Capacity strengthening for community health workers in trachoma prevention and referrals: a mixed-methods study in Al Sabra district, Ibb governorate, Yemen

1 Introduction

Trachoma, caused by the bacterium Chlamydia trachomatis, remains the leading infectious cause of blindness worldwide and continues to affect communities with limited access to clean water, sanitation, and primary health care services (1–3). Based on 2025 data, an estimated 97 million people live in trachoma-endemic areas and remain at risk of trachoma blindness globally (4). Repeated infection leads to conjunctival inflammation, eyelid scarring, trichiasis, and eventual vision loss if untreated (2). Transmission is closely linked to poor facial hygiene, overcrowding, and environmental conditions that facilitate fly–eye contact, highlighting the importance of both biomedical and behavioral components of control strategies (5). In Yemen, trachoma remains endemic in several governorates and ranks among the higher-burden settings in the region, with approximately 1.25 million people living in areas requiring interventions for trachoma elimination and multiple evaluation units exceeding the trachomatous inflammation–follicular (TF≥5%) threshold in children aged 1–9 years (6). Population-based surveys have consistently documented persistent trachomatous inflammation–follicular prevalence among children in multiple evaluation units, reflecting ongoing transmission in settings affected by displacement, limited infrastructure, and barriers to consistent hygiene practices (7, 8). National programs have implemented repeated rounds of mass drug administration (MDA) aligned with global elimination strategies, aiming to reduce infection while strengthening community-level prevention and primary health services (9, 10). The World Health Organization’s SAFE strategy, comprising surgery, antibiotics, facial cleanliness, and environmental improvement, provides the global framework for trachoma elimination (9). Programmatic experience suggests that biomedical interventions alone are insufficient where structural and behavioral drivers of transmission persist. While antibiotic distribution has reduced infection levels in many settings, evidence shows that sustained progress depends on improvements in facial cleanliness and environmental sanitation, supported by community engagement and behavior change interventions (5, 11). In Yemen and similar contexts, studies examining knowledge, attitudes, and practices indicate that awareness of transmission routes and preventive behaviors remains uneven, with gaps in consistent facial cleanliness and environmental hygiene reported despite ongoing campaigns (12). These behavioral challenges highlight the need for locally adapted approaches that strengthen community ownership and reinforce prevention messaging beyond MDA cycles.

Community health workers (CHWs) are increasingly recognized as key actors in bridging health services and communities, particularly in fragile settings. However, evidence on how CHW-led interventions influence behavioral determinants of trachoma transmission remains limited, and few studies have explored how capacity-strengthening activities and community dialogue shape prevention practices within endemic Yemeni contexts (11, 13). While national elimination efforts in Yemen have successfully implemented repeated rounds of MDA, programmatic focus has historically leaned toward the biomedical ‘S’ and ‘A’ components of the SAFE strategy (7, 10, 14). Evidence regarding the ‘F’ (Facial cleanliness) and ‘E’ (Environmental improvement) components remains disproportionately scarce within the Yemeni context. There is a critical research gap in understanding how community-based interventions and behavioral change strategies can effectively address the socio-environmental drivers of transmission in conflict-affected districts like Al Sabra. Given the behavioral and socio-cultural dimensions of trachoma prevention, a mixed-methods approach was used to examine both CHW-related outcomes and the contextual factors shaping community practices. This study seeks to address this gap by exploring the feasibility of a localized CHW capacity-strengthening model and by observing any short-term changes during this implementation.

2 Methods

This study used a mixed-methods pre–post implementation design to examine changes following a two-month CHW capacity-strengthening intervention in Al Sabra district, implemented between 7 May to 9 July 2025. Quantitative components included CHW knowledge surveys and a review routine facility data. Qualitative components included in-depth interviews (IDIs), community dialogues, and endline focus group discussions (FGDs). Quantitative and qualitative data were collected at baseline and endline, although the qualitative data-collection formats differed between the two stages.

Baseline activities were completed by 6 May 2025 and endline activities began on 10 July 2025. Routine facility records were reviewed through 30 September 2025. Because facility registers were available as calendar-month aggregates, May and July crossed the specified phase boundaries and could not be divided into exact baseline, intervention, and endline periods.

