Article Text

Original research
Impact of health system governance on healthcare quality in low-income and middle-income countries: a scoping review
  1. Joby George1,
  2. Susan Jack2,3,
  3. Robin Gauld3,4,
  4. Timothy Colbourn5,
  5. Tim Stokes1
  1. 1Department of General Practice & Rural Health, University of Otago, Dunedin, New Zealand
  2. 2Te Whatu Ora – Southern, National Public Health Service, Dunedin, New Zealand
  3. 3Department of Preventive & Social Medicine, University of Otago, Dunedin, New Zealand
  4. 4Otago Business School, University of Otago, Dunedin, New Zealand
  5. 5UCL Institute for Global Health, London, UK
  1. Correspondence to Mr Joby George; jobycare{at}gmail.com

Abstract

Introduction Improving healthcare quality in low-/middle-income countries (LMICs) is a critical step in the pathway to Universal Health Coverage and health-related sustainable development goals. This study aimed to map the available evidence on the impacts of health system governance interventions on the quality of healthcare services in LMICs.

Methods We conducted a scoping review of the literature. The search strategy used a combination of keywords and phrases relevant to health system governance, quality of healthcare and LMICs. Studies published in English until August 2023, with no start date limitation, were searched on PubMed, Cochrane Library, CINAHL, Web of Science, Scopus, Google Scholar and ProQuest. Additional publications were identified by snowballing. The effects reported by the studies on processes of care and quality impacts were reviewed.

Results The findings from 201 primary studies were grouped under (1) leadership, (2) system design, (3) accountability and transparency, (4) financing, (5) private sector partnerships, (6) information and monitoring; (7) participation and engagement and (8) regulation.

Conclusions We identified a stronger evidence base linking improved quality of care with health financing, private sector partnerships and community participation and engagement strategies. The evidence related to leadership, system design, information and monitoring, and accountability and transparency is limited.

  • health policy
  • quality in health care
  • patient participation
  • organisation of health services
  • primary health care
  • health services accessibility

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

STRENGTHS AND LIMITATIONS OF THIS STUDY

  • This review provides an extensive map of the studies that assess the impacts of a range of governance interventions on quality of care in low-/middle-income countries settings.

  • The findings highlight the need for more robust approaches to evaluating the impacts of governance interventions using comparable designs and measurement metrics.

  • In the absence of a commonly agreed framework for governance of quality, the interventions included in this study may not be an exhaustive list.

  • This review did not analyse the contextual, social and relational factors influencing the governance environment and its impacts on quality.

  • The heterogeneity of study designs and indicators measured in included studies makes comparisons across studies difficult.

Introduction

Achieving the health-related goals of the sustainable development goals (SDG) demands a renewed focus on improving the quality of healthcare services, particularly in the context of low-/middle-income countries (LMICs). The commitment to provide Universal Health Coverage (UHC) is an opportunity to give greater prominence to the agenda of quality of care (QoC). Mere expansion of access to health services, without intentional efforts to improve the quality, will compromise the prospects of UHC.1 Low QoC significantly contributes to excess mortality in developing countries. Improving health service quality will have a more significant overall effect on mortality than expanding service coverage alone.2 The 2018 Lancet Global Health Commission on High Quality Health Systems in the SDG era implored the national health systems to govern for quality.3

Several definitions and frameworks have been used to describe the different dimensions of quality.4–8 The Institute of Medicine described effectiveness, safety, people-centredness, timeliness, equity, integration and efficiency as the elements of quality.1 9 In this review, we adopted the Lancet Commission’s recommendation to evaluate health systems based on their impacts on people, such as competent care, user experience, health outcomes and confidence in the system.3

The Lancet Commission’s Framework describes governance as one of the foundations of high-quality health systems.3 There is no commonly agreed description of what constitutes health system governance in the context of QoC. The frequently mentioned health governance functions are leadership,3 10 11 formulating laws and policies,3 10–16 system design,10 12 accountability,3 12 14–16 transparency,13–15 information and monitoring,3 10 11 13 16 participation,13–16 regulation,10 12 16 partnerships3 10 12 and financing.3 11

In most LMICs, quality improvement initiatives primarily focus on clinical outcomes as opposed to addressing upstream governance and management practices.17 Identifying appropriate governance approaches to improve quality in LMICs is also challenging. Rigorous evaluations of the impact of interventions for non-facility determinants of quality, such as policies and management of healthcare organisations, are rare.18–21 The adaptation of available evidence from high-income countries to the unique contexts of LMICs is another barrier.21

