Document Type : Review Article
Authors
1 Centre for Health Policy, School of Public Health, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa
2 Department of Public Policy and Management, SD Dombo University of Business and Integrated Development Studies, Wa, Ghana
3 School of Community Health, Charles Sturt University, Orange, NSW, Australia
4 Warwick Medical School, University of Warwick, Coventry, UK
5 Department of International Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
6 School of Health Systems and Public Health, University of Pretoria, Pretoria, South Africa
7 Consortium for Advanced Research Training in Africa (CARTA), Nairobi, Kenya
8 School of Public Health, University of the Witwatersrand, Johannesburg, South Africa
Abstract
Keywords
All countries need to purchase healthcare in ways that ensure resources are used effectively and efficiently; the need for healthcare will always outstrip the finances available with ever improving medical technology and so the more expensive care options that are available.1 Purchasing (the allocation of funds to healthcare providers for services, on behalf of identified groups or a population),2 requires a continuous search for the best ways to maximise health system performance. It involves deciding which interventions to purchase, how to buy them, and from which providers, how providers will be paid, at what rates and under what contractual arrangements (eg, active or strategic purchasing [SP]).3,4
Given the international call for universal health coverage (UHC), many middle-income countries have started SP initiatives as part of established public insurance schemes.2,5-7 In this systematic qualitative review, we report on the experiences in nine middle-income countries (both lower- and upper middle-income countries) to understand what extent the activities that constitute SP have been established with public sector insurance schemes, what have been the challenges and facilitators, and to what extent SP is helping countries achieve their UHC goals.
SP requires the purchaser’s interaction with three key role players: the provider of healthcare services, citizens as the beneficiaries, and government as the regulator of both purchasing and provision of care. Figure 1 sets the actions associated with each key actor.
The aim of universal healthcare coverage is to provide quality healthcare and financial protection to all people in a given country.2,4,7,8 Pre-payment, either through taxation or insurance, is necessary to provide financial protection.6 While SP can be achieved through taxation systems, insurance schemes, with their separation of purchaser and provider roles into different organisations, is where SP is more visible. The core functions of insurance include pooling resources, enrolling members, defining the benefits package, contracting and paying providers and ensuring delivery of quality care that represents value for money. SP is then an essential building block to ensure that an insurance scheme remains financially viable, and that best use is made of available funds.2,9 Done well, these tasks, in theory, can amount to a virtuous circle, with risk and income cross-subsidisation providing protection from catastrophic expenditure for its members, new members joining as knowledge of the scheme grows, and a benefits package that increases as more resources become available.
We conducted a systematic search for literature in the following bibliographic databases: PubMed, CINAHL, Business Source Complete, Econlit, Web of Science, and Scopus (See Supplementary file 1, for the search syntax). We included the names of the 110 middle-income countries (as defined by the World Bank).10 Our search start date was 2011 as we found the rate of publications on SP increased at that time and the search was performed in November 2019. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) diagram presents the systematic search and screening process (Figure 2). In addition, 38 additional grey literature documents on SP in case study countries, identified through google and google scholar searches, were included (such as conference presentations and reports, from RESYST [Resilient and Responsive health systems] and the World Health Organization [WHO] Global Health Regional websites) to supplement our review.
Once duplicates had been removed, 691 articles were exported to Covidence (software for managing and streamlining systematic reviews) and screened following the predefined inclusion and exclusion criteria in Box 1. Non-English articles were excluded except Spanish, which ADK is able to read.11 Duplicate screening was conducted at title and abstract stage with differences resolved by a third person. The full-text screening was initially done in batches of 30 in duplicate. Differences were resolved through discussion, and if necessary, the inclusion and exclusion criteria were revised to clarify any uncertainty. We continued screening in batches of 30 until no more differences of opinion arose. A total of 91 articles were found relevant to SP in healthcare in addition to the 38 retrieved from grey literature search. Thus, a total of 129 articles were include in this review.
Inclusion Criteria
Describe a purchasing function (either in a public national health system or specific private insurance scheme) that aims to have characteristics associated with SP, ie, desires to obtain value for money through contracts or reimbursement mechanisms.
Includes evidence from a middle-income country as defined by the World Bank.
Evidence reviews if they include middle-income countries.
Exclusion Criteria
Those presenting evidence from either high- or low-income countries.
Those that describe reforms that aim to improve value for money without discussing contracts or reimbursement mechanisms.
Studies that discussed performance-based financing or results-based financing, or payment for performance.
If studies were conference abstract, opinion piece, systematic literature reviews or grey literature.
Studies published before 2011.
If they discussed only revenue mobilisation, contracting, health financing or health insurance other than SP functions.
We excluded opinion pieces and commentaries.
-------------------
Abbreviation: SP, strategic purchasing.
Listing the countries in order of the number of articles on SP, we found 21 countries with one or more articles about SP. We purposively 9 selected countries to include those with the most articles whilst ensuring geographical spread. The countries were: Iran (17 articles), Nigeria (10), China (10), Mexico (7), Ghana (6), Kenya (5), Thailand (3), Vietnam (3), and Indonesia (1).
We conducted additional searches in August 2020, to find articles on healthcare financing in each case study country. We searched for relevant articles in two main databases: Scopus and PubMed. We also conducted a grey literature search. We used the search terms: ‘healthcare financing’ OR ‘health financing’ AND ‘country’ for each country included. Screening was conducted using the following criteria: the article discussed at least one public healthcare financing mechanism, addressed any element of healthcare financing (eg, benefits package, service delivery platform, reimbursement mechanisms, provider-purchaser relationships, governance systems, etc) in one or more case countries, article was published in a peer-reviewed journal or was policy document that provides details of healthcare financing reforms in the case country. Time limitation was not applied to case study searches. For each case study we identified the following number of articles: Thailand (n = 45), Nigeria (n = 55), Ghana (n = 54), China (n = 91), Iran (n = 58), Mexico (n = 41), Vietnam (n = 42), Kenya (n = 52), and Indonesia (n = 38).
A data extraction template was designed and used to extract relevant information. Each team member focused on one or two case study countries. We read the supplementary articles to ensure we understood the healthcare financing system for each case study. We then focused on the SP articles. We extracted all results related to SP and made notes on each paper relating to the results to our understanding of healthcare financing in the country, and questions about SP relevant for our cross-country comparisons. We then reviewed the extractions and notes, and each team member wrote a case study summary covering the structure of healthcare financing, key health system and financing reforms, SP and the associated facilitators and challenges. We held twelve (n = 12) meetings to present the case study summaries to the rest of the team and to identify and discuss key experiences/issues, including the similarities and differences between countries’ and schemes. We used the extracted data, notes and case study summaries to finalise our analysis and write up. Samples of the data extraction files are attached (See Supplementary file 2).
While the RESYST framework emphasizes the relationships between the four sets of actors (purchasers, providers, citizens, and government), the detail provided is simply a list of specific tasks that each is responsible for executing. To develop an analytic framework within which to present the synthesis, we kept in mind the core elements of purchasing (deciding what to buy, from whom and how), as well as the importance of activities that maintain relationships between actors, and then organised the available data in the following sections with associated tables: healthcare financing country context; description of schemes (coverage, benefits etc); scheme performance; purchasing arrangements; and, governance of purchasing. (We were constrained by data availability; for example, there was data on reimbursement mechanisms but not the detail of specific contracts).
