Implementing RED/REC to Reach Zero-Dose Children in Zambia: A CFIR-Informed, Qualitative Study of Barriers and Facilitators

Abstract

Background: Childhood immunisation coverage remains suboptimal in Zambia, with persistent inequities, particularly among zero-dose and hard-to-reach populations. The Reaching Every District/Reaching Every Child (RED/REC) strategy was developed to strengthen equitable vaccine delivery and improve access to immunisation services. This study explored the barriers and facilitators influencing implementation of RED/REC components in selected zero-dose districts in Zambia. Methodology: An exploratory, descriptive qualitative study was conducted in 2025; of the 127 interviews, 36 were key informant interviews (KIIs), 42 were in-depth interviews (IDIs), and 49 were focus group discussions (FGDs) conducted with caregivers, health workers, community volunteers and programme officers. The data were analysed manually using an abductive thematic approach guided by the Consolidated Framework for Implementation Research (CFIR). Results: Implementation of RED/REC was facilitated by structured microplanning, community mapping, multidisciplinary coordination, outreach services, community engagement and the use of local data to identify and reach zero-dose and under-immunised children. Community-based volunteers and local leaders were particularly important for mobilisation and follow-up. However, implementation was constrained by staff shortages, inadequate transport and outreach infrastructure, limited refresher training, insufficient volunteer incentives, and weaknesses in data tools and capacity. Geographical barriers, seasonal farming and fishing activities, religious beliefs, misinformation and concerns about vaccine adverse effects further affected outreach and uptake. CFIR demonstrated that these barriers and facilitators operated across the Inner Setting, Outer Setting, Characteristics of Individuals and Implementation Process domains. Conclusion: RED/REC provides an interconnected platform for reaching zero-dose and under-immunised children, but effective implementation requires simultaneous strengthening of workforce capacity, outreach resources, community partnerships and data systems. Addressing geographical, seasonal and vaccine-confidence barriers is essential for sustaining implementation and improving equitable immunisation coverage.

1. Introduction

Childhood immunisation remains widely recognised as one of the most effective and cost-effective public health interventions for reducing morbidity and mortality from infectious diseases among children [1]. In sub-Saharan Africa (SSA), where childhood mortality remains high, strengthening vaccination against preventable diseases such as measles, polio, and hepatitis remains an essential public health priority towards attaining Sustainable Development Goal (SDG) 3 [2], which aims to reduce under-five mortality to 25 or fewer deaths per 1000 live births by 2030. Ensuring equitable access to routine immunisation is therefore central to global child survival and health equity agendas [3]. Under the Global Vaccine Action Plan (GVAP) 2012–2020, countries committed to achieving at least 90% national coverage and 80% district-level coverage of the third dose of the diphtheria–tetanus–pertussis vaccine (DTP3) among children aged one year [4], a widely used proxy for routine immunisation system performance. Despite progress in many settings, these targets have not been achieved uniformly, and substantial inequities in immunisation coverage persist [5].

In Zambia, the proportion of children aged 12–23 months who received all basic antigens has varied since 1992, declining from a peak of 78% in 1996 to a low of 65% in 2024. Evidence has shown that various factors influence childhood immunisation uptake, including socio-demographic, health-system and economic characteristics such as the gender of the caregiver, rudeness of the health worker, possession of a child immunisation card, understaffing at health facilities and household wealth level [6,7]. For clarity, a zero-dose child is defined, consistent with global usage, as a child who has not received the first dose of the pentavalent vaccine (Penta 1), while an under-immunised child is one who has begun but not completed the vaccine doses recommended for their age in the national immunisation schedule [8,9]. Reaching these children is the central equity objective of the RED/REC strategy examined in this study.

To address persistent inequalities in routine immunisation coverage, particularly in low-performing and hard-to-reach settings, Reaching Every District/Reaching Every Child (RED/REC) strategy was developed by the World Health Organization [10]. The strategy is organised around five core components: microplanning, outreach services, supportive supervision, strengthened community linkages, and data use for action. These components are intended to address commonly identified weaknesses in routine immunisation systems, including inadequate planning, weak supervision, limited accountability, poor community engagement, and underutilisation of local data [10]. RED/REC has been widely adopted across low- and middle-income countries and remains a cornerstone of routine immunisation system strengthening. Since its introduction in 2002, RED/REC has been widely acknowledged as a flexible and context-responsive strategy that can be adapted to local needs and high-priority populations. Its implementation has been linked to sustained gains in immunisation coverage, improvements in data quality, and enhanced engagement of community health workers [11].

However, despite its implementation, the barriers and enablers to the implementation of the core components of the RED/REC strategy are not well understood. Evidence to better understand how this strategy improves vaccine delivery implementation is limited. Moreover, few studies have applied an implementation-science framework to unpack the contextual, organisational and individual determinants that shape whether each RED/REC component is delivered as intended, particularly in high-burden, zero-dose settings. Addressing this gap is essential for translating RED/REC guidance into effective routine immunisation practice. Therefore, this study set out to explore barriers and enablers of implementation of the core components of the RED/REC strategy within the integrated primary health care system in selected Zambian districts with high numbers of zero-dose children. Specifically, the study objectives were to: (i) describe how the five core RED/REC components, namely, microplanning, outreach services, supportive supervision, community linkages, and data use for action, were implemented in selected zero-dose districts; (ii) identify the barriers and facilitators influencing implementation of each component; and (iii) map these factors to the domains of the Consolidated Framework for Implementation Research (CFIR) to explain why implementation was strengthened or constrained.

2. Materials and Methods

2.1. Study Setting

This study was conducted in six districts of Zambia, comprising three predominantly rural (Mazabuka, Lundazi and Mansa) and three predominantly urban study settings (Chingola, Solwezi, and Kabwe). The districts were selected in consultation with the Ministry of Health (MoH) based on their burden of zero-dose and under-immunised children and to capture variation in geographical and implementation contexts. At the time of study planning, 32 districts in Zambia had been identified as having a high burden of zero-dose and under-immunised children. Six of these districts were purposively selected for inclusion in this study.

