010 Council for Agricultural and RuralDevelopment

http://www.medicam-cambodia.org/publication/download.asp?publication_id¼55&pub_language¼English

Health Financing Policy MOH, 2012 Ministry of Health pannear@unimelb.edu.au

Reviews and evaluations:Review of mechanisms to improve

equity in access to health care,Cambodia

Crossland &Conway, 2002

DFID Health Systems ResourceCentre: London

http://www.odi.org.uk/resources/details.asp?id¼3455&title¼pro-poor-health-financing-equity-funds-cambodia

Summary Report of the HealthFinancing System Assessmentin Cambodia (OASIS)

Antunes et al., 2009 Korean Foundation for InternationalHealthcare and the Providing forHealth Initiative

http://www.who.int/health_financing/documents/cov-oasis_e_09-cambodia/en/index.html

Evaluation of subsidy schemesunder Prakas 809 to supportthe Ministry of Health ofCambodia to achieve universalsocial health protection coverage

Men et al., 2011 Belgian Technical Cooperation andthe Ministry of Health: Phnom Penh

pannear@unimelb.edu.au

Overall assessment for mid-termreview of the Health StrategicPlan 2008e2015.

Martinez, Simmonds,Vinyals, Hun, Phallyand Ir, 2011

HLSP: Phnom Penh, AugusteNovember2011

pannear@unimelb.edu.au

Synthesis assessment of medium-termissues and options for supply anddemand side-initiatives for improvingquality and access to health services,supported under HSSP2 (Draft Report)

MOH & WorldBank, 2011

Royal Government of Cambodia,Ministry of Health

pannear@unimelb.edu.au

P.L.Annear

etal./

SocialScience

&Medicine

96(2013)

223e231

226

P.L. Annear et al. / Social Science & Medicine 96 (2013) 223e231 227

implementers. While there is a potential policy-maker bias in ourdata, we are confident this does not affect our analysis regardingthe institutional and organizational structures proposed for socialhealth protection.

All interviews were conducted face-to-face in Cambodia in En-glish by the authors in September 2011 using a semi-structuredquestionnaire. Key informants were guaranteed anonymity toencourage an open expression of their views. Interview responseswere written by hand as previous experience tells us that by usingelectronically recorded interviews in the Cambodian culturalenvironment it is unlikely we would have heard frank and openviews from policy makers and officials about contentious policyissues. The key informants were asked questions specifically oninstitutional design, the proposed location of a national agency, theinstitutional and organizational structure, its mandate and re-sponsibilities, institutional and human resources capacity devel-opment, management information systems and procedures forfinancial management and accountability. As the interviews andwider research were carried out with informants in their profes-sional capacity on behalf and with the approval of the Ministry ofHealth, Cambodia, it was considered that formal ethics approvalwas not required.

Building on the work by Carrin and James (2004) and Mills(2007), we analysed the data using the Organizational Assess-ment for Improving and Strengthening Health Financing (OASIS)conceptual framework proposed by Mathauer and Carrin (2010,2011). We based our work also on an earlier OASIS analysis ofSHP strategies in Cambodia by Antunes et al. (2009). The frame-work identifies stewardship functions related to resource collec-tion, fund pooling and service provision at two levels: (i)institutional design and organisation practice; (ii) nine key healthfinancing indicators. The Framework also identifies six potentialbottlenecks in policy development and three feasibility conditions,which provide the basis for the Discussion section below.

Findings

In this section, we analyse institutional design and organiza-tional practice issues under the three health financing functions(collection, pooling and purchasing), consider the impact on healthfinancing indicators and identify apparent bottlenecks.

Design features

A common understanding among key informants was the needfor legal authorisation (a government decree), an independentboard and terms of reference for a SHP agency. During 2012, a newHealth Financing Policy was drafted by the MOH with the partici-pation of many of the policy makers interviewed for this research(MOH, 2012). The Policy proposed the establishment of a singleNational Social Health Protection Fund (NSHPF) to unify all schemesfor the poor and the informal sector. The NSHPF is to be a semi-autonomous body under the MOH, though the rules and regula-tions for the Fund are yet to be defined. From the institutionalperspective, a new Health Financing Charter, which will detail therules and regulations for implementation of health financing andsocial health protection arrangements, is due for release in 2013.