Participants included CHWs, healthcare facility personnel, and community members. In total, 15 CHWs, two district supervisors, and one governorate supervisor participated directly in the intervention, while community members participated in baseline community dialogues and endline FGDs.

2.1 Study setting and sampling strategy

The study was conducted in Al Sabra district, Ibb governorate, Yemen. Ibb was purposively selected based on its historical classification as a higher-risk area for trachoma and its operational accessibility. Al Sabra district was selected in coordination with the District Health Office (DHO) based on feasibility and willingness to participate.

The most recent district-level evidence was a 2024 population-based survey conducted after three rounds of MDA. It estimated TF prevalence among children aged 1–9 years in As Sabrah (Al Sabra) at 0.8% (1/129; 95% CI 0.02–4.24), below the WHO 5% elimination threshold (15). This estimate applies to the district as a whole. No recent prevalence estimate was identified for villages, subdistricts, or the three participating facility catchment areas.

Three primary health care facilities (Ainan, Bilad Al Shoaibi, and Najd Al-Jamai) were included to represent different catchment areas within the district. The research sample included 15 CHWs, two district supervisors, and one governorate supervisor. The 15 CHWs constituted the complete CHW roster engaged by the intervention across the three selected facility catchment areas; all who were invited agreed to participate. The sample size was determined by the operational scope of the intervention, available resources, and the number of CHWs engaged across the selected facility catchment areas. This purposeful sample was considered appropriate for an implementation-focused pilot intended to assess feasibility, acceptability, and short-term changes in knowledge, confidence, and reported practices, rather than to produce statistically powered estimates or district-wide generalizable findings.

Participants for qualitative components were purposively selected to capture perspectives from community members, healthcare facility personnel, and health workers within the study area. Participant-level CHW demographics (age, sex, education level, and years of experience) were not collected because the monitoring tool omitted identifying and demographic fields.

2.2 Data collection

2.2.1 Desk review of clinical records

A retrospective desk review covered routine clinical records from 1 January through 30 September 2025 at the three selected primary health facilities. As trachoma is not routinely differentiated in primary care due to the absence of specialized ophthalmic services and limited diagnostic capacity, all reported cases of ‘eye infections’ were reviewed as a proxy indicator. Facilities lack slit-lamp examination, laboratory confirmation, and trained ophthalmic personnel, and trachoma cases are therefore not distinguished from other eye infections in routine records. These data were used descriptively to examine patterns in reported eye infection cases, while acknowledging the limitations of potential diagnostic misclassification. The prespecified baseline period was 1 January–6 May, the intervention period was 7 May–9 July, and endline began on 10 July. Registers were available only as monthly aggregates: January–April were complete baseline months; May combined six baseline days with intervention days; June fell within the intervention period; July combined nine intervention days with endline days; and August–September were complete endline months. Monthly counts were therefore interpreted as contextual service data rather than directly comparable incidence measures or evidence of intervention effect.

Ibb has two typical rainy seasons, March–May and July–September (16). Baseline therefore spanned cooler/drier months and the first rains; the intervention overlapped the end of the first rainy season and the transition to the second; and endline coincided largely with the second rainy season. Changes in rainfall could affect household water availability, hygiene opportunities, fly density, non-trachomatous conjunctivitis, road access, and facility attendance. Weather was not measured during the study, so these factors were treated as potential seasonal confounders rather than observed explanations.

2.2.2 CHW knowledge surveys

Structured surveys were administered to CHWs at both baseline and endline. The survey comprised 18 items. Questions 1–11 assessed factual knowledge of trachoma, including transmission, symptoms, prevention, treatment, risk factors, the SAFE strategy, and environmental improvement. Questions 12–18 assessed prior training, risk perceptions, and self-reported confidence in case identification, referral, and community education. Knowledge items were analyzed individually as correct or incorrect, and no composite knowledge score was calculated. For multiple-response items, responses were counted when participants selected at least one correct option.

