Abstract
Introduction
Diarrheal disease is a significant public health concern, especially among children under the age of five. Its burden is particularly high in low-income countries, including Ethiopia. However, there is a lack of recent, nationally representative studies that adequately identify individual and community-level factors to inform effective interventions. Hence, this study aimed to fill this gap.
Methods
This study utilized data from the national representative Health Equity Survey, conducted in all regional states and two city administrations of Ethiopia. A total of 8,747 mothers with children under five were selected using a two-stage stratified cluster sampling. Data were collected through face-to-face interviews. Descriptive statistics were computed, and mixed- effect multilevel logistic regression identified factors associated with diarrheal disease. Significance was set at p-value of < 0.05, with results reported as AORs and 95% CI. Four models were fitted, and the best-fit model was selected based on the lowest value of Deviance Information Criterion (DIC), Akaike’s Information Criterion (AIC), and Bayesian Information Criterion (BIC).
Results
The prevalence of diarrheal disease among children under five years old was 7.8% and varied across different regions of the country, ranging from 21% (Afar region) to 1.9% (Southwest region). In the multilevel logistic regression analysis, children aged 6–12 months (AOR = 4.5; 95% CI: 2.86, 6.97), children aged over 12 months (AOR = 3.5; 95% CI: 2.31, 5.45), no formal maternal education (AOR = 1.8; 95% CI: 0.99, 3.13), unimproved water sources (AOR = 1.6; 95% CI: 1.07, 2.52), and unimproved sanitation facilities (AOR = 1.7; 95% CI: 1.34, 2.09) had higher odds of developing diarrheal disease, while mothers aged 30 years and older (AOR = 0.8; 95% CI: 0.59, 1.02), living at altitude of 2501–3500 m and 3501–5500 m (AOR: 0.6, 95%CI: 0.43–0.77) and (AOR: 0.5, 95%CI: 0.42–0.60), had lower odds of diarrheal disease.
Conclusion
The prevalence of diarrheal disease among children under five in Ethiopia remains high with regional variations. Contributing factors include younger age particularly aged < 12 months, maternal lack of formal education, limited access to improved water sources and sanitation facilities. To effectively decrease this burden, targeted public health strategies are needed focusing on interventions that will address educational disparities and expanding access to clean water and adequate sanitation.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1186/s12889-026-28388-9.
Keywords: Diarrhea, region, disparity, factors, multilevel, Ethiopia
Introduction
Diarrheal disease remains one of the most prevalent and fatal public health challenges impacting children under five years of age, mainly within low- and middle-income countries [1–4]. The World Health Organization (WHO) defines diarrhea as the passage of three or more loose or liquid stools in a day or an increase in frequency beyond what is normal for the individual [5, 6]. It is caused by contaminated drinking water, inadequate sanitation and hygiene practices, poor nutritional habits, lack of rotavirus vaccine, and low maternal education [7]. Despite being a preventable illness, diarrhea can lead to serious health complications in children, including dehydration, malnutrition, growth impairment, and even death [8]. According to the WHO 2024 fact sheet and a global systematic review study report, diarrhea annually claims the lives of approximately 443,832 children under five, with nearly 1.7 billion cases of childhood diarrheal disease reported globally each year [5, 9]. Globally, diarrheal disease is responsible for 15% of all daily child mortality cases [10, 11]. Although simple and effective treatments are available, over 2,195 children die from diarrhea every day [10]. Furthermore, children under five in low- and middle-income countries suffer from diarrhea about three times annually [12].
A study conducted in sub-Saharan Africa revealed that diarrhea causes more than 90% of deaths in children under five years old in the region [13]. Specifically, infants who are exposed to contaminated environments by anthropogenic activities and have compromised immune systems are at a higher risk of dying from diarrhea during their later infancy [14]. Studies show that children under five in Africa experience at least five episodes of diarrhea each year, resulting in approximately 800,000 deaths [15]. Ethiopia is among the Sub-Saharan countries with the highest rates of diarrheal morbidity and mortality [16]. The 2016 Ethiopian Demographic Health Survey (EDHS) report indicated that 12% of children under five experienced episodes of diarrhea in the two weeks prior to the survey [17]. The report highlighted regional differences in diarrheal episodes among children under five, with rates ranging from 6% in Addis Ababa to over 20% in Gambella and Benishangul-Gumuz [17]. Communicable diseases like diarrhea associated with unhealthy environmental conditions and poor behavioral practices pose a significant public health challenge and adversely affect children’s health [18]. By 2030, Sustainable Development Goal 3 aims to eliminate preventable deaths of infants and children under five, including those caused by diarrhea. Achieving this goal, requires a thorough understanding of regional disparities, as well as individual and community-level factors. Numerous studies support this. for instance, one study conducted in North East Ethiopia indicated that diarrheal disease in children under five results from a complex interplay of multiple factors [19]. It’s occurrence is frequently associated with socio-demographic, environmental, and behavioral conditions [20]. Studies also note that the type of mother’s occupation, educational level, household wealth index, age of the child, access to clean water, improved sanitation facilities, and waste disposal systems are considered significant contributors to diarrheal diseases among children.
Despite the importance of studying individual and community-level factors to design effective interventions, there is a lack of recent, nationally representative studies that adequately identify individual and community-level factors to inform effective interventions. This also helped to see the new regional disparities, as there are many regions established in Ethiopia. Hence, this study aimed to fill this gap. In addition, this study employs advanced statistical methods. Therefore, by identifying regional disparities and predictors of diarrheal disease in Ethiopian children under five years, this study measure the prevalence, assess regional disparities, and identify individual and community-level predictors of childhood diarrhea in Ethiopia. Thus contribute to the existing body of knowledge and inform public health policies aimed at reducing the burden of this preventable disease. The findings are expected to underscore the need for enhanced access to clean water, improved sanitation facilities, and educational initiatives. These improvements are expected to lead to, better health outcomes for children and to empower mothers across Ethiopia.