The draft Policy builds on previous documents and reflects theproposition offered, for example, in the Synthesis Assessment ofSupply- and Demand-Side Initiatives, where the MOH and WorldBank saw the highest priority as establishing a national fund toimplement HEFs and CBHI schemes. The draft Master Plan on SocialHealth Protection (Kingdom of Cambodia, 2009) recognises theneed for a single common framework to facilitate themove towardsuniversal coverage, but as one key informant argued, “we don’t

need to wait for a long-term arrangement for all insurance schemesat this stage; we should go ahead for the [poor and the] informalsector” (Key informant #11). The aim, said one key informant, “isnot to centralise management power to any single institution, but[to implement] a multi-sectoral approach within a common stra-tegic framework” (Key informant #4). Another recognised theconstraints:

Implementation of the Master Plan is not easy and may not befeasible in the short run. We must go step by step, and we know inother countries this process takes a very long time (Key informant#18).

Collection of funds

The proposed NSHPF will be tax-funded (with donor support),with subsidies to enrol all of the poor and the informal sectorpopulation. The Health Strategic Plan 2008e15 earlier proposed a‘mixed model’ of health financing that combines public and privaterevenue collection and service delivery together with fee-basedprepayment and social-transfer mechanisms; it called for anincreased health budget and the alignment of donor funding withMOH strategies. The Strategic Framework for Health Financingfocused on strengthening government funding (Strategic Objective1) and on reinforcing interim measures to protect households(Strategic Objective 3).

There are several challenges to organizational practice. Somerespondents preferred to place an SHP agency under the Council ofthe Ministers (COMethe executive body of the government) ratherthan the MOH to avoid an appearance of conflict of interest. Bothinsufficient funding from government sources and limited imple-mentation capacity were considered major challenges in trans-ferring responsibility for HEF and CBHI administration from thecurrent NGO implementers (Key informant #10).

The MOH does not have sufficient capacity to implement andmonitor the activities of a national HEF-CBHI agency; the MOHwillneed enough human capacity and other resources (Key informant#16).

Another concern was the lack of “a common national moni-toring and evaluation framework for SHP schemes” (Key informant#1), which reflected the lack of a central data management systemor adequate systems within ministries.

Pooling of funds

The Health Financing Policy supports the idea that commonrisk- and fund-pooling can be attained for various populationgroups under the NSHPF in order to address fragmentation. Keyinformants confirmed that “most of the donors are willing to pro-vide funds to a common pool” (Key informant #5), although:

Sometimes donors are not on the same ground, and disagreementsare found among them [and they] may not agree on commonfunding arrangements. Therefore, donors have to. be assured thatproper use of funds will be monitored effectively (Key informant#14).

More immediately, the Overall Assessment for Mid-Term Re-view of Health Strategic Plan recommended scaling-up HEFs tocover the poor population nationally, increasing governmentfunding to HEFs and transferring national HEF management andfunding functions to a government-led institution.

Organizationally, the Health Financing Policy proposes that boththe poor and the informal sector will be automatically enrolledwith the proposed NSHPF, which could address the concern that

P.L. Annear et al. / Social Science & Medicine 96 (2013) 223e231228

“often, the informal-sector population are unable to pay the pre-mium for CBHI due to their irregular income; they easily drop outfrom the scheme” (Key informant #6). Even though “identificationof the poor to receive free services and subsidies remains prob-lematic and challenging” (Key informant #4), key informants sug-gested it nonetheless could be extended to the non-poor informalsector.

Purchasing of services

The Health Financing Policy proposes that health services bepurchased from accredited health providers by the MOH and thethree SHP agencies (civil servants, formal sector, informal sector/poor), which together will define the content of the essential socialand medical benefit package and provider payment systems. Ser-vices will be purchased from the pooled funds, with regulationsand incentives to guarantee quality of care. Donors and imple-menters strongly supported the need for third-party status as apurchaser of health services from government facilities; otherswere less convinced:

The national agency must consider third-party arrangements inimplementation; these would facilitate smooth implementationand monitoring (Key informant #17).

I am not sure whether we need an autonomous HEF agency; from aSHP perspective, there is no need for the MOH to create anautonomous agency (Key informant #8).

The agency at the beginning could be located at the MOH and latercould be converted to an independent agency attached to the MOH(Key informant #16).

Revenues earned from user fees, HEFs and CBHI schemes havebeen an important source of incentives for health facility staff and amechanism for enforcing quality standards at contracted facilities.These procedures provide a foundation for new organizational ar-rangements in a unified fund together with administrative effi-ciency, avoiding duplication and implementing accountabilitymechanisms.

Health financing indicators

Policy documents and key informants anticipate that the moveto a national agency for the poor and informal sector would havepositive effects on key health financing indicators:

� Funding: Themove towards unification of various SHP schemesunder national administration has been made possible by sig-nificant increases in the health budget. During a period ofstrong economic growth, the annual budget allocation tohealth increased by 23% in 2008e2009 and a further 16% in2009e2010 (MOH, 2011).

