Article Text
PDF
Abstract
Background Drowning is the leading cause of death among children aged 0–17 years in rural Bangladesh, resulting in over 14 438 deaths annually—an average of 43 deaths per day. This study aims to identify socioeconomic, demographic and environmental factors linked to child drowning deaths in Northern Bangladesh—a region of high poverty, which is behind in overall socioeconomic indicators compared with other regions in the country.
Methods We conducted a cross-sectional survey through purposive sampling to identify child fatal and non-fatal drownings among a total of 18 004 households, comprising 71 185 people, in 2 unions in Northern Bangladesh. Interviews were conducted between January and March 2024 with the households that experienced child drownings in the region. We employed a mixed-methods approach to data collection, using quantitative analysis to examine socioeconomic, demographic and environmental factors, alongside qualitative analysis to explore situational factors associated with drownings in the region.
Results Through household visits, a total of 117 households were identified that faced child drowning incidents, comprising 84 fatal (71.8 %) and 33 non-fatal (28.2 %) drownings between 2018 and 2023. The households that faced drownings were comparatively of lower income groups, had lower rates of education and were mostly engaged in agriculture and other domestic work. In 2023, the number of drowning incidents was 34. Out of 117 drownings, 95% occurred between 9:00 and 15:00 hours, and more than 82% occurred between June and October. Out of 117 drowning incidents, approximately 97% of children did not know how to swim prior to the incident. Out of 117 respondents, 73.5% stated that they did not teach their child how to swim. Of those who taught their child to swim, the average age for learning to swim was 8.33 years. Out of 84 child drowning deaths, 75% were male and 25% were female, and the average age was 3.9 years. Out of the 84 fatal drowning deaths, 72.6% occurred in ponds.
Conclusion Identification of socioeconomic, demographic and environmental factors associated with child drownings will help to develop feasible prevention strategies and interventions in the region.
- Child
- Drowning
- Poverty
Data availability statement
Data are available in a public, open-access repository. Data are available on reasonable request. All data relevant to the study are included in the article or uploaded as supplementary information.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Child drowning in Bangladesh is a leading cause of death among children under 5, particularly in rural areas, due to the widespread presence of unprotected water bodies near homes, combined with inadequate supervision of young children.
WHAT THIS STUDY ADDS
Key factors contributing to child drownings in the region include allowing children aged 2–5 years to play unsupervised with peers between 9:00 and 14:00 hours, the prevalence of unfenced ponds near homes, low income, low parental awareness of drowning risks, delaying swim education until around age 8 when it could begin at 6, and the absence of effective drowning prevention measures.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Our findings highlight the urgent need for drowning researchers and practitioners in Bangladesh to act by fostering multisectoral coordination, increasing public awareness and urging the government to expand the integrated community-based childcare and swim-safe facilities project to cover more districts and subdistricts.
