Health worker capacity-building interventions and quality of primary healthcare delivery in sub-Saharan Africa: a systematic review protocol

Abstract

Background

Achieving high-quality primary healthcare (PHC) remains essential to improving health systems performance and advancing progress towards attainment of universal health coverage, particularly in sub-Saharan Africa (SSA), where gaps in service delivery persist. Capacity-building interventions, such as mentorship, in-service training and supportive supervision, have been widely employed to improve the quality of PHC delivery. The evidence base is, however, fragmented across diverse settings, intervention types and outcome measures. This systematic review will examine the evidence on how capacity-building interventions improve the quality of PHC services in SSA, with a focus on the domains of effectiveness, safety, people-centredness, timeliness, equity, integration and efficiency as propounded by the WHO.

Methods

We will conduct a systematic review using established evidence synthesis methods and report the findings in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines. Eligible studies will include peer-reviewed and grey literature published in English between 2000 and 2025, focusing on capacity-building interventions aimed at improving PHC quality in SSA. We will search databases including PubMed, Embase, Web of Science, Scopus, Google Scholar and the African Journals Online. Study selection will follow the Population (PHC providers), Exposure (capacity-building interventions) and Outcome (quality of PHC delivery across WHO domains) framework. Quality assessment will use the Mixed Methods Appraisal Tool (MMAT). Data will be narratively synthesised using Atlas.ti software.

Ethics and dissemination

Ethical approval is not required because there will be no collection of primary data. Only published studies/records available on peer-reviewed literature and grey sources will be included. Findings will be disseminated through a peer-reviewed journal, academic conferences and stakeholder platforms in SSA.

Registration

The protocol has been registered in PROSPERO (CRD420251131534) and reported according to PRISMA-P guidelines.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• Our protocol follows PRISMA-P 2015 guidance and has been registered in PROSPERO.

• Our systematic review will apply a systematic search approach across multiple electronic databases.

• Our use of the Population, Exposure and Outcome (PEO) framework will ensure inclusion of diverse study designs (quantitative, qualitative and mixed methods).

• We will assess methodological quality using the Mixed Methods Appraisal Tool to cover different study types.

• Our use of the PEO framework will ensure inclusion of diverse study designs (quantitative, qualitative and mixed methods).

Background

Global health systems acknowledge that a skilled, supported health workforce is essential for improving the implementation of primary healthcare (PHC) and achieving universal health coverage (UHC). The WHO, however, projects a global health worker shortfall of approximately 11 million by 2030.¹ Capacity building has proven to be an important tool in improving PHC implementation.² Capacity-building systems differ between high-income countries (HICs) and low- and middle-income countries (LMICs).³ HICs, for instance, integrate workforce development into education, regulation and incentives.^{4 5} By contrast, LMICs face underinvestment, limited job absorption and reliance on short-term donor training, leaving a gap between workforce supply and health needs.^{1 6 7}

In this review, capacity-building mainly refers to training, mentorship, supportive supervision and digital innovations that improve service delivery. Key capacity-building interventions for PHC workers include in-service training, mentorship, supportive supervision, job aids, decision support tools and clinical skill strengthening.8,10 Formal in-service training and workshops have repeatedly been shown to increase provider knowledge, skills and confidence, often translating into improved practices.^{11 12} Supportive supervision, involving regular on-site mentoring, feedback and problem-solving sessions, helps reinforce training.¹³ For example, integrated supportive supervision is a process designed to boost service quality by improving communication and mentoring health workers on-site.¹⁴

Various capacity-building interventions are being applied across sub-Saharan Africa (SSA) with a region-specific focus. In West Africa, for instance, Ghana and Nigeria have introduced programmes deploying community health nurses within structured supervisory systems. In East Africa, Ethiopia, Kenya and Uganda emphasise training and supervision, with mobile health (mHealth) modules and enhanced mentorship improving workforce capacity.¹⁵ In Southern Africa, Zambia has led in integrating digital supervision, with mobile-based support improving immunisation indicators.¹⁶ Together, these examples demonstrate how blended capacity-building strategies are being used to strengthen primary care in SSA.17,21 These efforts align with the Sustainable Development Goal (SDG) 3.8, which emphasises the achievement of UHC, including access to essential healthcare services and financial risk protection.^{6 16 17 22}