Firstly, we compare the structure of healthcare financing in each country based on the relative financial flows (ie, public, and private prepayment spending, external funds, out of pocket, etc) to provide a contextual understanding (Table 1).
Indicators 2017 | Iran | China | Mexico | Thailand | Vietnam | Indonesia | Ghana | Kenya | Nigeria |
GDP per capita ($) (2021) | 3000 | 8840 | 9900 | 6600 | 3743 | 4223 | 1960 | 1500 | 2360 |
Economic growth over 40 years (1981-2020) | 2.7 | 9.2 | 2.1 | 4.8 | 6.4 | 4.9 | 4.6 | 3.8 | 3.0 |
THE as a % of GDP | 8.7 | 6.4 | 5.5 | 3.8 | 5.9 | 2.9 | 3.3 | 4.8 | 3.8 |
Government health expenditure as % of GDP | 4.4 | 2.9 | 2.8 | 2.9 | 2.7 | 1.5 | 1.1 | 2.1 | 0.5 |
Government health expenditure as % of THE | 51.3 | 56.7 | 51.5 | 76.1 | 48.6 | 48.4 | 33.5 | 42.7 | 14.2 |
Private health expenditure as % of THE | 48.7 | 43.3 | 48.5 | 23.6 | 49.4 | 51.1 | 52.0 | 39.4 | 77.9 |
External resources on health as % of THE | 0.4 | 0.00 | N/A | 1.5 | 7.5 | 1.9 | 21.3 | 28.3 | 17.5 |
OOP expenditure as % of THE | 41.8 | 36.1 | 41.3 | 11.1 | 45.3 | 34.6 | 40.3 | 24.0 | 77.2 |
Government health expenditure per capita in Int$ | 901.4 | 474.8 | 552.2 | 525.8 | 187.4 | 163..8 | 48.5 | 67.2 | 31.36 |
Life expectancy | 76 | 77 | 75 | 77 | 75 | 72 | 64 | 66 | 54 |
Maternal mortality (per 100 000 live births) | 16 | 29 | 33 | 37 | 43 | 177 | 308 | 342 | 917 |
Abbreviations: THE, total health expenditure; GDP, gross domestic product; N/A, not available; OOP, out-of-pocket.
Source: The World Bank and WHO (2021): World Development Indicators, last updated on July 30, 2021.18
Secondly, we provide a description of the insurance schemes, including coverage and the benefits package, in order to compare the benefits packages (which are specified in different ways), we identified whether the packages included three particular treatments — HIV/AIDS treatment (as representative of chronic care), maternal care (including hospital delivery) because its need is widespread, and dialysis (because, its cost has a catastrophic effect on households).12-16 Three categories were created to represent the extent of the benefits package: (1) full package if it covers all three services, (2) partial package if it covers two of the services, and (3) limited package if only includes one of these services (Further details on membership eligibility criteria and contributions are provided in Table S1 of Supplementary file 2).
Thirdly, we provide an indication of the schemes’ performance using levels of out-of-pocket (OOP) by insured members, whether there were reports of purchases struggling to pay for care, and providers refusing to provide care (Table 2).
Iran | China | Mexico | Thailand | Vietnam | Indonesia | Ghana | Kenya | Nigeria | |
OOP by subscribers | 55.0% of OOPs by IHIO, SSO and IKRF members in 201763 | Cash informal OOPs (called ‘red envelope’) to physicians ranges between 54.4% (2011)64 to 76.1% (2015)65,66 by UEBMI, URBMI and NCMS |
Segura popular households have a lower proportion of OOP (42.95%) than households without insurance (57.05%)67 | 11% of OOP by UCS, SSS and CSMBS up to 202053 | Approx. 30.8% of OOP by VSS subscribers in 202068 34% OOP at commune health centres/stations 39% OOP at district hospitals69 |
18.0% OOP by JKN subscribers70 | 46.9% OOP by NHIS subscribers up to 202071 | 29.0% OOP by NHIF subscribers up to 201772 | 89.8% OOP FSSHIS subscribers73 |
Problems with payments | Delays reimbursing of claims | • Selective application of DRGs to selected disease conditions • No uniformity in application of the 4 payment methods for all hospitals in all provinces |
• State/REPSS has difficult paying providers | Not reported | Delayed reimbursing claims | Delays in reimbursing claims | Delay in reimbursing claims, which lead to comprised quality of care, and private accredited facilities refusing to provide care for members | Delays in payments due unavailability of funds | • Delayed payment due to complex payment system • Failure of NHIS to audit payments |
Providers refusing care to insured | Insured clients were not refused care | Yes, if providing care will lead to high cost beyond predefined service rate or cost ceiling | No evidence to suggest providers refuse care (The General Health Law prevents this) *Unclear what will happen under INSABI |
Providers do not refuse to provide health services to subscribers | There is no evidence to suggest that providers refused care to subscribers | • No evidence of providers refusing to provide care • However, there is evidence of some hospitals not being able to offer services under the benefits package because of lack of resourcesa |
Some providers especially the Catholic Health Association of Ghana refused care when NHIS delays in reimbursing claims | • Some members were refused or rationed services because of late payment of capitation and claims by NHIF • Members of the national scheme faced some discriminatory care compared to other schemes with higher reimbursement rates72 |
Patients were sometimes refused timely treatment because of delayed reimbursement of claims for previous services provided |
Abbreviations: UEBMI, Urban Employee’s Basic Medical Insurance; URBMI, Urban Resident’s Basic Medical Insurance; OOP, out-of-pocket payment; IHIO, Iran Health Insurance Organisation; SSO, Social Security Organisation; IKRF, Imam Khomeini Relief Foundation; NHIS, National Health Insurance Scheme; FSSHIS, Formal Sector Social Health Insurance Scheme; NHIF, National Health Insurance Fund; UCS, Universal Coverage Scheme; JKN, Jaminan Kesehatan Nasional; CSMBS, Civil Servants’ Medical Benefits Scheme; NCMS, New Cooperative Medical Scheme; INSABI, Institute for Health and Well-being; SSS, Social Security Scheme; DRGs, diagnosis related groups; REPSS, Regimen Estatal de Proteccion Social en Salud; VSS, Vietnam Social Security.
a The World Bank. Implementation completion and results report on a credit in the amount of special drawing right (SDR) 41.5 million (US$ 65.0 million equivalent) to the socialist republic of Vietnam for a central north region health support project. Hanoi: The World Bank; 2017.
Fourthly, we compare key elements of the purchasing arrangements, including reimbursement mechanisms, presence of gate keeping rules, whether there are caps on expenditure, evidence of provider-purchaser negotiations, research capacity, and whether purchasing was decentralised or not (We judged the latter to be important in enabling purchasers to be responsive, particularly in large countries) (Table 3 and Table S1).
