Prevalence of acute diarrhea and associated factors among children under five in semi-urban areas of northeastern Ethiopia

Background Diarrheal disease is still one of the most common causes of mortality and morbidity in children under five in developing countries, including Ethiopia. Lack of specific data on the prevalence of acute diarrhea and associated factors among under-five children in the semi-urban areas of Gelsha, found in northeastern Ethiopia’s South Wollo zone, remains a major gap. Therefore, this study was designed to provide data that is important for proper planning of intervention measures to reduce the problem in this area. Methods A community-based cross-sectional study was conducted among 340 systematically selected children under five in semi-urban areas of Gelsha from January to March 2019. The data was collected using a structured questionnaire and an observational checklist. Bivariable (crude odds ratio [COR]) and multivariable analysis (adjusted odds ratio [AOR]) were employed using binary logistic regression model with 95% CI (confidence interval). Variables with a p-value < 0.05 from the multivariable analysis were declared as factors significantly associated with acute diarrhea. Result The prevalence of acute diarrhea among children under five in the study area was 11% (95%CI: 7.8–14.3%). About two-thirds (63.60%) of study participants used water from improved sources. About half (54.90%) of study participants practiced poor handwashing and 45.10% practiced good handwashing. We found that factors significantly associated with acute diarrhea were a child’s age of 12–23 months (AOR = 4.68, 95% CI: 1.45–1.50), the presence of two or more under-five children in the house (AOR = 2.84, 95% CI: 1.19–6.81), unimproved water sources (AOR = 2.97, 95% CI: 1.28–6.87) and presence of feces around the pit hole/slab/floor of the latrine (AOR = 3.34, 95% CI: 1.34–8.31). Conclusion The prevalence of acute diarrhea among children under five was relatively high. To reduce the problem, various prevention strategies are essential, such as the provision of health education to mothers/caregivers that focuses on keeping sanitation facilities clean and child care, and construction of improved water sources. Furthermore, implementing a strong health extension program, advocating an open defecation-free environment, and practicing a community-led total sanitation and hygiene approach might be helpful to sustainably reduce childhood diarrhea. Supplementary Information The online version contains supplementary material available at 10.1186/s12887-021-02762-5.


Background
Acute diarrhea can be caused by different types of pathogens, among them the major five enteric pathogens Rotavirus (67.6%), Adenovirus (41.5%), enterotoxigenic E. coli (40.7%), Salmonella (38.4%), and Giardia (37.0%) [1]. Globally, acute diarrhea is the leading cause of morbidity and mortality among under-five children. In 2015, acute diarrhea and pneumonia accounted for almost one in every four deaths among under-five children worldwide [2]. During 2013, out of 6.3 million children who died before they reached their fifth birthday worldwide, about half (3.2 million) died from infectious diseases, with acute diarrhea killing more than 500,000 of them. It is estimated that in 2030, infectious diseases will cause 4.4 million deaths annually among under-five children, and 60% of those deaths will occur in sub-Saharan Africa [3].
A study that was conducted in the Eastern African countries of Burundi, Rwanda, and Tanzania among under-five children revealed that the prevalence of acute diarrhea was 24.80, 13.1 and 13.91%, respectively, and was associated with water-and sanitationrelated factors [4]. A recent meta-analysis in three East African countries, Ethiopia, Kenya, and Somalia, from 2012 to 2017 also revealed that the average prevalence of acute diarrhea among under-five children was 27% [5]. In another study conducted in Senegal, the prevalence of acute diarrhea among under-five children was 26%, with the highest acute diarrhea prevalence in the peri-central 44.8% and urban central zones 36.3% [6].
In Ethiopia, acute diarrhea is one of the major contributors to the morbidity of under-five children [7]. The recent Ethiopian DHS 2016 report also showed a 12% prevalence of acute diarrhea at the national level [8]. According to this report, the prevalence of acute diarrhea increases from 8% among children younger than 6 months, an age when many infants are exclusively breastfeed, to 23% among those between 6 and 11 months, the age children start consuming complementary foods and other liquids. Prevalence remains high (18%) at age 12-23 months since that is the time when children start walking and coming in contact with contaminants from the environment.
According to a study conducted in Wolita Soddo town in Southern Ethiopia, the prevalence of acute diarrhea was 11% [9]. In another study conducted in rural kebeles of Adama District, the prevalence of acute diarrhea was 14.7% [10]. In a study conducted in a rural area of North Gondar, the prevalence of acute diarrhea was 22.1% among under-five children [11]. Despite several studies being conducted on the prevalence of acute diarrhea among under-five children in different areas of Ethiopia [9][10][11][12], there was a scarcity of evidence from semi-urban areas; one study in Enderta district in northern Ethiopia found the prevalence of acute diarrhea to be high (35.6%) in a semi-urban area [13].
The information obtained from Gelsha Health Center indicates that acute diarrhea is the major public health problem locally and is among the top 10 causes of morbidity among under-five children in that area. However, there is a lack of research on the magnitude of and factors associated with the problem among this age group in Gelsha. Therefore, this study aims to address these gaps by determining the prevalence and associated factors of acute diarrhea among under-five children in semi-urban areas of Gelsha, northeastern Ethiopia.

