Ethics statement

Prior to participation, written informed consent was secured from each individual or their legally authorized representative. All eligible participants were asked to read and sign the Participant Information and Consent Forms outlining the scope of the study, the risks and benefits of participating in the study procedures and their role in the study. Ethical clearance was granted by the ethical review committee of the Bangladesh Medical and Research Council (no. 506 23 10 2022).

Study design, study setting

This study is a quasi-experimental study conducted over a 24-month period, from September 2022 to August 2024. It took place in two rural communities of Bangladesh: Kalapara upazila of Patuakhali district which was our intervention group and Shibganj upazila (subdistrict) of Chapainawabganj district was our control group. Intervention area is situated on the bank of the Bay of Bengal with twelve unions and thirty-three community clinics. The community clinic infrastructure facilitates awareness of community-based health risks and ensures a comprehensive essential health care services to 6,000 individuals in remote communities. Similarly, the control area is located in the northernmost part of Bangladesh and surrounded by rivers, and consisted of fifteen unions (Fig 2).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Map of the intervention and control area.

Map Data Source is: Local Government Engineering Department (LGED) & Bangladesh Bureau of Statistics (BBS), 2025. Source Link: https://data.humdata.org/dataset/cod-ab-bgd; Government of Bangladesh, via LGED/BBS. Processed by OCHA for HDX.

https://doi.org/10.1371/journal.pntd.0014180.g002

The intervention and control areas were indeed selected purposively, mainly due to operational feasibility and the higher burden of snakebite cases in the intervention site compared to the control site [2]. While Shibganj Upazila was selected as the control site due to its broad socio-demographic comparability and geographic distance, which minimized the likelihood of contamination. To ensure baseline comparability, we used both the Population and Housing Census 2022 [29]. Census data confirmed that Kalapara and Shibganj were broadly similar in terms of household size, sex ratio, access to electricity, mobile phone ownership, and youth NEET rates (Table 1). We also took specific steps to minimize contamination between areas, such as choosing geographically distant subdistricts (separated by several hundred kilometers).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Comparative table of Kalapara and Shibganj on key sociodemographic characteristics.

https://doi.org/10.1371/journal.pntd.0014180.t001

Study participants

The eligibility criteria of the study participants were people residing in the same household for the last six months. Additionally, to select the primary respondent, the household heads were considered.

Sample size

The sample size was estimated to detect a significant change in prevalence based on a two-sided significance level of 95%, a power of 80%, an allocation ratio of 1:1, and an expected effect size of 40% [30,31] with baseline and end line prevalence assumptions of 0.6% and 0.33%, respectively. This resulted in an estimated sample size was 120,315 individuals.

To achieve the sample size, 30,000 household from intervention and control area were surveyed. The baseline and end line survey were conducted following the intervention implementation using the same sampling strategy to evaluate changes in snakebite prevalence, knowledge, and preventive practices. However, to assess the knowledge and practices related to snakebite prevention and first aid measures, 10% of the total households were randomly selected.

Sampling strategy

A purposive selection strategy was employed to identify one subdistrict from the northern region and another from the southern region. Then we systematically identified community clinic catchment areas by compiling a list of all functioning community clinics. The eligibility criteria for clinic inclusion required that the community clinic demonstrate operational activity within the community, maintain a functional community clinic support group, and engage multipurpose health volunteers. Later, the community clinic catchment areas were then subdivided into primary sampling units (PSUs) to facilitate the household survey. A computer-generated randomization process was used to select 125 PSUs from each community clinic catchment area to reduce further selection bias. Within each PSU, 120 households were surveyed to estimate snakebite prevalence. However, systematic random sampling was applied to survey every tenth household for the assessment of knowledge and practices related to snakebite prevention and first aid measures (Fig 3).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. Methodological Framework.

https://doi.org/10.1371/journal.pntd.0014180.g003

Procedure

Intervention phase.