A total of 24 health facilities were included across the six districts. Facilities were selected to capture variation in geographical location, population served, immunisation performance, and implementation context. This study included stakeholders operating at community, health facility, district, and provincial levels, including caregivers, community-based volunteers (CBVs), healthcare providers, facility in-charges, immunisation programme managers, and other programme officers involved in immunisation service delivery.

2.2. Study Design

We employed an exploratory, descriptive qualitative design to explore implementation of the Reaching Every District/Reaching Every Child (RED/REC) strategy across the six study districts. Data were collected through key informant interviews (KIIs), in-depth interviews (IDIs), and focus group discussions (FGDs), allowing perspectives to be examined across different levels of the immunisation system.

The Consolidated Framework for Implementation Research (CFIR) informed data collection and analysis. CFIR provided an analytical framework for examining contextual and implementation factors across its five domains: Intervention Characteristics, Inner Setting, Outer Setting, Characteristics of Individuals, and Implementation Process [12,13]. CFIR was used to explain barriers and facilitators influencing implementation of the core RED/REC components. This study was designed and reported in accordance with the Consolidated Criteria for Reporting Qualitative Research (COREQ).

2.3. Study Participants

Participants included caregivers, healthcare providers, CBVs, health facility in-charges, and programme officers involved in immunisation activities at the district and provincial levels. Caregivers included mothers, fathers, grandparents, and other family members responsible for children classified as zero-dose, under-immunised, or fully immunised. Healthcare providers were eligible if they were involved in immunisation service delivery and had worked in the selected setting for at least six months. Facility in-charges and programme officers were included based on their roles in planning, managing, supervising, or monitoring immunisation activities. Individuals who met the eligibility criteria but did not provide informed consent were excluded.

2.4. Participant Recruitment and Sampling

Participants were purposively recruited across the six study districts using maximum variation sampling to capture diverse perspectives of RED/REC implementation across geographical settings, health-system levels, participant categories, and childhood immunisation status. Community mobilisers and CBVs supported identification and mobilisation of eligible caregivers. One community health volunteer per participating facility, selected based on familiarity with the local community, worked with the research team to identify potentially eligible caregivers. A total of 127 interviews and discussions involving 517 participants were conducted. Of these, 36 were KIIs, 42 were IDIs and 49 were FGDs (12 with community health workers and 37 with caregivers). Interviews were evenly distributed across the six districts. The distribution by participant category was: caregivers (n = 333), CBVs/community health workers (n = 106), healthcare providers (n = 24), facility in-charges (n = 12), district programme officers/managers (n = 12), provincial programme officers (n = 12), provincial-level stakeholders (n = 12) and implementing partners (NGOs) (n = 6). See Table 1 for a summary of the participants.

Table 1. Distribution of participants across different types of discussants.

Type of Discussant	Discussion Type	No. of Interviews Per Site	Number of Sites	Individuals/ Discussions	Total Interviews	Total Participants
Provincial-Level Stakeholders	Key Informant Interviews	2	6	1	12	12
Provincial Programme Officers (Provincial)	Key Informant Interviews	2	6	1	12	12
District Programme Officers/Managers	Key Informant Interviews	2	6	1	12	12
Implementing Partners (NGOs)	Individual Interviews	1	6	1	6	6
Facility In-Charges	Individual Interviews	1	12	1	12	12
Healthcare Providers	Individual Interviews	2	12	1	24	24
Community Health Workers	Group Discussions	1	12	8–10	12	106
Caregivers	Group Discussions	1	12	8–10	37	333
Total					127	517

Table 2 provides a summary of the participants recruited in each of the six districts. The recruitment of participants was even for most of the participant categories except for Mansa, Solwezi and Kabwe, which have slightly higher numbers of caregivers and community health workers.

Table 2. Distribution of participants by category and district.

Participant Category	Mazabuka	Chingola	Lundazi	Mansa	Solwezi	Kabwe	Total
Caregivers	18	18	17	19	19	19	333
CBVs/community health workers	17	17	19	16	18	19	106
Healthcare providers	4	4	4	4	4	4	24
Facility in-charges	2	2	2	2	2	2	12
District programme officers/managers	2	2	2	2	2	2	12
Provincial programme officers/managers	2	2	2	2	2	2	12
Implementing partners (NGOs)	1	1	1	1	1	1	6
Provincial-level stakeholders	2	2	2	2	2	2	12
Total							517

Data collection followed the planned sampling structure, with emerging findings reviewed throughout data collection to assess the adequacy and recurrence of information across participant groups and study sites. By the completion of data collection, substantial repetition of themes was observed, with no major new themes emerging from the later interviews and discussions.

2.5. Data Collection

Data collection consisted of three complementary qualitative approaches:

• KIIs with provincial-level stakeholders and with provincial and district programme officers and managers.
• IDIs with healthcare providers, facility in-charges and implementing partners.
• FGDs with caregivers and CBVs/community health workers.

Before data collection, the research team conducted advocacy and community-entry visits with Provincial and District Health Offices to obtain administrative support and facilitate entry into participating health facilities and communities, particularly communities with high numbers of zero-dose and under-immunised children.

Research Assistants (RAs) were recruited based on previous qualitative research experience, health-related training, and fluency in relevant local languages, including Bemba, Nyanja, Tonga, and Lozi. They received three days of training covering the study objectives, qualitative interviewing and facilitation techniques, study tools, research ethics, informed consent, confidentiality, audio recording, field-note documentation, and data-quality procedures.

Data were collected using semi-structured KII, IDI, and FGD guides. The guides were translated into relevant local languages, reviewed, and pretested in Chongwe District, which was not included among the study districts. Findings from the pretest informed minor revisions to improve clarity and comprehension.

Interviews and FGDs were conducted by pairs of trained RAs, comprising a moderator/interviewer and a note-taker. Interviewers had no prior relationship with participants before recruitment.

Data collection was conducted in locations that ensured privacy and were convenient for participants. Interviews and discussions were conducted in English or the participant’s preferred local language after written informed consent had been obtained. Sessions lasted approximately 45–90 min, were digitally audio-recorded with permission, and were supplemented with field notes documenting contextual observations and issues relevant to interpretation. Participants received transport reimbursement and refreshments in accordance with the approved study protocol and ethical guidelines.