Study area description
This study was conducted in semi-urban kebeles (Ethiopia's smallest administrative unit, each consisting of about 5000 people) of Gelsha, a town found in Dessie Zuria district. Semi-urban areas were defined as containing a population of 2000 or more, where roughly 20% of the total population engage in an occupation other than agriculture [13].
Gelsha is located 25 km from the city of Dessie. Dessie is the main town of South Wollo Zone, in the Amhara Region, 520 km north of Addis Ababa, Ethiopia. Gelsha has a total population of 8192, of which 4175 are men and 4017 women. Topographically, the town is mountainous (47%), plain (40%) and valley (13%). Administratively, the town is subdivided into 4 kebeles, of which one is semi-urban and the others rural. Agriculture is the major economic activity in the area. The town has one health center and one health post. Based on the data organized by health extension workers, the total number of under-five children in the town at the time of data collection was 770.

Study design and source populations
A community-based cross-sectional study was conducted from January to March 2019. The source population was all under-five children paired with their mothers/caregivers in semi-urban areas of Gelsha from which the study population was systematically selected. Mothers/caregivers and their under-five children who resided in semi-urban areas of Gelsha during the study period were included in the study. Under-five children who had been away from the study area during the 2 weeks prior to data collection but were available during data collection were excluded, because a child might develop acute diarrhea due to conditions in the other area where he/she had been before data collection.

Sample size determination and sampling procedure
The sample size for this study was calculated using a single population proportion formula: [Z-/2 ] 2 * P[1-P]/ d 2 [14] with the assumptions of 13.7% prevalence (P) of acute diarrhea among children under five in Amhara region, northern Ethiopia [8], 3% margin of error (d), Z 1 -/2 at 95% CI (confidence interval) = 1.96, and since source population of 770 under-five children in Gelsha was less than 10,000, sample size correction formula was used. We considered a 10% non-response rate and the final adequate sample size of the study was set at 340.
First, a preliminary assessment was conducted in order to identify households with at least one under-five child and the K interval was calculated. Then, the first household with an under-five child was selected by lottery method. Next, households with under-five children were selected using systematic sampling technique of every 2nd interval. In households with two or more children under five, one child was randomly selected and included into the study. Also, in a case where more than one caregiver of under-five children lived in the same household, only one caregiver was selected by a lottery method for the study.

Operational definitions Acute diarrhea
Diarrhea among children under five was identified by asking the participants' mothers/caregivers questions based on WHO-defined signs and symptoms of diarrhea [15] that had occurred during the previous 2 weeks. The WHO protocol [15] does not specify the recall period and the type of diarrhea. Because our study focused on acute diarrhea, we adopted a two-week recall period as specified in the World Gastroenterology Organization global guidelines for acute diarrhea surveys [16].

Good handwashing practice at critical times
Mothers/caregivers who reported washing their hands using water and soap at three or more of the five critical times: before preparing food, before feeding a child, before eating, after defecation and after cleaning a child who has defecated [17].
Poor handwashing practice at critical times Mothers/caregivers who reported washing their hands using water and soap at two or fewer of the five critical times [17].