The questionnaire was developed specifically for this intervention through collaboration between the project team and Yemen’s Ministry of Public Health and Population (MoPHP), using the agreed capacity-strengthening content and local program priorities. This consultation supported contextual relevance and clarity; however, the tool did not undergo formal pilot testing or a structured content-validity assessment. The tool assessed knowledge of trachoma transmission, symptoms, prevention, treatment, and risk factors, along with prior training and confidence in case identification, referral, and community education. To maintain comparability, the same survey was used at both stages. However, the endline survey included additional open-ended questions to capture CHWs’ perspectives on the training, perceived changes in their self-efficacy, and recommendations for future capacity-strengthening interventions. At baseline, CHWs and supervisors completed the survey anonymously, without names, personal identifiers, or a respondent-role variable. At endline, only CHWs completed the anonymous survey.

2.2.3 In-depth interviews

Semi-structured interviews were conducted with healthcare facility personnel at both baseline and endline. Endline interviewees were different individuals from those interviewed at baseline; therefore, the findings reflect differences between two participant groups rather than changes within the same individuals over time. While the interview guide maintained core thematic alignment to allow comparison, the endline tool was refined based on baseline findings to explore emerging gaps. These interviews examined community perceptions of trachoma, hygiene practices, access to water and sanitation, cultural influences on health-seeking behavior, and the evolving role of CHWs within the community.

2.2.4 Community dialogues and focus group discussions

At baseline, three community dialogues were held with participants recruited from facility waiting areas in the three selected villages. They captured cultural perceptions and hygiene practices, primarily to inform the design of CHW job aids and training materials.

At endline, three FGDs were conducted, each with eight to nine community members recruited from the waiting areas of the same health facilities, using the same recruitment approach as the baseline community dialogues.

The FGDs explored broader barriers to health-seeking behavior, access for internally displaced populations (IDPs), and socioeconomic influences on engagement with the SAFE strategy. Although the baseline community dialogues and endline FGDs used the same facility-based recruitment approach, they involved different participants and data-collection formats. Therefore, the endline FGDs were analyzed as a standalone exploratory component rather than as a direct baseline-to-endline comparison. The qualitative component was intended to complement the quantitative findings by exploring the feasibility and acceptability of the capacity-strengthening intervention from the perspectives of participants in the selected sites. It was not designed as a standalone qualitative investigation providing a comprehensive account of CHWs’ role in trachoma elimination; therefore, we did not seek or claim thematic saturation.

2.2.5 Intervention

The intervention was implemented from 7 May to 9 July 2025 and aimed to strengthen CHWs’ capacity for trachoma prevention and referral. It consisted of a one-day training workshop delivered to 15 CHWs, two district supervisors, and one governorate supervisor. Modules covered trachoma epidemiology, the WHO SAFE strategy, hygiene promotion, case identification, referral protocols, and communication techniques. Participatory methods including role-plays and case scenarios were used to reinforce practical application.

CHWs were provided with fact sheets and checklists to support case identification, referral, and hygiene promotion during community engagement.

CHWs were instructed to report and refer suspected trachoma cases using a standardized referral form developed collaboratively with MoPHP and the District health Office (DHO). Reported cases were compiled and cross-checked against facility health registries for validation and follow-up.

Supervisors conducted routine follow-up with CHWs. A WhatsApp group was established to facilitate real-time communication, problem-solving, and peer support throughout the two-month intervention. The digital group provided a low-cost supervision option because available funds for frequent in-person field supervision were insufficient.

2.3 Data analysis

Quantitative data from CHW surveys and routine facility records were entered into Microsoft Excel and analyzed descriptively. Frequencies and percentages were calculated to summarize baseline and endline outcomes. Qualitative findings were interpreted as exploratory rather than confirmatory.

Inferential statistics, including Wilcoxon signed-rank tests, were not performed. Baseline responses from 15 CHWs and three supervisors were anonymous and lacked both personal and role identifiers; endline responses from 15 CHWs were also anonymous. Individual responses could not be paired, and supervisor responses could not be separated from the baseline dataset.

Survey data conducted during the baseline included data from 15 CHWs and three supervisors which were anonymous, lacking both personal and role identifiers; endline responses from 15 CHWs were also anonymous. As a result, individual responses could not be paired, and supervisor responses could not be separated from the baseline dataset.

