Community health system capacities and capabilities within an evolving community health policy framework: mixed methods study of stakeholders in central Uganda
===============================================================================================================================================================

* Richard Kabanda
* Ronald Miria Ocaatre
* Diana Atwine
* Bounggui Kim
* Simon Erisa Waiswa
* Prichard Denzel Kavuma
* Yeni Lee
* Loretah Mutoni
* Solyi Kim
* Yein Park
* Monica Okuga
* Raymond Tweheyo

## Abstract

**Objectives** Empowering communities through identifying and unlocking community capacities and capabilities is vital for improving community health systems. This study assessed the community health system’s status quo and readiness for implementing a government-led, partner-supported community health worker project.

**Design** A mixed methods cross-sectional study.

**Setting** Two districts and one city in central Uganda.

**Participants** 21 key informants (KIIs) with district leaders, 4 focus group discussions (FGDs) with community health workers (CHWs) termed as village health teams in the Ugandan setting and a survey of 487 CHWs and 419 pregnant women who had childbirth 12 months before.

**Study measures** KIIs and FGDs explored community health system resources using the WHO health systems building blocks and the UNICEF health system strengthening maturation model. However, the surveys explored the work-related attributes and services delivered by the CHWs and to the community, respectively.

**Analysis** A framework analysis was used for qualitative data in NVivo 14. While descriptive and stratified analyses were conducted for quantitative data in Stata I/C 15.0: proportions for the varied geographical entities were compared using the t-test with p values <0.05 considered significant, one-way ANOVA was used to compare means.

**Results** Overall, all sites had relatively strong governance of community health only challenged by multiple implementing partners that were weakly coordinated. There was an exclusive paper-based information management system that linked to the national DHIS-2 software. Community reporting rates varied between 20% and 80%. Community health financing was weak, similarly to commodities availability. The mean age of the community health workforce was high at 50.9 (SD 11.9); the majority reported adequate skills in service delivery except for community first aid, Kangaroo mother care and noncommunicable diseases. Households’ most received CHW services included home visits, treatment for sick under-five children and child immunisation.

**Conclusions** The existing CHW system has governance and reporting strengths but could be enhanced through revitalization, enhanced coordination of stakeholders, providing appropriate training, work tools, supervision and engaging the community for accountability of results.

*   Community child health
*   Health policy
*   PUBLIC HEALTH

### STRENGTHS AND LIMITATIONS OF THIS STUDY

*   This cross-sectional study derives strengths from the use of mixed methods, multiple data collection tools and a triangulation of findings from a range of stakeholders (providers and beneficiaries) in a decentralised health system.

*   Stakeholder perspectives are explored using two theoretical frameworks that are widely used in low- and middle-income countries, thus offering an opportunity for transferable learnings.

*   The cross-sectional design limits any temporal inferences between potential predictors and the given findings.

## Introduction

Recent pandemics such as COVID-19 (2019–2022) and Ebola virus disease (EVD) in West Africa (2014–2016) demonstrated how fragile health systems are, particularly in low- and middle-income countries (LMICs), and the vital role of community health systems.1 2 Specifically, governments urgently borrowed from disease surveillance and response methodology to institute strategies, such as epidemiological investigations, risk profiling, risk targeting and risk communication,3 but critically found a need to depend on community engagement to enhance the adoption of healthier behaviours and cascade preventative interventions, such as handwashing and reducing vaccine hesitancy.4 Typically, community health worker programmes gained prominence in high-income countries and LMICs alike during these pandemics.5 6

While there is a growing consensus for optimising community health worker programmes in LMICs as the pathway to achieving Universal Health Coverage (UHC),7–9 by majorly increasing the delivery of Primary Healthcare (PHC) services at the community level.10 Health systems are weak in these settings, exhibiting poor planning, coordination and consequently failing to exploit partnerships that have potential for improving community health systems,11 mostly weaknesses in leadership and governance functions. A major limitation to exploiting the potential partnerships for community health programmes is the perception by actors that community health systems represent informal, complex adaptive systems12 where linkages to the formal health system are obscure. Additionally, several LMIC governments are concerned about the economic burden of fully financing large-scale CHW programmes,13 14 as the latter often hold larger staff numbers compared with the formal health workforce. Yet the health workforce, which is already in severe shortage, is estimated to consume over 60% of the health sector budgets in LMICs.15 Accordingly, the existing guidance supports the consideration of both financial and nonfinancial incentives for motivating the community health workforce.8 16 17

The Global Fund and the UNAIDS community health system frameworks include the components of leadership and governance; the community health workforce capacity and resources; community networks including linkages, partnerships and coordination; community service delivery activities; and enabling environments and advocacy encompassing community engagement.18 Community engagement espouses the nexus of provider-beneficiary interactions. The community engagement literature suggests that effectiveness revolves around multisectoral collaborations, community involvement, ownership, mutual accountability for results and solid government stewardship.19–21 Multisectoral collaborations involve, for example, the education, water sanitation and health sectors jointly framing problems, codeveloping and implementing solutions. The recent conditioning of national plans towards program-based budgeting in the health and related sectors in Kenya and Uganda is a step towards nationally led multisectoral collaboration.22 Further, for accountable delivery of health services, feedback loops to and from the community are vital, with examples of success from innovations such as the use of community scorecards, where communities rate the quantity and quality of health services received based on their values and expectations.23 24