Methods
Study setting
Ethiopia is a country located in the Horn of Africa. It is organized into 11 regional states and 2 city administrations. The regional states are Tigray, Afar, Amhara, Oromia, Somali, Southern Nations Nationalities and Peoples’ Region, Gambella, Benishangul Gumuz, Sidama, Southwestern Regions, and Harari. The two city administrations are Addis Ababa and Dire Dawa. The 2022/23 National Health Equity Survey (NHES) was conducted in all regional states and two city administrations of Ethiopia.
Study design and period
A national cross-sectional study was conducted from September 27, 2022, to December 20, 2022, in all regions of Ethiopia. However data collection in the Tigray region was from December 28, 2023, to February 14, 2024, due to security concerns.
Source population
All households within the EA having at least one under five children were the source population for this study.
Study population
Randomly selected mothers (caregivers) having at least one under five children were the study population.
Inclusion criteria
Mothers with at least one child under five years were included in this study. However, mothers with hearing problems or critical mental illnesses were excluded.
Sample size and sampling procedure
The sampling frame used for the 2022/23 National Health Equity Survey (NHES) is the Ethiopia Population and Housing Census, which was prepared in 2019 by the Ethiopia Central Statistical Services (CSS). It includes 147,602 EAs that cover the entire country, and each EA represents a manageable number of housing units. Two stages cluster sampling was used. EAs were considered as the primary sampling unit and households were considered as the secondary sampling unit. The sample was stratified into urban and rural areas, with the sampling strata. A fresh household listing was carried out in all the selected EAs in all regions and city administrations and the listed households were served as a sampling frame for the selection of households for the second stage. In the second stage of the selection, a fixed number of 22 eligible households per cluster were selected with a systematic random selection from the newly created household listing in ten regions. These fixed numbers of households from each EA were selected randomly through computer-generated methods. For this study, interviews were conducted with mothers who had at least one child under five years. Based on this process, 8,747 mothers were included.
Data collection and quality control
For this investigation, 82 data collectors and 16 supervisors were employed. To ensure data quality, an eight-day intensive training program was provided for both data collectors and supervisors. The training included instruction on interviewing techniques, field procedures, a detailed review of questionnaire content, guidance on administering both paper and electronic questionnaires, mock interviews between participants in the classroom, and practice interviews with real respondents in areas outside the survey sample. A questionnaire was developed and adapted from the 2016, Demographic Health Survey (EDHS) and the Expanded Programme for Immunization. It was modified to align with the study’s objectives for assessing population and health equity issues in Ethiopia, as outlined in Supplementary File 2. After that, the final questionnaire was translated into different local languages using both forward and backward translation by professionals to ensure consistency. Prior to the actual study, a pilot survey or pretest was conducted for two days before the actual data collection. Based on the pretest results, adjustments were made to improve the clarity, completeness, and consistency of the tool. Data collection was conducted through face-to-face interviews using the Census and Survey Processing System (CSPro) software. Regular supervision and spot-checking were conducted by survey coordinators, regional coordinators, and supervisors. Immediate feedback was given for each data collector accordingly. In addition, the office data manager monitored the accuracy of the data and took corrective action.
Data analysis and procedures
Prior to analysis in Stata 16, the dataset was checked for completeness. The data was weighted to ensure the sample accurately represents the country’s population, facilitating reliable statistical estimation. Both level-1 and level-2 weights were applied in the multilevel models, with level-2 weights accounted for the probability of selecting enumeration areas and level-1 weights reflected the probability of selecting households within those areas. All weights were derived using inverse probability weighting. Frequencies and percentages were computed as descriptive statistics and multicollinearity between the independent variables was checked using the Variance Inflation Factor (VIF).
Given the hierarchical structure of the data, traditional models may not accurately estimate the standard errors of effect sizes. The primary sampling units (enumeration areas) are stratified into urban and rural strata, with households nested within each stratum. Households in the same enumeration area tend to be more similar to one another than to those in other areas. The independence of observations and the assumption of equal variance across clusters are two fundamental tenets of traditional models that are violated by this. To overcome these problems and capture the heterogeneity between clusters, multilevel modeling was used. This method combines fixed effects, which account for features at the individual and community levels, with random effects, which account for changes among clusters.
Binary and multivariable multilevel logistic regression analyses were performed to determine the determinants at the individual and community levels. Variables with a p-value of less than 0.25 in the bi-variable analysis were included in the multivariable analysis. Four multilevel logistic regression models were developed: Model 1 (the empty or unconditional model) was analyzed without any covariates, estimated the random intercept at the cluster level and assessing the variation in diarrheal disease across communities; Model 2 included individual-level characteristics; Model 3 added community-level factors, and Model 4 combined both individual and community-level factors. These models account for the intra-class correlation of the outcome variable, ensuring valid standard error estimates. The final best-fit model was selected using Akaike’s Information Criterion (AIC), Bayesian Information Criterion (BIC), and Deviance Information Criterion (DIC). Variables with a p-value of < 0.05 were considered statistically significant, and the estimated parameters were presented and interpreted as AOR with 95% CI.
Study variables
Response variable
Diarrheal disease is defined as experiencing at least three loose or watery stools in a single day for at least one day during the two weeks leading up to the survey, based on reports from the child’s mothers [21]. Having diarrhea was classified as “yes” if the child had three or more loose stools per day in the previous two weeks before the survey and “no” if child had less than three loose stools per day.
Independent variables
The explanatory variables in this study were child age, age of mothers, mother’s educational status, marital status, occupation, family size, household wealth index/quintiles, number of children, residence, region, altitude, water source, use of water treatment, solid waste management, toilet facilities, and use of shared toilet.
Definitions of key terms
Improved water sources
Improved water sources are defined as those designed to effectively protect water from contamination, particularly from fecal matter [22]. These sources include piped water connections, public standpipes, boreholes, protected dug wells, protected springs, and rainwater collection systems [22].
Unimproved water sources
Defined as drinking water from an unprotected dug well or unprotected spring [23].
Improved toilet
Improved sanitation facilities are defined as those that hygienically separate human waste from human contact [24]. Improved sanitation includes flush or pour-flush to piped sewer systems, septic tank pit latrines, ventilated-improved pit latrines, or pit latrines with slab or composting toilets [24].