Background
Drowning, although a relatively neglected cause of death, is a major global killer, particularly among children and young adults. In 2019, over 236 000 people died from drowning, with 90% of them occurring in low-income and middle-income countries.1 Alarmingly, it is the third major cause of death worldwide for children ages 5–14, even higher than deaths from congenital anomalies, leukaemia, lower respiratory infections, epilepsy, dengue and meningitis.2 Globally, the highest drowning rates are among children aged 1–4 years, followed by children aged 5–9 years. Worldwide, males are twice as likely to drown as females.3 Drowning deaths are preventable by undertaking prevention interventions and policies that address known risk factors. However, this is still nascent in many developing countries. A better understanding of causes and circumstances of drowning is required through local data collection or surveillance in a particular region.4
The definition of drowning for this research has been taken from the consensus definition developed by the expert panel of the World Congress on Drowning in 2002 as ‘Drowning is the process of experiencing respiratory impairment from submersion/immersion in liquid’(1). This eliminates using separate definitions for fatal (death by drowning) and non-fatal cases. The drowning process begins with respiratory impairment as the person’s airway goes below the surface of the liquid (submersion) or water splashes over the face (immersion). If the person is rescued at any time, the process of drowning is interrupted, which is termed a nonfatal drowning. If a person dies at any time because of drowning, this is termed a fatal drowning. Any submersion or immersion incident without evidence of respiratory impairment should be considered a water rescue and not a drowning.4 5
Drowning deaths differ based on national and household income levels.6 7 Low-income and middle-income countries experienced 3.4 times more deaths than high-income countries. Specifically, drowning deaths are disproportionately higher in African and South-East Asian countries.3 Alarmingly, regardless of country-level economic development, child drowning deaths are disproportionately higher among the poorest and least educated people who live in rural settings, particularly around bodies of water, and in communities with minimal resources to avoid risks around water sources.8 Studies from Bangladesh, India, Vietnam and China have consistently attributed drowning as the major cause of death in children aged 1–4 years.9–11
The Bangladesh Health and Injury Survey (BHIS) 2003 was the first injury survey conducted in the country, with 171 366 households comprising 819 429 people. Drowning was identified as the highest cause of death in children ages 1–17 years, higher than victims from pneumonia, malnutrition and diarrhoea combined. The second BHIS survey, conducted in 2016, analysed survey results from 299 216 households, and drowning was rated as the third-leading cause of injury-related mortality, following suicide and traffic accidents.12 A report from Bangladesh revealed that drowning results in roughly 19 000 deaths each year, with about 77% of these victims being under the age of 18.9 Socioeconomic, behavioural, geographical and environmental factors are related to a disproportionate amount of child drownings in many developing countries, including Bangladesh.13 Male gender, children under 5 years of age, mother being illiterate, seasonality and being of lower socioeconomic status were associated with an increased risk of fatal and non-fatal drowning events. Lack of child supervision and inability to swim are considered major reasons for child deaths in Bangladesh.14 For young children, all water outlets including buckets, bathtubs, ponds or pools are sources for drowning.
To mitigate the risk of flooding, a common practice in Bangladesh involves constructing homes on elevated land masses formed by excavating the surroundings of the home sites. This process results in uncovered pits that subsequently transform into ditches and ponds adjacent to the homes. Unfortunately, a significant number of drowning incidents involving children under the age of 5 occur in these locations, which are generally within 20 m of homes, and particularly during morning hours (9:00–13:00 hours).9 14 15 Several studies from pilot drowning prevention interventions in the middle and southern region of Bangladesh highlighted these reasons and factors of child drowning deaths.14–16 Limited studies have focused on socioeconomic, behavioural, geographical and environmental factors of child drowning in Northern Bangladesh. The socioeconomic development in this region is considered to be lower compared with the other regions in Bangladesh17; this warrants to seek specific causes of child drowning and might have huge potential to contribute directly to drowning prevention and accelerate preventative interventions and strategies in Bangladesh. This study aims to identify socioeconomic, demographic and environmental risk factors associated with child drownings in Northern Bangladesh.
Methods
To achieve the main aim of the research and to answer the research question concerning how sociodemographic, geographical and environmental risk factors have contributed to child drowning in Northern Bangladesh, the study applied a mixed-methods approach18, collecting data through closed-ended survey questions (quantitatively) and open-ended interview questions (qualitatively). Each method has limitations of its own. However, mixed methods provide more robust results than relying on solely qualitative or quantitative approaches.19 20
A cross-sectional survey collecting drowning incidence through purposive sampling was conducted between January and March 2024 in two subdistricts, Debiganj in Panchagarh District and Taraganj in Rangpur District—regions of high poverty, which are behind in overall socioeconomic indicators compared with other regions in Bangladesh (figure 1). Drownings of children in ponds have been a leading killer in the region for generations. Debiganj is a subdistrict in Panchagarh district comprising 10 unions. A union is the lower-level local government unit and it is divided into nine wards. Out of 10 unions in this subdistrict (Debiganj), Chilahati Union was selected randomly for the field survey. The total number of households in the union was 7702 in 2023. The total population in the union was 30 100. Taraganj is a subdistrict in the Rangpur district comprising five unions, and Hariarkuti Union was selected randomly for the household survey. The total number of households in the Hariarkuti union was 10 302 in 2023. The total population in the union was 41 085. The total number of households and population in the two unions combined was 18 004 and 71 185, respectively. We developed a baseline of drowning data from the selected two unions. The choice of these two unions (moving from district to subdistricts and then to unions) was based on both the prevalence of drowning incidents and the availability of surveyors. The interview findings identified age and sex characteristics, timing and season, children’s activities prior to drowning, location factors, financial hardship, lack of supervision and environmental context as contributing factors to fatal and non-fatal drownings among children in the target households for the 12 months prior to 2024 and in the prior 5 years.