In sum, attributes of a strong community health system revolve around strong leadership and governance, which drive the other system building blocks. Uganda, in February 2023, saw the birth of the first-ever National Community Health Strategy (NCHS) whose priorities include increasing access to community health services, equipping and compensating the community health workforce, improving commodity availability, sustainable financing, quality data, multisectoral collaboration and community empowerment.25 To date, the strategy development and approval have provided a wake-up call for evaluating the feasibility of various incentives for the community health workforce,26 27 including piloting a new supervisory cadre of community health workers. Also, the community health information system has received substantive attention with urgency for shifting towards digital reporting and the adaptation of the Village Health Team tool kit linked to the national data archive DHIS-2,28 all supported by the development of the first national Health Information and Digital Health (HIDH) strategy.29 These advances point to policy shifts where the community health workforce is focal, including considerations for keeping up with technological advancements.

While these advances are vital and constitute major health sector reforms, key challenges remain, such as the need to assess existing community health system capacities and capabilities, learning from over two decades of working with a voluntary community health workforce in Uganda,30 31 for which this study aimed to contribute.

## Methods

### Context

This study was conducted as a baseline in central Uganda before exploring community health problems and for strengthening the community health system of three neighbouring local governments: Masaka City, Masaka District, and Bukomansimbi Districts. The three entities have a joint mid-year population in 2023 of 523 200 people served by 1,436 CHWs.

### Study design

We used a mixed methods approach to gain a broad set of providers’ and beneficiaries’ understanding of the existing community health system capacities and capabilities in two districts and one city of central Uganda. Mixed methods are known to unpack complex processes and systems in healthcare32 and were used in this study through the triangulation of design, data collection and analysis of results to enhance validity. As community health systems are not new but have seldom been studied holistically in Uganda, we aimed to provide a rich account using three data collection methods: (1) key informant interviews, (2) focus groups and (3) community health worker and household surveys.

### Data collection

Sources of data and data collection: Informed by a preceding desk review of policy-relevant documents for the local context, purposive qualitative key informant interviews were conducted with district/city health officials (online supplemental file 1) and focus groups with CHW members (online supplemental file 2). Qualitative research assistants used key informant interview and focus group guides that had a detailed section exploring context and using standard ice-breaking questions such as ‘(let every person) please describe yourself and what you do’. Such questions were followed by a reading of the participant information sheet, followed by a question-and-answer reflection session and then consenting before the interviews. Concurrent with the qualitative work, surveys were conducted for CHWs (online supplemental file 3) and for women who had childbirth in the preceding 12 months (online supplemental file 4). Table 1 summarises all participants by data collection method. A total of 12 research assistants with a minimum of a bachelor’s degree were trained in research ethics and survey methods to conduct the study between January and February 2023. Qualitative data were digitally audio-recorded, while quantitative data collection was aided by REDCap-enabled digital tools on tablets that were linked to a secure local server at Makerere University School of Public Health. Research ethics principles of consent, voluntary participation, autonomy, confidentiality and data privacy were followed. The study had received ethics approval from the Makerere University Research and Ethics Committee and the Uganda National Council for Science and Technology (MakREC Ref: SPH-2022–314; UNCST Ref: HS2545ES), respectively. It was not appropriate or possible to involve patients or the public in the design, conduct, reporting or dissemination plans of our research.

### Supplementary data

[[bmjopen-2023-082085supp001.pdf]](pending:yes)

### Supplementary data

[[bmjopen-2023-082085supp002.pdf]](pending:yes)

### Supplementary data

[[bmjopen-2023-082085supp003.pdf]](pending:yes)

### Supplementary data

[[bmjopen-2023-082085supp004.pdf]](pending:yes)

View this table:
[Table 1](http://bmjopen.bmj.com/content/15/1/e082085/T1)

Table 1 
Study participants and sample categorised by method

Sampling and sample size: Purposive sampling was done for the qualitative interviews guided by the study objectives, with samples determined by the saturation of data among participant categories. The CHW survey used sampling proportionate to size to derive from a listing of all eligible CHWs in the district/city. Simple random sampling of subcounties was conducted in Stata I/C 15.0 using random digit allocation. Thereafter, all accessible CHWs in the selected sub-counties were included to participate in the survey. For the household survey, we computed the sample size using access to community healthcare, specifically the proportion of under-fives treated within 24 hours by CHWs (43.3% nationally), the test statistic to achieve a two-tailed test of significance with 95% CIs and a 5% margin of error was used to arrive at a sample size. Participants were obtained by screening households for eligibility within the geographical locations of eligible CHWs.