Qualitative data from IDIs, dialogues, and FGDs were documented in detailed Arabic notes and translated into English where necessary. The translated notes were imported into NVivo 15 (QSR International) for coding and analysis. The process followed Braun and Clarke’s six-step thematic analysis framework: familiarization, initial coding, theme development, review, refinement, and interpretation (17). Coding was conducted systematically by the researcher using NVivo to organize and manage data, ensuring transparency and traceability of the analytic process. All qualitative data were coded by one researcher; therefore, formal inter-coder reliability was not assessed.

Quantitative and qualitative findings were then triangulated by comparing changes in CHW knowledge scores and routine health-facility records with themes identified in the qualitative data. Areas of convergence, complementarity, and divergence across the data sources were examined to develop an interpretation of the intervention’s feasibility and the observed changes.

3 Results

3.1 Baseline findings

3.1.1 Routine facility records at baseline (January – April)

A desk review of clinical data was conducted in three primary health facilities in Al Sabra district Ainan, Bilad Al Shoaibi, and Najd (Table 1). Most reported eye infection cases originated from Bilad Al Shoaibi, suggesting higher service utilization or a greater burden of reported symptoms compared to Ainan and Najd Al-Jamai. Cases were reported in all four months, with the highest number in March (17 cases), followed by April (15), February (11), and (January (10) (Table 1). Across baseline months, 40 of 53 reports (75.5%) involved children younger than 15 years; sex distribution was nearly even (27 male and 26 female).

Table 1. Monthly distribution of reported eye infection cases in Al Sabra health facilities at baseline (January–April 2025).

3.1.2 CHW knowledge survey

A structured knowledge and perception survey was conducted with all 15 CHWs and three supervisors who would be involved in the intervention (n = 18). The questionnaire was administered in Arabic, entered into Excel, and translated into English for analysis. It included 18 questions divided into two sections.

The first section (Q1–Q11) assessed factual knowledge of trachoma, including definition, transmission routes, preventive measures, clinical symptoms, risk factors, recommended treatment, components of the SAFE strategy, the role of environmental improvements, and key community health messages. The second section (Q12–Q18) explored experience and self-perception, including perceived community concern, population groups most affected, prior training exposure, and confidence in identifying, referring, and educating communities about trachoma prevention.

This structure captured both objective knowledge through correct responses and subjective capacity through attitudes, training history, and self-reported confidence. Table 2 summarizes findings. For multiple-response items, results reflect respondents selecting at least one correct option.

Table 2. Correct responses to knowledge items among CHWs and supervisors at baseline (n=18).

*Multiple-response item; classified as correct when at least one correct option was selected.

At baseline, participants perceived trachoma as a significant community concern, with none rating it as low. Training exposure varied, with some reporting recent trachoma-related training and others limited or no prior instruction. Confidence in identification, referral, and community education was generally high (Table 3).

Table 3. Self-reported perceptions, training history and confidence among CHWs and supervisors at baseline.

3.1.3 Qualitative data collection

Baseline qualitative data were collected through semi-structured in-depth interviews with three male healthcare facility personnel in Al-Mahwa, Ainan, and Najd Al-Jamai, and through three community dialogue sessions conducted in the same villages with participants recruited from facility waiting areas (Table 4).

Table 4. Summary of baseline community dialogue participants by location and sex.

Due to cultural sensitivities and security concerns, interviews and dialogues were not audio-recorded. Detailed notes were taken in Arabic, translated into English, and analyzed using NVivo 15. Data were coded inductively, with codes related to awareness, misconceptions, practices, barriers, CHW roles, and community engagement grouped into broader analytic themes.

Five overarching themes emerged. (Table 5).

Table 5. Thematic summary of qualitative findings (baseline vs endline).

Theme 1: Awareness and Misconceptions

Participants demonstrated a mix of correct knowledge and persistent misconceptions. While some recognized trachoma as a contagious disease that could lead to blindness, others attributed it to cultural causes rather than infection.

“Some people believe trachoma is caused by the evil eye or magic, so they use herbs or warm water to treat it.” (Healthcare facility personnel, Interview 2)

Confusion between trachoma and other eye conditions was common. The term Nafadhi was widely used to describe any eye redness or discharge, leading to underestimation of infection risk.