Iran | China | Mexico | Thailand | Indonesia | Vietnam | Ghana | Kenya | Nigeria | |
---|---|---|---|---|---|---|---|---|---|
PHC | FFS | Central government subsidies PHC through a line budget | Mix of historical-based funding, capitation, and activity-based funding | UCS and SSS: Capitation CSMBS: FFS |
Capitation | Performance-based capitation payment at community health centres | FFS; Capitation payment method was piloted but suspended | Capitation | Capitation, FFS |
Hospital | FFS and case-based payments using RVUsa or RBRVUs for outpatient and in-patient services, respectively | • Reimbursement mechanisms differ by province • DRGs in 20 of 32 provinces • Otherwise scale payment, FFS, capitation (in-patient and out-patient) |
Line budgets, case-based payments, FFS | • DRGs for in-patient services • FFS for some services with high levels needs of the population such as cataract surgery, hip replacement therapy, etc |
• DRGs is used in paying in-patient services • FFS payments • In-patient “hotel” care funded through a per-diem-payment with co-payments |
• Case-based group (a type of DRG) • FFS payment was applied to only small portions of care |
DRGs | • Case based payment for bundled care eg, maternity, renal and surgical care • FFS for radiology (MRI and CT scan capped) |
• FFS based on authorised referrals • Per case payment |
Medicine | FFS | FFS | FFS | FFS | FFS | FFS | FFS | • Outpatient – including in capitation • Inpatient • FFS |
FFS |
Purchaser-provider negotiation | • No platform for negotiations among stakeholders • Government independently determined the tariffs for purchaser organisations |
• No evidence of negotiation • Pricing was determined by the National Health Development Research Centre |
• Fees for physicians are negotiated through short-term contracts • Drugs are negotiated with suppliers at central level by a commission with diverse representation |
Purchaser, provider, and citizens were engaged in policy initiatives and negotiations, ie, policy-making, design of health benefits package, budgetary processes | • No evidence of negotiations between purchasers and providers in fixing INA-CBG rates, caps of certain health services, etc • BJPS and the MoH determined package prices |
• The MoH responsible for setting policy for both public and private providers including benefits package and setting reimbursement prices and co-payments • No negotiations with beneficiaries |
There were purchaser-provider negotiations. However, these were inadequate which led to stalling of the roll out of the capitation payment methods | NHIF board was mandated by law to determine the rates and claims in consultation with private and public providers | • Expert committee with representatives of HMOs, providers, the NHIS, civil society organisations, academia and the Federal MoH defines benefits packages • Evidence from actuarial study was used to determine rates for capitation and FFS |
Budget support to public providers | • Government allocates funds to public health services, although, inadequate • Salary and incentives paid to medical university staff based on the medical procedure or speciality |
Government subsidises public healthcare, although the subsidy level varied across the regions | Under Seguro Popular, providers were issued short-term contracts without benefits | There are adequate and regular budgetary allocations for public providers | Budgetary allocations are made to public providers | • There is annual budgetary support to public providers • Government pays salaries of public providers |
Annual budgetary allocation to public health providers | • Public facilities are government allocated budgets and salaries • Casuals and support staff in public facilities are paid from user feesb |
• The MoH pays staff salaries, mainly for tertiary hospitals • Budget support is inadequate for primary care |
Abbreviations: FFS, fee for service; DRG, diagnosis related groups; UCS, Universal coverage scheme; SSS, social security scheme; CSMBS, Civil Servants’ Medical Benefits Scheme; MoH, Ministry of health; BJPS, Badan Penyelenggara Jaminan Sosial; PHC, primary healthcare; RVUs, relative value units; RBRVUs, resource-based relative value units; MRI, magnetic resonance imaging; CT, computerized tomography; INA-CBG, Indonesian-Case Based Group; NHIF, National Health Insurance Fund; HMOs, health maintenance organizations; NHIS, National Health Insurance Scheme.
aRVUs is the real monetary value for health services.
b“user fees” is a charge imposed by the government for the primary purpose of covering the cost of providing a service, directly raising funds from the people who benefit from the care or service being provided.
Fifthly, we described the extent and type of governance because of its impact on SP (Table 4). We used the WHO framework on governance for SP to compare the case countries.4 The framework has three key areas: setting directions, coordination and alignment, and legal provisions and regulations. We used the following indicators for comparison: (1) the existence a 5- or 10-year policy/strategy document (setting direction); (2) existence of governance body (coordination and alignment); and (3) a legal framework (legal provisions and regulations). We supplemented this with evidence on accreditation, monitoring of the claims, evidence of corruption and strategies to reduce it, patient engagement and patient feedback channels.
Iran | China | Mexico | Thailand | Indonesia | Vietnam | Ghana | Kenya | Nigeria | |
---|---|---|---|---|---|---|---|---|---|
Governance bodies of insurer/purchaser | All 3 public insurance organisations (ie, IHIO, SSO, IKRF) and the national health security organisations has structures eg, HCHI | NHC, MoHRSS, and MoH | MoH, the COFEPRIS and the GHC, the REPSS, SHS | UCS, SSS, and CSMBS have clearly defined governance structures and their interrelationships with providers, NHSO and comptroller and accountant general | There exist governance structures for JKN, BPJS, and the provider | • The Department of Health Insurance and the VSS • There are provincial people’s committees that monitor revenue collection and payments at provincial level |
NHIA and NHIS has clearly mapped governance structures | NHIF has a governance structure and is regulated by the National Hospital Insurance Fund Act of 1998 | • The NHIS has formal structures for managing providers and HMOs • The NHIS runs the FSSHIS and statutorily oversees the HMOs |
Existence of provider governance body | MoHME | NHC, MoHRSS, the MoH of China supported by the various local and provincial government department | Multiple. • The primary one is Consejo de Salubridad General (GHC). Council has Executive Board made of heads of public institutions • Other councils, commissions and committees are represented on GHCa • MoH • The COFEPRIS • The REPSS and SHS |
MOPH, have clearly defined governance structures and their interrelationships with providers, NHSO and comptroller and accountant general | • Indonesian MoH • DJSN, the National Social Security Board. DJSN comprises both government officials, community members, and representatives of employee and employer associations |
• MoH, The Department of Health Insurance, VSS Agency, Social Affairs Committee of Vietnam National Assembly • There are Provincial People’s Committees that monitor revenue collection and payments at provincial level |
MoH, NHIA | • MoH, NHIF, Kenya Medical Practitioners and Dentists Board • NHIF has a governance structure and is regulated by the National Hospital Insurance Fund Act of 1998 |
• SMoH, HMOs • There are laws guiding the activities of government actors (eg, MoH), HMOs and providers • The NHIS has formal structures for managing providers and HMOs • The NHIS runs the FSSHIS and statutorily oversees the HMOs |
Accreditation and monitoring of quality of care | • The MoHME assesses and accredits providers • No monitoring of providers’ compliance with clinical guidelines |
• Local and international bodies accredit health facilities • Providers are accredited by JCI |
• Accreditation is mandatory for Seguro Popular providers, but not IMSS or ISSSTE • Accredited by Specialty Councils and MoH – COFEPRIS/DGCES to enable funding by CNPSSb • ISSSTE has developed a set of 44 quality and efficiency indicators for its hospitals |
Thai HAI provides accreditation to providers using the ISQua | • The Hospital Accreditation Committee assesses facilities every 3 years using ISQua standards • BPJS-Health conducts onsite supervision of providers and provides technical support to public providers • BPJS-Health Office conducts regular public reporting on providers; showcases each provider's performance under the commitment-based capitationc payment policy (KBK) payment system and benchmark each provider compared to their peers |
The MoH accredits service providers100 | Ghana Accreditation Board in collaboration with the MoH assesses and grants accreditation to private providers. However, public providers receive automatic accreditation | NHIF has a benefits and quality assurance management committee for monitoring the quality of services offered by providers. However, this was rarely done to inadequate capacity of NHIF to perform this function | • HMOs are accredited and registered by the NHIS to purchase healthcare services from providers on behalf of the NHIS. • NIHS rarely reviews/reaccredits HMOs and providers (lack of resources) • HMOs conducts quarterly review of providers to ensure provision of quality care |