### Data analysis

For the qualitative data, after verbatim transcription, coding followed in NVivo 14 using a framework analysis where a priori codes were adapted. The global fund community health systems strengthening framework formed the broader supraordinate themes, while within the study objectives, the WHO health system building blocks formed the subthemes. Both constant comparison and deviant cases were isolated within each of the themes. For both surveys, data from REDCap were directly imported to STATA I/C 15.0, where cleaning was done. Descriptive exploratory analyses were done to assess data distribution, normality, statistics of central tendency and dispersion, which were later followed by a stratified analysis between the geographic entities to provide locally relevant findings. For comparisons between the three groups, χ2 tests were run to assess for differences between proportions of the categorical variables for the varied geographical entities (eg, sick under-five children treated, women visited by CHWs during pregnancy). Since the data were normally distributed, χ2 test values are reported with p values <0.05 considered significant. To compare for differences of means between the three groups (such as for age, duration of work as a CHW and household visits), one-way ANOVA was used, providing an F-statistic and corresponding p values, with significance reported at p<0.05. Additionally, post hoc pairwise comparisons are conducted to assess whether differences in the ANOVA are significant, using the Bartlett’s test of equal variance through the Bonferroni method, which provided a χ2 value and corresponding p values.

### Patient and public involvement

None.

## Results and findings

### The community health workforce capacity and resources

Table 2 shows background characteristics of the CHWs, while table 3 shows similar characteristics of the household members (women who had childbirth in the past 12 months). The mean age of the CHWs was 50.7 (SD 11.9) years, higher compared with that of the household members, mean 27.8 (SD 6.9). Similarly, although purposively selected, the focus group participants had CHWs aged between 46 and 64 years. Some district health leaders voiced concerns about the advanced age of CHWs as quoted by a district manager: *(CHW) enrolment was done around 2011, the biggest number of the CHWs in the district were getting old and they were 70 and 80 years old. They were no longer seeing, not able to walk and so on (KII_11 District health team member, District 1*).

View this table:
[Table 2](http://bmjopen.bmj.com/content/15/1/e082085/T2)

Table 2 
Community health worker background characteristics

View this table:
[Table 3](http://bmjopen.bmj.com/content/15/1/e082085/T3)

Table 3 
Community health worker work resources and care processes

At least two-thirds of CHWs were female, and about 40% had attained some primary or completed primary level education, a similar proportion having had some secondary education. On average, CHWs had worked for at least 11 years (significantly higher for the districts compared with the city, p=0.032), and about half of them were periodically supported by an NGO-like implementing partner (higher in the rural location).

Related to training, nearly all CHWs reported prior training by the government before commissioning them 10 years ago. Two-thirds reported prior training on how to fill their register, but only one-third had received first-aid training, and less than 1% (09/487) had received any training relating to the care for patients with chronic conditions, including noncommunicable diseases and HIV.

Considering what resources CHWs had in place to conduct their mandate, table 3 shows that nearly all had household registers and reporting tools; two-thirds had nutrition assessment MUAC bands; and only half reported essential items such as uniforms and usually T-shirts. Further, about one-third had either gumboots or an essential commodity (such as oral rehydration salts, paracetamol and family planning items) or the integrated Community Case Management of Childhood illnesses (iCCM) medicines (oral rehydration therapy (ORT), amoxicillin and coartem) among others. Barely half of the CHWs (45%) received any form of compensation, and only one in six received financial compensation, highest in the rural locations compared with the city.

#### Leadership and governance

Table 3 shows that two-thirds of all CHWs uniformly reported supervision in the past quarter as required for their functionality, signalling support from the government or the implementing partner. Accordingly, CHWs reported work processes depicting positive work engagement, with about four in five reporting the performance in the last quarter of home visits, submitting reports, following up with pregnant women and newborns or reporting disease outbreaks. Only half the CHWs reported a role in following up HIV+ persons, and only one-quarter for providing first-aid services.

From the qualitative data, the district managers reported existing structures that support the community health systems to include staff within the health facilities, health assistants who are health facility staff that oversee CHW activities in the communities, and among the CHW member groups, there are informal parish (the administrative level that aggregates villages) level CHW coordinators who collectively form supervisory structures for CHWs. Specifically, district managers and stakeholders reported that almost all villages in the study area have at least two CHW members. Each village consists of about 100–200 households.

> Each village has two CHWs. Then in each sub-county, we have either a Health Centre II or a Centre III. And then at the district level, we have one Health Centre IV. We don’t have a hospital. Though they are, anyway, they are all part of the community health system because I know it is from the village to those facilities (KII_01 District health team, District 3).

Similarly, one of the implementing partners indicated that *I commend the districts because they have fully built the structures, they know the village health teams and they know their details (KII_09, Implementing partner 2*).

However, a majority of the CHWs had initial training spanning from over a decade ago. Despite this, the reporting structures for CHWs were well understood, as quoted for a district manager below, signalling strong accountability, considering that CHWs are a voluntary structure.

> At a parish level, we have a parish coordinator who coordinates all the CHWs in that parish. Her/his job is to relay information from the health facility to the CHWs by informing them what is happening, what we need and what they need to do and s/he also collects the reports from these people. So now, this parish coordinator reports to the health facility focal person, who could be any health worker; it could be a nurse or a health assistant attached to a health facility that has a passion for the CHWs. Hence, her/his job is to make sure that she/he links the whole CHWs system with our health facility (KII_03 District health team member, District 1).

A perhaps larger challenge for the governance of the community health system was inadequate government financing, where disease prevention and promotion activities are designed for delivery through PHC grants to the health facilities, but resources are meagre as indicated by one of the DHT members below:

> As a district we don’t have any facilitation to our CHWs since we can’t pay them, we don’t have funds for them, we depend on our Implementing partners to motivate them (KII_13 District health team member, District 3).