2.6. Data Management, Transcription and Translation

Audio recordings were transcribed verbatim. Interviews conducted in local languages were transcribed and translated directly into English by trained RAs fluent in both English and the respective local language. Translation quality was assured through comparison with audio recordings and back-translation of selected transcripts. Identifying information was removed from transcripts before analysis, and each participant was assigned a unique study identifier.

2.7. Data Analysis

Data were managed and analysed using NVivo (QSR International, Melbourne, Australia). We conducted thematic analysis using an abductive approach, combining deductive coding informed by CFIR with inductive coding to capture findings emerging from the data that were not adequately represented by predetermined CFIR constructs. An initial coding framework was developed from the study objectives, RED/REC components, CFIR constructs, and emerging concepts identified during familiarisation with the transcripts. Researchers independently coded transcripts during development and refinement of the coding framework. The coding team subsequently compared interpretations and refined code definitions to develop a common codebook. Coding disagreements were resolved through discussion and consensus, with consultation with a senior qualitative researcher where agreement could not initially be reached. The final codebook included code names, definitions, inclusion and exclusion criteria, and illustrative examples. Following agreement on the coding framework, the coding team coded the remaining transcripts.

Codes were compared within and across participant categories and study districts to identify similarities, differences, and recurring patterns. Related codes were grouped into categories and subsequently developed into themes and subthemes.

For the presentation of findings, themes were organised according to the five core RED/REC components: microplanning, outreach, supportive supervision, community linkages, and data use for action. Within each component, barriers and facilitators were identified and mapped to relevant CFIR domains and constructs. This allowed RED/REC to provide the primary structure for assessing implementation, while CFIR was used to explain the factors influencing implementation.

2.8. Trustworthiness and Reflexivity

Several measures were undertaken to enhance the credibility, dependability and confirmability of the findings. All data collectors received standardised central training, data collection tools were pretested before implementation, and audio recordings were supplemented by field notes.

Credibility was further supported by methodological triangulation across complementary data sources including key informant interviews, in-depth interviews and focus group discussions. Further, data triangulation across six districts, multiple health-system levels and diverse participant categories was conducted. Investigator triangulation was achieved through independent coding by more than one researcher, with discrepancies resolved by consensus, and emerging interpretations were, where feasible, discussed with field staff and stakeholders to confirm that they reflected participants’ accounts. Dependability and confirmability were strengthened by maintaining an audit trail of analytical decisions, an explicit and iteratively refined codebook, and reflective memos that distinguished participants’ accounts from the researchers’ interpretations. Transferability was supported by providing a thick description of the study setting, districts and participants, enabling readers to judge the applicability of the findings to other contexts. This study was reported in line with the COREQ checklist.

The research team used a reflexivity process to consider how researchers’ professional backgrounds, prior assumptions about RED/REC, and interactions with participants could influence data collection and interpretation. The multidisciplinary research team comprised members with backgrounds in public health, epidemiology, implementation research and immunisation programming. Team members reflected, through team discussion and reflective journaling, on how their professional roles, their prior involvement with or assumptions about the RED/REC strategy, and their positioning relative to the health system and the study communities could shape the questions asked, participants’ responses and the interpretation of the data and took deliberate steps to minimise the influence of these preconceptions on the analysis.

2.9. Confidentiality

All participants were informed that participation was voluntary and that their responses would be treated confidentially. Names and other direct personal identifiers were removed from transcripts and replaced with study identifiers.

Attention was given to protecting the identity of participants occupying specialised positions within districts, where combinations of job title, facility and location could make individuals identifiable. Accordingly, quotations were attributed using broad participant categories and/or anonymised study identifiers, and potentially identifying information including facility names, specific job titles or locations was removed or generalised where necessary. Audio recordings, transcripts, consent forms, and analytic datasets were stored securely and were accessible only to authorised research team members.

2.10. Ethical Considerations

The study protocol was reviewed and approved by the University of Zambia Biomedical Research Ethics Committee (Ref: 6551-2025) and the National Health Research Authority (Ref: 2153/23/04/2025). Administrative permission was obtained from the Ministry of Health headquarters, including relevant Provincial and District Health Offices, before commencement of field activities. Written informed consent was obtained from all participants before participation. Participants were informed of the purpose of the study, the voluntary nature of participation, their right to decline to answer individual questions or withdraw without penalty, arrangements for confidentiality, and the intended use of the findings.

3. Results

3.1. Participant Characteristics

This study included 517 participants, from different categories and levels, who were interviewed to capture different perspectives of the facilitators and barriers of implementation of the RED/REC strategy. The mean age was 30.4 years (SD 6.7; range 18–42 years). As shown in Table 3, most participants were female (68.7%, n = 202).

Table 3. Characteristics of participants.

Characteristic	Category	n	%
Age (years)	Mean (SD)	30.4 (6.7)	—
Sex	Female	202	69
	Male	92	31

3.2. Overview of RED/REC Implementation Across CFIR Domains

The findings identified multilevel facilitators and barriers of RED/REC implementation across four CFIR domains: Inner Setting, Outer Setting, Characteristics of Individuals, and Implementation Process. The strongest influences were evident within the Inner and Outer Settings. Inner Setting factors determined whether facilities had sufficient organisational capacity, human resources, infrastructure and logistical resources to implement planned RED/REC activities. Outer Setting factors, including geographical accessibility, seasonal livelihoods, population mobility, external partnerships, religious beliefs and community perceptions, influenced both service delivery and demand for immunisation. Characteristics of Individuals influenced implementation through the knowledge, motivation and beliefs of health workers, CBVs and caregivers, while the Implementation Process reflected how planning, engagement, outreach execution, supportive supervision and data use were operationalised.

These facilitators and barriers operated across the five interconnected RED/REC components: microplanning, outreach services, supportive supervision, community linkages, and data use for action. Table 4 summarises the CFIR determinants and the RED/REC components they influenced.

Table 4. Summary of CFIR determinants and their influence on RED/REC components.