Monitoring claims | No clearly established monitoring systems in place for monitoring claims | Have introduced electronic health records and e-claim processing | There are information systems for tracking monitoring quality of care | • NHSO had a thorough system of medical audits to prevent fraud. Use of global budgets prevented ‘DRG-Creep’ • Created unique identification number for each medicine to enable monitoring of use by prescribers |
• The district health offices conduct monthly monitoring of providers to track targets • However, on-site monitoring and technical support to providers is inconsistent • Evidence of fraud in provider claims processing • Although evidence of active monitoring and enforcement of capitation payment rules (eg, a private primary provider’s contract was retracted/terminated for violation of terms) |
• In 2015, VSS set up electronic claims management system and this provided insured persons with a smart card • In 2017 – 97% of service providers had access via a portal to the VSS claims management system, and 60% connected daily |
• There are provider monitoring units in all NHIS offices • Claim processing units are established in all health facilities to ensure proper completion of claims and adherence of providers to defined costing guidelines • Auditing of historical claims data to identify fraud • Claims centre has been able to detect fraud in claim processing |
NHIF responded to the rise of fraudulent claims by employing medically trained personnel to review medical claims before payment | • HMOs compile and send patient encounter data quarterly to NHIS • Facilities do not always send the data to HMO • No systematic approach and large variations in what is done and how often • NHIS is meant to audit payments to ensure timely reimbursement of claims, however, this is not done34 |
Evidence of corruption | • “Under table” payments for health services • Evidence of repeated visits to particular physicians in a short time • Dispensing prescriptions at a particular pharmacy suggests collusion between physician and pharmacy over prescriptions, inflates prescription claims by over 25% |
No evidence identified | • Mostly tied to state-level purchases of medicines and human resources • There is also a political side to corruption. For example, a former governor of Tabasco State was sent to jail in 2013 for mishandling funds for Seguro Popular. Other states have faced similar allegations |
Providers falsified DRGs called “DRG-Creep” where up-coding of diagnosis in favour of higher DRG weights | • Over prescriptions or clients lying to prescribers on specific health conditions, and afterwards go to sell those medicines at “underground market” • Professionals falsifying services that were never provided • Repeated submission of claims for a similar service • Changing the dates/medicinal records on patient records • Utilisation of unlicensed staff to provide substandard care |
• Civic organisations have documented corruption in Vietnam Besides documented inequalities due to misuse of resources and differences in services provided in provinces,101 there was no further evidence of corruption |
There were cases of corruption such as inflation of claims, overbilling of medicines, inappropriate use of tariff, duplication of claims, lack of diagnostic evidence to back claims, absence of linkage between treatment and diagnosis, and treatment outside benefits package | There were reports of corruption in the accreditation process and the processing of claims | • Many HMOs are owned by financially and politically affluent citizens, some of whom serve as members of the NHIS governing council • Currently no legislation that prohibit this arrangement |
Strategies to reduce corruption | • In the fifth development plan (2011-2015), teaching hospitals were granted autonomy to ensure staff satisfaction and decrease in provider fraud • A new payment system was introduced, to change the costs of clinical services and balance hospitals’ revenues and expenses |
The National Health and Family Planning Commission promulgated circulars that established “prohibitions” on corrupt practices among providers, and “blacklisted" system on pharmaceutical and medical device providers | • Adjustment of regulatory framework with CNPSS managing resources and times of transfer of resources throughout the different levels (Federal, state, REPSS) • The CNPSS and the REPSS had increased accountability for Segura Popular through new accounts created at the Federal Treasury, and new sanctions established |
NHSO manages malpractice by applying a global budget on top of the DRG system and a thorough system of medical audit administered | • Public Research Anti-Corruption Clearing House and the Corruption Eradication Commission established in 2015 to prevent provider fraud • An e-tendering policy for drugs and supplies was introduced in 2014 to expedite contractual arrangements and reduce corruption |
• In 2015, the government embarked on a project to enable electronic claims management • Provided insured persons with a smart card • By 2017 – 97% of service providers had access via a portal to the VSS claims management system, and 60% connected daily |
• Implementation of e-claim submission/processing platform • Introduction of clinical audit and historical auditing of claims through the claim processing centres |
• NHIF employed staff with medical training to review claims made by providers • There are also recent plans to cede the function of accreditation to the MoH |
No evidence of any |
Channels for feedback from members | There is a hotline “1690” to report informal payments and report patient complaints about providers | Patients can lodge complaints at the respective hospitals, or through complaint letters | • Members can lodge complaints through the CONAMED • The CNDH intervenes in high-profile cases • A Patient’s Charter was published by the CONAMED and replicated in human rights guidelines |
• A 24-hour call centre service (code “1330”) was established to create public awareness about members entitlements and for patients’ complaints and resolutions • The call centre was established for all three schemes, although, evaluations showed underutilisation |
• BPJN Kesehatan also mandates all providers to have a patient complaints and resolution box • Complaints were received through the box and monthly meetings by facility management to resolve the issues • Customers could also make complaints through the customer relations office |
The MoH has a dedicated division within the Department of Health Insurance that deals with issues reported by members | NHIS member complaints hotline was established to monitor provider-patient behaviours/relationships | A toll-free line and email address is indicated in their NHIF website and publicity materials but there were reports that the phone number is not functioning | • HMOS are mandated to conduct quarterly seminars with beneficiaries; few people engage as communication channels are not clear • HMOs hold forums to inform people of their benefits and entitlements, but there is limited awareness amongst users of the forums • A complaints system exists but is not fully implemented |
Abbreviations: IHIO, Iran Health Insurance Organisation; SSO, Social Security Organisation; IKRF, Imam Khomeini Relief Foundation; NHSO, National Health Security Office; HCHI, High Council of Health Insurance; NHC, National Health Commission; MoHRSS, Ministry of Human Resources and Social Security; MoHME, Ministry of Health and Medical Education; COFEPRIS, Federal Commission for the Protection Against Sanitary Risks; REPSS, Regimen Estatal de Proteccion Social en Salud; CSMBS, Civil Servants’ Medical Benefits Scheme; SSS, Social Security Scheme; UCS, Universal Coverage Scheme; BPJS, Badan Penyelenggara Jaminan Sosial; JKN, Jaminan Kesehatan Nasional; VSS, Vietnam Social Security; NHIA, National Health Insurance Authority; NHIF, National Health Insurance Fund; HMO, health maintenance organization; FSSHIS, Formal Sector Social Health Insurance Scheme; GHC, General Health Council; DJSN, Dewan Jaminan Sosial Nasional; SMoH, State Ministry of Health; JCI, Joint Commission International; IMSS, Instituto Mexicano del Seguro Social; ISSSTE, Institute for Social Security and Services for State Employees; DGCES, General Directorate for Health Quality and Education; CNPSS, National Commission for Social Protection in Health; ISQua, International Standards for Quality in Healthcare; HAI, Health Accreditation Institute; KBK, Kapitasi Berbasis Komitmen; CNDH, National Commission for Human Rights; DRGs, diagnosis related groups; REPSS, Regimen Estatal de Proteccion Social en Salud; MoH, Ministry of Health; SHS, State Health Secretariats.
aCouncils (against Addictions, for Accident Prevention, and for the Prevention and Control of HIV/AIDS), Commissions (Bioethics, Human Genome, Occupational Health and Safety, Human Resources Development, and Health Research), Committees (Oral Health, Care for the Aging, and Epidemiological Surveillance) and the Reproductive Health Group.
bUnder INSABI, plans are underway to transfer accreditation to the GHC.
cUnder the commitment-based capitation policy (2016) BPJS-Health employs indicators of staff commitment to decide on the capitation grant percentage to allocate to primary health provider.