Study variables
The dependent variable was acute diarrhea, denoted as yes [1] or no (0), where 'yes' indicated the presence of acute diarrhea and 'no' indicated the absence of acute diarrhea during the 2 weeks before to the survey.
The independent variables were socioeconomic/demographic, environmental and behavioral factors. Sociodemographic and economic variables that were selfreported by the study participants were mother's/caregiver's age (years), education status, marital status, household size (persons), house ownership, age of the child (months), child's sex, birth order of child, total number of under-five children in the household and household economic status (wealth status). Wealth status of the study participant household was estimated using principal component analysis (PCA).
Environmental factors included water, latrine and waste-related characteristics. Water-related variables that were self-reported were: main source of drinking water, round-trip time taken to obtain water (minutes), water consumption per capita per day (liters). Latrine-related factors that were observed by the data collectors were: type of latrine facility, presence of feces around the pit hole/slab/floor of the latrine, presence of feces in the house compound and presence of flies on the floor and/or around the latrine facilities. Latrine ownership was measured by selfreport of the study participants. Latrine distance from home (latrine proximity) was measured using GPS (geographical positioning system). Waste-related factors that were observed were presence of open raw sewage inside the house compound, presence of uncollected solid waste inside the house compound.
Behavioral factors that were self-reported by mothers/caregivers included child breastfeeding practices, child breastfeeding status (exclusive or partial), duration of breastfeeding, child given complementary food, age at which complementary food was started, rotavirus and measles vaccine status. Mothers'/caregivers' hand-washing practice was computed as good or poor based on self-reported practice at the five critical times.

Data collection and quality assurance
A structured questionnaire was prepared on acute diarrhea and associated factors in English and then translated to Amharic and then retranslated back to English, to ensure consistency of the questions. The questionnaire consisted of two parts. The first part included questions about the presence of acute diarrhea among the under-five children in the 2 weeks before the interview. The second part included questions about sociodemographic and economic, environmental, and behavioral factors.
Before the start of the actual data collection, the questionnaire was pretested on similar participants of 5% of the sample size in semi-urban areas of nearby Alansha. Some amendment was done based on the pre-test result before the actual data collection. The data were collected by face-to-face interviews and observations at the households from January to March, 2019. Four clinical nurses were recruited for data collection. They were given a day-long training session on study objectives, data collection tools, inclusion and exclusion criteria together with a work plan, and ethical issues.
Inter-observer reliability was ensured by providing clear definitions of variables to be recorded, by training data collectors, and by providing feedback about discrepancies during daily supervision. The principal investigator and trained supervisors were involved in the supervision of the overall data collection process. During the administration of the survey, each questionnaire was checked daily for completeness and consistency before the data entry. Unfilled or incompletely filled responses on a questionnaire were resought on the same or the next day by the same data collector.
In addition, to check the reliability of the information entered, 10% of the study participants were randomly selected and re-interviewed by another interviewer to check the reliability of the information entered by different interviewers. The qualifications of the interviewers and the training they received reduced the likelihood of interviewer bias. The data entries were checked for accuracy by reviewing randomly selected 10% of the questionnaires. In addition to this, descriptive statistics, such as cross-tabulations and frequency distributions were examined before performing statistical analysis.

Data management and analysis
The collected data were checked, coded and entered into EpiData version 3.1 and exported to SPSS version 25.0 for data cleaning and analysis. Descriptive statistics (frequency distribution and cross-tabulations) were calculated for categorical variables and the mean with standard deviation was estimated for continuous variables. Data about households' economic status was estimated by principal component analysis. The principal component analysis (PCA) was done to construct the household wealth index with the following considerations: communality value > 0.5, Kaiser-Meyer-Olkin (KMO) value > 0.5, and eigen-values greater than one [18]. Multicollinearity between independent variables was checked with the standard error of the coefficient with a cut-off point greater than 2 [19], which was not observed.
Bivariable (Crude Odds Ratio [COR]) and multivariable (Adjusted odds ratio [AOR]) analysis were employed using logistic regression analysis with a 95% confidence interval (CI). From the bivariable analysis, variables with p < 0.25 were considered for multivariable analysis. From the multivariable logistic regression analysis, variables with a significance level at p < 0.05 were taken as statistically significant and independently associated with acute diarrhea among under-five children. The model fitness was found to be good using the Hosmer Lemeshow goodness-of-fit test [19] and the model was fit at a p-value = 0.935.

Socio-demographic characteristics
Out of 340 under-five children in the study, 335 participated, for a response rate of 98%. The majority, 201(60.0%) of the mothers/caregivers were literate. Homes were owned by more than three-fourths of the households 297 (88.7%) ( Table 1).