CFIR Domain	RED/REC Component(s) Affected	Key Determinant	Facilitating Influence	Constraining Influence
Inner Setting	Microplanning; Supportive supervision	Organisational structure and coordination	Defined roles; multidisciplinary coordination	Competing responsibilities and limited capacity
	Outreach; Supervision; Data use	Human resources	Multidisciplinary involvement	Staff shortages, workload and postponed outreach
	Outreach services	Transport and logistical resources	Motorbikes, fuel and cold-chain equipment	Transport/fuel constraints
	Outreach services	Outreach infrastructure	Use of available community structures	Lack of appropriate shelters/outreach sites
	Data use; Microplanning	Information systems	Local data for targeting and tracking	Missing tools, incomplete records and limited data skills
Outer Setting	Outreach services	Geography and weather	Outreach brought services closer	Flooding, poor roads and impassable bridges
	Outreach; Microplanning; Community linkages	Seasonal livelihoods and mobility	Potential for locally adapted scheduling	Farming/fishing seasons and population movement
	Outreach; Community linkages; Supervision	External partnerships	Transport, fuel, training, incentives and mobilisation support	Reliance on externally supported resources
	Community linkages; Outreach	Community and religious context	Local leaders and community structures	Religious objections, misinformation and vaccine concerns
Characteristics of Individuals	Supportive supervision; Outreach	Health worker knowledge	Initial training and orientation	Limited refresher/in-service training
	Community linkages	CBV motivation	Commitment to identifying and mobilising children	Inadequate incentives and material support
		Caregiver knowledge and beliefs	Recognition of vaccination benefits	Misinformation and fear of adverse effects
Implementation Process	Microplanning	Planning	Target estimation, mapping and scheduling	Plans constrained by resources and context
	Community linkages	Engaging	CBVs, local leaders, schools and WDCs	Difficulty sustaining volunteer participation
	Outreach services	Executing	Community-based service delivery	Staffing, transport, infrastructure and seasonal constraints
	Supportive supervision	Implementation support	District/facility oversight and initial capacity building	Limited continuing/refresher training
	Data use for action	Reflecting and evaluating	Tracking coverage and unreached children	Incomplete data, missing tools and reporting discrepancies

3.3. Inner Setting: Organisational Capacity and Resources for RED/REC Implementation

3.3.1. Microplanning: Organisational Structure and Coordination

Structured microplanning emerged as an important facilitator of RED/REC implementation. Participants described RED/REC as introducing a more systematic planning process that included estimation of target populations, community mapping, identification of outreach locations and scheduling of immunisation activities. Knowing the expected number and location of eligible children enabled facilities to organise outreach more systematically.

“Previously, before we were trained in RED/REC strategy, we could just conduct under-five clinics without doing planning… But now, we’ve seen that it is always important that before you plan to go for immunisations, you need first of all to know the number of children you expect to find from that side.”

[KII No. 17, Healthcare Provider]

Community mapping complemented microplanning by identifying populations, influential community members and appropriate outreach sites.

“I think community mapping is knowing the people you are working with in the community. Then knowing also the people that are influential in that given community, and also identifying the outreach posts…”

[KII No. 8 DRO]

Clearly defined responsibilities across district, facility and community levels further strengthened planning. District nursing officers provided oversight, while facility in-charges coordinated outreach planning and scheduling.

“At district level, the one who plays a key role is the district nursing officer. Then at facility level the one who’s mainly responsible is the in-charge. So the in-charge has to come up with a schedule for the outreach places where we’d go in a particular month.”

[KII EPU District Officer]

Collaboration between nursing, environmental health, health promotion and monitoring and evaluation personnel supported multidisciplinary planning and coordination.

3.3.2. Outreach Services: Human Resources, Transport, Cold Chain and Infrastructure

The availability of motorbikes and cold-chain equipment facilitated implementation of outreach activities. Participants reported access in some settings to motorbikes, vaccine carriers, cooler boxes, ice packs and solar-powered cold-chain systems. However, human-resource shortages were a major barrier. Limited staffing increased workloads and created competing demands between routine facility-based services and community outreach. In some facilities, planned outreach sessions were postponed because insufficient personnel were available.

“If manpower is not there… we find that we postpone. Instead of going, we just say we’ll push it to the next week. Workload and manpower is an issuer.”

[KII No. 9, Facility In-Charge]

Facilities sometimes involved nutritionists and other cadres to compensate for shortages, but these arrangements did not fully address staffing constraints. Inadequate infrastructure at outreach sites presented an additional challenge. Some sites lacked permanent shelters, requiring health workers to provide services under trees or to use borrowed community structures such as churches.

“Facilities have outreach posts, but these outreach posts don’t have structures, like a building or a shelter. You find that people just go under a tree and they are doing their immunisation… during rainy season they can’t operate very well… most of our outreach services we do them either at a church…”

[KII, EPI District Officer]

These conditions made outreach particularly difficult during rain and strong winds.

3.3.3. Supportive Supervision: Organisational Oversight and Capacity

Supportive supervision benefited from defined organisational roles across district and facility levels. District nursing officers and other district personnel provided oversight, while facility in-charges coordinated implementation locally. The involvement of nursing, environmental health, health promotion, logistics, pharmacy and monitoring and evaluation personnel provided a multidisciplinary structure for supporting RED/REC activities.

Initial training and orientation also strengthened the implementation capacity. Health workers reported receiving training on vaccines, vaccine administration and cold-chain management. However, participants identified limited refresher and in-service training as a continuing barrier, particularly for newly recruited staff and for maintaining knowledge among previously trained personnel.

“We need also like the in-service kind of training where we have new staff and also just some refresher training to just remind people of how maybe important this activity is. Others may have forgotten certain principles or components.”

[KII, EPI District Officer]

The qualitative accounts provided comparatively less detail on the frequency, content and quality of routine supervisory visits; therefore, these aspects could not be characterised to the same depth as other RED/REC components.

3.3.4. Data Use for Action: Information Systems and Data Capacity

Local data were used to estimate target populations, monitor immunisation coverage and identify children who remained unreached. Data use therefore supported both microplanning and targeting of outreach activities.

“We analyse the data to know how many children we have reached and how many are remaining. This helps us to see whether we are on track.”

[KII, DHMT Member]

Despite its recognised importance, data use was constrained by staff shortages, heavy workloads and inconsistent availability of standard data tools. Participants reported missing registers and under-five cards, incomplete recording and limited data-management skills among some facility staff. Population mobility also contributed to discrepancies between administratively reported figures and the actual number of eligible children identified in communities.

“Sometimes, the system says we missed five children, but when they go to the community, they find more.”