Location of study areas, Chilahati and Harikuti unions in northern Bangladesh.
Data collection
Over a 3-month period (January–March 2024), a baseline database was created by estimating child drownings and identifying sociodemographic characteristics of child drownings and other risk factors including place, time and situation of drownings in Northern Bangladesh. To develop a full account of child drownings, a database and interview questions were developed adapted from the child drowning risk estimation index developed by Borse et al,21 the water incidence index by Hills et al,22 and the guidelines for community surveys on injuries and violence by WHO.23 Both qualitative and quantitative methods were applied to document and analyse drowning occurrences in the region. The pretested database format and interview tool were used consisting of a mixture of structured, semistructured and open-ended questions. All data were collected through face-to-face interviews. The interview tool consisted of two parts that included closed-ended and open-ended questions. The first part focused on the socioeconomic and demographic characteristics of the participants (age–sex of drowning child, mother’s age, profession of parents and monthly income of parents). The second part focused on place, time, activity before drownings, water depth, distance from home, activity of accompanying person, swim ability, specific location features, situations of drownings and preferences for prevention approaches. The interviews were conducted only with the households that faced drowning events. To identify a child drowning victim family, an investigator had to visit every household in the allocated ward during the survey period and ask if there had been a child drowning in the last 6 years. Then, interviews were conducted if a drowning incident, either fatal or non-fatal, occurred in the household in the last 6 years. Thus, a total of 117 households were identified that faced drowning incidents in the prior 6 years (figure 2). Each specific drowning case was further explored by narratives from drowning-experienced households. All questionnaires were written in Bengali, and a dedicated native Bengali-speaking person interviewed participants.
Flow chart of the data collection method.
18 data collectors, one for each ward in the unions were employed for data collection, and two supervisors, one for each union, were employed for the supervision and monitoring of the data collection process. All selected data collectors and supervisors were trained in the study objectives, interview questions and ethics protocol. During the visits, family members who were available and comfortable answering questions were interviewed. A household member was defined as someone living in the same house, including domestic helpers or long-term guests who shared daily meals and participated in regular activities within the household. Distance between household and drowning site was determined. If the site was near the household, then the data collector measured it visually. Repeat visits were made to the households where no respondents were available during the first visit. The excerpts of the interviews were translated and transcribed into English by a certified translator along with a researcher who also had bilingual expertise.
Patient and public involvement
Engaging patients or the public in the design, implementation, reporting or dissemination of our research was neither feasible nor suitable for this study
Data analysis
Data from the baseline census were used to estimate child drownings in the baseline year and in the prior 5 years, and these were categorised by several quantitative variables such as sex of child, age group of child, place, seasonality and time of drownings. The participants were categorised by sex, age group and sociodemographic characteristics. The responses to closed-ended, semistructured and open-ended questions were sorted, based on the theme of evidence and processed and analysed in sequence, both manually and automatically, using MS Word, NVivo V.12 software and MS Excel. For example, the ratio of male to female respondents, their income levels, the number of child drowning incidents and the mortality rate of male and female children were calculated using MS Excel. Furthermore, an inductive coding process was carried out in this study for the qualitative variables, to categorise the themes derived from similar narratives such as, ‘the activities of children prior to drowning,’ ‘the location of drowning,’ ‘financial hardship resulting in a lack of supervision and the inability to fence off bodies of water’. The codes were expanded to broader themes, in accordance with the study objective. The results were interpreted based on the relevant theoretical literature. Data were presented as averages calculated based on the responses provided by the participants. The quantitative findings were presented in the form of a frequency tally, in a hierarchical order, according to their weight values (tables 1–6). The qualitative findings were provided alongside the quantitative findings to gain a deeper understanding of the responses, in line with the questions addressed in this study.