In Table 1 we compare the case study countries’ health financing indicators (in 2017), life expectancy and maternal mortality as indicators of health system performance, as well as their average economic growth rates for the past 40 years (1981-2020) (The order of the countries in Table 2 is based on their maternal mortality rate).
Government expenditure on health was between 2%-3% of gross domestic product (GDP) for five countries (China, Mexico, Thailand, Vietnam, and Kenya) and below 2% in three countries (Indonesia, Ghana, and Nigeria); only in Iran was it above 4% (Evidence suggests that government expenditure needs to be above 5% of GDP to achieve UHC).17 Public and private expenditure amounted to roughly equal shares of total health expenditure in 6 countries, with the exception of Thailand (76%/23% split), Ghana (33%/52% split), and Nigeria (14%/77%). External donor resources were significant in Ghana, Kenya, and Nigeria (17%-28%). Out of pocket expenditure as percentage of total health expenditure was between 30%-45% of total expenditure in all the case study countries, except for Thailand (11%), Kenya (24%), and Nigeria (77%).18
Despite a GDP per capita of US$ 3000, and an average growth rate of 2.7% over the last 40 years, Iran had the highest government expenditure on healthcare as percentage of GDP (4.4%) as well as in terms of international dollars (purchasing power parity $); it was also the best performing with the lowest maternal mortality ratio (MMR) of 16/100 000 live births, and a life expectancy of 76 years, although OOP payments are high at 41%.18 Thailand, with its high government share of health expenditure (76.1%), manages to keep out of pocket expenditure low (11%), although its MMR is still double that of Iran. Nigeria is the worst performing country, with MMR of over 900, OOP over 70% and the government share expenditure of total expenditure at 14% (See Table 1).
Five countries had more than 70% population (including rural population) covered by public sector insurance schemes [Thailand (98.5%), China (96.9%), Iran (90%), Vietnam (87%), and Mexico (85%)]15,19-21 (Table S1). Indonesia, Ghana, Kenya, and Nigeria had less than 50% of the population covered by state insurance schemes with Nigeria having only 5% coverage of the population by the Formal Sector Social Health Insurance Scheme (FSSHIS).22-25 While most schemes’ membership were primarily formal sector employees (Kenya, Indonesia, and Nigeria),26-31 some governments provided insurance coverage for the poor in social health insurance (SHI) schemes such as the Universal Coverage Scheme (UCS) in Thailand, Imam Khomeini Relief Foundation (IKRF) in Iran, Urban Employee’s Basic Medical Insurance (UEBMI) in Chine, National Health Insurance Fund (NHIF) in Kenya, and SHI in Vietnam.32-38
Six of the countries (Iran, Thailand, Indonesia, Ghana, Vietnam, and Kenya) had a comprehensive benefits package by our categorisation, including maternity services with inpatient delivery, HIV/AIDS treatment, and at least part of the costs of dialysis in their benefits package.33,39-45 China, Mexico, and Nigeria’s schemes covered only maternity services, but neither HIV treatment nor dialysis14,45-48 (See Table S1).
We found evidence of delays in payments of providers in six of the countries (Iran, Mexico, Vietnam, Ghana, Kenya, and Nigeria). We found reports that providers in Nigeria, Vietnam, and Ghana failed to honour essential services contained in the package, often due to delays in payments by the purchaser.34,39,49-51 Christian Health Association of Ghana returned to full OOPs when national health insurance authority (NHIA) delayed reimbursing their facilities for services provided.52 In Indonesia there was evidence of some hospitals not being able to offer services under the benefits package because of lack of resources.29,51
In Thailand of pocket payments (OOPs) were below 15% of total health expenditure, suggesting catastrophic expenditure for households would be minimal.53 In Indonesia, OOP was also relatively low at 18%; Kenya 29% and Vietnam 30%-39% were the next lowest.
Only in Thailand, Mexico, Vietnam, and Indonesia was the OOP of members reported to be lower than, or equal, to the national OOP suggesting the scheme was achieving its goal of protecting members to some degree from catastrophic expenditure (Tables 1 and 4). (In Thailand OOP by members and nationally were the same, as coverage is close to 100%). In other countries OOP by members was reported to be higher than the national figure.
While OOP is complex to measure and therefore there is often considerable variation in its estimates, these figures suggest that the insurance schemes, other than Thailand and Indonesia, were failing to protect members against catastrophic expenditure, and that members were wealthier in comparison to the uninsured and so able to pay out of pocket when the insurance scheme fails them.29,39,49,51
For primary care, the case study countries predominately used capitation as the reimbursement mechanisms, occasionally with performance-based elements (China and Indonesia),54,55 and fee for service (FFS) for some specific services, or in specific schemes (Thailand).54 Indonesia expanded the use of capitation into commitment-based capitation (using indicators to assess health facility staff commitment to work), along with increased monitoring by BPJS (Badan Penyelenggara Jaminan Sosial)-Health. The introduction of capitation and gate-keeping led to a reduction in demand for hospital care.54 The exceptions were Ghana and Iran where FFS reimbursement is used.43 Capitation has been piloted in Ghana. As mentioned above, after protests by both providers and members, partly due to inadequate stakeholder engagement, and insufficient preparation and research, the capitation payment mechanism was abandoned.43,46,56
Most countries use a case-based payment of varying sophistication for hospital reimbursement, ie, Nigeria, China, Mexico, and Thailand use diagnosis related groups (DRGs),8,57-59 while the others used simpler forms, in combination with FFS for medication and for some specific procedures. Only Vietnam used a per-diem payment.60 China piloted its DRGs before scaling-up, but the implementation was negatively affected by insufficient information system capacity and lack of unified disease classifications. The pilots led to the adoption of mixed provider payment methods,55,57 with hospitals in some provinces allowed to choose FFS for older patients and those with complications, leading presumably to selection of the more profitable option by providers.57 Vietnam, Nigeria, and Kenya are using FFS for hospital care and had no process to reform provider payment methods.
Two countries have attempted to increase the supply of particular services by switching to FFS. Thailand changed the reimbursement methods for HIV and cataracts from capitation to FFS in order to increase supply,8,15 and China made specific payments for TB services, and paid for patients’ transport and a subsistence allowance in order to ensure better TB outcomes.61,62
Governments in all nine countries either paid salaries of staff in public facilities or provided other budgetary support (Table 2).