Unimproved toilet
Is defined as the use of pit latrines without a slab or platform, hanging toilets, or bucket latrines [3].
Wealth Index
The household wealth index reflects a family’s overall standard of living. It is derived from household asset data through principal components analysis. This index utilizes various asset types and follows the methodology established by the Ethiopian Demographic and Health Survey (EDHS) [17].
Results
Individual and community level characteristics of study participants
Of the 9,157 samples of households with children under five years, 8,747 were interviewed, resulting in response rate of 95.5%. The majority (73.6%) of the children were over 12 months old, while 12.9% were under 6 months. Based on the mean age of mothers, 51.7% (n = 4524) were under 30 years old, while 48.3% were aged 30 or above. Regarding the mothers’ educational status, over 47% of mothers had no formal education, and only 6.3% had education beyond the secondary level. Most (95.3%) of the mothers were currently married. The majority of mothers (71%) were housewives, and 56.6% of households had more than four family members. More than half of the participants fell into the lowest and second lowest categories of the wealth index. The majority (97.5%) of the participants had less than four children. In terms of residence, 73.2% of participants lived in rural areas. A majority (65.1%) resided at altitudes between 1,500 and 2,500 m, and 73.9% had access to improved water sources, although 95.4% did not treat their water at home. More than three-quarters (83.9%) of participants lacked proper solid waste management practices. Additionally, 68.8% used unimproved toilet facilities, and 27.1% used shared toilets (Table 1).
Table 1.
Demographic characteristics of study participants (n = 8747)
| Variables | Category | Diarrhea | |
|---|---|---|---|
| Yes; n (%) | No; n (%) | ||
| Child age in months | < 6 | 28(4.1) | 1105(13.7) |
| 6–12 | 114(16.6) | 1062(13.2) | |
| > 12 | 544(79.3) | 5894(73.1) | |
| Age of mothers | < 30 | 369(53.8) | 4155(51.5) |
| ≥ 30 | 317(46.2) | 3906(48.5) | |
| Mother’s educational status | No education | 348(50.7) | 3768(46.7) |
| Primary | 251(36.6) | 2803(34.8) | |
| Secondary | 65(9.5) | 959(11.9) | |
| Secondary and above | 22(3.2) | 532(6.6) | |
| Marital status | Currently Unmarried | 25(3.7) | 388(4.8) |
| Currently Married | 661(96.3) | 7673(95.2) | |
| Religion | Orthodox | 315(45.9) | 3831(47.5) |
| Protestant | 124(18.1) | 1479(18.4) | |
| Muslim | 241(35.1) | 2710(33.6) | |
| Catholic | 3(0.4) | 40(0.5)) | |
| Other | 4 (0.6) | 0(0) | |
| Occupation | Housewife | 444(64.7) | 5766(71.5) |
| Farmer | 187(27.2) | 1243(15.4) | |
| Other | 55(8.0) | 1051(13.0) | |
| Family size | ≤ 4 | 247(36.0) | 3550(44.0) |
| > 4 | 439(64.0) | 4510(55.9) | |
| Household wealth index/quintiles | Lowest | 232(33.8) | 2130(26.4) |
| Second | 193(28.1) | 2074(25.7) | |
| Middle | 136(19.9) | 1596(19.8) | |
| Fourth | 67(9.8) | 1231(15.3) | |
| Highest | 58(8.4) | 1030(12.8) | |
| Number of children | ≥ 4 | 3(0.5) | 212(2.6) |
| < 4 | 683(99.5) | 7848(97.4) | |
| Residence | Urban | 128(18.7) | 2212(27.4) |
| Rural | 558(81.3) | 5849(72.6) | |
| Region | Tigray | 10(1.5) | 76(0.9) |
| Afar | 18(2.6) | 65(0.8) | |
| Amhara | 286(41.7) | 2170(26.9) | |
| Oromiya | 146(21.3) | 3421(42.4) | |
| Ethiopian Somali | 16(2.4) | 255(3.2) | |
| Benishangul Gumuz | 3(0.5) | 78(0.9) | |
| SNNP | 128(18.7) | 997(12.4) | |
| Sidama | 37(5.4) | 268(3.3) | |
| Southwest | 7(0.9) | 360(4.5) | |
| Gambela | 3(0.5) | 31(0.4) | |
| Harari | 1(0.1) | 16(0.2) | |
| Addis Ababa | 27(3.9) | 292(3.6) | |
| Diredawa | 3(0.5) | 33(0.4) | |
| Altitude | < 1500 m | 102(14.9) | 1076(13.3) |
| 1500–2500 m | 481(70.1) | 5217(64.7) | |
| 2501–3500 m | 73(10.6) | 1387(17.2) | |
| 3501–5500 m | 30(4.4) | 382(4.7) | |
| Water source | Improved | 449(65.4) | 6013(74.6) |
| Unimproved | 237(34.6) | 2048(25.4) | |
| Water treatment | Yes | 25(3.7) | 378(4.7) |
| No | 661(96.3) | 7683(95.3) | |
| Solid waste | Proper | 101(14.7) | 1302(16.2) |
| Improper | 586(85.3) | 6759(83.8) | |
| Toilet | Improved | 155(22.6) | 2571(31.9) |
| Unimproved | 531(77.4) | 5490(68.1) | |
| Share toilet | Yes | 110(30.0) | 1509(26.9) |
| No | 257(69.9) | 4094(73.1) | |
Diarrheal disease and its regional disparities among children under five years
In this investigation, the two-week prevalence of diarrheal disease among children under five years old was found to be 7.8% (95% CI: 7.0%–8.8%). Of those who experienced diarrhea, 61.9% (95% CI: 56.0%–67.5%) sought treatment. The result indicates that 5.5% of urban children had diarrhea, compared to 8.7% in rural areas. The prevalence of diarrhea varied significantly across different regions. For instance, Afar reported the highest at 21.6%. Other regions like Tigray had a prevalence of 11.9%, Sidama 12%, and Amhara and SNNP showed rates of 11.7% and 11.4%, respectively. In contrast, Oromiya and Benishangul Gumuz had lower prevalence rates of diarrhea, at 4.1% and 3.9%, respectively. Altitude appeared to influence the prevalence as well: 8.7% of children living below 1,500 m had diarrhea, compared to 8.4% for those at altitudes between 1,500 and 2,500 m, with even lower rates at higher altitudes (Table 2; Fig. 1).