Earlier reviews have provided valuable insights into the linkages between governance mechanisms and healthcare quality in LMICs.11 22–24 A review of experiences from 25 countries highlighted promising practices such as the explicit inclusion of quality as a priority in health planning, establishing dedicated institutional structures, establishing mechanisms to monitor quality and allocating resources to improve quality.23 This paper did not analyse the impacts of those interventions on QoC at the service delivery level. An evidence-gap map of primary healthcare policy and governance in LMICs identified gaps in social accountability, public–private partnerships (PPP) and intersectoral collaboration.24 Other reviews have analysed the impacts of specific governance interventions such as demand-side and supply-side health financing strategies,25–34 stakeholder and community engagement,35 36 social accountability mechanisms,37–39 private sector partnerships40–46 and regulatory approaches.43 47–49 However, a broader mapping of the various governance interventions linked to the quality of healthcare services in LMICs is absent. This scoping review addressed this gap by mapping the available evidence on the impacts of health system governance interventions to improve healthcare quality in terms of care processes and quality impacts.

The review questions were: (1) What is known about the impacts of health system governance strategies or interventions on healthcare quality in LMICs? (2) What are the knowledge gaps regarding effective governance interventions to improve healthcare quality in LMICs?

Methods

Study design

This scoping review follows the established methodology for conducting and reporting scoping reviews.50–56 The study findings were analysed for effects on processes of care and quality impacts. We adopted the categorisation used by the Lancet Global Health Commission, which also incorporates the concepts from several other frameworks, to analyse and group the effects on QoC under care processes and quality impacts (table 1).3

Table 1

Definitions of quality healthcare applied in this review (adapted from Kruk et al.)3

The study protocol was registered with Open Science Framework (https://doi.org/10.17605/OSF.IO/BF75P). We present our findings in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews checklist (see online supplemental file 1).55

Eligibility criteria

We included English language research studies or programme publications presenting quantitative and/or qualitative data published on or before 31 August 2023 with no start date limitations. These studies investigated the impact of one or more governance interventions on healthcare quality in one or more LMICs. The following studies were excluded: those that did not report on healthcare quality in terms of care processes or quality impacts; those that reported effects only on the utilisation of services; those that implemented only clinical improvement tools (eg, clinical audit); and those that focused only on improving infrastructure, training of health human resources or pharmaceutical quality.

Data sources and literature search

The research strategy was developed based on a preliminary literature review to identify the essential governance functions and related interventions. The search strategy (see online supplemental file 2) used a combination of keywords and phrases relevant to health system governance, quality of healthcare and LMICs. The countries included were those classified as low-income, lower-middle-income and upper-middle countries based on the World Bank Atlas method for the 2021 fiscal year.57 We conducted searches of peer-reviewed journals and grey literature in seven electronic databases: PubMed, Cochrane Library, CINAHL, Web of Science, Scopus, Google Scholar and ProQuest Central. We also searched the reference list of included systematic and scoping reviews to identify additional primary studies relevant to this review. The websites of international development organisations were also searched to identify publications.

Study selection

All references searched from electronic databases were imported into Covidence (www.covidence.org), and duplicates were removed. Two reviewers (JG and TS) screened the titles and abstracts and reviewed the shortlisted full-text articles. Conflicts at both stages were resolved through discussion. Relevant primary studies from previous reviews and additional publications from grey literature were identified by JG and reviewed by TS before inclusion into the review.

Data charting and extraction

Data were extracted by JG using a data extraction template, which was refined and modified by JG and TS based on the experiences from the extraction of the first few papers. The information extracted included the study title, country, year of publication, objectives, study design, intervention(s), health services and the descriptive results on care processes, user experience, health outcomes, confidence in health systems and economic benefits. The country income classification was assigned using the World Bank list for 2021.57

Analysis and reporting

The analysis framework was developed by JG and reviewed by the TS. A numerical summary of the studies that reported any of the impacts on QoC was prepared. The descriptive findings were analysed using the content analysis method. The studies were grouped by the nine governance domains identified by the authors a priori based on the preliminary review of literature on the key functions of health sector governance and related interventions. Studies that used multiple governance interventions were grouped under a new category of ‘multiple domains.’ Content analysis was conducted for each of the governance domains and each type of quality impact.

