Measures and Models of Reach in Healthcare Interventions in LMICs:  A Systematic Review

Review methodology

The evidence base on reach within healthcare and health systems interventions in LMICs is methodologically and theoretically diverse, encompassing quantitative, qualitative, and mixed methods designs. To ensure a comprehensive and conceptually rigorous synthesis of this heterogeneous literature, we employed a mixed-methods systematic review (MMSR) approach, conducted in accordance with the PRISMA 2020 guidelines¹⁵. The synthesis process was guided by Whittemore and Knafl’s integrative review methodology, which provides a systematic yet flexible framework for combining empirical and theoretical evidence across multiple research traditions¹⁶. It combines conversations from different paradigms and researcher’s epistemological positions to make sense of a particular topic to present high impact findings¹⁷.

In contrast to traditional systematic reviews that emphasize methodological uniformity for meta-analysis, this review adopted a mixed-methods systematic review (MMSR) approach that embraces methodological diversity by integrating quantitative, qualitative, and mixed-methods evidence¹⁸. This approach supported three core aims: to bridge conceptual and empirical insights in order to advance theory-informed understandings of reach and its measurement across contexts; to ensure transparency and reproducibility through clearly defined inclusion criteria, comprehensive database searches, and detailed documentation of screening and analysis decisions; and to apply Whittemore and Knafl’s framework that allows the combination of quantitative, qualitative, and mixed-methods studies. This systematic yet flexible approach enabled a rigorous and contextually sensitive synthesis, consistent with the review’s aim to illuminate how reach is defined, conceptualized, and measured within the complex, equity-oriented landscape of LMIC healthcare interventions¹⁹. The structured data analysis process, which includes data reduction, display, comparison, and verification, facilitates the synthesis of different information into a unified conclusion^16,17,19.

Patient and Public Involvement

Patients and members of the public were not directly involved in the design, conduct, or reporting of this review. The study did not include primary data collection or direct participant recruitment. The research question and analytical framework were developed in response to priorities identified in the objectives of this study, which focuses on exploring reach of interventions aimed at improving care for people in South Africa and similar low- and middle-income contexts.

Findings from this review will be shared with stakeholder groups to inform future designs of implementation and dissemination.

Search and sampling strategy

The concept of reach under this review is conceptually grounded in the reach domain of the RE-AIM framework²⁰. To inform the search strategy, we first analysed RE-AIM’s definitions of reach, then expanded to its broader health systems conceptualizations. The research team (TT, JM, LF, RC) collaborated with an academic librarian to iteratively develop the strategy, refining terms via two consultative sessions. An initial test yielded limited results, necessitating broader search terms for sensitivity. We also utilized RE-AIM curated repository of framework-citing studies. Thus, two strategies were used 1) a database search (PubMed, Web of Science, EBSCOhost) with customized keywords and 2) A PubMed-specific search using RE-AIM.org terms. The overall review process, including the two complementary search strategies, is visualized in Figure 1. The review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to ensure transparency and reproducibility¹⁵.

Figure 1. Search strategy flow diagram.

Search terms, boolean combinations, search timeline and management

Search terms covered four main concept areas: Reach, Healthcare Interventions, Evaluation and Measurement, and Field Scope. Terms were adapted for each database syntax (e.g., Mesh for PubMed, TS for Web of Science, MH for EBSCOhost). The complete list of terms is presented in Supplementary Table 2. Two filters were applied: 1) Date range limitation to the past 10 years across all databases from the date of the search and 2). Setting: EBSCOhost search was filtered to return studies from low- and middle-income countries (LMICs) only. RE-AIM PubMed search was completed on the 11^th of February 2025, followed by the independent (non-RE-AIM) search on the 28^th of February 2025. The independent of RE-AIM search was update and an EBSCOhost search on the 3^rd of March 2025. An additional RE-AIM extraction was conducted on the 28^th of March 2025. The full list of search terms and Boolean combinations used across databases is presented in Table 1.

Study selection process

The study selection process followed a systematic approach involving de-duplication, title and abstract screening, and full-text assessment, as detailed below and illustrated in Figure 2 (PRISMA flow diagram). The criteria used during screening and full-text assessment are summarized in Table 2.

Figure 2. PRISMA flow diagram showing the selection of studies for inclusion.

A total of 1,776 records were identified across the two search strategies: 1,157 from the RE-AIM-specific PubMed search and 619 from the independent searches (PubMed and EBSCOhost).