Table 2.
Diarrheal disease and its regional disparities among children under five years (n = 8747)
| Variables | Category | Diarrhea | P-value | |
|---|---|---|---|---|
| Yes; n (%) | No; n (%) | |||
| Having diarrhea in the last two weeks | 686(7.8) | 8061(92.2) | ||
| Sought treatment for diarrhea | 490(61.9) | 301(38.1) | ||
| Residence | Urban | 128 (5.5) | 2212 (94.5) | < 0.001 |
| Rural | 558 (8.7) | 5849 (91.3) | ||
| Region | Tigray | 10 (11.9) | 76 (88.0) | < 0.001 |
| Afar | 18 (21.6) | 65 (78.4) | ||
| Amhara | 286 (11.7) | 2170 (88.3) | ||
| Oromiya | 146 (4.1) | 3421 (95.9) | ||
| Ethiopian Somali | 16 (6.0) | 255 (93.9) | ||
| Benishangul Gumuz | 3 (3.9) | 78 (96.1) | ||
| SNNP | 128 (11.4) | 997 (88.6) | ||
| Sidama | 37 (12.0) | 268 (87.9) | ||
| Southwest | 7 (1.9) | 360 (98.1) | ||
| Gambela | 3 (9.8) | 31 (90.2) | ||
| Harari | 1 (5.1) | 16 (94.9) | ||
| Addis Ababa | 26 (8.4) | 292 (91.6) | ||
| Diredawa | 3 (8.8) | 33 (91.2) | ||
| Altitude | < 1500 m | 102 (8.7) | 1076 (91.3) | 0.027 |
| 1500–2500 m | 481 (8.4) | 5217 (91.6) | ||
| 2501–3500 m | 73 (5.0) | 1387 (95.0) | ||
| 3501–5500 m | 30 (7.3) | 382 (92.7) | ||
Fig. 1.
Regional disparities of diarrheal disease in Ethiopia
Multilevel simple logistic regression analysis of diarrhea among children under five
Table 3 depicts the results of multilevel simple logistic regression analysis. In this analysis, child age, age of mother, mother’s education, mother’s occupation, household wealth index, residence type, altitude, type of water source, household water treatment, and type of toilet were significantly associated (p < 0.05) with diarrheal disease. In addition, variables such as family size and household solid waste management practices, which had a p-value of less than 0.25, were included in the final models, as they are known predictors of diarrheal disease (Table 3).
Table 3.
Multilevel simple logistic regression analysis of diarrhea among children under five with household and community level factors (n = 8747)
| Characteristics | Diarrhea | COR 95% CI | P-value | |
|---|---|---|---|---|
| Yes% | No% | |||
| Child age in month | ||||
| < 6 | 2.5 | 97.5 | Ref. | |
| 6–12 | 9.7 | 90.3 | 4.3 [2.92, 6.32] | < 0.01*** |
| > 12 | 8.5 | 91.6 | 3.4 [2.32, 5.11] | < 0.01*** |
| Age of mother | ||||
| < 30 | 8.2 | 91.8 | Ref. | |
| ≥ 30 | 7.5 | 92.5 | 0.8 [0.67, 0.99] | 0.04** |
| Mother’s educational status | ||||
| No Education | 8.5 | 91.5 | 2.5 [1.49, 4.22] | < 0.01*** |
| Primary | 8.2 | 91.8 | 2.5 [1.83, 3.56] | < 0.01*** |
| Secondary | 6.4 | 93.6 | 1.8 [1.24, 2.71] | < 0.01*** |
| More than secondary | 3.9 | 96.1 | Ref. | |
| Mothers occupational | ||||
| Housewife | 7.2 | 92.9 | Ref. | |
| Farmer | 13.1 | 86.9 | 1.4 [0.91, 2.22] | 0.12 |
| Other | 4.9 | 95.0 | 0.6 [0.39, 0.87] | < 0.01*** |
| Family size | ||||
| <=4 | 6.5 | 93.5 | Ref. | |
| > 4 | 8.9 | 91.1 | 1.3 [0.91, 1.82] | 0.15 |
| Household wealth index | ||||
| Lowest | 9.8 | 90.2 | Ref. | |
| Second | 8.5 | 91.5 | 0.8 [0.55, 1.13] | 0.19 |
| Middle | 7.9 | 92.1 | 0.8 [0.59, 0.99] | 0.04** |
| Fourth | 5.2 | 94.8 | 0.4 [0.19, 0.97] | 0.04** |
| Highest | 5.3 | 94.7 | 0.4 [0.31, 0.59] | < 0.01*** |
| Residence | ||||
| Urban | 5.5 | 94.5 | Ref. | |
| Rural | 8.7 | 91.3 | 1.8 [1.28, 2.63] | < 0.01*** |
| Altitude | ||||
| < 1500 m | 8.7 | 91.3 | Ref. | |
| 1500–2500 m | 8.4 | 91.6 | 1.1 [0.76, 1.47] | 0.74 |
| 2501–3500 m | 5.0 | 95.0 | 0.6 [0.39, 0.79] | < 0.01*** |
| 3501–5500 m | 7.3 | 92.7 | 0.5 [0.42, 0.71] | < 0.01*** |
| Type of household water source | ||||
| Improved | 6.9 | 93.1 | Ref. | |
| Unimproved | 10.4 | 89.6 | 1.9 [1.15, 2.98] | 0.01** |
| Household water treatment practice | ||||
| Yes | 6.3 | 93.7 | Ref. | |
| No | 7.9 | 92.1 | 1.4 [1.06, 1.79] | 0.02** |
| Household solid waste management practice | ||||
| Proper | 7.2 | 92.8 | Ref. | |
| Improper | 7.9 | 92.0 | 1.4 [0.89, 2.30] | 0.14 |
| Type of toilet | ||||
| Improved | 5.7 | 94.3 | Ref. | |
| Unimproved | 8.8 | 91.2 | 2.0 [1.55, 2.62] | < 0.01*** |
*** p < 0.01, ** p < 0.05
Multilevel logistic regression models
Random effects analysis
Table 4 presents the factors associated with diarrheal disease among children less than five years old, analyzed through multilevel logistic regression models. Model 4 is considered the best fit due to its lowest AIC (4480.46), DIC (4456.46), and BIC (4565.38) values compared to other Models. Consequently, interpretations are based on the results from Model 4. The measures of variation indicate notable differences in diarrhea prevalence across communities. The variance estimates show that the null model (Model 1) has a variance of 0.37, which increases to 0.49 in Models 3 and 4, suggesting that adding predictors introduces more variability rather than reducing it. All models yielded statistically significant p-values (< 0.05), highlighting the relevance of the included predictors. The Interclass Correlation Coefficient (ICC) increased from 10.2% in Model 1 to 13% in Model 4, indicating that community-level factors contribute to the variability in diarrhea prevalence. The Proportional Change in Variance (PCV) values were negative across all models, ranging from − 2.7% to -32.1%. This indicates an increase in unexplained between-community variance after adding predictors, suggesting that the included variables do not fully capture the factors driving differences in diarrhea prevalence across communities. This may reflect unmeasured community-level factors or residual heterogeneity that remains important. The Median Odds Ratio (MOR) values ranged from 1.8 to 1.9, indicating moderate variation in diarrhea prevalence between communities and confirming that geographic or community-level differences continue to influence child health outcomes (Table 4).
