This was a community-based cross-sectional study to assess the determinants of diarrhoeal diseases in children under five years in the Ndop Health District.

Ndop Health District is one of the 21 health districts of the North West Region of Cameroon. It is bordered to the north by Bangoren Health District, to the south by Tubah Health District, to the east by Kumbo East Health District, and to the west by Fundong and Oku health districts. The district is made up of 15 health areas: one urban health area, two semi-urban areas, and 12 rural health areas. The district is predominantly rural and has a population of 207,781 in 2025 according to the Center for Health Information (CIS). It is located about 60 km on the highway from Bamenda to Kumbo. It is linked to Kumbo by National Road Number 11 (RN-11). Fourteen out of fifteen health areas are accessible by road, and one health area, Mbissa, is accessible by water (Island) across the Bamenjim dam. It covers a total surface area of 1126 km². It has limited access to potable water and sanitation infrastructure. Ndop Health District covers the entire Ngoketunjia division, with Ndop, the headquarters, located in Bamunka. Ndop Health District is made up of 13 villages. Ndop is the administrative division of Ngoketunjia with three subdivisions, namely Ndop Central, Babessi, and Balikumbat Subdivisions, corresponding to three council areas: Ndop Central, Babessi, and Balikumbat councils, respectively. The area is lowland, surrounded by hills with streams that empty into the Bamenjim dam. The Ndop plain makes drainage difficult, and this causes slums around living homes and farmland, which facilitates waterborne diseases. Field data was collected from March to June 2025.

The study population was mothers/caregivers of children under five years living in households in Ndop health district. Mothers/caregivers of children under the age of 5 years from the selected households were included in the study. A household was considered as people living in the same dwelling and sharing common meals. To be eligible to participate in the study, mothers/caregivers in selected households must have children under 5 years. In households with more than two children under five years, a random child selection within eligible households was done. We excluded eligible mothers/caregivers who had mental cognitive problems or who were very sick and could not respond to the questions.

The variables included history of diarrhoea, water treatment, breastfeeding during diarrhoea episodes, purchase of food from roadside vendors, vaccination status, and drinking water sources, which included wells, springs, rainwater, piped water, and streams. Other variables were child’s stool disposal. Diarrhoeal occurrence was the outcome variable, which was defined as having diarrhoeal disease in the four weeks prior to the survey. Independent variables included non-water treatment, purchase of food from roadside vendors, non-vaccination with the Rotavirus vaccine, unimproved drinking water sources, water storage containers, stool disposal in open bushes, and poor garbage disposal.

Data collection tool:Data about the child were collected from their mothers/caregivers through questionnaires with the primary caregivers of the children. The survey instrument was developed based on established guidelines and previous research studies [10]-[12].

Data collection:Sociodemographic characteristics of the mother/caregivers and child included sex of child, age, number of siblings, dietary habits, vaccination status, and the occurrence of diarrhea episodes (defined as ≥3 loose stools within 24 hours). Mother’s information included age of mother, religion, location of family, mother’s education, number of children, marital status, age of mother at first pregnancy, relationship with the child, profession, and level of income. Data about prevalence of diarrhoea included the occurrence of diarrhoea and number of diarrhoea episodes and was categorized as follows: diarrhoea presence and diarrhoea absence. The practices used by mothers included sources of drinking water (pipe borne, well, stream), treatment given to water before drinking, utensils used in storing water, and treatment given to child at home with diarrhoea (ORS, bananas, antibiotics, nothing, etc.). The tool was developed in English, translated into the local language to respondents (where necessary), and pre-tested among 10 mothers/caregivers in a neighboring health area that was not part of the study sample. Feedback from the pre-test was used to refine and clarify ambiguous or culturally inappropriate items to ensure content validity and reliability. Trained field data collectors (Community Health Workers), fluent in both English and local dialects, administered the questionnaires through face-to-face interviews during household visits. Interviewers received training on the study objectives, ethical considerations, interviewing techniques, and procedures for obtaining informed consent. During data collection, the team visited each selected household and conducted interviews with eligible mothers/caregivers after obtaining their written informed consent. Data were collected using paper-based forms or phones, which were downloaded as Excel and checked daily by field supervisors for completeness and accuracy. Data were downloaded from the phone and merged as an Excel database and exported to SPSS version 26.0. The data have been stored in a password-protected computer and further data cleaning was done using the same software.

The sample size: The sample size was calculated using Cochran’s formula [20].

Sample Size(n) = Z²×P(1−P)/e²

The prevalence of diarrhoea of 29.25% from a study carried out by Mabvouna et al. [11] in Douala, Cameroon was used. Therefore, prevalence (P) of 29.25%, with a confidence level of 95% and a maximum error margin (e) of 5%. The standard value of Z at a 95% confidence interval is 1.960.

Therefore, applying the Cochran’s formula gives;

(1.960)² * 0.2925(1 − 0.2925/(0.05)) = 318

Assuming a response rate of 80%, the minimum sample size for this study was estimated to be 398 to compensate for non-response. However, a total of 507 mothers/caregivers and their children were recruited from nine health areas for this study.