� Population coverage: With 25% coverage of the total popula-tion, the HEFs provide a strong foundation for the proposedNSHPF; supported by automatic enrolment for the informalsector, up to 80% of the population will be protected.

� Equity in financing: Has been achieved through the HEFs, withproposed expansion to the informal sector; the MOH’s longstanding commitment to health equity (MOH, 2008a, 2008b,2009, 2012) reflects a recognition within government of thepolitical need to provide for rural communities in the wake ofthe Khmer Rouge tragedy of the 1970s.

� Financial risk protection: It has been shown that HEFs reduceout-of-pocket health expenditure by beneficiaries by anaverage 29% per household and debt by 25% (Flores et al., 2011)

and, where it is available, CBHI reduces the cost of healthshocks by 40% and household health-related debt by 75%(Levine, Polimeni, & Ramage, 2011).

� Pooling: Since the mid-1990s, the MOH has worked to create acommon platform with development partners throughmonthly Technical Working Group meetings and under pooleddonor-government funding arrangements known as a SWiM(Sector-Wide Management) approach.

� Administrative efficiency: While administration costs for HEFsare only 13% of total expenditures (MOH, 2011), CBHI costs arehigher due to low membership figures; it was argued thatachieving “administrative efficiency will depend on the [pre-cise] role of the autonomous agency; the current administra-tion of HEF and CBHI schemes through international andnational NGOs is expensive, and transaction costs are high .[and] we must minimise these costs” (Key informant #16).

� Benefit package: Subsidised by government through provisionof drugs and medical products (72% of government healthexpenditure) and salaries (6% of government health expendi-ture) (MOH, 2012); HEFs and CBHI currently reimburse userfees for access to a common standard package of services athealth facilities and hospitals. The Master Plan anticipates thata single national agency would allow for portability betweenschemes and proposes an expanded benefit package (toinclude reproductive, maternal, neonatal and child health andnon-communicable diseases). A concern was how to tailor anappropriate benefit package that would support the growth ofmembership but remain affordable:

Current benefit packages can be reviewed and based on a mini-mum package that can be introduced at the beginning. The secondoption could be a comprehensive benefit package, according to theavailability of funding and services at the different levels (Keyinformant #15).

Institutional and organizational bottlenecks

In the new round of policy development that commenced inCambodia in 2012, health policy makers confront a range of insti-tutional and organizational bottlenecks affecting further progresstowards UC (Mathauer & Carrin, 2011). There is no governinglegislation for social health protection, and the Master Plan is stillunder consideration by the government. At the centre of the policydebate is the third-party status of any agency that may emerge tocover the poor and the informal sector.

A number of conflicting rules or unresolved issues are evident.Official documents leave open the possibility of creating a singleumbrella agency for all SHP schemes or moving first to create aseparate structure for the informal sector and the poor alongsidethe existing national social security funds for civil servants andprivate employees. The documents also leave open the possibilitythat a national administration of the demand-side schemes (prin-cipally HEF and CBHI) could be placed under a semi-autonomousagency reporting to the MOH or in a completely autonomousstructure. One key informant expressed awidely held view that anyagency within the MOH would face a conflict of interest betweenservice provision and provider payment.

The enforcement of existing rules is in some cases incomplete.While MOH guidelines govern the implementation of HEF and CBHIschemes, newer schemes such as vouchers or CCTs lie outside theguidelines. At the same time, the HEF and CBHI implementers andoperators have an independent responsibility to implement theschemes efficiently and effectively. There is considerable variationin contracting arrangements, fee schedules and provider paymentsystems. Procedures for monitoring HEF and CBHI implementation

P.L. Annear et al. / Social Science & Medicine 96 (2013) 223e231 229

are in place, but, as one key informant observed, the MOH has hadlittle capacity to implement monitoring or to follow up on theresults.

Discussion and conclusions

In Cambodia as elsewhere, social health insurance initiativescommonly form one part of a broader health sector reform aimed atimproving equity and sustainability (McIntyre, Doherty, & Gilson,2003). Health policy makers in Cambodia have looked for ways toaddress a series of challenges: How to use the administrative andfinancial base created by the HEFs (the biggest single social pro-tection scheme in the country) to expand coverage into theinformal sector; how to transfer the responsibilities for HEFgovernance and management from development-partners to fullgovernment authority; and how to develop the institutional rulesand organizational structures required for a national UC system.The case of Cambodia suggests that establishing the most effectiveinstitutional and organizational structures needed to cover thepoor and the informal sector depends as much on context and ca-pacity as on identifying appropriate health financing settings.