Demographic and socioeconomic characteristics of respondents in northern Bangladesh, based on field survey held in January–March 2024
Drowning incidents per year in northern Bangladesh from 2018 to 2023 (n=117)
Age of fatal and non-fatal drowning in northern Bangladesh between 2018 and 2023
Time of drowning incidents among 117 drowning cases in northern Bangladesh between 2018 and 2023
Month of drowning incidents among 117 drowning cases in northern Bangladesh between 2018 and 2023
The activities of children prior to drowning in northern Bangladesh, based on field survey held in January–March, 2024
Results
Sociodemographic characteristics of the participants
Of the 117 respondents, 71 (60.7%) were male and 46 (39.3%) were female (table 1). As for the categories of the respondents by family type, 90 (77%) were nuclear families (ie, parents and children) and 27 (23%) were extended families (ie, grandparents, parents and children or siblings living with spouses and children). The average size of the family was 5.2. Approximately 92% of respondents were aged between 26 and 55. Approximately 95% of the respondents earned less than BDT15 000 per month (US$1=BDT116.99 as of May 2024). The average income of the respondents per month was BDT 7831. About 47% of our respondents were illiterate. Approximately 36.8% of our respondents were farmers and 30.8% were housewives. Other occupations included day labourers, small traders, service workers, unemployed, remittance earners and fishermen, amounting to 12.8%, 11.2%, 2.1%, 1.7% 0.9% and 0.9%, respectively. Approximately 69% of houses were made of corrugated iron sheets, locally known as tin-made houses. The remaining 31% of houses were semibrick-built (21%) or brick-built single-story houses (10%).
Geographical context of the respondents
Through visiting a total of 18 004 households, we identified 117 drowning incidents in Chilahati and Hariarkuti unions. Out of these 117 drowning incidents, 55.6% of respondents were in Chilahati Union and 44.4% were in Hariarkuti Union. The distribution of drowning incidents is presented by ward in table 2. In Chilahati Union, the highest number of drownings (23.1%) occurred in ward 2, where there was a high concentration of ponds in densely populated areas. In contrast, the lowest incidence (1.5%) was recorded in ward 9, where fewer ponds were present in sparsely populated areas. In Hariarkuti Union, the highest drowning incidence (17.3%) occurred in ward 2, largely due to the low-lying terrain where people had excavated deep ponds and built homes on plinths raised from the pond soil. The lowest incidence (3.8%) was in ward 1, where there were fewer ponds in sparsely populated areas.
Age and sex of child drowning
Out of 117 child drowning incidents between 2018 and 2023 in the 2 unions studied, 84 (71.8 %) were fatal and 33 (28.2 %) were non-fatal. In 2023, the number of drowning incidents was 34 (table 2). Out of 84 fatal drowning deaths, 45.2% of child drownings occurred in 2–3 years (table 3). The average age for drowning deaths was 3.9 years. Out of 84 child drowning deaths, 63 (75%) were male and 21 (25%) were female. The gender proportion of 33 non-fatal drownings, was 85% male and 15% female. Non-fatal drownings did not occur after 8 years old. Notably, four fatal drownings (4.8%) occurred among 12 years old.
Swimming ability prior to drowning
Out of 117 drowning incidents, approximately 97% of children did not know how to swim prior to the incident. In fatal drownings, the percentage ‘who knew how to swim’ and ‘did not know how to swim’ were 2.4% and 97.6%, respectively. In non-fatal drownings, the percentage ‘who knew how to swim’ and ‘did not know how to swim’ were 6.1% and 93.9%, respectively. We noted that out of 117 respondents, about half of them mentioned limited opportunities to teach children to swim, particularly because of pollution of the pond water. There were no designated swimming spots except traditional ponds and rivers for bathing and swimming purposes in the region.