Following de-duplication,

After de-duplication 1,618 unique records were screened. Title and abstract screening were conducted sequentially by two reviewers. To ensure consistency, the first 50 records were independently screened by both reviewers (TT and JM), after which TT completed the screening of the remaining 1,568 records. A total of 1,559 studies were excluded at the title and abstract stage.

A total of 59 full-text articles were retrieved and independently assessed for eligibility by TT. Following full-text assessment, 1,564 studies were excluded at either the abstract or full-text stages. The systematic database search and screening process resulted in 25 studies meeting the inclusion criteria. An additional 5 studies were identified through snowball sampling (citation searching of included studies), yielding a final total of 30 studies included in the integrated review.

Data collection process

Extraction of data was guided by the review's objectives to analyse how reach is defined, conceptualized, and measured in healthcare interventions. Data were extracted using two complementary matrices that captured both technical and conceptual variables (see Supplementary Tables 1–2): the first captured technical details (study context, design, frameworks, and reach measures), while the second facilitated interpretive analysis of epistemological positioning and assumptions about reach. Category development began with RE-AIM framework analysis, expanded through snowball sampling, and was informed by the study’s focus on the limited understanding of how health system interventions effectively reach individuals, and which components are impactful based on current measures. The research team (TT, JM, RC, LF) iteratively refined these categories through documented discussions during regular meetings, ensuring alignment with study objectives.

Data items

Data extraction focused on descriptive and conceptual variables. Study characteristics were summarised in a matrix capturing (see Supplementary Tables 1–2):

Conceptual variables were guided by a second matrix including:

Study characteristics

The reviewed studies span a wide range of low- and middle-income countries, particularly in sub-Saharan Africa and parts of Asia. A total of 30 studies were included, representing 15 countries, with South Africa (n=9) and Uganda (n=6) appearing most frequently. Other countries included Angola (n=2), Ghana, Malawi, Nigeria, Zimbabwe, Mozambique, India, Indonesia, Jordan, Zambia, Botswana, China, and multi-country LMIC initiatives (n=3). Most studies employed implementation science methodologies (n=27), with a strong emphasis on mixed methods approaches (n=25) that integrated both quantitative and qualitative data collection. Common data sources included interviews (n=25), surveys (n=18), routine health records (n=15), focus group discussions (n=10), and digital engagement tools (e.g., mobile phone apps) (n=9). Study designs were diverse, including randomized controlled trials (n=5), hybrid effectiveness-implementation trials (n=4), quasi-experimental evaluations (n=3), qualitative-only implementation evaluations (n=5), and observational or descriptive studies (n=7).

Sample sizes ranged from small qualitative cohorts (n=5), to large-scale interventions (n=22) involving thousands of participants, clinic-level actors, or full population coverage. For example, 17 key stakeholders in Mash 2024 GREAT Sustainability study, to large-scale evaluations involving tens of thousands of participants Ramadan 2023’s 18,644 antenatal visits and 23,262 sick child visits across 11 LMICs^21,22. Four studies did not report sample sizes in extractable sections. Participant demographics varied widely, encompassing adolescents and youth, adults, pregnant or postpartum women, and vulnerable or high-risk populations, including people living with HIV, tuberculosis, hypertension, depression, or disabilities. Study settings were equally varied, including urban, peri-urban, and rural clinics, community-based interventions, schools, mobile health platforms, and refugee camps. The most frequently used implementation and evaluation frameworks were RE-AIM (n=23), CFIR (n=6), and PHCPI (n=2), with 5 studies not explicitly naming a framework. Many studies emphasized the need to understand reach not just in terms of coverage but in relation to equity, representativeness, and contextual barriers to participation, offering critical insights for scale-up and health systems transformation.

Quality assessment

The methodological and relevance quality of included studies were assessed (see Supplementary Table 3) using a modified Hawker et al. (2002) framework scoring nine domains on a 1–4 scale (1=very poor, 4=good)²³. This enabled systematic comparison of rigor and relevance across qualitative, quantitative, and mixed-methods studies. Most studies scored highly, particularly in clarity of aims, methodological detail, and practical implications. RE-AIM and hybrid designs showed strong engagement with equity and representativeness. Overall, studies demonstrated methodological soundness and relevance, providing robust insights into reach measurement within LMIC healthcare interventions.

While recognized systematic review standards often recommend design-specific tools (e.g., Cochrane RoB or MMAT), the Hawker's tool was retained for its utility in integrative reviews, which encompassed the diverse¹⁵. The resulting quality scores were not used as an exclusion criterion but served to inform the certainty and weight given to the evidence during the final synthesis.