Our study reinforces the understanding that defining the mostappropriate settings for the health financing functions (collecting,pooling and purchasing) is essential. It is known that pro-poorstrategies may be either successful in achieving greater equity, asin Thailand (Limwattananon, Tangcharoensathien, & et al, 2012), ormay have unintended effects, as in Uganda (Nabyonga Orem &Muheki Zikusooka, 2010; Nabyonga Orem et al., 2012). Targetingof SHP schemes on different segments of the population is a com-mon approach in low- and middle-income countries, such asThailand (; Mills, 2007; Tangcharoensathien et al., 2011). Whileappropriate at early stages of development, the approach runs therisk of fragmenting risk pools and organizational structures, whichmay be ameliorated where risk-equalization methods of some sortare in place (Carrin & James, 2004). In Thailand, rapid progress wasmadewhen the government chose (for political reasons perhaps) toextend free-access to the informal sector, underpinned by manyyears of experimentation led by public health officials (Mills, 2007;Nitayarumphong, 2006). In Cambodia, risk equalisation for the poorhas been achieved through the subsidised HEFs, which pool fundsfinanced by transfers from taxation and donor funding.

The governance function provides the point of origin for theOASIS framework, and in practice is reinforced by challenges facingcountries like Cambodia. As Carrin et al. (2004) have noted, frag-mentation in the administration of different schemes raises theneed for a stronger stewardship role. Experience shows thateffective regulation and strong government intervention to allocatehealth investment are required for achieving UC (Allotey, Verghiset al., 2012; Allotey, Yasin et al., 2012). The challenge is to incor-porate the complex interrelations of health system componentsand the numerous contextual factors that may influence theeffectiveness of interventions, particularly their effects on disad-vantaged populations (Bosch-Capblanch et al., 2012). Understand-ably, as Kutzin (2008) notes, outcomes will depend on the nationalpriority given to the health sector.

The Cambodian MOH, together with development partners, hastaken a leading role in selecting an appropriate mix of regulatoryand financing mechanisms. Increasing funding and building ca-pacity for implementation of SHP schemes within the governmentstructure are the twomain challenges. Capacity building for changemanagement, human resources, administration and monitoringwill be needed at provincial and district as well as national levels(Antunes et al., 2009). Taking into account the population view ofthe quality, accessibility, value and appropriateness of health ser-vices is an additional challenge. Good governance is therefore

critical in achieving efficiency and developing the technical andadministrative capacity of public institutions and civil servants(Allotey, Verghis et al., 2012; Allotey, Yasin et al., 2012). From apolitical point of view, the government’s commitments toadvancing social protection makes reforms within the health sectortimely, politically feasible and desirable (according to measuressuggested by Oliver & Mossialos (2005).

Cambodia is one of many LDCs taking the initial steps on thepathway to UC, where the poor are a large proportion of the popu-lation and capacity for implementation of UC is yet to be developed.A system that depends mainly on mandatory insurance is likely toachieve a lower degree of equity in financing than one based ongeneral tax revenue, as in Thailand for example (Mills, 2007). HealthEquity Funds offer a promising approach (Hanson, Worrall, &Wiseman, 2008) despite the fact that such arrangements may leadto the creation of different schemes that must be harmonized overtime (Mills, 2007). The lessons from Cambodia indicate that, in thisprocess, a number of critical factors emerge: the need to expand thefiscal space for health care; a commitment to equity; specific mea-sures to protect the poor; building national capacity for SHP and UCadministration; and working within the specific national context.

The limitations of our study relatemainly to the small number ofkey informants, the lack of opportunity to canvass views from otherstakeholders and communities, and the limited resources availablefor the research. The research was informed by the long-terminvolvement in research and policy advice in Cambodia by threeof the authors, providing useful insights to the process. To reducethe subjectivity of the researchers, which is inherent in such anapproach, we consulted widely with a range of stakeholders aboutour findings. By seeking to integrate the largely donor-funded HEFsinto a broader social health protection system for the informalsector, the MOH has responded to the accepted need for strongnational ownership of the SHP process and a strong role for gov-ernment. Cambodia shows that an important part of the process isthe transfer of responsibilities from development partners to gov-ernment administration, using prepayment, risk pooling and sub-sidy mechanisms to create a uniform national structure. The way inwhich Cambodia faces these challenges and opportunities will beinstructive to other lesser-developed countries.