Finalization of intervention-community engagement: The community engagement approach incorporated multiple levels, including informing, consulting, involving, and empowering community members. Emphasis was placed on culturally tailored interventions that actively engaged the community and were adapted to the local cultural context, thereby enhancing their relevance and acceptability among participants. The design of the intervention was informed by the results of formative research, baseline surveys, and established guidelines, such as the National Guideline for Snakebite Management [32] and the WHO SEARO Guidelines (2016) [33]. The intervention package was improved through cooperative workshops with stakeholders and experts to successfully meet local needs. Additionally, necessary modifications were made in intervention delivery strategy after the midline process evaluation. Fidelity of the intervention was determined through midline process evaluation, including monitoring of planned vs. delivered activities (e.g., number of community meetings, seminars, conducted, and proportion of the target population reached). Supervisors used structured checklists to assess the quality and consistency of intervention delivery. For the assessment of the acceptability, we conducted two focus group discussions, and in-depth interviews with newly snakebite affected victims what caused the barriers/facilitators affected their treatment.

IEC material development and demonstration: Information, Education, and Communication (IEC) materials were developed by the research team by a series of workshop with clinicians, public health experts, and communication specialists. To guarantee cultural and contextual relevance, community health workers, educators, and local stakeholders also participated in the creation and presentation of these materials. Content was tailored through formative research, and pre-testing to ensure cultural appropriateness, clarity, and technical accuracy. Focus group discussions and in-depth interviews with community members, traditional healers, and health care workers helped identify misconceptions and locally relevant risk practices. Draft materials (posters, flipcharts, leaflets) were pre-tested among people with similar educational level to check clarity, language, illustrations, and acceptability. Revisions were made based on feedback. Messages were translated into Bangla and visuals were simplified to match literacy levels. The finalized materials were disseminated by field staffs, and facilitators of educational sessions through courtyard meetings, clinics, schools and community gatherings under supervision of project staff. (Insert S1–S4 Posters here).

Implementation of interventions: The implementation activity lasted from March 2023 to March 2024. Structured activities were incorporated through community engagement strategies to improve institutional and community awareness and readiness (Table 2). The Arnstein’s Ladder of citizen participation” could offer a useful lens to demonstrate how our intervention, engaged communities across different levels of participation [34]. The lower rungs of manipulation and therapy were absent, reflecting an intentional design to avoid symbolic or coercive involvement. Most health education initiatives such as flip-chart sessions, posters, folk songs, dramas, and social media content fell within the informing stage, ensuring wide knowledge dissemination. Activities like courtyard meetings, school seminars, and engagements with fishermen, imams, and bazaars reflected consultation, where community members were invited to receive information and share their views. However, decision-making authority remained within the research team. The signs of partnership were seen in the formation of snakebite support groups, collaboration with local NGO (CIPRB), and capacity building of healthcare providers, which fostered shared responsibility in awareness and preparedness. The intervention demonstrated a movement beyond tokenism toward partnership.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Community-Based Intervention Activities.

https://doi.org/10.1371/journal.pntd.0014180.t002

Outcomes

1. Knowledge of Preventive Measures (10 Questions): Evaluated knowledge of precautions like wearing boots in the field, keeping the environment clean, using mosquito nets, and refraining from dangerous practices like reaching into holes or keeping crops in bedrooms (Table 3).

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Different types of questions asked to respondent.

https://doi.org/10.1371/journal.pntd.0014180.t003

2. First Aid Knowledge (8 Questions): Evaluated participants’ knowledge of proper reactions to snakebite, including immobilization of the bite site and immediate hospitalization. Using tourniquets, sucking blood, cutting, applying ashes, or seeking advice from traditional healers are examples of harmful practices that should be avoided (Table 3).

3. Practice (11 Questions): Assessed practical implementation of first-aid and prevention practices, such as regular use of mosquito nets, keeping the environment clean, moving carefully at night, and safe practices when farming or fishing (Table 3).

Data collection

Data collectors holding at least a bachelor’s degree, were recruited following standard procedures for both baseline and end line data collection. In both phase, training of data collectors covered the research concept, objectives, key indicators, interview techniques, questionnaire formats, data collection, record-keeping, and confidentiality.