This study made use of a multistage technique to select health areas, health facilities, communities, households, and mothers/caregivers to participate in the study. In the first stage, health areas were classified into three categories: Urban, Semi Urban, and Rural health areas, which correspond to the classification by the district health service. The one Urban health area and the two Semi Urban health areas were included in the study, and six Rural health areas were randomly selected, giving a total of nine health areas that were considered for the study. The study population was distributed to the various health areas according to the weights of the population size. The second stage was the selection of communities. At the third stage of sampling, households within each of the 36 randomly selected communities were selected using a modified Lot Quality Assurance Sampling (LQAS) approach, maintaining the pre-defined lots-of-four sequence per health area. In each selected community, a rapid household listing was conducted with the assistance of the community health worker and village authorities to identify all households with at least one eligible mother/caregiver (defined as a woman residing in the household with a child in the target age group). Households were assigned unique identification numbers to constitute the sampling frame.

The total number of mothers/caregivers to be interviewed was allocated proportionally by area of residence as follows; 396 respondents in rural health areas, 100 respondents in Semi-Urban health areas, and 11 respondents in urban health areas. Within each selected community, the number of households to be surveyed was determined according to the LQAS lot size assigned to that health area. To minimize selection bias, systematic random sampling was applied within each community. The total number of eligible households in the community was divided by the required sample size to determine the sampling interval (4 households). A random starting number between 1 and 4 was selected using a random number table. Households were then selected at fixed intervals (every 4th household) until the required sample size was achieved. The household of the community/village head was used only as a geographic reference point for mapping and orientation and was not automatically included in the sample unless selected through the random procedure. If a selected household did not contain an eligible mother/caregiver, the next household in the predefined sampling sequence was approached. If no eligible respondent was identified after visiting two consecutive households, the team continued following the systematic interval sequence to preserve randomness. In cases of absence, the household was revisited up to two additional times on the same day at different hours before being classified as a non-response. Replacement was done only after confirming non-eligibility or non-response following the revisit protocol. A total of 507 mothers/caregivers were targeted for interview (396 rural, 100 semi-urban, 11 urban). Of these, 507 completed the interview, resulting in an overall response rate of 100%. Non-response was primarily due to absence at the time of visit or refusal to participate.

Data analysis:Data were entered and analyzed using SPSS for Windows version 26.0 statistical software. In order to ensure high-quality data, data collectors followed a restricted approach during the data collection phase and all entries were supervised for completeness and consistency. The dependent variable analyzed was the history of diarrhoeal disease in children, while the independent variables included demographic characteristics and Water, Hygiene, and Sanitation (WASH) variables. Descriptive statistics were used to present frequencies and percentages. We calculated the median and range of quantitative variables such as age. Chi-square tests assessed associations; bivariate logistic regression analyses were conducted to identify factors associated with diarrhoeal diseases. All measures of association were presented as odds ratios. Statistical significance was set at p < 0.05 at a confidence interval of 95% (CI).

Approval to carry out this study was obtained from the Institutional Review Board (IRB) of The University of Bamenda (Ref No 2022/0709H/Uba/IRB). Administrative clearance was obtained from the Regional Delegation of Public Health for the North West Region (Ref No 300/ATT/NWR/RDPH/BRIGAD). Written informed consent was obtained from all the quarter heads, and mothers/caregivers above 18 years of age signed a consent form before any data collection procedure started. Mothers/caregivers answered questions for children under five years about their health status. For eligible mothers who were less than 18 years old at the time of the survey, informed assent was given, and an assent form was provided for them to sign before participating in the study. The consent and assent forms were read and explained to participants who could not read or write. All data are stored in a computer protected with a password.

A total of 507 mothers/caregivers of children under five years were interviewed. For the children, 43% were aged 13 - 36 months and 22.9% were below 6 months. Slightly more than half (52.3%) were male, and 71.6% were not yet attending school as seen in Table 1. Most mothers/caregivers were female (82.6%) and young (mean age: 29.9 ± 8.6 years). Over half (61.9%) had their first child at or before 20 years of age. A majority (69.2%) were Christians, and almost half (45.2%) had only primary education. Most were unskilled workers (41%) or unemployed (21.5%), and 86% earned below 45,000 CFA per month. Over half (56.8%) were single as seen in Table 2.

Table 1. Sociodemographic characteristics of children under five years and their mothers/caregivers in Ndop Health District, Cameroon, recruited during a community-based survey from March to June 2025 (N = 507).

Table 2. Sociodemographic characteristics of mothers/caregivers of children under five years in Ndop Health District, Cameroon, recruited during a community-based survey from March to June 2025 (N = 507).