Environmental-related factors
Out of the total 335 households, two-thirds 213 (63.6%) of study participants used water from improved sources. Round-trip time taken to fetch water for two-thirds 224 (66.9%) was > 30 min (Table 3). On average, most 280 (83.6%) of the households used an improved latrine. Most of the latrines were private 281 (83.9%) and almost two-thirds 206 (61.5%) of the latrines were 15 m or more from home (Table 4).

Behavioral-related factors
Among 335 mothers/caregivers, the majority 184 (54.9%) practiced poor handwashing, while 151 (45.1%) practiced good handwashing. Of the total 335 children, about half 177 (52.8%) had never been breastfed and only 44 (13.1%) of the children were exclusively breastfed. One-fifth 65 (21.7%) of children started complementary food at less than 6 months of age. Less than one-tenth 24 (7.2%) of the children had not been given a rotavirus vaccine and one-fifth 70 (20.9%) of children also had not been given a measles vaccine (Table 5).

Prevalence and associated factors of acute diarrhea
The prevalence of acute diarrhea among the underfive children was 11.0% (95%CI: 7.8-14.3%). From the multivariable logistic regression analysis, child's age between 12 and 23 months, presence of two or more children under five in the house, unimproved water sources and presence of feces around the pit hole/ slab/floor of the latrine showed significant association with the occurrence of acute diarrhea among underfive children.
A child aged between 12 and 23 months was 4.85 times more likely to develop acute diarrhea than those in other-age categories (AOR = 4.68, 95%CI: 1.45-1.5). Age groups 0-5 months, 6-11 months, 24-35 months and 36-49 months did not show significant association with the occurrence of acute diarrhea. The presence of two or more children under-five in the house also showed a significant association with acute diarrhea. Children in households with two or more children under-five were 2.84 times more likely to develop acute diarrhea than those in households with only one child (AOR = 2.84, 95%CI: 1.19-6.81) ( Table 6).
On the other hand, children from households that used water from unimproved sources were almost three times more likely to develop acute diarrhea than those in households who used water from improved sources (AOR = 2.97, 95%CI: 1.28-6.87). Similarly, children in households with feces observed around the pit hole/slab/floor of the latrine were 3.34 times more likely to develop acute diarrhea than those in households with no feces seen around the pit hole/ slab/floor of the latrine (AOR = 3.34, 95%CI: 1.34-8.31) ( Table 6).

Discussion
This community-based cross-sectional study was conducted with the aim to determine the prevalence of and factors associated with acute diarrhea among under-five children in semi-urban areas of Gelsha. The prevalence of acute diarrhea among under-five children was 11% and was significantly associated with a child's age between 12 and 23 months, the presence of two or more under-five children in the house, unimproved water sources, and the presence of feces on the pit hole/slab/ floor around the latrine facility.
This study showed that the odds of contracting acute diarrhea were higher in children aged between 12 and 23 months. At this age, children begin walking and are at high risk of contamination from contact with different environmental sources. This finding is  consistent with the studies conducted in Sidama Zone [21], in Benishangul Gumuz Regional State [31], in Bangladesh [35], in Wolita Soddo town, in Mecha District [37], in Hadaleala District [33], in Arba-Minch District [32], a study conducted by Beyene et al. among agricultural and agro-pastoralist communities of Southern Ethiopia [40] and the Ethiopian DHS 2016 report [8]. However, this finding is in contrast with the studies conducted in Tigray [41], Wolita Soddo town [9] and Farta Wereda [42]. This difference might be due to the children's mothers/ caregivers different educational levels in these areas.
This study also showed that acute diarrhea among under-five children had a significant association with the presence of two or more under-five children in the house. Children in households with two or more children in this age group were 2.84 times more likely to develop acute diarrhea than those in households with only one such child. This might be due to reduced care given to the child by a mother with more under-five children. This finding is in agreement with studies conducted in Northeast Ethiopia, Tigray [41] Benishangul [31], Shebedino District [43], Yaya Gulele District [44], Wolita Soddo Town [9], in Hadaleala District [33]. However, a study that was conducted in Arba-Minch District [32] did not find any significant association between acute diarrhea and the number of children. Differences might be due to increased care given to the child by the mothers/caregivers in these areas.
In this study, children from households that got water from unimproved sources were almost three times more likely to develop acute diarrhea than those in households that got water from improved sources. Exposure to different pathogens from contaminated water might cause this acute diarrhea. This finding was supported by the studies conducted in Uganda [29], in Sheka Zone [24], Indonesia [45], Nigeria [46], Amhara Region [47], Tigray [9], Derashe District [48], Jigjiga District [49] but is in contrast to studies conducted in Farta Wereda [42], Benishangul Gumuz [31] and Debre Berhan town [50]. This difference might be due to the use of water from mixed improved and unimproved sources and the use of different water treatment practices.
Acute diarrhea among under-five children showed significant association with the presence of feces around the pit hole/slab/around the latrine facility. Children from households with feces seen around the pit hole/slab/around the latrine were three times more likely to develop acute diarrhea than children from households with a clean latrine. This may be due to the presence of feces around the pit hole/slab/ around the latrine creating conditions for the breeding of flies that allows the transmission of diarrheacausing pathogens. This finding was consistent with former studies in rural area of North Gondar [51], Debre Berhan town [50], Nekemte town [36], and Gummer Wereda [52] but is in contrast with studies conducted in Bahirdar [53], and Derashe District [48]. This difference might be due to a difference in keeping the cleanliness of the sanitation facility.