“People call it Nafadhi and rub their eyes with basil leaves, not realizing it is contagious.” (Healthcare facility personnel, Interview 3)

Theme 2: Community Practices and Reliance on Traditional Remedies

Participants described improving hygiene practices alongside continued reliance on traditional remedies.

“People wash their faces often and avoid sharing towels or soap.” (Healthcare facility personnel, Interview 1)

However, risky practices persisted.

“Some still use herbs to wipe the eyes and wash together in one bowl.” (Healthcare facility personnel, Interview 2)

Dialogues confirmed frequent use of informal care before seeking formal services, often resulting in delayed treatment.

Theme 3: Structural and Systemic Barriers

Water scarcity, poor sanitation, distance to facilities, and treatment costs were repeatedly identified as major barriers.

“Some areas have no water or sanitation, and people still fetch water by donkeys.” (Healthcare facility personnel, Interview 2)

“The hospital is far and expensive. We try eye ointment from the shop first.” (Participant, Ainan)

Limited resources and insufficient training for health workers further constrained prevention efforts.

Theme 4: Role of Community Health Workers

CHWs were consistently described as trusted actors in awareness raising and referral.

“The community health worker plays an active role in preventing illness and providing education.” (Healthcare facility personnel, Interview 3)

At the same time, participants highlighted workload constraints and the need for greater support, including training, transport, and supplies.

Theme 5: Impact on Daily Life and Community Priorities

Trachoma was reported to disrupt household productivity and children’s schooling, with women experiencing a disproportionate burden due to caregiving roles. Participants proposed practical solutions, including visual educational materials, use of local language, and involvement of religious and community leaders to improve trust and uptake of health messages.

3.2 Endline findings

3.2.1 Routine facility records during the intervention and endline (May –September)

Follow-up records were extracted from the same three primary health facilities for May–September 2025. As trachoma is not routinely diagnosed at the primary care level due to the absence of specialized ophthalmic services and laboratory confirmation, all reported eye infection cases were reviewed as a proxy indicator.

Across May to September, 48 eye infection cases were recorded; 18 in May, 15 in June, 9 in July, and 3 each in August and September (Table 6). Females accounted for 27/48 cases (56.3%), and children aged 1–14 years accounted for 37/48 (77.1%). These calendar-month patterns span different program phases and seasons and cannot be interpreted as a pre–post incidence comparison or attributed to the intervention.

Table 6. Demographic distribution of reported eye infection cases during the intervention and endline periods (May–September 2025).

3.2.2 CHW knowledge survey

At endline, the structured knowledge and perspectives survey was administered to the 15 CHWs who participated in the intervention. Unlike baseline, which included 15 CHWs and three supervisors (n=18), the endline survey included CHWs only (n=15). As noted above, supervisor responses could not be separated from CHW responses during analysis. Therefore, baseline and endline knowledge levels should be interpreted descriptively rather than as a direct pre-post comparison. The same survey tool was used at both stages, with four additional open-ended questions (Q19–Q22) included at endline to capture reflections on the training, perceived changes in knowledge and confidence, and support needs. Data were collected in Arabic, entered into Excel, translated into English, and analyzed. Endline CHW perceptions, training history, and self-reported confidence are summarized in Table 7.

Table 7. CHW perceptions, training history, and self-reported confidence at endline.

The first section (Q1–Q11) reassessed factual knowledge of trachoma. All CHWs answered these items correctly except Q10, which required identification of the least effective intervention. The first section (Q1–Q11) reassessed factual knowledge of trachoma. All CHWs answered these items correctly except Q10, which required identification of the least effective intervention. CHW knowledge was high across most items (Table 8).

Table 8. CHW knowledge scores at endline (n=15).

The baseline survey included 15 CHWs and three supervisors, whereas the endline survey included 15 CHWs only. Because baseline responses were anonymous, supervisor responses could not be excluded, and the findings should not be interpreted as a direct paired comparison.

All CHWs rated trachoma as a very high community risk. All participants had received training within the previous three months and reported prior exposure to trachoma-related content. Confidence in case identification, referral, and community education was uniformly high. These findings suggest strong self-reported capacity among CHWs at endline.