Statistical analysis

Stata version 20 was used for the analysis of the data. Descriptive analysis was conducted to summarize the characteristics of the study population. Independent two-sample t-tests were performed to examine differences in mean scores between the two study areas for Preventive Measures, First Aid, and Practice. Additionally, Chi-square tests were used to assess associations between categorical variables. Logistic regression models were applied to evaluate the association between the dependent variables (Preventive Measures Knowledge, First Aid Knowledge, and Practice) and the study areas, adjusting for potential confounders. Three separate logistic regression models were conducted: one for Preventive Measures Knowledge (coded as 1 for Good Preventive Measures Knowledge and 0 for Poor Preventive Measures Knowledge), one for First Aid Knowledge (coded as 1 for Good First Aid Knowledge and 0 for Poor First Aid Knowledge), and one for Positive Practice (coded as 1 for Negative Practice and 0 for Positive Practice). Knowledge scores were deemed “Good” if participants obtained >5/10 in Preventive Measures Knowledge or >4/8 in First Aid Knowledge. A practice scores greater than 5/11 was considered “Positive,” while scores of ≤50% in any segment were classified as poor [36].

Visualizations included a confidence interval (CI) plot depicting changes in mean scores and a line and point chart previously uploaded, providing a graphical overview of trends. Statistical significance was determined at a p-value of < 0.05. During pre and post intervention data collection process, four supervisors were employed to ensure quality and compliance with study guideline.

Impact of intervention

The interventions at the intervention area led to a significant reduction in snakebite cases and incidence rates, reflecting the efficacy of community-based strategies in addressing snakebite, an important public health issue. The number of reported cases in the intervention area dropped from baseline to endline, corresponding to a 43.4% reduction in the incidence rate. In contrast, in the control area, there is an increase in snakebite cases of 10.6%. These findings highlight the impact of targeted interventions that combined education, behavior modification, and local engagement to foster preventive practices and reduce risks. The results are very important in an area where snakebites happen often, as they show that health education programs in rural Bangladesh can lower snakebite cases by increasing awareness and promoting safer behaviors [8,35]. Another study on newborn care in the same country emphasized the importance of community mobilization for improving health outcomes in resource-limited settings, further underscoring the necessity of localized, participatory approaches [17]. Several key drivers effectively addressed the community’s specific risk factors, leading to the success of the intervention in the intervention area. Enhanced preventive practices, such as using torches at night, avoiding the storage of poultry and food grain in bedrooms, and tapping the floor with sticks while walking, significantly reduced human-snake interactions, a primary cause of snakebite in rural areas. The intervention also fostered improved knowledge about snakebite prevention and first aid, empowering individuals with critical information to adopt safer behaviors and effectively manage emergencies. Moreover, the participatory approach of the intervention ensured high levels of community engagement, leading to widespread acceptance and adherence. These efforts align with findings from India and Sri Lanka, which indicated that health programs designed for specific cultures and led by the community are essential for creating lasting changes in health behaviors among high-risk groups [21,22,36]. In the Brazilian Amazon, a modest decrease in snakebite incidents has been found 3 years following a multifaceted program for community health education but not quantified in the paper by Vaiyapuriet et al. [22]. Collectively, these elements demonstrate how comprehensive, community-centered programs can mitigate the burden of neglected tropical diseases like snakebite.

Improvement in preventive knowledge

One of the most notable outcomes of the interventions was the improvement in knowledge regarding various preventive measures. Educational sessions, such as bi-weekly health meetings at community clinics, courtyard gatherings, traditional songs, and interactive workshops, could achieve this. During these activities, emphasis was given to practical preventive measures, such as using mosquito nets, wearing boots while working in fields, and avoiding the storage of grains and poultry in bedrooms—practices that reduce the risk of snake encounters. Visual aids like flip charts, combined with the distribution of posters and leaflets, reinforced these messages and extended their reach beyond the immediate participants. In contrast, the control area experienced rather a decline in preventive knowledge, from 13% at baseline to 10% at end-line, which underscores the necessity of continuous education and active engagement to sustain awareness. This finding aligns with prior studies indicating that consistent reinforcement of knowledge is essential to maintaining and improving preventive behaviors [8,9]. A significant improvement in awareness in communities that attended the educational outreach program was found by the Madras Crocodile Bank Trust but not objectively measured/quantified [22].