Time and seasonality of drowning
The findings suggested a pattern in time and seasonality of drowning in the region. Out of 117 drownings, 83% occurred between 9:00 and 14:00 hours (table 4). These were the periods when households engaged heavily with household chores. 47% of drownings occurred between 11: and 12:00 hours which is the highest peak period of drowning incidents in the region. Those that occurred in the afternoon between 17:00 and 18:00 hours were mostly related to water sports. More than 82% of drowning incidents occurred between June and October—the monsoon season, when the level of water through seasonal rainfall and flooding rise to the household level in the region (table 5).
The activities of children prior to drowning
The findings suggest that prior to drowning, children were doing a range of activities, which were grouped into eight categories (table 6). The first five activities included (1) playing in the courtyard alone; (2) playing with peers; (3) staying inside the house; (4) playing in the ponds and (5) having a bath in the river. As noted earlier, the average age of drowning was 3.9 years, which is an active playing age—an age of huge inquisitiveness about one’s surroundings. Except for two children who were under their grandparent’s supervision, the remaining children were without direct supervision of adults. Those 13 children who were inside the house had the scope to escape from it and attempted to explore outside alone.
A parent residing in ward number 8 of the Hariyarkuti Union shared a distressing account:
On the morning, my spouse and I left for work around 9 am, leaving our five year-old son playing in the backyard. When we returned home at noon, our hearts sank as we could not find him. Despite an exhaustive search, he was nowhere to be found nearby. It was a harrowing moment when we eventually discovered him in the pond.
Locational factors of child drowning
Most ponds were around 5–7 feet deep, and their proximity to houses, typically 20–30 m, heightened the risk of drowning for children. Out of 84 fatal drowning deaths, 72.6% occurred in ponds, 19.1% occurred in rivers and 8.3% occurred in ditches. Out of 33 non-fatal drownings, 75.8% occurred in ponds and 24.2% were in rivers. The average depth of water in the ponds where the fatal drownings occurred was 7.177 feet. Out of 117 drownings, the average distance from the place of the incident to the house was 23.5 m.
A participant explained the issue of the proximity between houses and ponds causing a child to drown(figure 3). A mother, aged 33, from the Hariyarkuthi Union in Debiganj subdistrict, stated; I am blessed with two sons and a daughter. A tragedy struck when one of my 3-year-old sons tragically drowned in a nearby pond. It was a usual good morning for my family which later filled with sorrowful memories. The child was playing at the main house while I was occupied with household chores in a separated little room around 9 am. I usually hear the child whilst it plays but sometimes I could not because of noise arising from cooking. On occasions, I also could not heed it whilst heavily busy with household chores. On that day, upon realising I had not heard its noise for a while, I went to search for him in the house. I could not find the child by tirelessly searching here and there. Later, I found its lifeless body floating in an adjacent pond, located just 10 meters away from the house.
Location of ponds and houses in Debiganj and Taraganj subdistricts in Northern Bangladesh in February 2024. The absence of barriers between ponds and nearby houses also posed a significant risk for child drowning.
Out of 117 drownings, a total of 93 (79.4%) occurred in ponds and ditches, which were without any barriers. This open access to reservoirs allowed children an easy entrance. The ponds where child drownings occurred were most often owned by more than one family and needed everyone’s consent and financial contributions for fencing. This challenge was described by one participant in Debiganj: The grandfather of a dead child stated:
“I got my granddaughter drowned eight or nine months ago. The mother of the child went to a relative’s house. A brother of the child also went with her mother. We all know that the child was accompanied by her mother. The granddaughter did not follow her mother towards the relative’s house and was playing with friends close to a pond. When the child fell into the water, peers got scared and hid the drowning incidence. After a while, when I was walking along the pond, I found her floating on pond. I picked her up and pressed on her stomach, she was turned on her head. Some water also came out of the stomach. But I could not save my granddaughter. Her life was up to this point. The lack of fencing around the pond exacerbated the tragedy, as it could have potentially prevented the child’s untimely death. We could not put fencing around the pond because it was owned by others. The devastating loss serves as a stark reminder of the urgent need for protective measures, such as fencing, to safeguard against such tragic incidents in the future. What happened has already happened, we must accept it. We can only pray for her departed soul.”