Similarly, one of the city health team members narrated:

> One of the biggest challenges that we face is the low resources that we have. Even at times, we may end up not getting any or having any budget for CHW activities. Hence, we are always just there waiting for any implementing partner to come in and implement (KII_10 City health team member).

#### Community health services delivery

Antenatal care attendance was near universal; however, only one-third of pregnant women had a home visit by a CHW. Also, for the entire preceding year, about one-sixth (16.2%) of all households had no home visit by a CHW.

As observed in table 4, about 18% of women who had childbirth in the past 12 months reported ever being diagnosed with hypertension, nearly all during their last pregnancy. Similarly, about 15% of the women respondents reported ever being diagnosed with diabetes; nearly all reported it within their last pregnancy.

View this table:
[Table 4](http://bmjopen.bmj.com/content/15/1/e082085/T4)

Table 4 
Household member (women who had childbirth ≤12 months) key characteristics

Concerning child health, nearly all households had more than one under-five child, and about half of these children were sick in the preceding 2 weeks to the survey (higher in Bukomansimbi, the most rural district). Of the sick children, four-fifths suffered from either fever or cough with or without fast breathing, while one-third had diarrhoea. Altogether, Masaka City, the most urban locality, significantly had the least child morbidity, depicting rural-urban disparities in health status. However, in relation to child immunisation, Bukomansimbi, the most rural, had significantly higher immunisation completion rates (80% vs 62%), depicting difficulties in immunisation programme penetration within urban localities.

From a CHW’s perspective, it appears that a huge burden of demands from multiple implementing partners, in addition to government, stifles their effectiveness as narrated below:

> Once, my supervisor clearly gave me a role to make sure that I register pregnant mothers and register new births. I am supposed to treat the children that come in and along the way maybe at 10:00 AM someone comes from [NGO named] and wants me to arrange or mobilize for him people going for family planning and then shortly after someone comes from [NGO2 named] and wants to know the people defaulting on ART [anti-retroviral treatment]. They are all coming to me to look [execute tasks] for them. Then maybe shortly someone [Government official] calls and says that they received information that there is a problem of cholera and that I should investigate for them (FGD 04, Participant 3).

While the delivery of community health services could be enhanced, the limitations revolve around a weak community health system and workforce governance, including irregular training, a mismatch between job demands and job resources, threatening job strain for CHWs, which are all critical prerequisites for a functional system, as discussed subsequently.

## Discussion

Summary of findings: For this study which assessed existing community health system capacities and capabilities in three local governments of central Uganda, we found relatively high levels of community health leadership and governance for all entities with existing structures, clear roles and reporting lines but only majorly undermined by limited community health financing.11 Existing community networks, including NGO partnerships, were being harnessed and, to an extent, supporting majorly the most rural district for improving reporting rates, but also service delivery, such as immunisation completion rates.13 The community health workforce capacity and resources were limited7 9: while an adequate community health workforce existed as per recommended ‘staffing’ norms, the majority were of advanced age. Also, two-thirds were women, suggesting gendered roles in the workforce. For the resources, there was limited equipping and tooling of the CHWs, inadequate stocking with iCCM supplies and commodities and minimal evidence of compensation for their work,25 yet with a burgeoning list of work demands from both government and the implementing partners. Regarding community service delivery, CHWs had the capacity to complete routine screening of pregnant women and newborns, treat sick children, support child immunisation and investigate disease outbreaks;33 however, they weakly supported first-aid response34 35 and chronic disease care or follow-up,36 37 despite the high prevalence of diabetes and hypertension.

To harness the strong leadership and governance of community health systems and to confront scaling challenges,14 a district-led coordination of partners is required to harness resources for strengthening community health workforce capacity, information management, commodities and coordinated services delivery, where community scorecards can be used to track progress for results.23 24 Existing evidence suggests that community health partnerships with the numerous actors at this level promise stronger community health systems,11 18 which enhance ownership and accountability for results. While partnerships are key for all community health actors, a convergence point when demanding results is the community health worker whose job demands need to be supported with adequate training, supervision, equipment, tooling and compensation for supporting PHC activities which drive the UHC agenda.17 33

In light of the advanced age of the community health workforce, a bold recruitment plan is required alongside a retention plan that adequately attracts the most suitable people willing to serve their communities. There has been a relative neglect in the literature to confront the issue of the advancing age of CHWs, when proposing recruitment, succession planning and training for professionalising CHWs in LMIC settings.38 39 Evidence has shown that CHWs’ community connections and embeddedness build trust, which supports the onward provision of quality health services, accessible to those in most need.10 20 Implicit in the connections and embeddedness with community members is not only the professionalism but also the trust accorded to CHWs of a given age, due to their experience and commitment to the job.39 Therefore, findings in a setting where recipients of CHWs’ care are concerned about the effective performance of CHWs aged 50 years and above suggest a need for succession planning and the provision of fair compensation that motivates both the new entrants and the existing workforce.