Conflict of interest

Dr Peter Annear is Senior Research Fellow at the Nossal Institutewith 15 years’ experience as a researcher and health policy analystin Cambodia and has previously contributed to the development ofthe Strategic Framework for Health Financing and the Master Planfor Social Health Protection. Dr Shakil Ahmed is an independentresearcher from the Nossal Institute with extensive experience inBangladesh and elsewhere and has played no previous role inpolicy development in Cambodia. Mr Chhun Eang Ros is the di-rector of the Bureau for Health Economics and Financing at theMinistry of Health and Dr Por Ir is a researcher at the NationalInstitute of Public Health; both contributed to this project as in-dependent researchers and not in any way as representatives oftheir respective organizations. None of the researchers have avested interest in the outcome of the research process. The authorsdeclare no conflict of interest.

Acknowledgements

A longer report on which this paper is based was prepared forthe Department of Planning and Health Information at the Ministryof Health, Cambodia. We would like to thank Dr Lo Veasnakiry,Director of the DPHI, for his participation and support. Many peoplefrom the MOH, the Council of Ministers, the Ministry of Economy

P.L. Annear et al. / Social Science & Medicine 96 (2013) 223e231230

and Finance, development partner organisations and non-government agencies participated willingly in the study and wethank them all for their insight and wisdom. The initial draft of thispaper was improved following comments by Net Neath from theCambodia Development Resource Institute. The research was fun-ded by the AusAID Health Policy and Health Finance KnowledgeHub at the Nossal Institute for Global Health, The University ofMelbourne. The preliminary results of this study were presented tothe Health System Reform in Asia Conference, Hong Kong,December 2010, sponsored by Social Science & Medicine.

References

Ahmed, S., & Khan, M. (2011). Is demand side financing equity enhancing? Lessonsfrom a maternal health voucher scheme in Bangladesh. Social Science & Medi-cine, 72, 1704e1710.

Allotey, P., Verghis, S., Alvarez-Castillo, F., & Reidpath, D. (2012). Vulnerability, eq-uity and universal coverageea concept note. BMC Public Health, 12(Suppl. 1), S2.

Allotey, P., Yasin, S., Tang, S., Chong, S. L., Cheah, J. C. H., & Reidpath, D. (2012).Universal coverage in an era of privatisation: can we guarantee health for all?BMC Public Health, 12(Suppl. 1), S1.

Annear, P. L. (2008). Cambodia: developing a strategy for social health protection. InPromoting sustainable strategies to improve access to health care in the Asian andPacific region. Bangkok: UNESCO.

Annear, P. L. (2010). A comprehensive review of the literature on health equity funds inCambodia 2001e2010 and annotated bibliography. Health Policy and HealthFinance Knowledge Hub Working Paper No. 9. Melbourne: Nossal Institute forGlobal Health.

Annear, P. L., Bigdeli, M., & Jacobs, B. (2011). A functional model for monitoringequity and effectiveness in purchasing health insurance premiums for the poor:evidence from Cambodia and the Lao PDR. Health Policy, 102, 295e303.

Annear, P., Bigdeli, M., Ros, C. E., & Jacobs, B. (2008). Providing access to healthservices for the poor: health equity in Cambodia. In B. Meessen, X. Pei, B. Criel,& G. Bloom (Eds.), Health and social protection experiences from Cambodia, Chinaand Lao PDR. Antwerp: ITG Press.

Antunes, A. F., & Saksena, P. (2009). Health financing system review 2008 e optionsfor universal coverage. Geneva: MOH Rwanda and the World HealthOrganization.

Antunes, A. F., Wanert, S., Bigdel, M., & Ros, C. E. (2009). Summary report of thehealth financing system assessment in Cambodia. Korean Foundation for Inter-national Healthcare and the Providing for Health Initiative.

Bosch-Capblanch, X., Lavis, J. N., Lewin, S., Atun, R., Røttingen, J.-A., Dröschel, D.,et al. (1 March 2012). Guidance for evidence-informed policies abouthealth systems: rationale for and challenges of guidance development. PLoSMedicine, 9(3).

Brenzel, L., & Naimoli, J. (2009). Health results-based financing in practice: teninstitutional challenges. Technical Brief, World Bank.

Canagarajah, S., & Setharaman, S. (2001). Social protection and the informal sector indeveloping countries: Challenges and opportunities. Washington: World Bank.

CARD. (2010). National social protection strategy for the poor and vulnerable.Cambodia: Council for Agricultural and Rural Development.

Carrin, G., & James, C. (2004). Reaching universal coverage via social health insurance:Key design features in the transition period. Health Financing Policy Issue Paper.Geneva: World Health Organization.