Data analysis and synthesis

Our analysis incorporated Whittemore and Knafl integrative review method. We employed rigorous thematic and interpretive processes to analyse: 1) how reach was measured, 2) participant characteristics and recruitment methods, and 3) representativeness of target populations - particularly addressing gaps in assessing differential reach among vulnerable group^16,19.

Initial thematic analysis revealed reach data's diverse applications, from implementation refinement to equity assessment. The final synthesis integrated technical and conceptual data extractions with external frameworks, informed by epistemological discussions on positionality and reach's contextual meanings. This approach highlighted how reach intersects with relations, equity, and health system responsiveness beyond technical measurement alone.

Reach within a coverage-focused community of practice

Reach defined through a coverage-oriented lens emphasizes quantifiable participation, using metrics like proportions, enrolment rates, or coverage rates. These are shaped by eligibility criteria (e.g., enrolled patients, trained providers, or treatment cascades). Rooted in a positivist epistemology, this approach prioritizes standardised measurement, comparability, and generalisability. It aims to assess participation, guide scalability, and ensure accountability. Coverage-based approaches rely on administrative data, registers, digital records, or surveys, to inform decisions across settings, populations, and time. Frameworks like RE-AIM often guide these efforts through multidimensional reach assessment.

For example, in the implementation of the GREAT for diabetes program in South Africa, reach was defined as the percentage of trained facilities that implemented the program (31%), and number of patients reached (n=625) based on eligibility criteria, and trained facilitators (n=23)²². Similarly, a review of COVID-19 vaccine uptake in LMICs measured reach through reported proportions of the eligible population willing to receive or those who are actually receiving the vaccine, using secondary data from household surveys and national reports²⁴. In another example, a multi-site study in Indonesia and India defined reach as the proportion of the target population who engaged with intervention activities. In Depok, Indonesia, this was calculated using attendance-based estimates, where total attendance at health camps and community events was divided by the district population, resulting in a reach of 32%. In Ballabgarh, India, reach was assessed through community surveys asking about exposure to program messages or participation²⁵. Together, these examples illustrate how coverage-focused definitions of reach are operationalized across different contexts using varying data sources, from routine records to population-level surveys.

Reach within a relational-focused community of practice

Relational definitions of reach focus on understanding program engagement through concepts like acceptability, willingness, barriers, and depth of participation. Rooted in interpretivist or constructivist epistemologies, they value context, meaning making, and interactions between implementers and participants²⁶. These approaches rely on qualitative methods such as in-depth interviews, focus group discussions, or participatory inquiry to explore how participants navigate choices about engagement²⁷. Relational studies extend beyond coverage metrics, capturing how people experience interventions within their realities²⁸. Reach may include diverse forms of involvement, participant priorities, and how meaning emerges over time. Success is seen in how well participants connect, feel heard, and find the intervention relevant to their daily lives. Meaning is shaped contextually and co-constructed with those directly involved^29,30.

For example, in a study evaluating uptake of 3HP tuberculosis preventive therapy among people living with HIV, reach was assessed through interviews exploring participants’ motivations and barriers. Decisions to participate were shaped by perceived TB risk, trust in peers and providers, and the convenience of the weekly dosing³¹. Similarly, in a CHW-led screening program in Soweto, South Africa, qualitative interviews revealed how home visits made participants feel seen and valued. One noted, “I was so happy, because it shows that we also matter,” highlighting how the intervention extended reach by fostering a sense of inclusion among those often missed by clinic-based care³². These studies demonstrate that relational approaches aim to understand the nuanced components influencing participation and acceptance.

Reach within a systemic-focused community of practice

Structural or systemic definitions of reach frame it as a function of broader issues such as accessibility, equity, and inclusivity, emphasizing how health systems, policies, geography, or social determinants shape who benefits from interventions^33,34. This orientation draws from positivist, realist, or critical epistemologies, depending on how structural influences are theorized, and aligns with goals such as universal health coverage or service transformation^29,34. The purpose of systemic-oriented measurement is to evaluate structural factors that enable or hinder access and participation, focusing on equity, inclusivity, and broader system transformation goals. These approaches often use disaggregated quantitative data, geographic mapping, or policy analysis to assess the equity and inclusiveness of interventions³⁴. They view reach not only in terms of numeric coverage but also in how interventions proactively address population-level inequities, infrastructure limitations, and access constraints^29,33. These approaches emphasize how reach reflects the inclusiveness of service delivery, particularly for vulnerable or underserved groups. While some systemic studies incorporate measurable indicators, they also reflect broader structural concerns and commitments to inclusive and equitable care delivery.