Community health service delivery related to maternal and child health can be improved by strengthening iCCM components such as commodity availability and including community first-aid response.34 36 40 Gauging from the high burden of NCDs, particularly diabetes and hypertension in association or in co-existence with other chronic conditions such as TB and HIV, there is a need to equip CHWs with simple but highly engaging health talking points that can support community health education for early screening, provision of supportive care, follow-up of specific persons and linkage to appropriate care.36 37 In other programmes in several countries, the landscape of community health services delivery is gradually shifting from a predominant disease focus, where PHC services are delivered, to include advice on behavioural modification of risk factors for chronic disease. A new era of programming (eg, training of CHWs in chronic care support), tooling (eg, developing simple case definitions for screening) and monitoring of care quality is required. This approach aims at integration into the existing design of UHC, essentially rethinking efficiencies from CHW programme designs.36 37 41 In this study, self-reports of noncommunicable disease commencing during pregnancy appear slightly lower for hypertension but much higher for diabetes than the average for Ugandan females (18% vs 22.9% and 15% vs 1.1%, respectively) compared with the WHO STEPS survey on NCDs for Uganda.42 Findings suggest an important role of the community health system in the surveillance and screening of probable cases of noncommunicable diseases, as well as the possible improvement in surveillance to increase the validity of self-reports.43

Programmatically, a revised recruitment plan for the CHWs workforce, coupled with a revised training plan that incorporates aspects of chronic disease care, first-aid response and potentially other skills such as technological adoption, requires a major reform. This requires a rethink about the adequacy of the community health workforce or the workload ratios,14 the appropriate age and qualifications at selection,31 39 the full range of training modules, duration and models of training38 and the retention strategy.44 45 The existing national community health strategy for Uganda is silent on issues of succession planning, evolving skill needs and the requisite toolkit.25 There is a need for an increased policy-practitioner debate on these evolving CHW policy needs where CHW programmers, who are mostly the non-governmental actors, should actively engage.

Methodologically, this study has inherent limitations. First, the cross-sectional design limits any temporal inferences46 between potential predictors, such as community resources and outcomes such as community health system’s resilience. Second, while mixed methods studies have considerable explanatory power for providing nuanced accounts and for theoretical immersion, this study used a concurrent design where both qualitative and quantitative methods were deployed simultaneously, and triangulation was completed at analysis. In this regard, there is limited exploratory or explanatory capacity of the methods in theoretical building or validation as there would have been if the mixed design used a sequential method with qualitative or quantitative stages preceding, respectively.47 48 Finally, this study had three populations: the households, community health workers and district health leaders; however, the households were not necessarily matched to the respective community health workers, and likewise, the latter were not mapped to their respective health facilities that resource them, limiting comparisons across contexts for the populations.49 In fact, the community health system in Uganda is still regarded as an informal establishment, requiring stronger linkages to the formal health system. Therefore, the learning from this study of the community capacities and capabilities needs to be interpreted within an evolving community health system that is weakly governed.

In conclusion, this mixed methods study set out to examine the community health system capacities and capabilities within an evolving local policy framework in three decentralised settings in Uganda. The study found that the existing public governance structures and leadership were able to sustain a poorly coordinated community health system comprising a voluntary workforce, weakly resourced and poorly skilled for noncommunicable diseases and first-aid but demonstrating effective performance for PHC services and with varied reporting rates. Several local NGO partners existed but were perceived to provide fragmented support that overburdened the community health workforce. Effective district leadership in the coordination of existing partners is vital for harnessing scarce resources to better the community health system. Future studies exploring improvements in community health worker programmes should examine how the various community health system inputs affect the motivation of the workforce. Also, consideration is needed for the evolving landscape of health services delivery, especially regarding CHWs’ capacities and capabilities to offer noncommunicable disease and first-aid services. Further, as noted from this study, future work should design and test simple chronic disease behaviour change messaging targeting averted risk behaviour and improving care processes such as adherence to, and continuity of care.

## Data availability statement

Data are available upon reasonable request.

## Ethics statements

### Patient consent for publication

Not applicable.

### Ethics approval

This study involves human participants and received ethics approval from: (1) the Makerere University Research and Ethics Committee (MakREC Ref. No.: SPH-2022-314) and 2) the Uganda National Council for Science and Technology (UNCST Ref. No.: HS2545ES), respectively. Participants gave informed consent to participate in the study before taking part.

## Footnotes

*   RK and RT are joint first authors.

*   X @RTweheyo

*   Contributors RK, RMO, DA and RT conceptualised the study. RT led the tool development, with the participation of RK, RMO, DA, BK, SEW, PDK, YL, LM, SK, YP and MO. All authors participated in data analysis, iterations of the report, reviewing all versions of and approving the final manuscript. RT acted as the guarantor.

*   Funding This work was supported by Korea Foundation for International Healthcare (KOFIH) under the project title of 'the capacity building for community health workers (CHW) in the central region of Uganda'.

*   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.

[http://creativecommons.org/licenses/by-nc/4.0/](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/](http://creativecommons.org/licenses/by-nc/4.0/).