[IDI, Healthcare provider]

These limitations affected both data quality and the ability of facilities to use information consistently for planning and implementation.

3.4. Outer Setting: Contextual Influences on RED/REC Implementation

3.4.1. Outreach Services: Geography, Weather and Seasonal Livelihoods

Geographical and environmental conditions strongly influenced the implementation of outreach. Heavy rainfall, flooding, poor road networks and impassable bridges restricted both caregiver access and health workers’ ability to reach some communities. In certain areas, neither motorbikes nor vehicles could access outreach sites during the rainy season.

“Like geographical factors, I can say during the rainy season, we have one outreach post where you can’t pass. With a motorbike, you can’t, with a vehicle, you can’t.”

[KII, Facility In-Charge]

Seasonal livelihood activities, particularly farming and fishing, further affected outreach attendance and the ability of health workers to locate eligible children. During planting and harvesting periods, caregivers frequently prioritised agricultural activities or temporarily relocated to farms, while some households moved to fishing areas. As one health worker explained,

“When it’s time for planting… most people are farmers there. So when it’s farming season, you find that they are usually in the farm… So even for outreach, you find the turnout is a bit low compared to the other seasons”

[IDI, Healthcare provider]

Population mobility also complicated the follow-up of missed children:

“When you find that this child is a missing child, you send the CBV to go and check up on the household… you find that these people have gone into farms and it’s very difficult for us now to track them.”

[IDI, Healthcare provider]

Caregivers similarly described missing immunisation activities because they were working in distant fields.

At the same time, outreach itself addressed an important Outer Setting barrier by bringing services closer to communities. Caregivers valued outreach because it reduced travel and allowed them to access vaccination services closer to their homes while managing competing household responsibilities.

3.4.2. Outreach and Community Linkages: External Partnerships

External partnerships including UNICEF, WHO, Churches Health Association of Zambia (CHAZ) and Centre for Infectious Disease Research in Zambia (CIDRZ) facilitated implementation by supplementing health-system resources. Participants described partner contributions to transport, fuel, staff incentives, volunteer training and social mobilisation. These resources expanded facilities’ capacity to conduct outreach and support community engagement.

3.4.3. Community Linkages: Religious Beliefs, Misinformation and Vaccine Concerns

Community-level beliefs and perceptions influenced demand for immunisation. Participants reported that some religious beliefs discouraged vaccination and contributed to refusal among some caregivers. Parental authority and faith-based values were described as important influences on decisions regarding childhood vaccination.

Misinformation and rumours further contributed to vaccine hesitancy. Caregivers described misconceptions that vaccines could cause infertility or that multiple injections could weaken children.

“There are rumours that injections make children infertile when they grow up. This makes some mothers afraid to attend.”

[FGD Caregiver No. 4]

Concerns about actual or perceived adverse effects also influenced caregiver attitudes towards immunisation. Caregivers described fever, prolonged crying, disrupted sleep and injection-site reactions following vaccination, with some reporting that these experiences discouraged subsequent attendance. One caregiver explained,

“Children cry a lot and [have] high fevers. We don’t sleep at night after those injections are given. That is what sometimes discourages us to take our children for injections.”

[FGD: Caregiver]

However, negative perceptions were not universal. Other caregivers viewed vaccination as protective against childhood illness, with one explaining that

“the children don’t get sick anyhow because they are protected already from these diseases.”

[FGD; Caregiver]

These contrasting accounts demonstrate that concerns about adverse effects coexisted with strong recognition of the protective benefits of childhood vaccination.

3.5. Characteristics of Individuals: Knowledge, Motivation and Beliefs

Individual knowledge, motivation and beliefs influenced RED/REC implementation across health workers, CBVs and caregivers. Health workers’ knowledge of vaccination, vaccine administration and cold-chain management supported implementation. Participants described training and orientation as important for developing the knowledge and skills required to deliver immunisation services.

However, limited opportunities for refresher training and staff turnover were reported to create potential knowledge and skills gaps, particularly among newly recruited personnel.

“The issue is lack of continuous training of staff especially the newly employed staff….this creates a challenge as many are not conversant with some of these guidelines and practices.”

[IDI, Facility In-Charge]

Among CBVs, personal commitment to improving childhood immunisation motivated continued participation in identifying and mobilising zero-dose and under-immunised children. This commitment sometimes sustained their involvement despite limited financial and material support. However, inadequate incentives and material support could undermine motivation where the demands placed on CBVs were perceived to exceed the support provided.

“We are always there to support the facilities and to find children who are not vaccinated, but we do this without any support, not even transport or something for food. This becomes difficult for us.”

[FGD: CBV]

Caregiver knowledge and beliefs also influenced engagement with immunisation services. Some caregivers understood vaccination as important for protecting children against vaccine-preventable diseases:

“These immunisations prevent our children from having diseases… they don’t get sick of measles because they were already protected from such illness when they were still babies.”

[FGD: Caregiver]

Conversely, concerns about adverse effects, misinformation and religious beliefs contributed to hesitancy or reluctance among some caregivers. These beliefs were shaped by family, religious and wider community influences.

“We still have parents that fear taking their children for vaccinations because they believe their children will get sick after receiving the vaccine.”

[FGD: Caregiver]

3.6. Implementation Process: Operationalising the Five RED/REC Components

3.6.1. Microplanning: Planning, Mapping and Targeting

Planning was central to RED/REC implementation. Facilities estimated target populations, mapped communities, identified outreach sites and scheduled immunisation activities. This provided an operational foundation for subsequent community mobilisation and outreach.

However, the ability to execute microplans depended on resources and local context. Staffing shortages, transport constraints, weather, geography and population movement could prevent planned activities from being implemented as scheduled.

3.6.2. Community Linkages: Engagement and Mobilisation

Community engagement linked facility planning with households. CBVs, traditional and religious leaders, schools, Ward Development Committees and other community structures supported the identification and mobilisation of eligible children. Community meetings and informal dialogue also provided mechanisms through which caregivers could contribute to decisions regarding outreach timing and location.

“We have got foot soldiers we call CBVs that are dotted in some zones… we explain to them who a zero-dose child is, then they mobilize children, and we do go there and give them vaccines in their respective zones…”

[KII, Facility In-Charge]

Positive relationships between health workers and caregivers strengthened this process. Caregivers described providers as approachable and reported feeling comfortable asking questions both at facilities and during outreach.