Carrin, G., Mathauer, I., Xu, K., & Evans, D. B. (2008). Universal coverage of healthservices: tailoring its implementation. Bulletin of the World Health Organization,86(11), 857e863.

Carrin, G., Waelkens, M., & Criel, B. (2005). Community-based health insurance indeveloping countries: a study of its contribution to the performance of healthfinancing system. Tropical Medicine and International Health, 10, 799e811.

Chan, M. (22 November 2010). The path to universal health coverage e Statement tothe press at the launch of the World Health Report: Health Systems Financing, ThePath to Universal Coverage. Berlin.

Chankova, S., Sulzbach, S., & Diop, F. (2008). Impact of mutual health organizations:evidence from West Africa. Health Policy and Planning, 23, 264e276.

Chua, H. T., & Chee, J. C. H. (2012). Financing universal coverage in Malaysia: a casestudy. BMC Public Health, 12(Suppl 1), S7.

Crossland, A., & Conway, T. (2002). Review of mechanisms to improve equity in accessto health care, Cambodia. London: DFID Health Systems Resource Centre.

Ekman, B., Nguyen, T. L., Ha, A. D., & Axelson, H. (2008). Health insurance reform inVietnam: a review of recent developments and future challenges. Health Policyand Planning, 23, 252e263.

El-Idrissi, D., Miloud, K., & Belgacem, S. (2008). Constraints and obstacles to socialhealth protection in the Maghreb: the cases of Algeria and Morocco. Bulletin ofthe World Health Organization, 86, 902e904.

Evans, D., & Etienne, C. (2010). Health systems financing and the path to universalcoverage. Bulletin of the World Health Organization, 88, 402.

Flores, G., Ir, P., Men, R. C., O’Donnell, O., & van Doorslaer, E. (2011). Financial pro-tection of patients through compensation of providers: The impact of health equityfunds in Cambodia. The Netherlands: Tinbergen Institute Discussion Paper TI2011-169/3.

Gilson, L. (Ed.), (2012). Health policy and systems research: A methodology reader.Geneva: Alliance for Health Policy and Systems Research and the World HealthOrganization.

Gilson, L., & Raphaely, N. (2008). The terrain of health policy analysis in low andmiddle income countries: a review of published literature 1994e2007. HealthPolicy and Planning, 23(5), 294e307.

Gottret, P., & Schieber, G. (Eds.), (2006). Health financing revisited. A practitioner’sguidebook. Washington: World Bank.

Gwatkin, D. R., & Ergo, A. (2011). Universal health coverage: friend or foe of healthequity? The Lancet, 377(9784), 2160e2161.

Hanson, K., Worrall, E., & Wiseman, V. (2008). Targeting services towards the poor:a review of targeting mechanisms and their effectiveness. In S. Bennett,L. Gilson, & A. Mills (Eds.), Health, economic development and household Poverty:From understanding to action. Routledge.

Hardeman, W., Van Damme, W., van Pelt, M., Ir, P., Heng, K., & Messen, B. (2004).Access to health care for all? User fees plus a health equity fund in Sotnikum,Cambodia. Health Policy and Planning, 19(1), 22e32.

Health Messenger, Issue 47, JulyeSeptember 2011. Phnom Penh.Ir, P. (2009). Assessment of HEF beneficiary identification in three operational health

districts in Kampong Cham. Phnom Penh: Provision of Basic Health Services-Kampong Cham (PBHS-KC).

Ir, P., Heang, K. S., & Sim, S. (2010). Good practices in health equity fund imple-mentation in Cambodia from the perspective of operators and implementers.Phnom Penh: Medicam.

Ir, P., Horemans, D., Souk, N., & Van Damme, W. (2010). Using targeted vouchers andhealth equity funds to improve access to skilled birth attendants for poorwomen: a case study in three rural health districts in Cambodia. BMC Pregnancyand Childbirth, 10.

Ir, P., Kristof, D., Hardeman, W., Horemans, D., & Van Damme, W. (2008).Challenges in identifying the poor: an assessment of household eligibility forhealth equity fund after four years of pre-identification in Oddar Meanchey,Cambodia. In B. Meessen, X. Pei, B. Criel, & G. Bloom (Eds.), Health andsocial protection experiences from Cambodia, China and Lao PDR (pp. 385e407).Antwerp: ITG Press.

Jacobs, B., Bigdeli, M., Pelt, M., Ir, P., Salze, C., & Criel, B. (2008). Bridging community-based health insurance and social protection for health care e a step in thedirection of universal coverage? Tropical Medicine and International Health, 13,140e143.