Table 4.
Factors associated with under five diarrhea through multilevel logistic regression models (n = 8747) Weighted
| Characteristics | Model 1 (Null Model) | Model 2 | Model 3 | Model 4 | |
|---|---|---|---|---|---|
| AOR(95% CI) | |||||
| A. Individual-level characteristics | |||||
| Child age in month | |||||
| < 6 | Ref. | Ref. | |||
| 6–12 | 4.3 [2.93, 6.42]*** | 4.5 [2.86, 6.97]*** | |||
| > 12 | 3.6 [2.41, 5.28]*** | 3.5 [2.31, 5.45]*** | |||
| Age of mother | |||||
| < 30 | Ref. | Ref. | |||
| ≥ 30 | 0.8 [0.57, 0.99]** | 0.8 [0.59, 1.00]** | |||
| Mother’s educational status | |||||
| No education | 2.6 [1.46, 4.80]*** | 1.8 [1.01, 3.13] ** | |||
| Primary | 2.5 [1.79, 3.53]*** | 1.9 [1.41, 2.73]*** | |||
| Secondary | 1.7 [1.18, 2.55]*** | 1.6 [1.04, 2.45]** | |||
| Above Secondary | Ref. | Ref. | |||
| Mother’s occupation level | |||||
| Housewife | Ref. | Ref. | |||
| Farmer | 1.4 [0.93, 2.12] | 1.1 [0.75, 1.64] | |||
| Other | 0.7 [0.53, 1.03] | 0.9 [0.61, 1.23] | |||
| Family size | |||||
| <=4 | Ref. | Ref. | |||
| > 4 | 1.5 [0.99, 2.17] | 1.4 [0.98, 1.89] | |||
| B. Community-level characteristics | |||||
| Household wealth index | |||||
| Lowest | Ref | Ref | |||
| Second | 0.9 [0.71, 1.18] | 0.9 [0.69, 1.20] | |||
| Middle | 0.9 [0.81, 1.22] | 1.0 [0.82, 1.24] | |||
| Fourth | 0.7 [0.41, 1.12] | 0.7 [0.46, 1.14] | |||
| Highest | 0.8 [0.66, 0.86]*** | 0.9 [0.70, 1.26] | |||
| Type of residence | |||||
| Urban | Ref. | Ref. | |||
| Rural | 1.2 [0.89, 1.51] | 1.1 [0.86, 1.52] | |||
| Altitude | |||||
| < 1500 m | Ref. | Ref. | |||
| 1500–2500 m | 1.1 [0.89, 1.28] | 1.1 [0.89, 1.34] | |||
| 2501–3500 m | 0.6 [0.43, 0.75] *** | 0.6 [0.43, 0.77] *** | |||
| 3501–5500 m | 0.5 [0.42, 0.63]*** | 0.5 [0.42, 0.60]*** | |||
| Type of household water source | |||||
| Improved | Ref. | Ref. | |||
| Unimproved | 1.6 [1.08, 2.37]** | 1.6 [1.07, 2.52]** | |||
| Household water treatment practice | |||||
| Yes | Ref. | Ref. | |||
| No | 1.2 [0.82, 1.69] | 1.1 [0.81, 1.44] | |||
| Households solid waste management practice | |||||
| Proper | Ref. | Ref. | |||
| Improper | 1.3 [0.75, 2.19] | 1.3 [0.75, 2.09] | |||
| Type of toilet | |||||
| Improved | Ref. | Ref. | |||
| Unimproved | 1.6 [1.32, 1.87]*** | 1.7 [1.34, 2.09]*** | |||
| Measures of variation | |||||
| Variance (SE) | 0.37(0.18) | 0.38 (0.19) | 0.49 (0.25) | 0.49 (0.25) | |
| 95% CI | (0.14, 0.98) | (0.14, 1.02) | (0.18, 1.33) | (0.18, 1.35) | |
| ICC | 10.2% | 10.4% | 12.9% | 13% | |
| PCV | Reference | -2.7 | -30.8 | -32.1 | |
| MOR | 1.8 | 1.8 | 1.9 | 1.9 | |
| Model fit statistics | |||||
| AIC | 4655.79 | 4567.37 | 4548.06 | 4480.46 | |
| BIC | 4669.95 | 4623.97 | 4625.90 | 4565.38 | |
| DIC(-2log-likelihood) | 4651.79 | 4551.37 | 4526.06 | 4456.46 | |
Abbreviations: SE Standard error, ICC Intra-class correlation coefficient, PCV Proportional Change in Variance, MOR Median Odds Ratio, AIC Akaike’s information criterion, BIC Bayesian information criteria, DIC deviance information criterion. Model 1 (Empty model) was fitted without determinant variables; Model 2 is adjusted for Individual-level variables; Model 3 is adjusted for community-level variables; Model 4 is the final model adjusted for individual- and community-level variables
***p < 0.01,** p < 0.05,
Fixed effect analysis
In Model 4, various individual and community-level characteristics were assessed to determine their impact on the prevalence of diarrhea among children under five. Significant associations were found for child age, age of mother, maternal education, altitude, type of water sources, and type of toilet (p-value < 0.05). Children aged 6–12 months had 4.5 times higher odds of experiencing diarrhea (AOR = 4.5; 95% CI: 2.86, 6.97) compared to those under six months. Similarly, children over 12 months had 3.5 times higher odds (AOR = 3.5; 95% CI: 2.31, 5.45) compare to those under six months. Conversely, mothers aged 30 years and older had 20% lower odds of their children experiencing diarrhea (AOR = 0.8; 95% CI: 0.59, 1.00) than those aged less than 30 years. Maternal education significantly influenced child health outcomes: mothers with no education had 1.8 times higher odds (AOR = 1.8; 95% CI: 1.01, 3.13) of having children with diarrhea compared to those with higher education levels. Regarding altitude, mothers who lived at an altitude of 2501–3500 m and 3501–5500 m had 40% and 50% lower odds of their children experiencing diarrheal disease compared to those who lived in an altitude of < 1500 m (AOR: 0.6, 95% CI: 0.43–0.77) and (AOR: 0.5, 95% CI: 0.42–0.60), respectively. Community-level factors further impacted diarrhea prevalence. Households relying on unimproved water sources had 1.6 times higher odds of diarrhea (AOR = 1.6; 95% CI: 1.07, 2.52) compared to those with improved sources. Additionally, children in households with unimproved sanitation facilities had significantly higher odds of experiencing diarrhea (AOR = 1.7; 95% CI: 1.34, 2.09) than those who used improved sanitation facilities. (Table 4)