Despite the recognition of capacity-building as a strategy to improve the quality of healthcare delivery, it remains largely unexplored and fragmented in SSA across contexts. Recent evidence shows that while capacity-building initiatives such as mentorship, skills transfer and digital supervision are increasingly adopted, they often remain siloed within disease-specific programmes and lack system-wide integration.²³ Health Technology Assessment (HTA)-related activities are uncoordinated, poorly documented and disconnected from policy. This reflects a broader issue of fragmented capacity-building efforts and a lack of consolidated evidence to guide strategic planning ²⁴ . This fragmentation is partly due to the absence of standardised evaluation frameworks and the short-term nature of many donor-funded initiatives, which limit longitudinal assessment and cross-country comparability.^{25 26} Moreover, most interventions are not designed with scalability or sustainability in mind, further weakening their impact on broader health system quality. A recent landscape assessment also highlights persistent gaps in institutional capacity, gender equity and long-term investment planning, underscoring the need for more coordinated and inclusive approaches to capacity-building.²⁷

The present review, therefore, seeks to synthesise the existing evidence of what strategies work best in improving healthcare quality. It builds on earlier systematic reviews of capacity-building in public health practice^{16 28} while contextualising lessons for sub-Saharan settings. By linking global evidence with local innovations, the review aims to guide policymakers and programme designers in shaping sustainable and context-appropriate workforce strategies for universal primary care and achievement of SDG 3.

Conceptual framework

The WHO’s Quality of Care Framework has been adapted to underpin this systematic review. The framework was developed in 2018 as part of the WHO’s agenda to accelerate progress towards UHC.²⁹ The framework was specifically designed to provide countries, particularly LMICs, with a comprehensive model for assessing and improving the quality of PHC services.

The WHO framework sets out a multidimensional understanding of healthcare quality, identifying key tenets that must be present for services to be considered high-quality.³⁰ These attributes include safety, which ensures that care minimises harm and risk to patients; accessibility, which guarantees that services are available in a timely manner and without financial or geographic barriers; acceptability and patient-centredness, which emphasise respect, responsiveness and cultural appropriateness of care; equity, which ensures that all population groups fairly receive the same standard of care irrespective of demographic or socioeconomic differences; efficiency which requires the optimal utilisation of resources; integration, which stresses the need for services to be connected across levels of care; continuity, which emphasises follow-up and sustained care over time; and timeliness, which highlights the importance of prompt service delivery.³¹ Collectively, these domains provide a broad, practical framework for evaluating and strengthening the delivery of PHC.

The framework has been applied across various contexts, particularly in LMICs, to guide health system reforms by strengthening the measurement of effective coverage and service quality, thereby enabling countries to target resources and implement quality improvement initiatives^{31 32} and evaluate interventions aimed at enhancing PHC delivery.³³ For instance, in several SSA countries, it has been adapted to monitor maternal and child health programmes, assess continuity of care for chronic conditions and evaluate patient-centred approaches within community health systems.³⁴ Beyond SSA, countries in Asia³⁵ and Latin America have also employed the framework to measure progress in integrating PHC with broader national health agendas.³⁶ Its flexibility and comprehensiveness make it adaptable to diverse health system contexts, whether resource-rich or resource-constrained.³³

For the purposes of this study, the WHO Quality of Care framework is considered appropriate because it directly addresses the complexity of PHC delivery in SSA.³⁶ It encompasses broad and contextually relevant dimensions such as accessibility, continuity and integration, which are factors that are particularly critical in the region’s health systems.³¹ Furthermore, the framework’s alignment with SDG 3, which calls for ensuring healthy lives and promoting well-being for all at all ages, makes it a powerful tool for analysing how capacity-building interventions contribute to both health worker performance and broader system outcomes.²² By employing this framework, we will be able to situate our systematic review within a globally recognised model while also producing evidence that is immediately relevant to policymakers, health managers and practitioners working to strengthen PHC systems in SSA.³²

Figure 1. 36: WHO Quality of Care Framework.