Jacobs, B., & Price, N. (2006). Improving access for the poorest to public healthservices: insights from Kirivong Operational Health District in Cambodia. HealthPolicy and Planning, 21, 27e39.

Jacobs, B., Price, N., & Sam, S. O. (2007). A sustainability assessment of a healthequity fund initiative in Cambodia. International Journal of Health Planning andManagement, 22, 183e203.

Kingdom of Cambodia. (2009). Draft master plan on social health protection. PhnomPenh.

Kingdon, J. W. (1984). Agendas, alternatives and public policies. Boston and Toronto:Little, Brown and Company.

Kutzin, J. (2008). Health financing policy: A guide for decision-makers. Copenhagen:World Health Organization Regional Office for Europe.

Levine, D. I., Polimeni, R., & Ramage, I. (2011). Insuring health or insuring wealth? Anexperimental evaluation of health insurance in Cambodia. Phnom Penh.

Li, C., Yu, X., Butler, J., Yiengprugsawan, V., & Yu, M. (2011). Moving towards uni-versal health insurance in China: performance, issues and lessons fromThailand. Social Science & Medicine, 73, 359e366.

Limwattananon, S., Tangcharoensathien, V., et al. (2012). Why has the UniversalCoverage Scheme in Thailand achieved a pro-poor public subsidy for healthcare? BMC Public Health, 12(Suppl. 1), S6.

McIntyre, D., Doherty, J., & Gilson, L. (2003). A tale of two visions: the changingfortunes of Social Health Insurance in South Africa. Health Policy and Planning,18(1), 47e58.

McIntyre, D., Garshong, B., Mtei, G., Meheus, F., Thiede, M., Akazili, J., et al. (2008).Beyond fragmentation and towards universal coverage: insights from Ghana,South Africa and the United Republic of Tanzania. Bulletin of the World HealthOrganization, 86, 871e876.

Martinez, J., Simmonds, S., Vinyals, L., Hun, A., Phally, C., & Ir, P. (2011). Kingdom ofCambodia e Overall assessment for mid-term review of Health Strategic Plan2008e2015. First Draft Report 30 October. Phnom Penh: HLSP.

Mathauer, I., & Carrin, G. (2010). OASIS user manual. A tool for health financing ePerformance assessment e Options for improvement. Geneva: WHO, Departmentof Health Systems Financing.

Mathauer, I., & Carrin, G. (2011). The role of institution design and organizationalpractice for health financing performance and universal coverage. Health Policy,99, 183e192.

Mathauer, I., Cavagnero, E., Vivas, G., & Carrin, G. (2010). Health financing challengesand the role of institutional design and organizational practice for health financingperformance in Nicaragua. Discussion Paper No. 2 2010. Geneva: World HealthOrganization.

Mathauer, I., Schmidt, J., & Wenyaa, M. (2008). Extending social health insurance tothe informal sector in Kenya. An assessment of factors affecting demand. In-ternational Journal of Health Planning and Management, 23, 51e68.

Men, R. C., Ir, P., Annear, P. L., & Sour, I. (2011). Evaluation of subsidy schemes underPrakas 809 to support the Ministry of Health of Cambodia to achieve universalsocial health protection coverage. Phnom Penh: Belgian Technical Cooperationand the Ministry of Health.

P.L. Annear et al. / Social Science & Medicine 96 (2013) 223e231 231

Mills, A. (2007). Strategies to achieve universal coverage: are there lessons from middleincome countries? A literature review commissioned by the Health SystemsKnowledge Network. Health economics and financing programme, LondonSchool of Hygiene and Tropical Medicine.

MOH. (2008a). Strategic framework for health financing 2008e2015. Ministry ofHealth: Department of Planning and Health Information.

MOH. (2008b). Health strategic plan 2008e2015 e Accountability, efficiency qualityequity. Ministry of Health.

MOH. (2009). Implementation of health equity funds guideline. Ministry of Health.MOH. (2011). Annual health financing report 2010. The Kingdom of Cambodia,

Ministry of Health: Department of Planning and Health Information, Bureau ofHealth Economics and Financing.

MOH. (2012). Health financing policy. Submitted Draft, November 26, 2012.Cambodia: Ministry of Health.

MOH, & World Bank. (2011). Synthesis assessment of medium-term issues and optionsfor supply and demand side-initiatives for improving quality and access to healthservices, supported underHSSP2 (Draft Report).MinistryofHealth andWorldBank.

Morris, S., Flores, S., Olinto, P., & Medina, J. M. (2004). Monetary incentives in pri-mary health care and effects on use and coverage of preventive health careinterventions in rural Honduras: cluster randomized trial. The Lancet, 364,2030e2037.