Financial hardship leading to lack of child supervision and fencing around ponds
A significant proportion (82.1%) of respondents highlighted the proximity of ponds to houses as a predominant cause of child drownings. Furthermore, respondents reported that the absence of protective fencing around these ponds contributed to the occurrences of child drowning. Approximately 17.9% of respondents identified poverty as a key factor, attributing it to the prevalent dependence on agriculture, day labour and small businesses, leading to an inability to erect protective fencing around ponds. The financial constraint leading to a lack of child supervision was also responsible for child drownings. This account from an observation of a participant helped elucidate this point:
A participant housewife by profession, approximately 45 years old (Debiganj subdistrict), completed her education up to lower secondary. She has been married for about 25 years and has a total of six children, though currently, only four are alive. Unfortunately, two of her sons passed away—one due to cancer and the other after falling into the nearby pond. The pond, located roughly 20 m from their house, sits amidst a courtyard where children frequently play. The child who drowned…was just 3 years old. The child wandered too close to the pond’s edge and fell in. Despite extensive search efforts, he was discovered submerged in mud, with his body inverted. Tragically, he could not breathe due to the mud and water, leading to his demise. The family had previously experienced another drowning incident before 2020, but they were fortunate to rescue the child promptly by administering first aid techniques applying pressure to expel water from his lungs. The father, a farmer by profession, supports the family by leasing land. However, the loss of two children has significantly strained their financial resources. The expenses incurred for the cancer treatment of one son were particularly burdensome and exacerbated their financial challenges. The participant expressed regret, acknowledging that if they could afford it, they would have fenced the pond to prevent such tragedies. However, due to financial constraints, this preventive measure was not implemented. The participant underscores the importance of vigilance in supervising children, lamenting the lapse in oversight that led to the tragic incident.
Discussion
Out of 18 004 households surveyed in 2 unions in Northern Bangladesh, 117 households were found to have experienced childhood drownings. Through face-to-face interviews and interactions with the households that faced child drownings, this study identified sociodemographic, locational and environmental factors related to child drowning deaths. The findings gleaned from these region-specific, local-level studies carry significant implications for the prevention of child drowning. In our studies, the number of drowning deaths in the baseline year of 2023 was significantly higher than in the previous 5 years. The decrease in reported deaths during those earlier years could be attributed to the longer recall period for drowning incidents. This could also happen if households had moved from that place to elsewhere. Overall, our results show that the households that faced drowning were comparatively from lower income groups, had lower rates of education and mostly engaged with agriculture and other domestic work. The average income of 117 respondents per month was BDT7831 compared with the national average per household of BDT32 422 in 2022.24 Although literacy rates in Bangladesh in 2022 were 74% (2), they were only 60% among the studied households. Overall, this region has high rates of poverty and it is behind in overall socioeconomic indicators, compared with other regions in the country17; special attention is warranted to understand the role of these contextual factors in child drowning prevention.
This study identifies specific risk factors associated with child drowning including male gender, children aged 2–5 years (74% fatal drowning), the monsoon season, time (9:00–14:00 hours: 83% of drownings) while parents are engaged with household chores, the abundance of a pond or ditch without fencing around the household, inability to swim and limited child supervision. These findings are comparable with previous studies in other regions in Bangladesh14 21 25 and other studies in India, Vietnam and China.26–28 These studies concurrently present that the child’s age–sex, types of supervision, child activities, the caregiver’s literacy level and the household income level have a significant impact on the risk of childhood drowning.