## References

1.  Lotta G, Wenham C, Nunes J, et al. Community health workers reveal COVID-19 disaster in Brazil. The Lancet 2020;396:365–6. [doi:10.1016/S0140-6736(20)31521-X](http://dx.doi.org/10.1016/S0140-6736(20)31521-X)
    
    

2.  Koroma O, Chen Y, Wang P, et al. Community health workers’ job satisfaction in Ebola-stricken areas of Sierra Leone and its implication for COVID-19 containment: a cross-sectional mixed-methods study. BMJ Open 2021;11:e051645. [doi:10.1136/bmjopen-2021-051645](http://dx.doi.org/10.1136/bmjopen-2021-051645)
    
    

3.  Fall IS, Rajatonirina S, Yahaya AA, et al. Integrated Disease Surveillance and Response (IDSR) strategy: current status, challenges and perspectives for the future in Africa. BMJ Glob Health 2019;4:e001427. [doi:10.1136/bmjgh-2019-001427](http://dx.doi.org/10.1136/bmjgh-2019-001427)
    
    

4.  Engelbrecht M, Heunis C, Kigozi G. COVID-19 Vaccine Hesitancy in South Africa: Lessons for Future Pandemics. Int J Environ Res Public Health 2022;19:6694. [doi:10.3390/ijerph19116694](http://dx.doi.org/10.3390/ijerph19116694)
    
    

5.  Haines A, de Barros EF, Berlin A, et al. National UK programme of community health workers for COVID-19 response. The Lancet 2020;395:1173–5. [doi:10.1016/S0140-6736(20)30735-2](http://dx.doi.org/10.1016/S0140-6736(20)30735-2)
    
    

6.  Ballard M, Johnson A, Mwanza I, et al. Community Health Workers in Pandemics: Evidence and Investment Implications. Glob Health Sci Pract 2022;10:e2100648. [doi:10.9745/GHSP-D-21-00648](http://dx.doi.org/10.9745/GHSP-D-21-00648)
    
    

7.  Schneider H, Okello D, Lehmann U. The global pendulum swing towards community health workers in low- and middle-income countries: a scoping review of trends, geographical distribution and programmatic orientations, 2005 to 2014. Hum Resour Health 2016;14:65. [doi:10.1186/s12960-016-0163-2](http://dx.doi.org/10.1186/s12960-016-0163-2)
    
    

8.  WHO. WHO guideline on health policy and system support to optimise community health worker programmes. Geneva, Switzerland World Health Organisation; 2018.
    
    

9.  McCollum R, Taegtmeyer M, Otiso L, et al. Healthcare equity analysis: applying the Tanahashi model of health service coverage to community health systems following devolution in Kenya. Int J Equity Health 2019;18:65. [doi:10.1186/s12939-019-0967-5](http://dx.doi.org/10.1186/s12939-019-0967-5)
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=31064355&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 

10. Perry HB, Chowdhury M, Were M, et al. Community health workers at the dawn of a new era: 11. CHWs leading the way to 'Health for All'. Health Res Policy Syst 2021;19:111. [doi:10.1186/s12961-021-00755-5](http://dx.doi.org/10.1186/s12961-021-00755-5)
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.1186/s12961-021-00755-5&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=34641891&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 

11. Afzal MM, Pariyo GW, Lassi ZS, et al. Community health workers at the dawn of a new era: 2. Planning, coordination, and partnerships. Health Res Policy Syst 2021;19:103. [doi:10.1186/s12961-021-00753-7](http://dx.doi.org/10.1186/s12961-021-00753-7)
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=34641912&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 

12. Sarriot E, Kouletio M. Community Health Systems as Complex Adaptive Systems: Ontology and Praxis Lessons from an Urban Health Experience with Demonstrated Sustainability. Syst Pract Action Res 2015;28:255–72. [doi:10.1007/s11213-014-9329-9](http://dx.doi.org/10.1007/s11213-014-9329-9)
    
    

13. Kamya C, Abewe C, Waiswa P, et al. Uganda’s increasing dependence on development partner’s support for immunization - a five year resource tracking study (2012 - 2016). BMC Public Health 2021;21:160. [doi:10.1186/s12889-021-10178-0](http://dx.doi.org/10.1186/s12889-021-10178-0)
    
    

14. Hodgins S, Kok M, Musoke D, et al. Community health workers at the dawn of a new era: 1. Introduction: tensions confronting large-scale CHW programmes. Health Res Policy Syst 2021;19:109:109. [doi:10.1186/s12961-021-00752-8](http://dx.doi.org/10.1186/s12961-021-00752-8)
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=34641886&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 

15. WHO. World health report 2010: health systems financing: the path to universal coverage. Geneva, Switzerland World Health Organisation; 2010.
    
    

16. Glenn J, Moucheraud C, Payán DD, et al. What is the impact of removing performance-based financial incentives on community health worker motivation? A qualitative study from an infant and young child feeding program in Bangladesh. BMC Health Serv Res 2021;21:979. [doi:10.1186/s12913-021-06996-y](http://dx.doi.org/10.1186/s12913-021-06996-y)
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.1186/s12913-021-06996-y&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=34535147&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 

17. Colvin CJ, Hodgins S, Perry HB. Community health workers at the dawn of a new era: 8. Incentives and remuneration. Health Res Policy Syst 2021;19:106:106. [doi:10.1186/s12961-021-00750-w](http://dx.doi.org/10.1186/s12961-021-00750-w)
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.1186/s12961-021-00750-w&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=34641900&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 

18. Global Fund. Community systems strengthening framework. 2014. Available: [https://www.theglobalfund.org/media/6428/core\_css\_framework\_en.pdf](https://www.theglobalfund.org/media/6428/core_css_framework_en.pdf)
    
    

19. WHO. WHO community engagement framework for quality, people-centred and resilient health services. World Health Organization, 2017.
    