Materials and methods

Protocol registration and reporting

This systematic review protocol has been developed in line with established international standards for evidence synthesis. The protocol has been registered with the International Prospective Register of Systematic Reviews (PROSPERO: CRD420251131534) database to ensure transparency, avoid duplication and enable peer scrutiny. The protocol has been reported following the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) 2015 statement, which provides guidance for transparent reporting of systematic review protocols.³⁷ The conduct of the review will follow methodological guidance from the Joanna Briggs Institute (JBI) Manual for Evidence Synthesis, and the final report will be prepared in accordance with the PRISMA-P reporting guidelines.³⁸

Given that we will adopt a narrative synthesis approach in our results evidence synthesis, the review will also comply with the SWiM (Synthesis Without Meta-analysis) guidelines,³⁹ which provide structured guidance for transparently reporting narrative syntheses. The JBI Manual for Evidence Synthesis⁴⁰ will guide the processes of study selection and data extraction. Any amendments to the registered protocol during the course of the review will be documented, justified and reported transparently in both the PROSPERO record and the final publication.

We acknowledge that restricting eligible publications to English may introduce language bias and limit inclusiveness, particularly for studies conducted in Francophone and Lusophone African countries. To mitigate this, we will: (1) screen English abstracts of non-English articles where available, (2) search regional repositories such as African Journals Online (AJOL) and WHO Africa databases for English translations or executive summaries and (3) document all studies excluded solely for language in an online supplemental table. The potential direction and magnitude of this bias will be discussed when interpreting findings, especially regarding regional representativeness.

Review process and operational steps

We will follow a 10-step workflow adapted from the JBI Manual for Evidence Synthesis to ensure rigour and transparency.⁴¹

1. Develop and pilot the search strategy in PubMed, then adapt the syntax for other databases.

2. Import and deduplicate records in reference-management software.

3. Screen titles and abstracts independently by two reviewers using a calibrated form.

4. Retrieve and assess full texts against the eligibility criteria, documenting reasons for exclusion.

5. Pilot and refine the data-extraction template on a sample of studies.

6. Extract all eligible data independently in duplicate.

7. Appraise methodological quality using the Mixed Methods Appraisal Tool (MMAT) V.2018.

8. Synthesise findings thematically in ATLAS.ti, grouping studies by intervention and WHO quality domain.

9. Consider limited meta-analysis if we identify homogeneous quantitative data.

10. Compile evidence tables and a PRISMA 2020 flow diagram for transparent reporting.

Two reviewers will perform every independent task. We will resolve disagreements through discussion and, if needed, arbitration by a third author (HA or CA). We will use several complementary tools to enhance methodological consistency. PRISMA-P (2015) and PRISMA 2020 guide transparent reporting. The JBI Manual for Evidence Synthesis provides operational direction for screening, extraction and synthesis. We will apply the MMAT V.2018 to appraise study quality across different designs, use SWiM guidance to structure our narrative synthesis, and analyse qualitative and mixed-method data in ATLAS.ti (V.X). Together, these tools ensure reproducibility, transparency and rigour.

Study dates

The review commenced on 22 August 2025 with protocol registration and database searches. We anticipate completing study screening, data extraction and analysis by 27 November 2025, with final manuscript submission expected by 31 January 2026.

Eligibility criteria

We will define eligibility using the Population, Exposure and Outcome (PEO) framework.⁴² Inclusion and exclusion criteria based on PEO are explained below.

Population

We will include any study among health workers delivering PHC in SSA. Eligible health workers will include medical doctors, nurses, midwives, community health officers, physician assistants, public health officers and other cadres of health workers actively engaged in the provision of essential services at the PHC level. Studies that do not clearly indicate health workers in PHC settings or that focus exclusively on secondary or tertiary-level healthcare providers will be excluded.