Discussion
This study aimed to explore regional disparities and predictors of diarrheal disease in Ethiopian children under five years. The two-week prevalence of diarrheal disease among Ethiopian children under five was 7.8%, with significant variation across regions and between rural and urban areas. Key individual-level predictors of diarrheal disease included child age (especially 6–12 months), low maternal education, and younger maternal age. At community level, factors such as unimproved water sources, poor sanitation, lower altitude, and rural residence were also associated with increased risk.
The 2-week prevalence of diarrheal disease among children under five years was 7.8%. This prevalence is lower than the EDHS 2016 finding and cross-sectional studies conducted among Ethiopian children under five years in 2019, 2021, 2023, and in 2025 [2, 3, 17, 25–30]. It could be due to the study timeline as also reflect the positive impact of national WASH interventions and improved healthcare access implemented in Ethiopia over the past few years. On the other hand, this recent prevalence of diarrhea is nearly consistent with similar studies conducted in Myanmar and India [31, 32]. The difference in prevalence of diarrhea in this recent study compared to similar studies conducted in different parts of Ethiopia may be attributed to several factors, method of data collection, sample size, and participant selection. Additionally, geographical variability across the country might also play a role, as certain regions may face greater environmental or infrastructural challenges. Socioeconomic factors, including income levels, education, and access to clean water and sanitation, further influence health outcomes. Cultural practices regarding hygiene and food safety, seasonal factors affecting disease transmission, and variations in health-seeking behavior may also contribute to these discrepancies.
The results of this study highlight critical individual and community-level characteristics associated with diarrheal disease among children under five years old in Ethiopia. Regarding individual factors influencing diarrheal disease, child age is notably impactful, with children aged 6–12 months showing 4.5 times higher odds of experiencing diarrhea compared to infants under six months. This finding is similar to the other study done in Ethiopia and the systematic review done across 25 countries [33, 34]. These studies highlight that the age of the child is a crucial factor. Infants, particularly those in their first year of age, are mostly vulnerable. This increased risk may be attributed to the transition to complementary feeding, which exposes children to new pathogens. Additionally, children over 12 months also showed elevated odds of 3.5 times. A study conducted in Awi Zone, Northwestern Ethiopia, along with another review of observational studies in the country, revealed that children over 12 months are 16 to 23 times more susceptible to diarrheal disease compared to infants [35, 36]. This might suggest that as children grow and become more mobile, their exposure to disease-causing agents increases. Maternal age appears to be a protective factor, with mothers aged 30 years and above having lower odds of their children experiencing diarrhea. A systematic review done across 25 African countries highlighted that higher maternal age is associated with a decreased prevalence of diarrhea among children under five years old [34]. This may reflect greater experience and knowledge in childcare among older mothers, as well as potentially better health-seeking behaviors. Maternal education also significantly influences child health outcomes, as mothers with no formal education had 1.8 times higher odds of having children with diarrhea compared to those with higher education levels. This finding is consistent with previous research conducted in Nigeria and Ethiopia [37, 38].
On the other hand, community factors also contribute significantly to the prevalence of diarrhea. Access to improved water sources is crucial; households relying on unimproved sources had 1.6 times higher odds of diarrhea. The other study done in Ethiopia also showed that those households who use an unprotected water source had 1.3 times higher odds of diarrheal disease [39].This emphasizes the role of water quality, and adequate access to clean water is consistently linked to lower rates of diarrheal diseases. The use of unimproved sanitation facilities further increases the risk, with odds of 1.7 times higher compared to those using improved toilets. A similar study done in Ethiopia reflected that those who use unimproved latrine have 1.3 higher odds of having diarrheal disease [39]. In addition, another systematic review highlighted that the presence of improved sanitation facilities is a significant protective factor against diarrheal diseases [34]. This highlights the urgent need for investments in water, sanitation, and hygiene (WASH) infrastructure to mitigate the risk of diarrhea.