In Ghana’s EMBRACE intervention, reach was assessed by analysing equitable access to maternal and newborn care across rural and urban areas, highlighting demographic disparities needing tailored strategies³⁵. In Uganda, a study on nutrition services framed reach to include uptake and equity-focused delivery, integrating hypertension care into routine HIV visits to reduce travel and cost barriers, benefiting rural and disadvantaged groups²¹. The Angola Maternal and Child Health Handbook study measured reach via equitable distribution and uptake across facility types, revealing urban hospitals had higher coverage (83.3%) than rural health posts (52.7%). Transport challenges and staff shortages limited supervision and consistent handbook use in remote areas. Recommendations included improving transport, increasing training, and enhancing community engagement to boost equitable reach³⁶.

Hybrid and multidimensional measures of reach

While many studies adopt a single dominant orientation of measuring reach or participation others employ multidimensional approaches that blend coverage-based measures with relation and systemic based measure to broaden their perspectives³⁷. These studies conceptualize reach in more integrated terms, recognizing that meaningful engagement requires attention not only to how many are reached but also under what conditions participation occurs³⁸. In these hybrid approaches, reach is measured using proportion-based indicators alongside relational insights like acceptability, contextual relevance, and perceived barriers^14,39. Frameworks such as RE-AIM or other implementation science models often guide these studies, encouraging the combination of standardized metrics with qualitative dimensions of engagement and equity^10,40.

A mixed-methods evaluation of the Health Scouts trial in Rakai, Uganda used the RE-AIM framework to measure reach through both quantitative and qualitative data⁴¹. The intervention involved Community Health Workers (CHWs) delivering HIV services in a hyperendemic fishing community. Coverage metrics came from CHW logbooks, a community survey, and a mobile app, while 72 in-depth interviews with clients, CHWs, and community leaders provided relational insights. While 95% of residents were aware of the program, only 30.7% received counselling. Multivariable analysis showed that men and HIV-negative individuals were less likely to be reached. Qualitative data revealed that perceived CHW usefulness promoted reach, while busy lifestyles and stigma hindered engagement. This approach moved beyond participation rates to illuminate contextual barriers and enablers of reach.

In another instance, a study by Stoutenberg et al. (2024) on the feasibility of home-based hypertension and physical activity screening in South Africa illustrates a hybrid approach to understanding reach by blending coverage with relational insights⁴². Community Health Workers (CHWs) conducted home visits in Soweto to perform standardized health screenings, providing coverage-based data on households and residents reached. To capture relational insights, consenting residents were recontacted within three days for semi-structured interviews, aiming to deepen understanding of their experiences with the CHW-led assessments at home.

Underlying assumptions

A recurring assumption is that populations are both passive recipients and active participants in health systems. Some approaches treat participation as passive, e.g., receiving services during home visits focusing on quantitative metrics. Others emphasize community action, empowerment, and co-production. Viewing participants as active shifts attention to systemic and contextual enablers or barriers, promoting more equity-focused approaches. This framing shapes method and framework choices, influencing how strengths and gaps are understood. Interventions involving multiple stakeholders often assume that once individuals are identified, informed, and supported, they will engage. Community Health Workers, peer supporters, schools, and civil society groups are seen as both service channels and community voices, enabling more meaningful engagement⁴³.

There is a nuanced belief that structural and contextual barriers such as stigma, digital gaps, language, or system inefficiencies limit engagement more than individual willingness. Thus, participation is often viewed as a function of system design, leadership, and resourcing, rather than inherent population disinterest. For example, the PRIDE mental health program in Mozambique highlighted how stigma and limited community awareness made people reluctant to seek help, even when mental health services were available and individuals were open to support⁴⁴. Likewise, in iCARE Nigeria, youth expressed strong interest in digital health tools, but participation was limited by poor network coverage, lack of smartphones, and high data costs, barriers that overwhelmed personal willingness⁴⁵. In the Health Scout intervention in Uganda, community members were receptive to HIV counselling, yet the program’s reach was undermined by high mobility, fear of stigma, and distrust in CHW confidentiality. Similarly, in the Maternal and Child Health Handbook program in Angola, motivated mothers struggled to access services due to transportation difficulties, low literacy, and insufficient health worker support⁴⁶.