Exposure

We define capacity-building interventions as structured activities that strengthen the knowledge, skills, attitudes or organisational support of PHC workers with the explicit aim of improving service quality. Eligible interventions include in-service or competency-based training, mentorship and coaching, supportive supervision, digital or mHealth learning, peer learning, and blended approaches that combine these elements. To qualify, an intervention must target PHC providers and evaluate at least one WHO quality-of-care domain (effectiveness, safety, people-centredness, timeliness, equity, integration or efficiency). We will, however, exclude any study that does not address any of the tenets of the WHO Healthcare Quality framework: safety, accessibility, acceptability and patient-centredness, equity, efficiency, integration, continuity and timeliness. Interventions must explicitly target improvements in quality care in PHC settings. Studies where capacity-building is not a central intervention will be excluded.

Outcome

We will include any study that measures the quality of PHC delivery through a capacity development intervention. We will include any study that includes the six domains of effectiveness, safety, people-centredness, timeliness, equity and efficiency as propounded by the WHO Healthcare Quality framework. All empirical study designs addressing these elements will be eligible (including experimental randomised control trials (RCTs), quasi-experimental and observational studies). Secondary outcomes include provider knowledge, competence and process indicators such as adherence to clinical standards. For each study, we will capture the exact definitions and instruments used, map outcomes to the corresponding WHO domains, and standardise metrics where possible. We will discuss the remaining variability in outcome measurement when interpreting findings.

Because operational definitions of quality domains often vary across studies (eg, measures of patient-centredness, timeliness or safety), we will create an outcomes codebook that maps each extracted indicator to its corresponding WHO Quality of Care domain. During data extraction, we will capture the exact definition, measurement instrument and unit for every outcome. This mapping helps us group conceptually similar outcomes, standardise metrics where possible (eg, by converting counts to proportions) and report heterogeneity in outcome measurement transparently. We will discuss these variations explicitly in the synthesis to clarify how they influence interpretation and comparability across studies.

Information sources

We will primarily source relevant studies from electronic databases such as PubMed, Embase, Web of Science, Scopus, Google Scholar and the AJOL. To capture grey literature, we will also search organisational websites (eg, WHO, African Union, Ministries of Health) and repositories of theses and reports. We will document search dates and database coverage. We will include studies published in English between 2000 and 2025. Our justification is to enable us to focus on contemporary interventions. We will also check reference lists of included studies and consult experts in the field to identify additional possible sources.

Search strategy

Our search strategy will combine keywords for each of the PEO elements. We will use terms such as “health worker”, “health personnel”, “health professionals”, “community health worker”, “nurse”, “public health officer”, “medical doctor”, “capacity building”, “training”, “mentorship”, “supportive supervision”, “peer-learning”, “primary care”, “health”, “quality of primary healthcare delivery”, “healthcare service quality”, “patient satisfaction”, “effectiveness of primary healthcare”, “continuity of patient care”, “safety of care”, “equitable healthcare delivery”, “timeliness of care”, “efficiency of healthcare services”, “people-centred care” and the names of all the SSA countries. We will use the Boolean operators “AND” and “OR” to combine our searches. A detailed draft search strategy for PubMed, including full Boolean strings, Medical Subject Headings, filters and limits, is provided in online supplemental file 1. The strategy will be adapted for other databases using database-specific controlled vocabulary and syntax.

Selection process

We will screen records in two phases. First, two authors (RMG and ALAED) will independently screen titles and abstracts for potential relevance. Second, the two authors will retrieve the full texts of all potentially eligible records and independently assess them against our inclusion (eligibility) criteria. Any disagreements at either stage between the two authors will be resolved through discussions between the two of them. However, if they still do not agree, the HA or CA of the team will be consulted to make a final determination. We will use a standardised screening form (such as Covidence) and keep a log of reasons for exclusion at full text. The screening process will be summarised in a PRISMA 2020 flow diagram whose sample has been added as online supplemental appendix 1.