Open-ended responses consistently described the training as useful, practical, and easy to apply. CHWs reported improved ability to distinguish trachoma from other eye conditions, increased confidence in responding to community questions, and greater clarity regarding prevention messages. They highlighted improved understanding of disease transmission, the role of facial cleanliness and environmental sanitation, and the limitations of antibiotics alone (9). Participants described feeling more confident and trusted within their communities.

Despite these reported gains, CHWs identified continued needs, including information and education materials, access to soap and clean water, basic screening tools, improved referral and supervision systems, transport support, refresher training, and stronger engagement with community leaders.

3.2.3 Qualitative data collection

The endline qualitative assessment maintained core thematic alignment while expanding questions in response to baseline findings. Healthcare facility personnel participated in in-depth interviews, and three FGDs were conducted in Al-Mahwa, Ainan, and Najd Al-Jamai (Table 9).

Table 9. Summary of Endline FGD participants.

Seven overarching themes emerged (Table 5).

Theme 1: Community Awareness and Beliefs

At endline, participants responses suggested a greater biomedical understanding of the etiology of trachoma. Healthcare facility personnel reported that community members more often identified specific symptoms, distinguished trachoma from other eye conditions, sought care earlier, and relied less on traditional remedies.

“Previously, redness and swelling were seen as normal eye problems. Now, these symptoms are recognized as indicators of trachoma.” (Healthcare Facility Personnel, Interview 1)

Awareness efforts were attributed primarily to CHW outreach, mosque sermons, and community meetings, with mothers consistently identified as key drivers of change in household hygiene practices.

Theme 2: Reported Hygiene Practices

Interview and FGD participants reported more frequent face and hand washing with water and soap and perceived these practices as increasingly embedded in daily routines, particularly among children. They described mothers as reinforcing hygiene practices and men as acting as motivators. These perceptions suggest that mothers and children were involved in reported household hygiene changes, but the study did not directly observe practices at the household level or assess their sustainability. Participants also reported improvements in environmental cleanliness, including latrine use.

Theme 3: Water Access and Environmental Conditions

Despite reported behavioral improvements, water scarcity remained the dominant constraint. Households relied on multiple inconsistent water sources, and limited access to clean water continued to restrict sustained hygiene practices. Participants described adaptability but emphasized that environmental conditions remained a major barrier.

Theme 4: Access to Healthcare and Health-Seeking Behavior

Participants perceived increased reliance on primary health centers for eye problems, with reduced dependence on traditional healers and pharmacies. Walking was described as the main mode of transport and was generally not perceived as a barrier to seeking care. FGDs also highlighted strong community solidarity in supporting basic health costs. Displaced families were reported to have access to services, although participants noted that additional attention was needed to ensure equitable access.

Theme 5: Role and Capacity of Community Health Workers

Community health workers were consistently described as central to prevention, education, case identification, and referral. Their visibility and trust within communities were widely acknowledged, with participants associating reported behavioral changes with their efforts. However, constraints persisted, including lack of supplies, transport, and financial incentives.

Theme 6: Local Initiatives and Community Engagement

Endline participants reported increased community-led initiatives to improve water sources and reinforce hygiene practices. Community gatherings, mosque sermons, storytelling, and visual tools were viewed as effective platforms for delivering health messages. Participants emphasized the importance of combining antibiotic treatment with sustained hygiene promotion.

Theme 7: Ongoing Challenges and Required Support

Persistent challenges included water scarcity, limited resources for CHWs, and lack of specialized eye care. Participants stressed the need for continued engagement between health authorities, CHWs, and community leaders, alongside sustained awareness efforts and improved access to clean water to maintain progress.

4 Discussion

This mixed-methods pilot implementation study examined a community health worker capacity-strengthening intervention for trachoma prevention and referral in Al Sabra district, Yemen, across three phases. The baseline phase included CHW and supervisor knowledge surveys, a review of routine facility records, interviews with local health staff, and community dialogues. The intervention phase consisted of a one-day training for 15 CHWs and three supervisors, supplemented by job aids, referral tools, supervisory follow-up, and WhatsApp-based peer support over two months. At endline, the knowledge survey and facility-record review were repeated, alongside interviews and focus group discussions. Endline findings showed high item-level knowledge and self-reported confidence among CHWs, although baseline and endline survey samples could not be directly compared. Qualitative participants reported greater awareness, increased emphasis on facial cleanliness and hygiene, and reduced reliance on traditional remedies, while routine facility records showed no clear or sustained change in reported eye infections.