Additionally, altitude appears to influence diarrhea prevalence, with those living below 1,500 m showing higher rates of diarrhea. This evidence is supported by multiple studies done in Gonder and Oromya [35, 40]. These studies found that lower-altitude areas had a significant association with under five diarrheal disease. The possible explanation can be due to higher population density, poor sanitation, and warmer temperature helping in pathogen survival. Additionally, this might be related to environmental factors affecting water quality and sanitation practices.
Furthermore, economic disparities are evident, with households in lower wealth quintiles exhibiting higher odds of diarrhea, as shown in model 3, despite it not being significant in the final model 4, indicating that inadequate financial resources can limit access to healthcare and sanitation. The prevalence of diarrhea was notably higher in rural areas (8.7%) compared to urban areas (5.5%), indicating that rural households encounter greater barriers to healthcare access and sanitation services. The regional disparities were also evident, with Afar reporting the highest prevalence at 21.6%, compared to lower rates in Oromiya (4.1%) and Benishangul Gumuz (3.9%). These differences may be attributed to variations in access to healthcare, sanitation facilities, and environmental factors that affect water supply services. Overall, these results underscore the multifaceted nature of diarrheal disease and the need for targeted public health strategies that consider both individual and community-level factors. Addressing educational gaps, improving access to clean water and sanitation, and enhancing healthcare access, particularly in rural areas, are essential steps toward reducing the prevalence of diarrhea among children under five in Ethiopia.
Conclusion
Despite the current prevalence of diarrheal disease among children under five years old being relatively low compared to national statistics, it remains a significant public health concern. The study identified several individual- and community-level factors that influence the occurrence of diarrhea. The study identified significant regional variations in the prevalence of diarrheal disease. The study’s findings highlight the critical need for targeted public health strategies that address educational disparities and improve access to clean water and sanitation. To effectively reduce diarrhea prevalence among children under five in Ethiopia, it is recommended that Water, Sanitation, and Hygiene (WASH) initiatives, along with maternal education, be integrated into national child health programs.
Strengths and limitations of the study
The strengths of this study are its national representativeness, the validated questionnaire, and the large sample size. However, since the data from the Tigray region were collected at a different time than in other regions, seasonal variation in diarrhea occurrence may have influenced the findings. Additionally, recall bias could be a problem in this study’s use of survey data, and the study’s cross-sectional design limits the ability to draw causal inferences. It recommended that future researchers to plan to include more variables to explain the variability of diarrhea in Ethiopia more.
Supplementary Information
Acknowledgements
We extend our heartfelt gratitude to the Ethiopian Ministry of Health for their financial support. We also appreciate the hard work of the data collectors, study supervisors, and participants, whose invaluable contributions made this research achievable.
Authors’ contributions
HA was responsible for the conceptualization, methodology, and writing of the original draft. AZ, AMG, GT, GM, FG, TS, HT, and Daniel Alemayehu participated in designing the methodology, data collection, coordination, and supervision. HA and AMG conducted the formal analysis. HA, AMG, TG, DA, WT, MS, SAB, and Tefera Tadele contributed to the interpretation, investigation, validation, and visualization of the data. Getachew Tollera and AZ contributed to project administration. HA, AMG, AT, TS, TD and Girum Taye contributed to carefully editing, revising, reviewing, and approving the final manuscript.
Funding
This study received financial support from the Ministry of Health, Ethiopia. Nonetheless, the funder had no involvement in the conception, study design, data collection, analysis, or in the decisions related to publication and data interpretation.
Data availability
The corresponding author makes the dataset for the study’s conclusion available upon reasonable request.
Declarations
Ethics approval and consent to participate
The study received approval from the scientific and ethical review board of the Ethiopian Public Health Institute, with reference number EPHI-6.13/906. The research methods adhered strictly to the principles outlined in the 1964 Helsinki Declaration and its subsequent amendments, along with other relevant ethical standards. Written informed consent was obtained from all participants before data collection began. Throughout the study, confidentiality and privacy for all participants were diligently protected.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Getachew Z, et al. Diarrheal disease and associated factors among children aged 6 to 59 months in Oda Bultum District, Eastern Ethiopia: a community-based cross-sectional study. BMC Infect Dis. 2024;24(1):303. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Alemu ZA, et al. Prevalence of Diarrhea Disease and Associated Factors Among Children Under 5 Years in Geshiyaro Project Implementation Sites in Ethiopia: A Cross-Sectional Study. Health Sci Rep. 2025;8(1):e70380. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Girmay AM, et al. Associations of WHO/UNICEF Joint Monitoring Program (JMP) Water, Sanitation and Hygiene (WASH) Service Ladder service levels and sociodemographic factors with diarrheal disease among children under 5 years in Bishoftu town, Ethiopia: a cross-sectional study. BMJ open. 2023;13(7):e071296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Ugboko HU et al. Childhood diarrheal diseases in developing countries. Heliyon, 2020;6(4):e03690. 10.1016/j.heliyon.2020.e03690 [DOI] [PMC free article] [PubMed]
- 5.Organization WH. Diarrheal disease. 7 March 2024.
- 6.Girmay AM, et al. Diarrheal disease and associated behavioral factors among food handlers in Addis Ababa, Ethiopia. AIMS Public Health. 2020;7(1):100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ngugi Mwaura J. Preventive Measures for Diarrhea in Communities Lacking Clean Water Supply: A Scientific Review.
- 8.Mokomane M, et al. The global problem of childhood diarrheal diseases: emerging strategies in prevention and management. Therapeutic Adv Infect disease. 2018;5(1):29–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Manetu WM, M’masi S, Recha CW. Diarrhea disease among children under 5 years of age: a global systematic review. 2021.
- 10.UNICEF, Committing to child survival: a promise renewed. 2015, eSocialSciences.
- 11.Hug L, Sharrow D, You D. Levels and trends in child mortality: report 2017. The World Bank; 2017.