    

20. Odugleh-Kolev A, Parrish-Sprowl J. Universal health coverage and community engagement. Bull World Health Organ 2018;96:660–1. [doi:10.2471/BLT.17.202382](http://dx.doi.org/10.2471/BLT.17.202382)
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.2471/BLT.17.202382&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=30262948&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 

21. Brunton G, Thomas J, O’Mara-Eves A, et al. Narratives of community engagement: a systematic review-derived conceptual framework for public health interventions. BMC Public Health 2017;17. [doi:10.1186/s12889-017-4958-4](http://dx.doi.org/10.1186/s12889-017-4958-4)
    
    

22. Tsofa B, Musotsi P, Kagwanja N, et al. Examining health sector application and utility of program-based budgeting: County level experiences in Kenya. Int J Health Plann Manage 2021;36:1521–32. [doi:10.1002/hpm.3174](http://dx.doi.org/10.1002/hpm.3174)
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.1002/hpm.3174&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=33955046&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 

23. Gullo S, Kuhlmann AS, Galavotti C, et al. Creating spaces for dialogue: a cluster-randomized evaluation of CARE’s Community Score Card on health governance outcomes. BMC Health Serv Res 2018;18:858. [doi:10.1186/s12913-018-3651-3](http://dx.doi.org/10.1186/s12913-018-3651-3)
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.1186/s12913-018-3651-3&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=30428881&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 

24. Mupara LM, Mogaka JJO, Brieger WR, et al. Scorecard metrics for assessing the extent of integration of community health worker programmes into national health systems. Afr j prim health care fam med 2021;13. [doi:10.4102/phcfm.v13i1.2691](http://dx.doi.org/10.4102/phcfm.v13i1.2691)
    
    

25. MoH. National community health strategy 2021/22 - 2025/26. Kampala Republic of Uganda, Ministry of Health; 2022.
    
    

26. Agarwal S, Tweheyo R, Pandya S, et al. Impact of a recognition package as an incentive to strengthen the motivation, performance, and retention of village health teams in Uganda: a study protocol for a cluster randomized controlled trial. Trials 2023;24::428. [doi:10.1186/s13063-023-07426-6](http://dx.doi.org/10.1186/s13063-023-07426-6)
    
    

27. Tweheyo R, et al. Evaluating the pilot of the Community Health Extension Worker (CHEW) strategy in Uganda: an implementation science protocol. Trials 2023.
    
    

28. Kirk K, McClair TL, Dakouo SP, et al. Introduction of digital reporting platform to integrate community-level data into health information systems is feasible and acceptable among various community health stakeholders: A mixed-methods pilot study in Mopti, Mali. J Glob Health 2021;11:07003. [doi:10.7189/jogh.11.07003](http://dx.doi.org/10.7189/jogh.11.07003)
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=33791098&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 

29. MoH. The Uganda health information and digital health strategic plan 2020/21 - 2024/25. Kampala: Republic of Uganda, Ministry of Health, 2021.
    
    

30. Musoke D, Atusingwize E, Ndejjo R, et al. Enhancing Performance and Sustainability of Community Health Worker Programs in Uganda: Lessons and Experiences From Stakeholders. Glob Health Sci Pract 2021;9:855–68. [doi:10.9745/GHSP-D-21-00260](http://dx.doi.org/10.9745/GHSP-D-21-00260)
    
    [Abstract/FREE Full Text](http://bmjopen.bmj.com/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NDoiZ2hzcCI7czo1OiJyZXNpZCI7czo3OiI5LzQvODU1IjtzOjQ6ImF0b20iO3M6MjY6Ii9ibWpvcGVuLzE1LzEvZTA4MjA4NS5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 

31. Turinawe EB, Rwemisisi JT, Musinguzi LK, et al. Selection and performance of village health teams (VHTs) in Uganda: lessons from the natural helper model of health promotion. Hum Resour Health 2015;13:73. [doi:10.1186/s12960-015-0074-7](http://dx.doi.org/10.1186/s12960-015-0074-7)
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.1186/s12960-015-0074-7&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=26346431&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 

32. Fetters MD, Curry LA, Creswell JW. Achieving integration in mixed methods designs-principles and practices. Health Serv Res 2013;48:2134–56. [doi:10.1111/1475-6773.12117](http://dx.doi.org/10.1111/1475-6773.12117)
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.1111/1475-6773.12117&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=24279835&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 
    
    [Web of Science](http://bmjopen.bmj.com/lookup/external-ref?access_num=000327391900002&link_type=ISI) 

33. Glenton C, Javadi D, Perry HB. Community health workers at the dawn of a new era: 5. Roles and tasks. Health Res Policy Syst 2021;19:128. [doi:10.1186/s12961-021-00748-4](http://dx.doi.org/10.1186/s12961-021-00748-4)
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=34641903&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 

34. Diango K, Mafuta E, Wallis LA, et al. Implementation and evaluation of a pilot WHO community first aid responder training in Kinshasa, DR Congo: A mixed method study. Afr J Emerg Med 2023;13:258–64. [doi:10.1016/j.afjem.2023.09.001](http://dx.doi.org/10.1016/j.afjem.2023.09.001)
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=37790995&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 