Nabyonga Orem, J., Mafigiri, D. K., Marchal, B., Ssengooba, F., Macq, J., & Criel, B.(2012). Research, evidence and policymaking: the perspectives of policy actorson improving uptake of evidence in health policy development and imple-mentation in Uganda. BMC Public Health, 12, 109.

Nabyonga Orem, J., & Muheki Zikusooka, C. (2010). Health financing reform inUganda: how equitable is the proposed National Health Insurance scheme?International Journal for Equity in Health, 9, 23.

Nguyen, H., & Knowles, J. (2010). Demand for voluntary health insurance indeveloping countries: the case of Vietnam’s school-age children and adolescentstudent health insurance program. Social Science & Medicine, 71, 2074e2082.

Nitayarumphong, S. (2006). Struggling along the path to universal coverage of healthcare for all. Thailand (cited in Mills, 2007): National Health Security Office.

Noirhomme, M., Meessen, B., Griffiths, F., Ir, P., Jacobs, B., Thor, R., et al. (2007).Improving access to hospital care for the poor: comparative analysis of fourhealth equity funds in Cambodia. Health Policy and Planning, 22, 242e262.

Oliver, A., & Mossialos, E. (2005). European health systems reforms: looking back-ward to see forward? Journal of Health Politics, Policy and Law, 30, 7e28.

OECD. (2009). Promoting pro-poor growth social protection. Paris, France: Organi-zation for Economic Cooperation and Development Publishing.

Pauly, M., Blavin, F., & Meghan, S. (2008). Is there a market for voluntary health in-surance in developing countries?. NBER Working Paper # 14095, Cambridge, MA.

Poletti, T., Balabanova, D., Ghazaryan, O., Kocharyan, H., Hakobyan, M., Arakelyan, K.,et al. (2007). The desirability and feasibility of scaling up community healthinsurance in low-income settings e lessons from Armenia. Social Science &Medicine, 64, 509e520.

Ros, C. E. (2011). Community based health insurance and microfinance programs inhealth: Cambodia. Report prepared under the Australian Leadership AwardsFellowship program, Nossal Institute for Global Health, Melbourne.

Saneth, V., Sum, S., Ros, C. E., Mok, C., Sok, S., & Ir, P. (2010). Towards consolidating theexisting social health protection schemes in Cambodia: Assessment of best practiceson provider payment methods. Capacity Building for Effective Health: HealthFinancing 2009e2010 e Cambodian country project implementation, FinalReport. 10 July 2010.

Seddoh, A., & Akor, S. (2012). Policy initiation and political levers in health policy:lessons from Ghana’s health insurance. BMC Public Health, 12(Suppl. 1), S10.

Tangcharoensathien, V., Patcharanarumol, W., Ir, P., Aljunid, S. M., Mukti, A. G.,Akkhavong, K., et al. (2011). Health-financing reforms in Southeast Asia: chal-lenges in achieving universal coverage. The Lancet, 377, 863e873.

Tangcharoensathien, V., Prakongsai, P., Limwattananon, S., Patcharanarumol, W., &Jongudomsuk, P. (2007). Achieving universal coverage in Thailand: What lessonsdid we learn? A study commissioned by the Health Systems Knowledge Network.International Health Policy Program, Ministry of Public Health, Khon KaenUniversity, National Health Security Office.

Thornton, R. L., Hatt, L. E., Field, E. M., Islam, M., Diaz, F. S., & Gonzalez, M. A. (2010).Social security health insurance for the informal sector in Nicaragua: a ran-domized evaluation. Health Economics, 19, 181e206.

Van Damme, W., van Leemput, L., Ir, P., Hardeman, W., & Meessen, B. (2004). Out-of-pocket health expenditure and debt in poor households: evidence fromCambodia. Tropical Medicine and International Health, 9(2), 273e280.

Walt, G., & Gilson, L. (1994). Reforming the health sector in developing countries:the central role of policy analysis. Health Policy and Planning, 9, 353e370.

Walt, G., Shiffman, J., Schneider, H., Murray, S. F., Brugha, R., & Gilson, L. (2008).‘Doing’ health policy analysis: methodological and conceptual reflections andchallenges. Health Policy and Planning, 23(5), 308e317.

WHO. (2005). Achieving universal health Coverage: Developing the health financingsystem. Technical Briefs for Policy Makers, Number 1. Geneva: World HealthOrganization.

WHO. (2008). The world health report 2008: Primary health Care e Now more thanever. Geneva: World Health Organization.