- 12.Ghosh K, Chakraborty AS, Mog M. Prevalence of diarrhea among under five children in India and its contextual determinants: A geospatial analysis. Clin Epidemiol Global Health. 2021;12:100813. [Google Scholar]
- 13.Demissie GD, et al. Diarrhea and associated factors among under five children in sub-Saharan Africa: Evidence from demographic and health surveys of 34 sub-Saharan countries. PLoS ONE. 2021;16(9):e0257522. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Siddiqui FJ, Belayneh G, Bhutta ZA. Nutrition and diarrheal disease and enteric pathogens, in Nutrition and infectious diseases 2021, Springer. pp. 219–41.
- 15.Dairo MD, Ibrahim TF, Salawu AT. Prevalence and determinants of diarrhea among infants in selected primary health centres in Kaduna north local government area, Nigeria. Pan Afr Med J. 2017;28(1):151–151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Wolde D et al. The Burden of Diarrheal Diseases and Its Associated Factors among Under-Five Children in Welkite Town: A Community Based Cross-Sectional Study. Int J Public Health, 2022:239. [DOI] [PMC free article] [PubMed]
- 17.Csa I. Central Statistical Agency (CSA)[Ethiopia] and ICF. Ethiopia Demographic and Health Survey, Addis Ababa. Central Statistical Agency; 2016.
- 18.Yang K, et al. Global distribution of outbreaks of water-associated infectious diseases. PLoS Negl Trop Dis. 2012;6(2):e1483. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Workie GY, Akalu TY, Baraki AG. Environmental factors affecting childhood diarrheal disease among under-five children in Jamma district, South Wello zone, Northeast Ethiopia. BMC Infect Dis. 2019;19(1):804. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Daffe ML, et al. Water, sanitation, and hygiene access in Senegal and its impact on the occurrence of diarrhea in children under 5 years old. J Water Health. 2022;20(11):1654–67. [DOI] [PubMed] [Google Scholar]
- 21.Bauhofer AFL, et al. Intestinal protozoa in hospitalized under-five children with diarrhoea in Nampula–a cross-sectional analysis in a low-income setting in northern Mozambique. BMC Infect Dis. 2021;21(1):1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.WHO. and UNICEF, Core questions on water, sanitation and hygiene for household surveys–2018 update. JMP Water Supply Sanitation Hygiene, 2018:1–24.
- 23.Girmay AM, et al. Access to water, sanitation and hygiene (WASH) services and drinking water contamination risk levels in households of Bishoftu Town, Ethiopia: A cross-sectional study. Health Sci Rep. 2023;6(11):e1662. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Unicef, Organization WH. Progress on household drinking water, sanitation and hygiene 2000–2017. Special Focus on Inequalities; World Health Organization: Geneva, Switzerland,; 2019.
- 25.Dagnew AB, et al. Prevalence of diarrhea and associated factors among under-five children in Bahir Dar city, Northwest Ethiopia, 2016: a cross-sectional study. BMC Infect Dis. 2019;19(1):1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Feleke DG et al. Two-week prevalence of acute diarrhea and associated factors among under five years’ children in Simada Woreda, South Gondar Zone, Northwest Ethiopia, 2021: a multi-central community based cross-sectional study. Pan Afr Med J, 2022. 42. [DOI] [PMC free article] [PubMed]
- 27.Tafere Y et al. Diarrheal diseases in under-five children and associated factors among Farta district rural community, Amhara regional state, north central Ethiopia: a comparative cross-sectional study. Journal of environmental and public health, 2020. 2020. [DOI] [PMC free article] [PubMed]
- 28.Motuma A, et al. Prevalence and risk factors of diarrhea among under-five children in Ethiopia: A systematic review and meta-analysis. BMC Public Health. 2025;25(1):1815. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Natnael T, Lingerew M, Adane M. Prevalence of acute diarrhea and associated factors among children under five in semi-urban areas of northeastern Ethiopia. BMC Pediatr. 2021;21(1):290. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Ayalew AM, et al. Assessment of Diarrhea and Its Associated Factors in Under-Five Children among Open Defecation and Open Defecation‐Free Rural Settings of Dangla District, Northwest Ethiopia. J Environ public health. 2018;2018(1):4271915. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Soe TK, et al. Hygiene practice and diarrhea prevalence among underfive children in Myanmar: a cross-sectional study. BMC Pediatr. 2024;24(1):675. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Ghosh K, Chakraborty AS, SenGupta S. Identifying spatial clustering of diarrhoea among children under 5 years across 707 districts in India: a cross sectional study. BMC Pediatr. 2023;23(1):272. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Atnafu A, et al. Geographical disparities and determinants of childhood diarrheal illness in Ethiopia: further analysis of 2016 Ethiopian Demographic and Health Survey. Trop Med health. 2020;48(1):64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Azanaw J, et al. Determinants of diarrhoeal diseases among under-five children in Africa (2013–2023): a comprehensive systematic review highlighting geographic variances, socioeconomic influences, and environmental factors. BMC Public Health. 2024;24(1):2399. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Adam Birhan N, et al. Prevalence of diarrhea and its associated factors among children under five years in Awi Zone, Northwest Ethiopia. BMC Pediatr. 2024;24(1):701. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Sahiledengle B, et al. Burden of childhood diarrhea and its associated factors in Ethiopia: A review of observational studies. Int J Public Health. 2024;69:1606399. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Desmennu AT, et al. Maternal education and diarrhea among children aged 0–24 months in Nigeria. Afr J Reprod Health. 2017;21(3):27–36. [DOI] [PubMed] [Google Scholar]
- 38.Melese B, et al. Prevalence of diarrheal diseases and associated factors among under-five children in Dale District, Sidama zone, Southern Ethiopia: a cross-sectional study. BMC Public Health. 2019;19(1):1235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Fenta SM, Nigussie TZ. Factors associated with childhood diarrheal in Ethiopia; a multilevel analysis. Archives Public Health. 2021;79(1):123. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Gelan E et al. Spatial disparities and associated factors of Under-five diarrhea disease in Ilubabor Zone, Oromia Regional State, Ethiopia. 2022.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The corresponding author makes the dataset for the study’s conclusion available upon reasonable request.