At endline, CHWs demonstrated high item-level knowledge and self-reported confidence in their understanding about trachoma, while qualitative findings suggest shifts participants perceived trends toward hygiene-centered prevention consistent with the facial cleanliness and environmental components of the SAFE strategy (9, 18). (9, 18).

(16). Seasonal changes could influence water availability, eye-seeking flies, non-trachomatous conjunctivitis, road access, and facility attendance. Referral volume, service utilization, and detection rates were not systematically measured, and general ‘eye infections’ lacked diagnostic specificity. The routine data therefore provide contextual background only and cannot indicate trachoma transmission or intervention effect.

Reported eye-infection cases fluctuated over the observation period, increasing from 10 cases in January to 17 in March, declining slightly to 15 in April, and reaching 18 in May before falling to 15 in June, nine in July, and three cases in both August and September. Although the decline after May occurred during and following the intervention, it may reflect seasonal influences rather than an intervention effect. Seasonal rainfall and temperature can affect household water availability and hygiene practices, the abundance of eye-seeking flies, road accessibility, and attendance at health facilities. Evidence from other endemic settings makes such an explanation plausible, although not conclusive. A study among children in Pakistan found significant seasonal variation in active trachoma prevalence, while research in Ethiopia documented seasonal differences in water availability, face washing, and other hygiene practices relevant to trachoma control (19, 20). These findings suggest that the fluctuation observed in Al Sabra could reflect seasonal environmental and behavioral conditions; however, differences between settings and the non-specific nature of the facility records prevent direct attribution. (2).

A notable pattern was prominent role attributed to mothers and children in translating trachoma-prevention messages into household hygiene changes. This is consistent with evidence from highly endemic areas of Yemen, where only about half of surveyed households reported regular hand and face washing with soap and water, and children’s face cleanliness was strongly associated with better preventive practices (12). Previous trachoma elimination activities in Yemen have therefore targeted families with children through hygiene kits and health-promotion messages delivered through brochures, radio, mobile announcements, and mosques; these activities accompanied reductions in trachomatous inflammation-follicular prevalence in three of four evaluated areas, although their independent contribution could not be separated from mass drug administration (14).

Evidence outside Yemen similarly supports engaging caregivers. In Ethiopia, mothers who had recently received health education were more likely to report good trachoma-prevention practices (21). CHWs and other frontline health workers have been used to deliver these messages through household and community education, face-washing demonstrations, school activities, distribution of educational materials, and engagement of community leaders. In Zambia, health workers who had received training and had access to information, education, and communication materials, were more likely to undertake facial-cleanliness and environmental-hygiene promotion activities (13). The pattern observed in Al Sabra therefore aligns with evidence that mothers and children are important targets for household-level prevention and that trained, locally trusted CHWs may provide an appropriate channel for reinforcing these practices. However, the present findings support the plausibility of this pathway rather than demonstrating that CHWs education independently produced the reported change.

The intervention highlighted critical structural constraints. Despite high levels of awareness and motivation, water scarcity remains the dominant barrier to sustaining the ‘F’ and ‘E’ components of elimination. Households in Al Sabra rely on inconsistent water sources, which may restrict the frequency of hygiene practices regardless of knowledge levels (3, 8). Similar findings have been reported elsewhere in Yemen, where TF was more common among households using unimproved or surface water for washing than among those with improved washing-water sources (14). In Ethiopia, seasonal variation in water availability, competing household demands, and soap access have also been shown to influence facial-hygiene practices (20). However, the WASH upgrades for health in Amhara (WUHA) cluster-randomized control trial found that an intervention combining water points, wash stations, soap provision, and hygiene promotion did not reduce ocular Chlamydia trachomatis transition over three years (22). This suggests that behavioral interventions may need to be coupled with tangible environmental improvements to achieve long-term impact, while recognizing that improved infrastructure alone may not be sufficient and should be integrated with other components of the SAFE strategy.