Gate keeping has been implemented in some countries (ie, Vietnam, Ghana, Iran, Indonesia, and Thailand (except the Civil Servants’ Medical Benefits Scheme [CSMBS]), although implementation varied across countries.8,74-76 Ghana, Vietnam, and Thailand had a defined list of health services for which referral was permitted under the gate keeping system.13,15,76 In Indonesia and Mexico, there were no clearly defined services or the cost that are covered during the referral. In the literature reviewed there was no mention of a gate keeping system in Nigeria, Kenya, and China (Table 2).
Although reimbursement rates for particular services are part of each contract, purchasers need to control overall expenditure, to ensure that it does not exceed the scheme’s income, and so its sustainability. This can be done by setting an overall limit and adjusting the payments rates per service, should the quantity of services push threaten to push expenditure above the limit (for example UCS in Thailand).19 Alternatively, expenditure limits per person can be set; this is used in China, Indonesia, and for certain services in Kenya.54,77-81 There is no cap on overall expenditure for the CSMBS in Thailand, and in Iran, Mexico, Vietnam, Ghana, or Nigeria we found no mention of caps on expenditure (Table 2).
Provider-purchaser negotiations are important in determining prices, the affordability of services and the sustainability of a scheme. Purchaser-provider engagement can enable sharing relevant information and building relationships based on trust and collaboration. Lack or limited engagement with providers is more likely to encourage provider opportunism to meet income targets.19,33,82
In two countries (Indonesia and Thailand) the purchaser has effective means of engaging with providers,8,83 with well-structured forums. In Thailand, providers were involved in the national health security board of UCS. The National Health Security Office (NHSO) uses its substantial purchasing power to negotiate for lower prices for some selected high-cost medicines and medical devices, leading to cost savings, increased affordability, and access to essential services.15,84 For example, when cataract services were on high demand, the NHSO used its central bargaining capacity to negotiate an affordable price for soft lens for providers. Therefore, hospitals could then choose to reimburse on an agreed rate or to use the lens supplied by the NHSO-negotiated vendors.8,15,84
Indonesia has reaped substantial benefits from engaging with providers. Performance indicators, on which price revisions and capitation payments were benchmarked, enabled the purchaser to hold primary health providers accountable and minimised opportunistic behaviour (such as using lower cadre of staff despite regulations), despite the administrative burden associated with its execution.83,85 This was possible through the PPJK agency (Perusahaan Pengurusan Jasa Kepabeanan-health Insurance Directorate) under the Ministry of Health (MoH) to evaluate and calculate prices, simulated with expected revenues.54
In six countries (Ghana, Nigeria, Vietnam, Kenya, Mexico, and China), we found limited evidence of active engagement, and prices, benefits, and modes of payments are often fixed by either boards or committees (Kenya, Mexico, and Ghana) and the health ministries (Vietnam). In Kenya, the private sector, represented in the NHIF governing board, had a strong voice, and influence, leading to a ‘purchaser capture,’ demonstrated by the favourable reimbursement rates and terms extended to private health facilities.86
Inadequate provider-purchaser engagement and protests have led to Ghana’s capitation policy being suspended; the policy was decided at the level of “elite stakeholders.”43 However, patients needed to know what interventions were included in the benefits package.41,87 Health professional associations and providers need to assess the likely effect on service delivery, the services covered in health benefits package, and their income levels.43 Likewise, in Iran the Ministry of Health and Medical Education (MoHME) independently determined the prices for benefits and the revision of relative value units has faced multiple challenges because of lack of active negotiations with relevant stakeholders.19,88
Purchaser-provider tensions were a key contributor to the unravelling of Mexico’s decentralised, SP units within Seguro Popular in favour of a centralized system under the Institute for Health and Well-being (INSABI). Decentralised purchasing units struggled to negotiate effectively with centralised and highly influential medical unions. This led to increased contracting which distorted state-level fund allocations (eg, some states were spending nearly 70% of costs on contracts, causing purchasers to introduce a 40% cap). Efforts to “regularise” contracts significantly increased costs (eg, by over 30% in one year). However, better contracts did not produce better performance from providers. This suggests that federal-state coordination was poor, and oversight of purchaser provider negotiations was insufficient.89,90 However, INSABI, the new centralised purchasing body, emerged without adequate stakeholder consultations91 (Table 2).
Five countries (Thailand, China, Indonesia, Vietnam, and Mexico) have established health technology assessment (HTA) units, key for deciding what services and technologies to purchase.54,92-94 Ghana has some internal capacity within government to conduct research and has initiated the process to establish HTA. In Indonesia the National Basic Health Research Unit, established in 2013, has improved the availability and quality of data on which to make purchasing decisions. The government has also started to collect village level data to guide planning and policy decision-making.83 Nigeria, Kenya, and Iran had little internal capacity, and relied on publications produced by university academics (Table 2).
Although decentralization of healthcare service delivery has been accepted globally as a means to improve the responsiveness of the health system, decentralisation of the purchasing function is less common. In three of our case study countries (Ghana, Iran, and Vietnam), purchasing was carried out at national level.19,22,33,95-97 Three countries (Thailand, Indonesia, and China) have decentralised the purchasing function to the local level.84,98
Thailand implemented decentralisation by increasing budgetary allocation from 9% to 26% to local governments between 1999 and 2012, to increase their purchasing capacity.8 However, the government’s action plan has not been fully implemented (such as the involvement of community committees in purchasing) due to a frequent change in governments. The largest public health insurance scheme in China — New Cooperative Medical Scheme has been decentralised to the state level, which give local governments vast autonomy in system design, leading to varying degrees of local government subsidies for premiums, levels of coinsurance and deductibles and reimbursement procedures.98 Given that Indonesia’s sizeable population, spread over many islands with diverse ethnic and religious groupings, the country has adopted high levels of decentralisation, with district level schemes having considerable autonomy in terms of scheme design and purchasing of healthcare.35,99
Under the Seguro Popular in Mexico, the central government provided funds to state-level autonomous purchasing units called Regimen Estatal de Proteccion Social en Salud (REPSS).102 The REPSS were designed to exist outside of the state health agencies, in order to separate financing from provision, as a mechanism to improve the efficiency and quality of service delivery.102 However, recent reforms have eliminated REPSS, effectively (re)centralising purchasing to a national level under INSABI91,102 (Table 2).
Indonesia, Iran, Vietnam, Kenya, and Thailand had policy frameworks guiding health system and SP reforms. Specifically, Thailand began health reforms in 1942 with an evidence-based National Economic and Social Development Plan (NESDP), which contained six separate reforms.95 Indonesia had a detailed policy document that provided the sequence for health reforms starting from 1945. Iran transitioned through five sequential reforms from 1964 to the 2014 Health Transformation Plan.19 Four countries have no explicit policy framework (Ghana, Nigeria, Kenya, and Vietnam).