3.6.3. Outreach Services: Execution of Planned Activities

Outreach translated microplans and community mobilisation into service delivery. Bringing vaccination services closer to underserved communities reduced travel-related barriers and increased convenience for caregivers.

“They make it easy because they come with their transport and they follow you to your community… for us women maybe we are busy with chores at home so when they follow us in the community it is easy.”

[FGD, Caregivers]

However, execution was vulnerable to staffing shortages, inadequate transport and infrastructure, poor road access, weather and seasonal livelihood activities.

3.6.4. Supportive Supervision: Coordination and Implementation Support

Supportive supervision provided organisational oversight and implementation support across RED/REC activities. Defined district and facility responsibilities and initial training facilitated implementation. However, limited refresher training indicated a need for more sustained capacity strengthening.

3.6.5. Data Use for Action: Monitoring, Reflection and Adaptation

Data completed the RED/REC implementation cycle by enabling facilities to monitor performance, identify missed children and adapt subsequent activities. Routine registers, monitoring charts and community-level records were used to assess children reached and those remaining unvaccinated.

At the district level, performance review could trigger corrective action:

“If you see maybe there is a high dropout in a certain vaccine… from there we come up with activities to just address that one… We also do data reviews… to show what are their challenges and what are their recommendations.”

[KII, Mazabuka EPI District Officer]

At the facility and community levels, registers enabled more targeted follow-up. One participant explained:

“they’ll know which child is fully vaccinated, which child is missing some doses and which child hasn’t received any… community registers… [are] helping us to ensure that we identify the zero doses.”

[KII, Healthcare Provider]

Data use also linked outreach to subsequent follow-up. Following outreach sessions, CBVs reviewed community registers to identify children who had not been vaccinated and initiated tracing in preparation for subsequent sessions. However, the reliability and utility of these data were constrained by incomplete documentation, inadequate data tools and staff workload. As one participant explained,

“You may give the vaccine, but if you didn’t document, it’s a challenge.”

[Mazabuka]

In some settings, improvised registers omitted information considered necessary for monitoring, while high workloads limited supervision of register completion and data quality.

3.7. Cross-Domain Interactions Across the Five RED/REC Components

The findings in this study demonstrated that microplanning, outreach services, supportive supervision, community linkages and data use for action functioned as an interconnected implementation system rather than as discrete activities. Microplanning identified target populations and outreach locations; community linkages enabled CBVs and community leaders to identify and mobilise eligible children; outreach brought vaccination services closer to underserved populations; supportive supervision provided organisational oversight and implementation capacity; and data use enabled facilities to monitor progress and refine subsequent planning.

The CFIR helped explain why this implementation chain functioned effectively in some circumstances and was disrupted in others. Inner Setting constraints, particularly staffing, transport, infrastructure and data-system limitations, reduced facilities’ capacity to execute planned activities. Outer Setting conditions, including geography, seasonal livelihoods, population mobility and community beliefs, influenced both accessibility and demand. Individual knowledge, motivation and beliefs affected implementation by health workers, CBVs and caregivers, while Implementation Process factors determined how effectively planning, engagement, outreach, supervision and data use were linked.

Notably, although there was broad convergence across stakeholder groups regarding the major barriers and facilitators to RED/REC implementation, the emphasis differed by role: managers and supervisors more often highlighted programme coordination and supervision, frontline providers and CBVs emphasised operational and outreach constraints, while caregivers predominantly described accessibility, service experience and household- and community-level influences on vaccination.

These findings indicate that strengthening RED/REC requires coordinated attention to the implementation context and the linkages between its five core components rather than isolated strengthening of individual activities.

4. Discussion

This study examined the implementation of the RED/REC strategy through its five core operational components, namely, microplanning, outreach services, supportive supervision, community linkages, and data use for action, using CFIR to explain the multilevel determinants that facilitated or constrained their implementation. Four CFIR domains were particularly prominent: the Inner Setting, Outer Setting, Characteristics of Individuals, and Implementation Process. Rather than operating independently, these domains interacted across the five RED/REC components. Implementation was strongest where structured planning, organisational capacity, adequate resources, community engagement, motivated implementers and local data use reinforced one another. Conversely, staffing shortages, inadequate transport and infrastructure, geographical and seasonal barriers, insufficient support for community volunteers, community beliefs and weaknesses in data systems disrupted multiple components simultaneously. These findings suggest that effective RED/REC implementation depends not only on the delivery of individual components but also on the strength of the implementation system connecting them.

4.1. Inner Setting: Organisational Capacity as a Foundation for RED/REC Implementation

The Inner Setting emerged as one of the strongest influences on RED/REC implementation, particularly through organisational structure, human resources, transport and logistics, infrastructure, training arrangements and information systems. These determinants affected several RED/REC components simultaneously but were particularly evident in microplanning, outreach services, supportive supervision and data use for action.

For microplanning, participants described a clear improvement over previous, less-structured approaches to organising immunisation activities. Estimating target populations, mapping communities and outreach sites, identifying influential community members and scheduling activities enabled facilities to move from relatively reactive service delivery towards more systematic planning. This reflects the original rationale for RED when introduced by the WHO and partners in 2002 [10,14]. These findings also support previous evidence suggesting that microplanning functions less as a stand-alone product than as an organisational mechanism for coordinating the other RED/REC components [15,16]. Importantly, our findings indicate that effective microplanning depended not merely on the existence of a written plan but on clearly defined responsibilities and coordination across district, facility and community levels. This is consistent with evidence that embedding microplanning within routine organisational processes is important for programme performance [17].

However, the findings also demonstrate an important distinction between planning and execution. A well-developed microplan did not guarantee that planned activities would occur. Human-resource shortages, transport constraints and inadequate infrastructure frequently limited facilities’ ability to translate plans into outreach activities. This was particularly evident for outreach services, where insufficient staffing sometimes resulted in the postponement of scheduled sessions. Evidence from Zambia and elsewhere in sub-Saharan Africa similarly shows that regular outreach is highly vulnerable to shortages of health personnel, transport, fuel and other operational resources [18,19]. The implication is that improving microplanning without simultaneously addressing implementation capacity may produce technically sound plans that cannot be consistently executed.