The results were reported to show a numerical mapping of the availability of evidence on QoC under each of the governance domains. The findings from the content analysis were summarised by governance domain and type of interventions. JG conducted the analysis with support from TS. SJ, RG, and TC reviewed the analysis reports and provided comments.

Consultation with stakeholders

A summary of the findings of this review was shared with 19 stakeholders who have expertise in supporting LMIC health systems for improving healthcare quality, seeking their perspectives, and exploring any additional studies for inclusion. The inputs from the consultation were incorporated into the final report.

Patient and public involvement

None

Results

Search results

A total of 8299 articles were retrieved, 7407 articles underwent abstract screening, and 203 papers were selected for full-text review, of which 109 were excluded after the review. Snowballing identified an additional 107 primary studies. A total of 201 articles were selected for the final review (see figure 1). A list of excluded papers and reasons for exclusion are presented (see online supplemental file 3).

Figure 1

Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram.

Characteristics of included primary studies (n = 201)

Overall, 64% (n=128) of the studies were conducted in lower-middle-income countries, followed by 25% (n=49) in low-income countries, 9% in upper-middle-income countries (n=18) and 3% (n=6) in multiple countries. Study designs included 39 randomised controlled trials, 65 quasi-experimental, 77 observational or descriptive studies and 20 qualitative studies. Most studies evaluated the quality impacts on reproductive, maternal, newborn and child health (RMNCH) services (n=135). General primary care (n=52), hospital inpatient care (n=11) and communicable diseases (n=3) accounted for the remaining studies.

Findings from included studies

This section presents a numerical summary of the types of impacts on QoC reported by the included studies. Most studies reported impacts on competent care and systems (n=149), and confidence in health systems, primarily evidenced by improved utilisation of services, were reported by 105 studies. Evidence on user experience (n=94), better health (n=62) and economic benefits (n=41) were reported by fewer studies (see table 2). It is pertinent to highlight that several studies reported using composite QoC scores, which are presented here under competent care and systems.

Table 2

Summary of studies reported impacts on healthcare quality

Impacts on quality by governance domains (n=201)

This section summarises the descriptive findings from all primary studies (n=201). Health financing interventions (n=114) accounted for the majority (57%) of studies analysed, followed by engaging the private sector (n=24) and regulation (n=11). Leadership (n=3), information and monitoring (n=8), accountability and transparency (n=7), system design (n=8), and participation and engagement (n=9) had fewer studies (see online supplemental file 4).

Leadership

Three studies analysed interventions related to leadership, policies and strategic plans.58–60 Government stewardship of private clinics was associated with improved overall QoC and care processes.60 Formulating a national quality policy and strategy in Lebanon and Jordan led to improved licensing and regulatory systems.58 No patient-level impacts were reported. Developing and implementing governance action plans in Afghanistan resulted in mixed effects on care processes and quality impacts.59

System design

We included eight studies61–68 that analysed system design interventions. The interventions included decentralised prioritisation and planning, initiatives to improve managerial efficiency and local-level policy initiatives. Positive effects included reduced medication errors,61 increased screening of pregnant women for HIV,63 high patient satisfaction with the services61 62 64 65 68 and trustful relations between the providers and users.61 62 Negative effects were reduced access and affordability of services and client satisfaction,67 decreased utilisation of services and additional costs incurred in procuring the drugs from private providers.68

Decentralisation

Decentralisation of decision-making, planning and implementation had varied impacts on QoC.65–68 In Sudan, decentralisation resulted in the deterioration of the overall QoC and utilisation of services. Availability and affordability of services were reduced, and inequity in service accessibility increased.67 In Nigeria, the low availability of equipment, drugs and supplies left many clients dissatisfied with the services. 68 Increased autonomy of hospitals in Indonesia for decentralised planning, budgeting, and management showed overall positive results in structural quality and client satisfaction.65 In Kenya, decentralisation was helpful to improve accessibility and availability of services, but the QoC remained low owing to low investments in infrastructure and staffing. Stakeholders also believed that quality needs to be emphasised better in the priority-setting process.66

Strengthening management systems

In Tanzania, strengthening management systems and community engagement resulted in improved availability of reproductive health services, structural readiness, client-perceived QoC and high levels of client satisfaction.64 The integrated supervision system in Nigeria improved a few aspects of clinical quality but no changes in the coverage of services like immunisation.63

Local-level policy initiatives

The treatment-before-deposit policy in China62 and no-class-wards in Indonesia61 reduced perceptions of discrimination and improved trust in providers. The perceptions of quality and satisfaction also improved.