35. Sun JH, Wallis LA. The emergency first aid responder system model: using community members to assist life-threatening emergencies in violent, developing areas of need. Emerg Med J 2012;29:673–8. [doi:10.1136/emermed-2011-200271](http://dx.doi.org/10.1136/emermed-2011-200271)
    
    [Abstract/FREE Full Text](http://bmjopen.bmj.com/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiZW1lcm1lZCI7czo1OiJyZXNpZCI7czo4OiIyOS84LzY3MyI7czo0OiJhdG9tIjtzOjI2OiIvYm1qb3Blbi8xNS8xL2UwODIwODUuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

36. Jeet G, Thakur JS, Prinja S, et al. Community health workers for non-communicable diseases prevention and control in developing countries: Evidence and implications. PLoS One 2017;12:e0180640. [doi:10.1371/journal.pone.0180640](http://dx.doi.org/10.1371/journal.pone.0180640)
    
    

37. Mishra SR, Neupane D, Preen D, et al. Mitigation of non-communicable diseases in developing countries with community health workers. Glob Health 2015;11:43. [doi:10.1186/s12992-015-0129-5](http://dx.doi.org/10.1186/s12992-015-0129-5)
    
    

38. Schleiff MJ, Aitken I, Alam MA, et al. Community health workers at the dawn of a new era: 6. Recruitment, training, and continuing education. Health Res Policy Syst 2021;19:113. [doi:10.1186/s12961-021-00757-3](http://dx.doi.org/10.1186/s12961-021-00757-3)
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.1186/s12961-021-00757-3&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=34641898&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 

39. Jaskiewicz W, et al. Recruitment of community health workers. In: Developing and strengthening community health worker programs at scale: a reference guide and case studies for program managers and policymakers. Washington, DC: MCHIP, 2014.
    
    

40. George A, Young M, Nefdt R, et al. Community health workers providing government community case management for child survival in sub-Saharan Africa: who are they and what are they expected to do? Am J Trop Med Hyg 2012;87:85–91. [doi:10.4269/ajtmh.2012.11-0757](http://dx.doi.org/10.4269/ajtmh.2012.11-0757)
    
    [Abstract/FREE Full Text](http://bmjopen.bmj.com/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoidHJvcG1lZCI7czo1OiJyZXNpZCI7czoxMzoiODcvNV9TdXBwbC84NSI7czo0OiJhdG9tIjtzOjI2OiIvYm1qb3Blbi8xNS8xL2UwODIwODUuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

41. Kok MC, et al. Which intervention design factors influence performance of community health workers in low-and middle-income countries? A systematic review. Health Policy Plan 2014;30:1207–27.
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=25500559&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 

42. WHO. Uganda steps survey 2014 fact sheet. Geneva, Switzerland World Health Organisation; 2014.
    
    

43. Siddharthan T, Kalyesubula R, Morgan B, et al. The rural Uganda non-communicable disease (RUNCD) study: prevalence and risk factors of self-reported NCDs from a cross sectional survey. BMC Public Health 2021;21::2036. [doi:10.1186/s12889-021-12123-7](http://dx.doi.org/10.1186/s12889-021-12123-7)
    
    

44. Colvin D. What motivates community health workers? designing programs that incentivize community health worker performance and retention. In: Developing and strengthening community health worker programs at scale: a reference guide for program managers and policy makers. Washington, DC: USAID/Maternal and Child Health Integrated Project (MCHIP), 2014.
    
    

45. Ludwick T, et al. Poor retention does not have to be the rule: retention of volunteer community health workers in Uganda. Health Policy Plan 2014;29:388–95.
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.1093/heapol/czt025&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=23650334&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 

46. Savitz DA, Wellenius GA. Can Cross-Sectional Studies Contribute to Causal Inference? It Depends. Am J Epidemiol 2023;192:514–6. [doi:10.1093/aje/kwac037](http://dx.doi.org/10.1093/aje/kwac037)
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.1093/aje/kwac037&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=35231933&link_type=MED&atom=%2Fbmjopen%2F15%2F1%2Fe082085.atom) 

47. Creswell JW, Creswell JD. Mixed methods research: Developments, debates, and dilemmas. Res in orgs Found and meth of inq 2005;2:315–26.
    
    

48. Creswell JW, Fetters MD, Ivankova NV. Designing a mixed methods study in primary care. Ann Fam Med 2004;2:7–12. [doi:10.1370/afm.104](http://dx.doi.org/10.1370/afm.104)
    
    [Abstract/FREE Full Text](http://bmjopen.bmj.com/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6ODoiYW5uYWxzZm0iO3M6NToicmVzaWQiO3M6NToiMi8xLzciO3M6NDoiYXRvbSI7czoyNjoiL2Jtam9wZW4vMTUvMS9lMDgyMDg1LmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

49. Venkatesh V, Brown SA, Bala H, et al. Bridging the Qualitative-Quantitative Divide: Guidelines for Conducting Mixed Methods Research in Information Systems. MISQ 2013;37:21–54. [doi:10.25300/MISQ/2013/37.1.02](http://dx.doi.org/10.25300/MISQ/2013/37.1.02)