Only four countries had independent governance structures to provide oversight. In Kenya, there is a 12-member NHIF board,38 31-member UCS board with diverse membership (including non-governmental and civil society organisations in health, and members) in Thailand,15,103 Ghana’s NHIA has a governing council,22 and BPJS in Indonesia has a 2-member board of director and commissioner.104 In Mexico, the General Health Council, a collective decision-making body composed of various stakeholders [representatives from the National Commission for Social Protection in Health (CNPSS), MoH, REPSS], defined and updated the package of high-cost interventions and assists with provider accreditation.102 Many of these boards have patient and private provider representation.8,105 However, the management board members (ie, CSMBS) in Thailand,15 Ghana and Kenya perform functions that were unrelated to their expertise.33,38,105 Moreover, large boards (eg, 31 for national health security board of UCS in Thailand) can also delay decision-making.15
All case study countries had legal and regulatory frameworks establishing the schemes and setting directions for purchasing and service provision, as well as established institutions for implementing SP and associated reforms.21,33,38,91,95,106-109 However, diversities exist in how they were constituted and operationalised, for example in Nigeria, although the NHIS was overseeing the activities of health maintenance organizations (HMOs), the activities required of the HMOs were not properly defined, leaving lapses for opportunism.34,108
Only four countries had functioning/integrated information management system, including e-databases of clients and e-claims/tendering processes (China, Thailand, partly implemented in Ghana, and Mexico).
All the case study countries have some form of mechanism for accreditation of health facilities. Three countries use standards provided by international bodies (Thailand, Indonesia, and China), such as the International Society for Quality in Health Care (ISQua) (Thailand, Indonesia) and Joint Commission International Standards (China).104,110,111 For most countries there were reports of either infrequent assessments, with considerable variation in terms of what is done in which province or state (Nigeria, China, Vietnam, Mexico, Ghana, and Iran). Key challenges reported were inadequate and fraudulent assessment and ranking of providers (Iran and Indonesia),104,107,111 as well as inadequate and non-strict criteria for credentialling providers (Ghana, Kenya, Indonesia, and Iran).40,104,111-113 In some countries, public providers are given automatic accreditation (Ghana and Kenya).113,114
Monitoring of claims is important to identify fraud. All case study countries had instituted some strategy for monitoring claims, but they vary in complexity and the extent to which they were implemented. Three countries (Ghana, Vietnam, and Thailand) have implemented e-claim processing to reduce processing time and ensure timely payment of providers (only some health facilities in Ghana). Ghana and Indonesia have established paper-based claim processing units in all health facilities.
Electronic health records are a significant milestone in being able to check for service over-supply (Ghana, China, Indonesia, and Thailand). An integrated health information management system has been developed in China (One Health Information Management) with unique patient and prescriber identification numbers.115,116 Thailand uses its electronic system to track claims of accredited providers.15 For Ghana, the system allows for auditing of historical claims data.22,117
Clinical audits of the services covered in claims were conducted in Ghana, Thailand, China, and Mexico.15,22,118,119 In Mexico, Instituto Mexicano del Seguro Social scheme tracks and audits providers for services provided under the SHI.91 Indonesia utilises a taskforce against fraud to clampdown retrospectively on financial wastage, through on-site monitoring of community health centres and district heath offices at given time periods.120 Provider performance is publicly showcased through quarterly hearings. These strategies have identified ‘under the table’ payments from patients, excessive treatment, and fraudulent use of insurance cards by non-members.121
Although HMOs in Nigeria are mandated by the law to check claims sent from accredited health facilities on a quarterly basis, this is not often done, thereby creating variations in claim amounts versus services provided.34,108 In Iran, the MoHME was responsible for monitoring providers and purchaser organisations for compliance with clinical guidelines and audit regulations,122 however, this is not done.
Key explanations for the failure in countries where monitoring was inadequate were limited capacity within NHIF (Kenya),123 influence of politicians and other beneficiaries of the dysfunctional system (Nigeria).124
In two countries, patients’ rights are enshrined in law (Mexico, China), and there are channels for patients’ complaints in six countries (Ghana, Thailand, Mexico, Nigeria, China, and Kenya). In five countries, health insurance schemes have hotlines for patients to report complaints (Thailand, Mexico, Ghana, Iran, and Indonesia).15,22,91,122 In Mexico, complaints can be lodged at complaints units or health facility user orientation offices, which are part of the State Health Secretariat.91 In Indonesia there are two mechanisms for patients to voice concerns: patients’ complaints and resolution box (paper-based system) and the customer relations office of the scheme (either in person or by phone), where monthly meetings are held to resolve clients’ issues.120,125 However, across countries, the channels functioned poorly, and information gained from patients are not fed back into the system to improve things. Moreover, patients’ awareness of their rights, their entitlements and how to access the benefits was poor.120,121
We found reports of corruption in eight countries (Iran, Mexico, Thailand, Vietnam, Indonesia, Ghana, Kenya, and Nigeria). Typical of these were:
Over prescriptions (Iran, Indonesia), inflation of prices of medicines (Mexico, Ghana),91,126,127 collusion (physicians had certain patients make repeated visits to them within short intervals, and those patients were directed to a particular high-cost pharmacies) between dispensaries and patients (Iran).111,122,128
Falsification of DRGs coding (called “DRG-Creep’) (Thailand),15 inflation of claims (Ghana),126 inappropriate use of tariffs (Ghana),127 claiming services that were not provided (Indonesia, Ghana),126,127 and ‘under table’ payments (Iran).129
Falsification of accreditation documents (Kenya),112,113 or employing unlicensed staff (Indonesia).104,120,121
Conflict of interest where HMOs were owned by political elites who controlled the national health insurance scheme (Nigeria).34,108,124
Misuse or embezzlement of health sector funds (Mexico, Vietnam).100,101,130
Several measures were introduced in various countries to prevent fraud. Iran granted autonomy to teaching hospitals to manage staff motivation to reduce fraud and introduced new payment reforms to manage clinical services and balance hospital revenues.15,131 Mexico adjusted its regulatory framework, increasing accountability of the CNPSS and the REPSS through new accounts created at the Federal Treasury and new sanctions.91,119 Thailand’s NHSO introduced a global budget to augment the DRG payments and established a rigorous medical audit system.15
In Indonesia, the Public Research Anti-Corruption Clearing House and the Corruption Eradication Commission was established to prevent provider fraud.104,121 This was supported by e-tendering for drugs and supplies, introduced to expedite contractual arrangements and reduce corruption.120,121 Vietnam introduced an electronic claim management system and provided smart cards for members. Ghana piloted an electronic claim submission and processing system for some providers, provided identity cards for insured members, introduced clinical audit and historical claim auditing of services provided.126,127 The Kenya’s MoH ensures that adequately trained medical staff conduct clinical reviews/audits in Kenya.112,113
While evidence of corruption exists in Nigeria’s healthcare purchasing system, there were no descriptions of efforts to reduce it, in the literature we reviewed (Table 3). While there were no reports of corruption in China’s healthcare purchasing system, the National Health and Family Planning Commission introduced laws that prohibited corrupt practices. Providers and pharmaceuticals who were found culpable were blacklisted.