Accountability and transparency

All eight studies analysed the effects of community scorecards or other social accountability tools.69–76 Studies in the Democratic Republic of Congo75 and Malawi74 analysed the use of community scorecards to strengthen the accountability of health service providers. Positive effects included improved quality of antenatal counselling, provider attitudes, and rapport and fairness, and reduced incidence of bribes. Community meetings and tools for citizen voices and action were implemented in Uganda, India, Cambodia, Afghanistan and Ghana. Positive effects included a better screening of children,69 71 improved provider courtesy and provision of information,70 72 improved nutritional status among children and decline in child mortality,69 and enhanced trust and confidence in providers and higher utilisation rates of services. In Kenya, the clients reported being treated respectfully, though some negative provider behaviours and long waiting times persisted.76 In India, the results were not so positive, with no changes in maternal and newborn care service quality or outcomes.73 None of the studies reported any negative effects.

Demand-side financing

A total of 55 studies77–131 examined interventions such as the removal or exemption of user fees (n=18), health insurance (n=19), vouchers (n=9) and conditional cash transfers (CCTs; n=9).

Complete or partial removal of user fees

Eighteen studies reported the effects of total or partial removal of user fees.77–92 95 110 All except three studies86 87 90 involved user fee exemptions specifically for maternal healthcare services. Partial or complete removal of user fees generally did not significantly alter the technical quality of services.78 80–82 85 87 89 90 110 However, a few studies reported increased waiting times, reduced duration of consultations, worsening provider attitudes and inadequate drugs, supplies and equipment.83 85 89 92 95 The utilisation of services, particularly for antenatal care, delivery, child vaccinations and management of delivery-related complications improved in most instances. Client satisfaction with the free services was high.79 82 87 88 90 95 110 One study reported that the free services were only partially free as patients continued to pay for certain services, such as investigations or medicines, informal payments, and the indirect costs of seeking care.85 Low staff morale among providers resulting from increased workload has also been documented.84 Four studies reported better health outcomes.80 82 87 110

Some of the negative consequences include low privacy, poor hygiene, lack of compliance with obstetric care standards, neglect by the providers, longer waiting times, reduced availability of drugs and decreased trust due to unofficial payments.83–85 89

Insurance or risk-pooling

The effects of health insurance were analysed in 19 studies.93 94 96–109 129–131 The results ranged from improved perceived and actual QoC93 94 104 107 to the perceived worsening of quality.96 100 105 Increased service utilisation has been reported by several studies, particularly for antenatal care, outpatient visits and facility births. Client satisfaction levels show wide variations across schemes. The study from China concluded that insurance increased unnecessary service provision and substantial costs for the poor.105 Better health outcomes were reported by two studies102 107 and economic benefits by four studies.93 97 103 107 In contrast, the study from Costa Rica challenged the notion that insurance can lead to significant reductions in infant and child mortality.99 There was no difference between insured and uninsured clients in the perceived responsiveness of outpatient services in Ethiopia.130 In Zambia, insurance was associated with greater confidence in health systems and improved care experiences.129 131 As in the case of user fee exemptions, QoC may need more investments to improve structural quality in health facilities.93 101

Vouchers

Nine studies examined the effects of voucher schemes on the quality of reproductive health or maternal healthcare services.111 112 115 116 118 119 124 127 128 Generally, voucher schemes improved the demand for family planning and maternal health services and reduced inequities in the utilisation.111 112 115 116 124 127 128 The effects on clinical QoC were varied, with four reporting positive effects.111 115 118 119 Client satisfaction levels were generally positive.118 124 127 128 Two studies reported better health outcomes.112 115 None of the studies reported adverse effects on care processes or quality impacts.

Cash transfers

Nine studies analysed the effects of CCTs on the quality of maternal, newborn and child health services.113 114 117 120–123 125 126 Three studies from India, which examined the effects of the Janani Suraksha Yojana, reported that the intervention led to a substantial increase in the uptake of maternal health services.117 121 123 Two of them suggest a possible decline in neonatal mortality.117 123 Three studies from Brazil, which analysed the impact of Bolsa Familia on the quality of child health services, reported a significant reduction in child mortality resulting from postneonatal conditions. The effects were more prominent for mortality related to malnutrition and diarrhoea.122 125 126 One study also reported increased child vaccination coverage and antenatal care and reduced admission of under-5 children to hospitals.122 The study on Oportunitades in Mexico showed significant improvements in the quality of antenatal care for low-income rural women,113 and that of Progresa showed a significant reduction in infant mortality rate, particularly in areas with low socio-economic indices.114 The programme in Nigeria120 positively affected the utilisation and quality of maternal health services, but there was no reduction in mortality. None of the studies reported any negative effects of CCT on care processes or quality impacts.