Data extraction process

We will extract the relevant studies using a standardised form based on the JBI Manual for Evidence Synthesis.⁴² RMG and DAA will pilot the form on a sample of studies to ensure consistency, then extract the relevant studies independently from all included studies using an Excel template of the form. We will collect study background information related to the studies (author(s), year, journal/publisher, location (country of publication)), study design and sample characteristics, details of the intervention (type, duration and content), comparison conditions, contextual factors (country, health system level), outcome definitions/measures, results and sources of funding. Any discrepancies between the two authors’ extractions will be resolved by discussion between them. However, if this fails, then a third author (HA or CA) will come in for arbitration. If necessary, we will contact study authors for missing or unclear data. The use of dual independent extraction ensures accuracy and reproducibility.

Quality assessment

We will assess the methodological quality (risk of bias) of the included studies using the MMAT, V.2018.⁴² This tool is designed to assess the quality of diverse study designs, including qualitative research, RCTs, non-randomised studies, quantitative descriptive studies and mixed-methods research, which makes it the most suitable tool for our review (see online supplemental appendix 2 for a template of the tool). Each included study will be evaluated using the five criteria relevant to its design category. Two authors (RMG and ALAED) will independently conduct the assessments. Any disagreements will be resolved through discussion between them. If this does not work, a third coauthor (HA or CA) will be consulted to make a final determination. We will summarise MMAT judgements item-by-item in tables and use these results to describe evidence quality, perform sensitivity analyses and weigh interpretations toward higher-quality studies. We will not compute an overall numeric MMAT score; instead, we will present domain-level ratings to make limitations transparent. Online supplemental appendix 3 includes the MMAT 2018 checklist.

Data synthesis and analysis

We will use a narrative synthesis approach, structured around key comparisons and contexts. We will group studies by intervention type, outcome domain and context (eg, country or facility type) and summarise findings in text and tables. Tabulation of study characteristics and results (eg, in evidence tables) will be provided to ensure transparency. We will use a narrative approach to synthesise data collected from all our included studies for analysis. We will explore themes across studies using qualitative thematic analysis of extracted findings with ATLAS.ti software to identify recurring concepts. The synthesis will follow SWiM guidance for non-meta-analytic reviews, ensuring that we describe grouping criteria, compare results across groups and discuss limitations.³⁹

Although we primarily plan a narrative synthesis, we will evaluate whether limited quantitative analyses are feasible when we identify a sufficient number of homogeneous studies. We will conduct a meta-analysis when studies use comparable intervention types, similar population characteristics and consistent outcome definitions with available effect estimates (eg, risk ratios, ORs or mean differences). When meta-analysis is appropriate, we will apply a random-effects model, calculate statistical heterogeneity using the I² statistic, and perform subgroup or sensitivity analyses. If the data remain too heterogeneous, we will present the findings within the SWiM-structured narrative synthesis framework.

Patient and public involvement

Patients and members of the public were not directly involved in the design, conduct, reporting or dissemination plans of this systematic review. However, the topic is of broad relevance to health workers, policymakers and communities in SSA, and dissemination will target these stakeholders. We will prepare policy briefs for Ministries of Health, share summaries with frontline health managers and WHO country offices through webinars, and seek feedback from practitioner networks to strengthen knowledge translation and policy uptake.

Ethics and dissemination

This systematic review does not involve the collection of primary data and therefore does not require formal ethical approval. All data used will be obtained from publicly available peer-reviewed literature and grey sources. The findings of this review will be disseminated through publication in a peer-reviewed journal, presentations at academic and policy conferences, and knowledge-sharing platforms with stakeholders in PHC and health systems strengthening in SSA.

In addition to publishing the review in a peer-reviewed journal and presenting it at academic conferences, we will actively share the findings through tailored knowledge-translation activities that reach decision-makers and frontline health professionals. We will produce policy briefs for Ministries of Health across SSA, host webinars for health system managers and WHO country offices, and organise stakeholder workshops in collaboration with regional health networks. We will also develop a plain-language summary and a technical brief, which we will deposit in institutional repositories to promote uptake among practitioners and implementers.