Significant gains in first aid knowledge

The improvement in first aid knowledge was even more striking in the intervention. In one area, the proportion of individuals with good knowledge of first aid practices rose dramatically from 50% at baseline to 73% at end-line (OR: 2.62, 95% CI: 2.25–3.06). Training sessions at community clinics and the sub-district hospital provided critical guidance on first aid measures, such as immobilizing the bite site, avoiding harmful practices like using tourniquets or making incisions, and seeking immediate medical care. Educational seminars in schools and colleges further amplified the reach of these messages, embedding correct first aid knowledge among younger generations. Culturally resonant activities, including “Boyati (folk)” song performances and drama displays, also played a pivotal role in addressing common misconceptions about first aid. By integrating these methods, the interventions ensured that the information was accessible, engaging, and relatable to diverse community members. Conversely, the control area exhibited no significant change in first aid knowledge, which remained at 57% from baseline to endline (OR: 1.03, 95% CI: 0.89–1.19). This lack of progress in the control area reflects the absence of structured interventions and highlights the need for proactive health education to counter harmful traditional practices, as documented in other rural settings with low resources in the region, including Bangladesh [12,13].

Limited progress in positive practices

The improvement of practices at the intervention area was significant, even though it was modest over the 20-month period. Positive practices, such as wearing boots, using torches at night, and tapping the ground with a stick while walking, increased significantly. These behavioral changes were supported by activities like village market (‘bazar’) sessions, meetings with religious leaders, and parent education initiatives led by ‘Anchal’ officers of the Centre for Injury Prevention Research, Bangladesh (CIPRB). These approaches addressed cultural and social barriers, encouraging the community to adopt safer practices gradually. In contrast, the control area experienced a slight decline in positive practices, which underscores the challenges of promoting behavior change in the absence of targeted interventions. In a study conducted in Sri Lanka, a mismatch was found between knowledge and practice among the farmers; despite adequate knowledge, they do not practice appropriate prevention and first aid measures for snakebite [36]. Previous studies have similarly highlighted the difficulty of sustaining positive practices without continuous engagement and reinforcement [5,6].

The data reveal significant differences between the intervention and the control area in terms of age, gender, occupation, and education, which influenced the outcomes of snakebite prevention interventions. The intervention area’s younger demographic (32% aged 18–30 vs. 25% in the control area) facilitated greater engagement with educational activities, while the control area’s older population faced challenges in adopting new behaviors. Higher female participation in the intervention area (51% vs. 47% in the control area) amplified the impact of interventions due to women’s roles as caregivers and homemakers. Occupation patterns showed the intervention area’s high-risk farming and fishing occupations aligned with targeted preventive measures, whereas the control area’s diverse occupations like driving and business required tailored strategies. Also, the intervention area had a lower illiteracy rate (7.4% compared to 30% in the control area), which helped people adopt preventive and first aid practices more easily, highlighting the importance of using simple and culturally relevant strategies in communities with less education [9,18,22].

This study has important limitations. The intervention and control districts were purposively selected, which may have introduced selection bias and limits the generalizability of findings to other settings with different socioeconomic or ecological characteristics. Especially, the findings may not have been as comprehensive due to the exclusion of children and underrepresentation of diverse occupational groups. On the other hand, the study did not follow the same cohort over time but rather employed repeated cross-sectional surveys, which may have influenced the measurement of changes in snakebite-related knowledge. Furthermore, the outcomes were based on self-reported knowledge and practices, making them susceptible to recall and reporting biases that could have resulted in either underestimation or overestimation of the true effects. The main challenges we encountered during community engagement processes were resources and time constraints, managing expectation and conflicting priorities, and superficial enagement. Despite these limitations, the results provide a scalable framework for improving immediate care and prevention of snakebite in environments with limited resources. However, higher rungs of community engagement such as delegated power and citizen control were not achieved through this research but these activities laid the foundation for stronger, citizen-driven participation in future scale-up and sustainability efforts.