Managed well, public insurance schemes, with risk and income cross-subsidisation, can provide financial protection of their members. In theory, new members join as knowledge of the scheme grows, and the benefits package can be increased as more resources become available. However, for schemes to grow beyond the mandatory enrolment of the formal sector and include the poor, the use of public funds is required; resource constraints and need for care are then at their highest, and the need for ensuring value for money even more important.33,91,95
As coverage increases, a scheme needs to keep costs at an affordable level, while ensuring that OOP (ie, costs borne by members) do not escalate. Various elements of SP enable control of a schemes’ costs, such as limiting the benefit package, a cap on expenditure, using reimbursement mechanisms that enable cost control (capitation and DRGs), gate-keeping, as well as monitoring claims and minimising corruption. However, containing OOP also requires that appropriate services are provided and paid for an at appropriate rate ie, that the members do not have to seek care elsewhere,133 or the providers do not charge a co-payments or informal fees.7,133 Other elements of SP are focused on this task, such as research capacity to ensure the most appropriate services are purchased, the auditing of facilities and quality of care, as well as engagement with providers and members to understand if the scheme is meeting both of their needs.
The schemes in Thailand and Indonesia have managed to keep OOP relatively low (11% and 18%, respectively). Both have a comprehensive benefit package, a cap on expenditure and some gate-keeping mechanism. They both have forums for systematic engagement between purchaser and providers, HTA research capacity, and have successfully reformed reimbursement mechanisms to change provider incentives. They both use international standards for accrediting facilities and conduct clinical audits of services. In sum, both have effective SP, although Indonesia has not managed to substantially include the poor (coverage is 32%) unlike Thailand (99%) where public funds subsidise membership of the poor.
In the Vietnamese scheme, coverage is high (87%), and while benefits are limited, there is HTA capacity, a performance capitation mechanism at primary healthcare as well as a gate keeping mechanism, but FFS is used for hospital care and there is no cap on expenditure. As a result, 20% co-payments were introduced. Provider accreditation is inadequate, and there are no clinical audits. OOP are between 30%-39%. More effective cost control through a cap on expenditure and the use of DRGs might have prevented the need to introduce co-payments, and better accreditation and clinical audits might have led to better quality care, reducing the need to seek care elsewhere, both of which would have lowered OOP. Similarly, in Ghana there are limited elements of SP (FFS for hospital care, no cap on expenditure, and limited HTA capacity), and with a comprehensive benefit package, OOP is high at 50%, even though only 60% of the population is covered. In Mexico, moves toward SP that accompanied expanded coverage, did not translate to large reductions in OOPs and raised governance concerns, causing officials to revert to passive purchasing arrangements.
If SP is to play its role in ensuring the sustainability of an insurance scheme (including fending off politically motivated demands), there needs to be considerable institutional and organisational capacity, both at the purchaser and in government.19,134 These include research capacity to assess health needs and which services are affordable and best value of money, the capacity to accredit facilities, monitor the quality of care and the claims submitted, and governance capacity to provide stewardship and regulation.104,110,111
The literature has demonstrated that insufficient regulation leads to lack of trust between providers, purchasers, and service users, and so a failure of financial protection.26,135 Liu and colleagues reported that unregulated marketization of healthcare provision and inadequate financial protection by purchasers induced unhealthy competition among patients and so “red envelope” payments to secure care.66 Similar relationships were reported in Iran to skip long queues. Provider opportunism was fuelled by loose/non-existent regulations, disempowerment of patients to make choices,136 and lack of appropriate incentives for providers. For SP to ensure resources are used wisely, government regulation is also necessary.
In their realist review of SP, Sanderson et al conclude that SP requires national government purchasers to build close, trusting relationships with providers to facilitate access to local knowledge about needs and priorities.138 While ‘provider decision autonomy may drive innovation and efficient resource use, it may also create scope for opportunism; interdependence [of purchasers and providers] is likely to be the best power structure to incentivise collaboration needed to drive performance improvement.’138 Only in two countries did we find evidence of consistent engagement between purchasers and providers.
Ensuring patients’ rights in doing SP could be achieved through encouraging their participation in committees and boards, creating some awareness, implementing community verification of health benefits packages,139 ascertain population views and values.6,15,22,91,122 These require a degree of decentralisation and institutional purchaser capacity to engage with providers and patients that is not available in most middle-income countries.85,113
In a comparison of SP in 10 European countries, Klasa et al9 argue that SP has not been fully implemented in any one of their case study countries; they conclude that SP is unlikely to work elsewhere and an ‘idea too perfect to exist in reality.’ Similar critiques have been raised by others.140 The review includes the requirement that there are sufficient providers in any location such that purchaser and patient can choose where to purchase/seek care for SP to be able to occur; we have not included this requirement, because of the idle and so wasted capacity it requires.9 However, there are a myriad of additional reasons why SP is hard. In practice, purchasers, often lack the data,83,115 expertise,91,131 policy capacity and negotiating power to shape an effective purchasing strategy that is focused on the quality of care and the actual needs of the population,4 instead of historical utilization patterns, prices and volumes.
One of the limitations of this study is the use of heterogenous information available on the case study countries in published sources. Often information was incomplete and difficult to interpret (for example the details of the contracts and the contracting processes). Some countries may have separate HIV programmes, that are not part of the public insurance scheme, however our focus was the purchasing carried out by the insurance schemes. Interviews with key informants would have provided useful additional information and an opportunity to confirm the published sources, however, this was not possible given the resources available. Two authors were from case study countries (Ghana and Kenya) and so had a greater degree of insight, however, insights from China were sometimes limited due to certain source and government documents being available in Mandarin only.
In middle income countries, with relatively limited formal employment, managing resources well is particularly important when public funds are needed to provide cover for the poor. Schemes need to control their costs (through, for example, a cap on expenditure, capitation, DRGs, gate keeping, limiting corruption), as well as ensuring appropriate services are available and paid for at an appropriate rate which requires research capacity, audits and engagement, so that OOP do not escalate.
While SP appears to be working well in both Thailand and Indonesia, it is only Thailand that has managed to provide a comprehensive package, include the poor, and keep OOP low. In Vietnam and Ghana, the combination of partial implementation of SP and relatively high levels of coverage is accompanied by higher levels of OOP.
We recommend greater investment in purchaser and research capacity, and a focus on strong governance including regular engagement between purchaser, provider and citizens, that enables the building of trusting relationships. Improvements in these areas will allow countries to leverage the potential of SP more fully, thereby progressively expanding financial protection, and furthering movement towards UHC.
The evidence from nine countries suggests that purchasing reforms, while crucial, remain difficult to enact and sustain.
Benjo Delarmente (a PhD Candidate) of The John Hopkins University, Baltimore, participated in the article deliberation sessions. We are grateful for his insightful contributions.
Beryl Maritim was supported by the Consortium for Advanced Research Training in Africa (CARTA). CARTA is jointly led by the African Population and Health Research Center and the University of the Witwatersrand and funded by the Carnegie Corporation of New York (Grant No. G-19-57145), Sida (Grant No:54100113), Uppsala Monitoring Center, Norwegian Agency for Development Cooperation (Norad), and by the Wellcome Trust [reference no. 107768/Z/15/Z] and the UK Foreign, Commonwealth & Development Office, with support from the Developing Excellence in Leadership, Training and Science in Africa (DELTAS Africa) programme. The statements made and views expressed are solely the responsibility of the Fellow.
Ethical approval was not necessary because extracted data from peer-reviewed publications.
Authors declare that they have no competing interests.
This project received funding from South African Chairs Initiative (SARChI) at the Centre for Health Policy (CHP), University of the Witwatersrand, Johannesburg, South Africa.
Supplementary file 1. Search Strategy.
Supplementary file 2 contains Table S1.