Supply-side financing

Fifty-nine studies132–190 examining supply-side financing interventions are included.

Financial incentives

Incentivised payments linked to healthcare performance are known by several names—performance-based incentives (PBI), pay-for-performance, results-based financing (RBF), performance-based financing (PBF), and results-based incentives.

Though there are wide variations in the design of the schemes in different countries, the intervention works by providing additional financial payments linked to a set of performance indicators. In most designs, quality indicators are part of the performance criteria set for the payment. Incentives are often paid as bonuses to the service providers proportionate to achieving performance targets. Indirect incentives, which operate at higher levels of the system, are also included in many of the designs. This paper uses the term PBF to describe all these interventions.

Most of the supply-side financing studies (n=48) fall under this category. Most of the studies reported a positive impact of the incentives on service utilisation for the incentivised services. The improvements were greater for services with the highest incentive payment and those that required the least effort from the provider.135 Most of the studies reported improvements in clinical QoC,133 135 139–142 144 149 151 155 156 159 162 163 166 167 170–176 178 182 184 185 187 and client satisfaction levels.132 133 143 144 146 149 154 159 160 166 170–172 176 178 183 185–187 189 Three showed a reduction in mortality rates,145 184 187 and two reported improvements in child nutritional status.151 168 In Tanzania, the incentives significantly reduced bypassing of facilities.137 The quality of non-incentivised services either improved or remained unchanged, resulting from the quality-multiplier effect.132 179 183 Instances of disrespect, abuse and neglect were reported by women seeking care from PBF facilities in Malawi.157 No studies reported negative consequences on quality impacts.

One study reported a decline in the use of non-targeted services for children,138 and one reported no effects on non-targeted services. 154 The availability of demand-side incentives and addressing the weaknesses in the service delivery and management capacities are essential to the effectiveness of supply-side incentives.141 148 157 161 163

Non-financial incentives

Two studies from Malawi180 and El-Salvador136 reported the effects of non-financial incentives on RMNCH services. Both studies reported improvements in clinical QoC and utilisation of services.

Performance incentives, in combination with other interventions

A study in Bangladesh, which compared the effects of PBF, CCT and a combination of both,172 reported that PBF and CCT led to significant improvements in clinical QoC, client satisfaction and utilisation of maternal healthcare. The clinical quality for antenatal and delivery care was better in the PBF group, while utilisation was higher in the CCT group. A combination of both interventions resulted in significantly higher client volumes. In China, a policy initiative to pay a capitated budget proportionate to the number of clients to township health centres and village posts, along with PBF, improved drug prescribing practices and reduced cost per visit.182 In the Philippines, the health insurance scheme combined with PBF for physicians resulted in sustained improvements in clinical quality scores.170 The study in Burkina Faso, Ghana and Tanzania to implement an electronic decision-support system coupled with PBF did not show improvements in antenatal and childbirth care quality despite high acceptance of the new technology.147 The RBF and CCT for maternal healthcare in Malawi led to improved clinical practices during childbirth and a corresponding decline in in-facility maternal mortality.145 In Zimbabwe, implementing continuous quality improvement in facilities under the PBF scheme improved compliance with maternal healthcare services at the primary healthcare level. In contrast, no improvements were reported for other services or hospital services.190

Cost recovery approaches

All three studies on the introduction of user fees were from low-income countries. Three studies from Zaire, Niger and Eritrea examined the effects of user fees on QoC.134 152 181 The results show a mix of positive and negative effects. The positive outcomes include improved interpersonal aspects during the antenatal period and high levels of patient satisfaction. The negative consequences were more, which concerns about widening inequities and low QoC,134 decreased client satisfaction related to waiting time,152 bypassing of lower-level facilities,134 reduced utilisation of services and increased cost of consultations.152

Engaging the private sector

A total of 24 studies which examined different types of health systems engaging with the private sector for health service delivery are included.