Programmatically, these findings align with recent systematic review evidence from fragile and conflict-affected settings showing that CHW-based approaches may expand healthcare coverage, improve disease detection and case management, and strengthen access to care where formal health systems are disrupted (23, 24). Evidence from conflict-affected settings in the Central African Republic and South Sudan further demonstrates that community-based health service delivery can be feasible when supported by community engagement, continued training, and collaboration with local actors (25). Similar approaches have been implemented in high-burden endemic settings such as Ethiopia, where CHWs integrated within the national Health Extension Program play a central role in delivering preventive health services at the household level. Evidence from national program evaluations suggests that this model may have contributed to improved coverage of primary healthcare and uptake of hygiene and sanitation practices critical for trachoma control (26). At the community level, studies indicate that health education and behavior change interventions are strongly associated with improved trachoma prevention practices, with caregivers exposed to such interventions demonstrating significantly higher adoption of hygiene behaviors (21).

Routine primary healthcare records could not measure trachoma specifically because facilities in Al Sabra recorded all eye infections under one broad category and lacked personnel who are trained in standardized trachoma grading. Consequently, changes on recorded cases could represent trachoma, other forms of conjunctivitis, or differences in recording practices rather than changes in trachoma transmission.

Despite these measurement limitations, the community trust and movement towards formal healthcare suggest that a CHW capacity-strengthening approach may be feasible and warrants further evaluation. Together, the variation in prior training identified at baseline and CHWs’ request for continued support suggest that one workshop is insufficient. A standardized capacity-strengthening approach should combine initial training with refresher sessions, regular supervision, appropriate training materials, and referral support.

Several limitations should be interpreted in relation to the exploratory purpose of this pilot. The study included the complete intervention roster of 15 CHWs across three facility catchments, which was appropriate for testing the feasibility and acceptability of the approach but not for estimating effectiveness or generalizing beyond the participating sites. The absence of a comparison district, the short follow-up, and the single one-day workshop also prevent attribution of the observed patterns to the intervention or assessment of their durability. In addition, the anonymous baseline survey included CHWs and supervisors, whereas the endline survey included CHWs only, precluding paired or CHW-only comparisons. The absence of CHW demographic information further limits assessment of transferability.

The locally developed questionnaire was suitable for assessing content delivered during an initial program pilot. Nevertheless, the close alignment with the workshop, limited item difficulty, ceiling effects, and reliance on self-report mean that the findings reflect short-term, item-level knowledge and perceived confidence rather than demonstrated competence. Routine facility records and qualitative findings were therefore used primarily to provide contextual and implementation insights. The facility category of “eye infections” lacked diagnostic specificity and was collected across unequal and seasonally different periods, while the qualitative components involved different participants and data-collection formats at baseline and endline. Note-based documentation, translation, and single-researcher coding were pragmatic decisions in a fragile setting but may have introduced information, selection, and social-desirability bias. The audit trail and triangulation across sources improved transparency, but the findings should ultimately be understood as evidence to refine the approach and inform a larger evaluation rather than evidence of intervention effectiveness.

5 Conclusion

This exploratory study in Al Sabra suggests that there is a need to move beyond awareness toward integrated structural support. Programmatically, three strategic priorities emerge from this research. First, CHW-led behavior change must be linked to tangible water and sanitation improvements, including reliable household water access and latrines in homes and schools; without these enabling conditions, hygiene practices are unlikely to persist despite high knowledge levels. Second, CHW support systems should be institutionalized through structured supervision, transport mechanisms, updated information and education tools, and focused training to strengthen referral quality. Third, monitoring systems should move beyond proxy indicators toward measures that better reflect impact. This includes tracking referral completion, timeliness of care, and household hygiene assessments. Where feasible, future evaluations should utilize trachomatous inflammation prevalence surveys conducted using WHO Tropical Data standards, ensuring the inclusion of equity considerations for displaced populations and female-headed households.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fitd.2026.1854052/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Esmael Habtamu Ali, University of London, United Kingdom

Reviewed by: Mekonnen Moges, Debre Markos University, Ethiopia

Krishna Jasani, All India Institute of Medical Sciences, India