These underlying assumptions are critical to understanding the strengths and limitations of current reach measurement approaches. While many methods capture the presence or absence of participation, fewer account for the layered conditions that enable it. Future approaches could benefit from more reflexive and equity-oriented designs that interrogate these assumptions, integrate contextual and relational data, and attend to the diverse ways in which participation is enabled or constrained.

Implication and usage

The ways in which reach data is used across the studies reveal a strong focus on operational learning, system accountability, and implementation refinement. For example, in the Friendship Bench intervention in Zimbabwe, reach was quantified by measuring the percentage of clinic attendees who were screened for common mental disorders and subsequently received psychological services⁴⁷. These measures uncovered significant gaps in uptake between larger urban clinics and smaller rural ones, prompting targeted adjustments in resource allocation, supervision structures, and CHW deployment to enhance equitable access across settings. Similarly, the Ghana EMBRACE study monitored the proportion of eligible women and newborns receiving continuity of care services across antenatal, delivery, and postnatal stages³⁵. This revealed urban and rural inequities, informing program improvements and advocacy to strengthen accountability and implementation. Information on who was and wasn’t reached is used to assess feasibility, refine delivery models, and guide scale-up, adaptation, and policy. In community and peer-led models, reach metrics shape training, supervision, and resource allocation. Identified gaps prompt targeted investments or redesigns to engage vulnerable groups. In trials, reach informs interpretation, generalizability, and cost-effectiveness. Many interventions use reach in continuous quality improvement (CQI), emphasizing real-time adaptation to context. Whether through coverage stats, participation logs, or digital tools, reach serves as a key metric for equity, sustainability, and health system strengthening.

Reflexivity

Reflexivity is a common thread across studies, though its depth varies. Many acknowledge challenges in measuring reach, such as incomplete data, self-selection, and contextual variability. Discussions often highlight the limits of quantitative-only approaches and the need to engage marginalized voices. Qualitative studies show greater reflexivity, addressing researcher positionality, trust, and broader social influences like stigma and migration. Some authors reflect on their insider or outsider roles and how these shape interpretations of findings. For example, in a South African study on home-based hypertension and physical activity screening by community health workers (CHWs), the authors reflect on how weekday-only visits likely excluded employed individuals and students, while 82.1% of participants were women, highlighting gendered and temporal biases in reach. They also note that CHW trainees (rather than full-time staff) conducted the visits, which may have made the encounters more relaxed and thorough than in routine conditions⁴². These reflexive insights show that reach is shaped not just by who is contacted but by broader patterns of access, workforce structure, and relational dynamics.

Similarly, in the Ghana EMBRACE study on maternal and newborn care, the authors acknowledge that formal reach metrics, like participation rates, do not fully reflect accessibility or inclusion. They identify that women outside formal monitoring structures were likely underrepresented, and that factors like health-seeking behaviour, facility choice during pregnancy, and population mobility influenced group assignment and data completeness³⁵. This reflexivity leads them to interpret reach figures within the wider social context, pointing to the need for more participatory, context-sensitive indicators that account for exclusion and structural inequities.

Challenges of measuring reach

Measuring reach remains difficult across five key domains: data limitations, contextual barriers, methodological constraints, participant-level issues, and systemic gaps. Unreliable or missing data due to incomplete records, poor documentation, or inconsistent participation tracking undermines accuracy. External factors like community events, weather, and personal obligations further affect attendance and data completeness. Many studies rely on self-reports, provider perceptions, or observational data, often without standardized tools, introducing bias and limiting comparability.

Systemic and contextual factors also hinder equitable inclusion. Financial hardship, transport challenges, stigma, and logistical issues, like weekday-only outreach or shared phones skew participation. Health worker shortages, high turnover, and space constraints often disrupt implementation fidelity.

Methodologically, small or convenience samples, lack of control groups, and reliance on attendance metrics limit representativeness and analytic depth. Double counting, recall bias, and exclusion of non-participant data compromise validity. Translation issues and inconsistent definitions across settings further hinder cross-study synthesis.

While mixed-methods approaches are common, integration is often weak. Quantitative data dominate, with qualitative insights treated as supplementary rather than analytical. Few studies combine these to explain variation in reach or link participant experiences to broader system dynamics. A shift toward hybrid approaches signals a move toward more contextually grounded, multi-layered evaluations that recognize both measurable outcomes and lived experiences.