Franchising

The fourteen studies from Pakistan, Nepal, India, Vietnam, Myanmar, Cambodia, Ghana and Kenya evaluated various franchising models.191–204 One analysed the results from the social franchising programmes in 17 Asian and African countries.197 Almost all franchises provided reproductive health or maternal and child health services, except the one from Myanmar,195 which implemented tuberculosis control interventions. Overall, franchising interventions improved the availability of services and supportive products and supplies, counselling and provision of information. Client satisfaction and client loyalty were found to be high, resulting in increased patient volumes. The quality of maternal and newborn care services in Pakistan and India remained low.194 199 200 203 Studies on Sky Health, Merrygold, and Matrika franchising in India reported overall low quality of services, though improvements in provider behaviour and client satisfaction were reported.194 199 203 An increase in the cost of care was reported as a negative effect.192

Contracting

Six studies examined the contracting of government-funded health services to non-governmental providers.200 205–209 The results show a high degree of variation across countries. In Cambodia, there was improved management of diarrhoea, but the clients’ perceptions of provider attitudes were negative.206 In Pakistan, the clinical quality of services improved, but the overall quality remained low. However, the client’s perception of quality and satisfaction levels were high. Utilisation rates and client-reported quality improved.200 208 209 The results from Bangladesh were generally very positive, with improved availability and quality of services, higher utilisation rates, reduced mortality, improved nutritional status and high levels of client satisfaction.205 207 One study from Bangladesh reported a reduction in mortality rates,205 while the study from Cambodia reported reduced morbidity among children.206

Contracting negatively affected staff attitudes, provider competence and availability of equipment and supplies.200 206

Public–private partnership

PPP models are examined in four studies from India210–212 and Lesotho.213 The Chiranjeevi Yojana in India was associated with more and better clinical services, reduced waiting times and high patient satisfaction. The intervention was not associated with a change in the uptake of maternal and newborn care services or the management of complications.211 212 The PPP hospital networks in Lesotho resulted in better health outcomes, such as lower hospital mortality rates, lower stillbirth rates, and improved survival of low birth weight newborns.213 Poor equipment availability and inappropriate staff attitudes led to low user satisfaction.210

Private sector capacity strengthening

The only study from Kenya analysed the effects of an intervention to strengthen the capacity of the private sector to improve the quality of general healthcare services. The results were mixed, with improved interpersonal aspects of care and reduced unnecessary procedures and waste. However, the intervention was associated with a reduction in the correct management of outpatient cases and showed major deficiencies in laboratory quality. There were no changes in the patient perceptions of quality or client satisfaction.214

Information and monitoring

Eight studies that analysed various information and monitoring interventions are included in this review.215–222 In Mexico, the study involved benchmarking by measuring effective coverage to monitor progress, foster accountability and create a culture of evidence. This improved effective coverage of maternal and child health services, though some inequities remained unchanged.217 Balanced scorecards used in Afghanistan as a national health service performance assessment tool showed improved availability, quality and equity of service provision and client and provider satisfaction.216 Quality-of-care audits of perinatal mortality in South Africa did not establish an effect of the intervention on perinatal mortality. However, more facilities were able to identify modifiable factors and take remedial actions.215 A dashboard-driven patient safety programme in India showed significant improvements in composite quality scores and compliance with patient safety protocols.220 Healthcare performance evaluation in three districts of Ethiopia, Tanzania and Uganda showed improved quality and better governance in decision-making, accountability and allocation of resources.222

Holding quality contests among health facilities in Morocco improved the quality scores of the primary healthcare facilities participating.219 In Kenya, the intervention to apply data-driven prioritisation at the health facilities contributed to significant improvements in structural readiness, better infection prevention and control, compliance with clinical protocols, reduced waiting times and reduced neonatal mortality.218

Participation and engagement

Four randomised controlled trials in Nepal, Bangladesh and Malawi evaluated the effects of participatory women’s groups and strengthened health services for maternal and newborn care.223–226 Three of these studies reported a significant reduction in neonatal mortality rates.224–226 One study showed a decrease in maternal mortality,226 while two trials did not affect maternal mortality.223 224 One trial in Bangladesh, which did not show a reduction of neonatal mortality rate, highlighted the importance of appropriate design to reach the coverage of the intervention and address the contextual factors.223

Four other studies from India, Ghana and Malawi examined the effects of community mobilisation without inputs for health service improvements.227–230 The results were mixed, with one study reporting improved care processes such as information provision and respectfulness, and one study showed a reduction in the rates of stillbirths, neonatal mortality and perinatal mortality. 230 The effects on the utilisation of health services varied. None of the studies reported negative effects on care processes or quality impacts. A study from Indonesia reported the low willingness and readiness of service recipients to engage in patient safety initiatives.231