Outreach nevertheless remained an important mechanism for improving equitable access. Caregivers valued outreach because it reduced the financial, geographical and logistical burden associated with travelling to facilities, particularly for women managing competing household responsibilities. This is consistent with the equity rationale underpinning RED/REC, which seeks to extend routine immunisation to populations inadequately served through fixed-site services [10,20]. The availability of motorbikes, fuel, vaccine carriers, cooler boxes and solar-powered cold-chain systems enabled facilities to reach communities and maintain vaccine integrity. These findings reinforce the importance of sustained investment in the operational infrastructure required to translate outreach plans into service delivery.

Inner Setting determinants were also evident in supportive supervision. Clear coordination between district and facility teams and initial training in vaccine administration and cold-chain management strengthened implementation. Similar findings have been reported in immunisation programmes where district oversight, mentoring, on-the-job training and multidisciplinary coordination have strengthened health worker capacity and routine immunisation planning and monitoring [21,22]. However, participants highlighted inconsistent refresher and in-service training. Staff trained previously could gradually lose familiarity with RED/REC principles, while newly recruited personnel could enter the programme without adequate orientation. Sustained RED/REC implementation therefore requires continuous capacity strengthening rather than reliance on one-off training.

The Inner Setting also shaped data use for action. Health workers used routine immunisation information to estimate target populations, track children reached, identify those remaining unvaccinated and inform subsequent activities. However, heavy workloads, staff shortages, missing or inconsistent registers and under-five cards, and limited data-management skills constrained these processes. These findings reinforce the evidence that strengthening immunisation information systems requires more than improving data availability. Frontline workers require sufficient time, appropriate tools, data-management skills and organisational processes that enable them to interpret and act on information [20]. Evidence from the Better Immunisation Data Initiative in Zambia and Tanzania similarly showed that improving data use required attention to health worker practices, confidence, workflow and the organisational environment within which decisions were made [23].

Collectively, these findings indicate that the Inner Setting provides the operational backbone of RED/REC. Microplanning, outreach, supervision and data use are unlikely to function optimally where basic organisational capacity, staffing, transport, infrastructure and information systems remain inadequate.

4.2. Outer Setting: Geography, Livelihoods, Partnerships and Community Context

Outer Setting determinants strongly influenced both the supply of and demand for immunisation services. Geographic accessibility, seasonal livelihoods, population mobility, external partnerships, religious beliefs and community perceptions affected outreach services and community linkages particularly while also influencing the feasibility of microplans.

Geographical and seasonal barriers were particularly important for outreach services. Heavy rainfall, flooding, poor roads and impassable bridges restricted health workers’ ability to reach some communities and limited caregiver access to services. Farming and fishing activities also competed with immunisation during planting and harvesting periods, while temporary relocation made some children difficult to identify and follow up. These findings are consistent with evidence showing that geographical barriers, opportunity costs and competing livelihood demands contribute to inequities in immunisation coverage and the persistence of zero-dose populations [24,25].

Our findings highlight the importance of considering geographic context when implementing RED/REC. In rural and remote settings, distance, poor road conditions, seasonal accessibility and livelihood-related mobility presented prominent barriers to outreach and vaccination access. In more urban settings, geographic proximity to facilities may reduce some transport-related barriers but does not necessarily ensure effective access. These contextual differences reinforce the central RED/REC principle of locally responsive microplanning and suggest that implementation strategies should be adapted to the specific barriers experienced by communities rather than applied uniformly across districts [15].

Importantly, these findings challenge interpretations of missed vaccination as primarily reflecting caregiver choice or unwillingness to use services. In some contexts, caregivers may be willing to vaccinate but face structural conditions that make attendance difficult at particular times of the year [26,27]. RED/REC implementation should therefore be responsive to local conditions rather than relying on uniform outreach schedules. Seasonally adjusted microplans, flexible outreach schedules and deployment of staff and transport according to changing accessibility and livelihood patterns may improve the programme’s ability to reach underserved populations.

External partnerships were an important Outer Setting facilitator. Partners supported outreach, community linkages and implementation capacity through transport, fuel, training, incentives and social mobilisation. Such support expanded the capacity of facilities to implement RED/REC where routine health-system resources were constrained. However, reliance on externally provided resources raises questions about sustainability. Where transport, fuel or volunteer support depend heavily on partner funding, implementation may weaken when external support changes or ends. Strengthening RED/REC therefore requires consideration of how externally supported activities can progressively be institutionalised within routine district and facility systems.

The Outer Setting was equally important for community linkages. CBVs worked alongside traditional and religious leaders, schools, Ward Development Committees and other community structures to identify and mobilise children and connect households with immunisation services. Similar approaches in other settings have strengthened population mapping, the identification of missed children, community mobilisation and linkages between health facilities and communities [17,28].

However, religious beliefs, misinformation and concerns about vaccine side effects constrained these linkages by influencing caregiver demand. These findings demonstrate that physical access alone is insufficient for reaching all zero-dose and under-immunised children. Immunisation programmes must simultaneously address the social and informational environment in which vaccination decisions are made. Sustained engagement with trusted community and religious leaders, coupled with context-specific communication addressing misinformation and concerns about adverse effects, should therefore remain integral to RED/REC implementation.

4.3. Characteristics of Individuals: Capability, Motivation and Community Trust

The Characteristics of Individuals domain intersected with supportive supervision and community linkages, highlighting the importance of health worker capability, CBV motivation and caregiver beliefs in RED/REC implementation. Initial training strengthened health worker knowledge and skills; however, limited refresher training and staff turnover suggest that implementation capability requires continuous reinforcement through orientation, mentorship and supportive supervision. Evidence from LMICs similarly shows that continuous training and supportive supervision are important determinants of community-based health worker performance [29,30].

The CBVs provided a critical bridge between health facilities and communities by identifying zero-dose and under-immunised children, mobilising caregivers and supporting tracing and follow-up. Their contribution therefore linked microplanning directly to service delivery. Similar evidence from Cameroon demonstrates the value of community health workers in household identification, referral and linkage of zero-dose children to routine and outreach vaccination services [31]. However, reliance on CBVs despite limited and inconsistent incentives raises sustainability concerns. Evidence from LMICs indicates that predictable remuneration, non-financial incentives, recognition, training, supportive supervision and integration within the formal health system are important for sustaining CHW motivation and performance [29,30]. Strengthening RED/REC should therefore include systematic investment in CBV training, supervision, recognition and appropriate financial and non-financial support.