The number of drowning deaths was three times higher in males than in females in the region. This finding supplements previous studies where, among many others, Rahman et al29 and Wallis et al,30 provided case findings from southern regions of Bangladesh, which found that male drowning was double that of female drowning. For the extremely high cases of male drowning deaths, we found that limited child supervision and allowing male children to roam around outside and play alone with peers played a role. The average age for the 84 drowning deaths was 3.9 years and occurred between 9:00 and 14:00 hours. These are the periods that a household engages heavily with household chores. This age group at that time should have been in childcare, but there were none present in the region. Nonetheless, an earlier pilot intervention called ‘Saving of Lives from Drowning’ demonstrated the large-scale effectiveness and cost-effectiveness of crèches as primary prevention strategies for drowning deaths among children under 5 years old in Bangladesh.31 32 This project has since been scaled up, handed over to the government of Bangladesh, and is currently being implemented in selected districts by the Bangladesh Shishu Academy. The results of this study highlight the importance of conducting further research in other districts of Bangladesh, as well as implementation studies to assess the effectiveness of creches in these regions. Over 82% of drownings occurred between June and October, which is the monsoon season, when the level of water through seasonal rainfall and flooding rise to the household level in the region. Since there were no drowning prevention interventions in the region, enhanced activities considering seasonality and timing of drowning are needed. Approximately 72.6% of drowning deaths occurred in ponds, which were without fencing, allowing children to easily fall into the water.
A vast majority, approximately 73.5% of families, did not teach their child how to swim. For those who taught their kids to swim, the average age for the child to learn how to swim was 8.33 years. This implied that households should be encouraged to teach children about the dangers of water and how to swim from earlier ages. The ability to swim is a basic survival skill from water-related incidences.33 Rahman et al29 noted that considering health conditions, Bangladeshi children can start learning to swim from the age of six. However, a review of studies by Olivar and Moreno-Murcia suggested that for children of 10–14 years, knowing how to swim did not significantly reduce drownings compared with those who did not know how to swim.34 In our study, we found children in that age group died from drowning while enjoying swimming and bathing activities in tidal rivers in the region. The varying evidence regarding the effectiveness of swimming skills underscores the need for standardised survival swimming programmes to be integrated into school curricula, along with regular refresher training to ensure children retain these skills.
The findings have significant policy implications about organising fencing around walkways of ponds and ditches, launching water safety skills, conducting CPR training, active child supervision, awareness in the population about not allowing children to play alone and encouraging children learn to swim from 6 years of age. Along with individual and community level awareness and capacity building, to address the entire issue of child drowning prevention, it will require the involvement of local and central governments in developing appropriate prevention packages, considering localised contextual factors and including consultations with communities.
There are several limitations in this study that should be acknowledged. First, the reliance on participants’ recollections of drowning incidents over the past 6 years may introduce recall bias. Participants might not accurately remember the details of past events, potentially leading to incomplete or inaccurate data. Additionally, the findings are specific to two unions in Northern Bangladesh and may not be generalisable to other regions with different socioeconomic and environmental conditions. Future research should include a broader geographical area to enhance the generalisability of the findings. The study employed a cross-sectional design, which limited the ability to infer causality. Longitudinal studies are needed to better understand the causal relationships between sociodemographic factors and drowning incidents. The data on socioeconomic status, education levels and other demographic factors are self-reported, which can introduce social desirability bias. Participants might over-report socially desirable behaviours or underreport undesirable ones.
For future research, several directions are suggested. Longitudinal studies should be conducted to track changes in drowning incidents over time and to better understand the long-term effects of interventions aimed at reducing drowning rates. Specific interventions, such as swimming lessons for young children, community awareness programmes and construction of barriers around bodies of water, should be implemented and evaluated for their effectiveness in reducing drowning incidents. The study should be expanded to include multiple regions across Bangladesh, including urban and coastal areas, to identify region-specific risk factors and develop tailored prevention strategies. Detailed assessments of environmental factors, such as water quality, vegetation around water bodies, and the presence of safety features like fences or warning signs, should be conducted to aid in designing targeted environmental interventions. Incorporating qualitative research methods, such as in-depth interviews and focus groups, will help gain a deeper understanding of the cultural and behavioural factors influencing drowning incidents, providing valuable context to the quantitative findings. Analysing the impact of existing policies and regulations related to water safety and drowning prevention will help identify gaps in policy implementation and recommend improvements based on empirical evidence. Additionally, exploring the use of technology, such as mobile apps and GIS mapping, to monitor drowning incidents in real time and to enhance community-based surveillance systems, could be highly beneficial.