In this study, 334 out of 507 children (65.8%) had diarrhoea in the past one month. The prevalence was highest among children aged 6 - 12 months (75.7%), followed by 13 - 59 months (67.3%), and lowest in <6 months (56.1%). Males (67.5%) had slightly higher rates than females (64%). Children living in rural areas had a significantly higher prevalence (71.5%) compared to urban (43%) and semi-urban (27.3%) areas (p < 0.001), as seen in Table 3. The findings also reveal that more than half of mothers reported treating water at home (56.2%), while a significant proportion (43.3%) still consumed untreated water. Over half (54.8%) continued breastfeeding during diarrhoeal episodes, while almost half (45.2%) did not. Risky behaviors were prevalent, including frequent purchase of food from roadside vendors (65.5%), inadequate handwashing with soap (only 41.6% practiced it), and poor toilet hygiene, with just 20.5% cleaning toilets very often and 37.4% not cleaning them at all. Although 62.5% of children were vaccinated against rotavirus, over a third were not. Feeding during diarrhoea was mostly reduced (57%). Unsafe water sources, like wells (29.1%) and streams (13.8%), were used, and over half of households used unimproved toilets (58.1%) and disposed of children’s stool in open spaces (61.7%). Waste disposal was mainly in nearby bushes (66.5%). Storage of water in utensils with lids was practiced by 92.3%, as seen in Tables 4-6.

Table 3. Prevalence of diarrhoea in the past one month among children under five years in Ndop Health District, Cameroon, by age group, sex, and place of residence, March-June 2025 (N = 507).

Table 4. Mothers’ practices in the prevention and management of diarrhoeal diseases among children under five years in the Ndop Health District, Cameroon, March-June 2025 (N = 507).

Table 5. Association between the occurrence of diarrhoea and the age, sex, and place of residence of children under five years in the Ndop Health District, Cameroon, March–June 2025 (N = 507).

Table 6. Predictors of diarrhoea occurrence among children under five years in Ndop Health District, Cameroon, March–June 2025 (N = 507).

There was a statistically significant association between diarrhoea in the past one month and child’s age (χ² = 11.31, p = 0.004) as well as place of residence (χ² = 29.01, p < 0.001). No significant association was observed between diarrhoea and sex (χ² = 1.44, p = 0.23). The prevalence of diarrhoea was 56.1% among children aged < 6 months and 75.7% among those aged 6 - 12 months. Additionally, 71.5% of children residing in rural areas experienced diarrhoea in the preceding month compared to 43.0% of children in urban areas.

Children residing in rural areas had significantly higher odds of diarrhoea compared to those in urban areas (COR = 3.28; 95% CI: 2.03 - 5.29; p < 0.001). Consumption of food from roadside vendors was associated with increased odds of diarrhoea (COR = 1.60; 95% CI: 1.23 - 2.08; p < 0.001). Children from households that did not practice regular handwashing with soap had higher odds of diarrhoea (COR = 2.41; 95% CI: 1.75 - 3.32; p < 0.001). Non-vaccination against rotavirus was strongly associated with diarrhoea (COR = 2.76; 95% CI: 1.98 - 3.85; p < 0.001). Infrequent toilet cleaning was associated with increased odds of diarrhoea (COR = 1.69; 95% CI: 1.15 - 2.49; p = 0.008). Unsafe water storage without lids increased the odds of diarrhoea (COR = 1.54; 95% CI: 1.08 - 2.21; p = 0.018). Disposal of children’s stool in open spaces was significantly associated with diarrhoea (COR = 1.82; 95% CI: 1.34 - 2.48; p < 0.001). Garbage disposal in nearby bushes was also associated with increased odds (COR = 1.71; 95% CI: 1.22 - 2.40; p = 0.002). Breastfeeding during diarrhoeal episodes was not significantly associated with diarrhoea occurrence (COR = 0.93; 95% CI: 0.68 - 1.27; p = 0.64).

After adjustment for potential confounders, Rural residence remained independently associated with diarrhoea (AOR = 2.54; 95% CI: 1.49 - 4.33; p = 0.001). Non-vaccination against rotavirus remained a strong independent predictor (AOR = 2.21; 95% CI: 1.52 - 3.22; p < 0.001). Non-practice of handwashing with soap remained significant (AOR = 1.98; 95% CI: 1.36 - 2.89; p < 0.001). Open disposal of children’s stool remained independently associated with diarrhoea (AOR = 1.57; 95% CI: 1.11 - 2.23; p = 0.011). Consumption of roadside vendor food showed attenuated but still significant association (AOR = 1.42; 95% CI: 1.05 - 1.94; p = 0.023). Garbage disposal in nearby bushes remained marginally significant (AOR = 1.46; 95% CI: 1.01 - 2.12; p = 0.044). However, unsafe water storage without lids (AOR = 1.29; 95% CI: 0.88 - 1.89; p = 0.19) and infrequent toilet cleaning (AOR = 1.33; 95% CI: 0.89 - 1.99; p = 0.16) were no longer statistically significant after adjustment. Breastfeeding during diarrhoeal episodes remained non-significant (AOR = 0.95; 95% CI: 0.69 - 1.31; p = 0.75).