Caregiver knowledge and beliefs further shaped implementation. Recognition of vaccination as protective facilitated uptake, whereas concerns about adverse effects, misinformation and religious beliefs contributed to reluctance. These perceptions were influenced by family, religious and broader community norms, demonstrating the interaction between the CFIR Characteristics of Individuals and Outer Setting domains [30]. Addressing hesitancy therefore requires not only improving individual knowledge but also sustained engagement with trusted community actors and social networks.

4.4. Implementation Process: Connecting the RED/REC Components

The findings suggest that the five RED/REC components functioned as an interconnected implementation cycle: microplanning identified target populations and locations; community linkages identified and mobilised eligible children; outreach translated plans into vaccination services; supportive supervision strengthened implementation; and data use assessed progress and informed subsequent targeting and planning. Implementation failure at any point could therefore weaken the entire cycle. For example, identifying a zero-dose child had limited value if transport or staffing constraints prevented outreach, while outreach without complete documentation weakened subsequent tracing and microplanning.

Data provided an important feedback mechanism within this cycle. Facilities and CBVs used registers and monitoring information to identify missed children, assess coverage and target subsequent activities, consistent with evidence that community-level tracking can strengthen zero-dose identification and microplanning [32]. However, there was less evidence of systematic use of these data for broader programme learning and adaptation. Strengthening the CFIR process of reflecting and evaluating could therefore help transform routine data use into a continuous learning cycle in which implementation challenges identified during outreach, supervision and community engagement routinely inform subsequent microplans [13].

Community engagement similarly cut across RED/REC components rather than functioning solely as a community-linkage activity. CBVs, community leaders, schools, churches and Ward Development Committees helped translate facility plans into community action and facilitated trust, mobilisation and follow-up. Overall, these findings suggest that strengthening RED/REC requires attention not only to individual components but also to the connections between them, particularly the feedback loops linking community identification, outreach, data use, supervision and subsequent microplanning.

4.5. Implications for Policy and Practice

The findings have several implications for policy and programming. First, because the most prominent constraints were located within the Inner and Outer Settings, rather than being intrinsic characteristics of RED/REC itself, strengthening implementation is likely to require greater attention to the health-system and contextual conditions required to operationalise the strategy. Participants did not identify the RED/REC approach itself as a major barrier. Instead, implementation was constrained by staffing, transport, infrastructure, seasonality, volunteer support, data systems and community beliefs. This suggests that further refinement of RED/REC guidance alone is unlikely to address the principal implementation bottlenecks. Second, because the five components function as an interdependent cycle rather than as parallel, independent activities, implementation strategies and monitoring frameworks should track cross-component linkages explicitly—for example, whether community-identified zero-dose children are subsequently reached through outreach, and whether outreach coverage data feed back into supervision priorities and subsequent microplans—rather than assessing each component in isolation [15]. Such efforts will be essential for reducing the number of zero-dose and under-immunised children and advancing equitable immunisation coverage [8,9] in line with the goals of Immunisation Agenda 2030 and the broader commitment to leave no one behind. This also supports the WHO’s repositioning of RED as a platform for strengthening PHC delivery more broadly, since the same systemic constraints identified here for immunisation are likely to affect the delivery of other PHC services co-located with RED/REC activities [10].

Third, implementation should be more context-responsive. High-burden districts should consider: (i) seasonally adjusted outreach schedules that account for rainy season accessibility and agricultural calendars; (ii) predictable financial or non-financial support, supervision and recognition for CBVs; (iii) routine refresher and in-service training, particularly for newly recruited staff; (iv) consistent availability of standard data tools alongside strengthened data-management and use-for-action capacity; and (v) sustained engagement with traditional and religious leaders to address misinformation, vaccine concerns and hesitancy.

4.6. Strengths and Limitations

This study has several strengths: it used a multi-district, multilevel design spanning rural and urban settings and diverse participant categories, triangulated three complementary qualitative methods, and applied an established implementation-science framework (CFIR) to interpret the findings, enhancing the analytical rigour and the transferability of the insights. Some limitations should nonetheless be noted. First, this study was conducted in selected districts and health facilities, which may limit the transferability of the findings to other settings with different contextual, socio-cultural, and health-system characteristics. Secondly, the qualitative nature of this study relied on participants’ self-reported experiences and perceptions, which may be subject to recall bias and social desirability bias, particularly among health workers and community stakeholders involved in immunisation activities. Participants may have provided responses that they perceived to be socially acceptable or aligned with programme expectations. Additionally, because community-based volunteers supported both participant identification and served as study participants themselves, there is potential for selection bias in caregiver recruitment and for social desirability bias in CBVs’ own accounts of their role; this should be considered when interpreting the community-linkage findings. Despite these limitations, this study provides important insights into the multilevel factors influencing RED/REC implementation and offers practical recommendations for strengthening immunisation service delivery, community engagement, and equitable vaccine access in underserved settings.

5. Conclusions

This study demonstrates that implementation of RED/REC was shaped less by the design of the strategy itself than by the interaction of determinants across the Inner Setting, Outer Setting, Characteristics of Individuals and Implementation Process domains. Staffing, transport, infrastructure, geographical and seasonal accessibility, volunteer support, community beliefs and data-system capacity influenced multiple RED/REC components simultaneously. The findings further demonstrate that microplanning, outreach services, supportive supervision, community linkages and data use for action operate as an interconnected implementation cycle. Strengthening individual components in isolation may therefore be insufficient. Implementation strategies should reinforce the connections through which communities and zero-dose children are identified, mobilised, reached through outreach, supported through supervision and subsequently tracked through routine data. Strengthening RED/REC consequently requires investment in implementation capacity alongside institutionalisation of data use as a feedback mechanism linking the five components. Addressing these systemic determinants will be critical to reaching zero-dose and under-immunised children, improving equitable immunisation coverage and strengthening RED/REC as a platform for broader primary health care delivery.