Incorporating these suggestions into future research will help address the current study’s limitations and provide a more comprehensive understanding of the factors contributing to child drowning in Bangladesh. This can ultimately lead to more effective prevention strategies and interventions.
Conclusions
The study addressed significant gaps in knowledge by going beyond the state-of-the-art public health management and by constructing new knowledge on child drownings in remote rural areas in Northern Bangladesh. The findings indicated that inadequate child supervision and allowing children to play unsupervised with peers, especially for 2–5 years old and between 9:00 and 14:00 hours, a time when households in rural Bangladesh were heavily engaged in chores, were factors associated with child drownings. Overall, the community’s risk perception was low, which implied the need for enhancing cognition of risk, considering seasonality and timing of drownings in the region. Since most drowning deaths occurred in ponds without fencing, action at the community level, with support from local government, should be considered to improve fencing around ponds. We found that there was a reluctance to teach swimming at the household level, with most children learning to swim at an average age of 8.33 years. However, the WHO has recommended providing basic swimming skills and water safety training to children starting from the age of six. Therefore, this study suggests that encouraging swimming skills for everyone from an earlier age can significantly enhance drowning prevention. Evidence from other regions of Bangladesh demonstrates the success of childcare centres, suggesting the need to implement and assess a similar intervention in Northern Bangladesh. The findings call for immediate action from drowning researchers and practitioners in Bangladesh to initiate multisectoral coordination, raise awareness and advocate for the government to expand the coverage of the integrated community-based centre for childcare, protection and swim-safe facilities project to reach more districts and subdistricts.
Data availability statement
Data are available in a public, open-access repository. Data are available on reasonable request. All data relevant to the study are included in the article or uploaded as supplementary information.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and the Research and Ethics Committee of Rabdan Academy approved human subject consultation for this study (#RAREC00040 dated 2 June 2023). Participants gave informed consent to participate in the study before taking part.
Acknowledgments
The time and effort of field data collection assistants and the research participants are graciously acknowledged. We sincerely appreciate Mr. Mohammad Abdul Matin for his efforts in typing and translating the questionnaire into Bengali. We are grateful to the two anonymous reviewers for their valuable contributions, which helped enhance the context and rigour of this research article.
References
Footnotes
Correction notice This article has been updated since it was first published online. The author Fahim Sufi was incorrectly listed as Sufi Sufi. In addition to this, affiliation 4 has been updated.
Contributors The authors’ responsibilities were as follows: EAlam conceptualised and designed the research; KSAH, HA, JA, EAlam, EAlvarez, ARMTI and MKI implemented the research; EAlam, HA, JA and SS conducted the analysis; EAlam wrote the draft manuscript. All authors contributed to the article and approved the submitted version. EAlam accepted full responsibility for the work and/or the conduct of the study, had access to the data and controlled the decision to publish. EAlam is the guarantor for the overall responsibilities including this statement.
Funding Funding for this research was received from the Rabdan Academy, Abu Dhabi, UAE. This work was financially supported by the Deanship of Scientific Research, Vice Presidency for Graduate Studies and Scientific Research at the King Faisal University, Saudi Arabia (Grant: KFU241347).
Map disclaimer The inclusion of any map (including the depiction of any boundaries therein), or of any geographical or locational reference, does not imply the expression of any opinion whatsoever on the part of BMJ concerning the legal status of any country, territory, jurisdiction or area or of its authorities. Any such expression remains solely that of the relevant source and is not endorsed by BMJ. Maps are provided without any warranty of any kind, either express or implied.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.