Regulation

Eleven studies evaluated the impacts of various forms of regulation of health facilities on QoC.232–242 In South Africa, hospital accreditation was associated with improved compliance with quality standards. It also improved patient satisfaction with care.239 Both the studies from Egypt showed improved compliance with clinical protocols.233 235 One study also reported reduced morbidity among children and improved family planning and maternal care services uptake.235 In Turkey, accreditation improved quality management scores, improved infection prevention practices and improved hospital patient handling and medication practices.242 Hospital accreditation in Thailand reduced hospital mortality related to stroke and sepsis and significantly increased client volumes.240 Client satisfaction levels were high in two studies.233 239 The quality improvement support for private hospitals through the Manyata certification resulted in significant improvements in the overall quality scores of the facilities.241

The accreditation of health service providers linked to insurance payments in the Philippines improved quality scores among physicians in both public and private hospitals.238 Studies based on quality certification programmes were found in Egypt,236 India232 and Tanzania.237 In Ethiopia, clinical and administrative standards implementation improved quality in all areas assessed.234 The results from these four studies were heterogeneous, with Gold Star certification of family planning clinics in Egypt showing significantly improved availability of family planning products and quality of service provision.236 The SafeCare certification in Tanzania did not lead to improved clinical quality.237 The overall impact of National Quality Assurance Standards (NQAS) certification in India was low, though structural aspects of the quality, such as infrastructure, human resources and supplies, improved.232

Multiple governance domains

Sixteen studies included in this review implemented interventions across multiple governance domains.243–258 Studies from the Philippines,250 South Africa243 and Madagascar247 show remarkably positive effects on clinical quality, utilisation and health outcomes. CCT and the expansion of the Family Health Programme in Brazil resulted in a significant reduction in perinatal mortality rate over 12 years.249 Decentralisation and regulatory changes in Indonesia did not significantly improve the QoC for prenatal and adult care. They led to inequities in the distribution of health human resources.245

The effects of several health governance interventions, including hospital accreditation, the introduction and subsequent withdrawal of PBIs, and the introduction of user fees, are reported in the study from Egypt.246 Health insurance combined with local-level leadership for health sector reforms such as improved infrastructure, supplies, human resource development, service delivery, accountability and regulatory oversight in the Philippines resulted in increased uptake of institutional deliveries and a reduction in maternal mortality.250 In Nigeria, the combination of health insurance and facility upgrades increased hospital deliveries but did not significantly impact maternal mortality.244

A social franchising intervention, along with free vouchers, training and accreditation of providers in Pakistan, reported high levels of user satisfaction, increased uptake of family planning services, and trust in providers.254 A participatory community-led health system intervention and quality improvement initiative in Tanzania significantly reduced the proportion of women experiencing disrespect and abuse during childbirth and improved client satisfaction.252

A combination of health insurance and franchise midwife clinics in the Philippines showed increased prenatal care, early initiation of prenatal care and facility births. The visits to franchise midwife clinics did not improve prenatal care standards.251

Contracting health services to the private sector and PBF in Cambodia led to a shift from delivery at home and private clinics to public health facilities. This change, however, did not translate into improved neonatal health outcomes due to deficiencies in the QoC at public facilities.253 Supervision and incentives in the form of salary top-ups and housing arrangements to improve the quality of integrated management of childhood illnesses showed that the incentives were more effective in improving quality and patient satisfaction.248

Implementing SafeCare accreditation standards and health system improvements improved overall QoC standards and compliance with clinical protocols for surgery, anaesthesia and overall outpatient services in Nigeria.255 Regulation of fees and health insurance reduced hospital admission rates for ambulatory care-sensitive conditions for hypertension in Ghana.256 At-scale implementation of Every Mother Every Newborn Quality Improvement standards, including better information and monitoring and community engagement, improved compliance with maternal and newborn clinical standards, improved patient rights and better health outcomes for mothers and newborns in Bangladesh, Ghana and Tanzania.257

Discussion

There is growing support for enhancing the quality of healthcare delivered in LMICs. Some recent global guidelines and publications supported LMICs in developing health policies and reform initiatives emphasising quality and safety in healthcare.3 9 21 259 260 However, the efforts to establish a strong evidence base to inform and evaluate such initiatives remain inadequate. This review was able to map the available evidence linking governance interventions to impacts on QoC. The study has also pointed to areas where more robust research may be required.