As this study did not undergo review by an ethics committee, oversight for research integrity at the authors’ institution is provided through the Department of Population and Behavioural Sciences, Fred N Binka School of Public Health, University of Health and Allied Sciences, Ghana. The Head of Department serves as the appropriate contact for any research integrity-related enquiries.

Research integrity contact

Department of Population and Behavioural Sciences, Fred N Binka School of Public Health, University of Health and Allied Sciences, Hohoe, Ghana.

Discussion

PHC quality is globally recognised as central to the achievement of the health-related SDGs, especially goal 3.8 on UHC.⁴⁰ WHO estimates that up to 75% of projected health gains from the SDGs depend on strong PHC systems.¹ The quality of care provided to clients is critical for equity.¹ Services are, therefore, expected to be effective, safe, people-centred, timely, equitable, integrated and efficient, and tailored to people’s needs.³⁶ Our review will synthesise the available evidence on PHC service quality interventions in SSA. The PEO framework guides our scope toward real-world PHC contexts and outcomes relevant to equity and system performance. By addressing these aims, we seek to emphasise the relevance of PHC quality to global health equity and SDG targets.⁴⁰

Given the expected heterogeneity in study designs, contexts and outcomes, we will use a narrative (thematic) synthesis approach, facilitated by qualitative analysis software (eg, ATLAS.ti). Gibbs et al⁴³ provide guidance on this approach for reviews lacking data amenable to meta-analysis. Specifically, we will employ the SWiM reporting framework to ensure transparency.⁴³

Using ATLAS.ti, we will code extracted findings into thematic categories and identify patterns across studies. This method allows us to integrate diverse evidence types (qualitative, quantitative and mixed) systematically, guided by PEO-derived themes (eg, indicators of PHC quality, patient and provider perspectives).

We anticipate that our study will make substantial contributions to policy and practice, particularly for SSA. Synthesising the best available evidence on PHC quality interventions will inform health policy and system strengthening in SSA. For example, findings could guide national PHC quality improvement initiatives and UHC strategies by identifying what works in similar contexts. We envisage that this review will provide actionable insights for stakeholders such as Ministries of Health, WHO country offices and NGOs. It could also reinforce commitments made in the 2018 Astana Declaration and inform implementation of the WHO’s PHC frameworks by highlighting effective, equity-promoting practices.⁴⁰

By mapping gaps in the evidence, this review will also highlight research priorities, for instance, on user experience or outcomes that are currently neglected.³³ These contributions align with global calls to place quality at the core of health systems, since poor-quality care accounts for a majority of preventable deaths in LMICs.⁴⁰ We acknowledge that SSA is heterogeneous in terms of health system structures, governance, financing mechanisms and workforce organisation. As such, the findings of this review will not assume uniform applicability across all countries. Instead, we will interpret results within contextual groupings and highlight how implementation considerations may vary by setting. The purpose of this review is to synthesise transferable principles and patterns rather than prescribe a single model for all countries. Country-specific adaptation will remain essential.

Strengths and limitations

Our protocol has several strengths. We followed PRISMA-P guidance⁴³ and used a systematic search strategy across databases to capture all eligible studies on PHC quality. The PEO framework ensures that all relevant study designs (quantitative, qualitative and mixed) are included. We will use the MMAT to assess study quality comprehensively.⁴² Having two independent reviewers at each stage will enhance rigour.

We, however, acknowledge possible limitations inherent in our review. For instance, our restriction of the search to English-language publications may exclude relevant studies from Francophone or Lusophone Africa. Heterogeneity in definitions of ‘quality’ and in intervention details will limit direct comparisons. The MMAT, while useful for mixed evidence, provides only a broad quality score and may not capture nuanced biases. Finally, the use of narrative synthesis means findings may be subject to interpretive bias, though we will mitigate this by transparent coding and linking themes to data. Despite these limitations, our approach is appropriate given the complexity and context-dependence of PHC quality, and it will yield a comprehensive narrative overview rather than an underpowered meta-estimate. Given the diversity of health systems across SSA, implementation implications drawn from the review may require contextual adaptation at the country level.