This review identified a stronger evidence base linking improved QoC with health financing, private sector partnerships and community participation and engagement strategies. Studies related to leadership, system design, information and monitoring, and accountability and transparency are limited. Though one of the earlier reviews highlighted the potential links between leadership interventions and healthcare quality,11 the evidence for a causal relationship between governance initiatives and health system performance is lacking, which makes it difficult for governments and donors to make investment decisions.11 59 As previously noted by a review, there needs tobe more research evidence on the effectiveness of strategies involving legislation and regulatory mechanisms to improve health service quality.22 Approaches involving an appropriate constellation of governance interventions involving demand and supply-side financing, accreditation and accountability mechanisms offer better prospects of improving QoC.

A wide range of factors influenced the achievement of positive impacts on quality. These include health system context, levels of quality at baseline, contextualisation of the intervention designs, acceptability of the intervention by the community, providers and stakeholders, quality of implementation, availability of other health systems inputs such as additional donor support, duration of implementation, technical support available, completeness and reliability of data, and the design and robustness of the evaluations. An important observation from this evidence-mapping exercise is the heterogeneity of results from similar interventions in multiple contexts. The broader social, political and economic contexts and the overall national governance environment also have considerable influence on the governance of QoC, which remains underinvestigated. Therefore, it is essential for studies evaluating the impacts of governance approaches on QoC to analyse and report the organisational and contextual factors influencing the outcomes. The causal pathways of how governance interventions lead to positive impacts on QoC also need to be better analysed and explained.

The findings presented are consistent with the conclusions of some of the earlier reviews on specific interventions. Previous reviews on the impacts of PBF have noted improvements in utilisation but inconclusive evidence on its effects on quality outcomes.34 A review of women’s groups’ participation and learning significantly impacted maternal and newborn survival.35 Other reviews reported inconclusive evidence for the effectiveness of hospital accreditation on quality and patient safety outcomes.47 261

Studies from high-income country contexts similarly indicate the importance of upstream governance and management practices influencing QoC.262 The findings from a review of governance and leadership in seven developed countries show greater emphasis on evidence-based priority setting and performance monitoring. At the same time, uncertainty remains on optimal mechanisms for accountability.263

The main strength of this review is that it has strung together a range of governance strategies and interventions to analyse their impacts on healthcare quality at the service delivery level. This review has several limitations. First, there is no agreed-upon definition or framework that identifies all the governance functions or interventions to improve QoC. The governance interventions identified in this review are not an exhaustive list of potential interventions to improve quality. Second, the health systems operate in complex and highly variable environments. The impact of the interventions on QoC is also influenced by organisational, social and relational factors such as social norms, trust and values.17 264 This review did not analyse the interactions and relationships of these factors and their influence on quality, which is one of the limitations. Third, the heterogeneity in the design and metrics used to measure quality limits the comparability across studies. This has been identified as a challenge in a recent systematic review of the impacts of PPP on QoC.46 Most studies limit their evaluations to changes in structural aspects of quality or utilisation of services, while many others use composite scores for overall quality of services.

The study findings will be valuable to inform future research priorities, including the need for a harmonised approach to selecting indicators for measurement while evaluating the impact of interventions on healthcare quality. This review is of broader relevance to governments, policy-makers and programme managers, donors and other development partners working to improve healthcare services in LMICs.

Conclusion

We identified more robust evidence linking improved QoC with health financing, private sector partnerships and community participation and engagement strategies. The evidence base related to leadership, system design, information and monitoring, and accountability and transparency needs to be improved. More robust evaluations of policy and health reform initiatives intended to impact QoC are required. Studies could use a more harmonised measurement framework, which incorporates aspects of care processes and quality impacts in their evaluations.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Ethics statements

Patient consent for publication

References

Supplementary materials

Footnotes

  • Twitter @geojo925, @timcolbourn, @StokesTim63

  • Contributors JG conceptualised and designed this study with inputs from TS, RG, SJ and TC. JG conducted the literature search. JG and TS screened the articles and conducted a full-text review of the included studies. JG conducted the data extraction and analysis with inputs from TS, RG and SJ. JG drafted the manuscript with inputs from TS, RG, SJ and TC. All authors revised the manuscript and approved the contents before submission. JG is the guarantor for this article.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.