Limitations of the study and gaps for future research
There are several limitations to this study. As data were collected from self-reported questionnaire or direct observation, there was risk of response bias with over/ under reporting of acute diarrhea. This has also happened in other studies from Ethiopia [54]. Our findings do not reveal seasonal variation in prevalence since the study was conducted from January to March, which is a dry season. Other studies have shown that during the rainy season diarrhea is increased [55]. Besides, the prevalence was determined based on the reports of mothers/caregivers without confirmation by laboratory examination. Thus, further study including laboratory examination of stool and water samples is needed to measure factors related to acute diarrhea more accurately.
Because there are several residual confounding factors due to unmeasured variables, additional studies are highly recommended that consider water, sanitation, and hygiene management status, which will enable collection of strong evidence. The other limitation of the study was that the study did not investigate household, individual, and community factors using multilevel analysis and the fact that there was a scarcity of literature on semi-urban areas; thus, the discussion compared findings with urban and rural community studies. Despite these limitations, the study provides new insight into the extent of acute diarrhea among under-five children in semi-urban areas.

Conclusion
This study identified socio-economic, environmental, and behavioral factors associated with acute diarrhea among under-five children in semi-urban areas of Gelsha. The study revealed that the prevalence of acute diarrhea among these children was 11%, a result that indicates a relatively high prevalence in these areas. The main factors that were significantly associated with acute diarrhea were a child's age between 12 and 23 months, the presence of two or more under-five children in the house, unimproved water sources, and the presence of feces around the pit hole/slab/floor of the latrine. Implemention of various prevention strategies is essential, such as the provision of health education to mothers/caregivers that focuses on keeping sanitation facilities clean and child care, and construction of improved water sources. Furthermore, implementing a strong health extension program (HEP), advocating open

Acknowledgements
We would like to express our special thanks to Dessie zuria District Health Office, Gelsha Health Center, and data collectors for their valuable contribution during data collection. We are also very grateful to the mothers/caregivers for their willingness and commitment to providing information. We also thank Lisa Penttila for the language editing of the manuscript.

Funding
Wollo University funded the research. The funder had no role in study design, data collection, and analysis, decision to publish, or preparation of the manuscript.

Availability of data and materials
The datasets collected and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations
Ethics approval and consent to participate All study methods were performed in accordance with the ethical principles of the Declaration of Helsinki [56]. An ethical clearance letter was obtained from the Institutional Ethical Review Committee of the College of Medicine and Health Science of Wollo University. A formal letter of cooperation was written from the university to the authority of Dessie zuria District and Dessie Zuria District Health Office and then the authority of Dessie Zuria District and Dessie Zuria District Health Office wrote a similar letter of cooperation for Gelsha town administration and Gelsha Health Center respectively. Necessary permission for data collection was received from the authority of Gelsha. Before starting the interviews, the data collectors explained the purpose of the study to all the participants and obtained written informed consent from the mothers/caregivers. They were informed that participation was voluntary and if they felt discomfort during the interview they could stop at any time.
The confidentiality and anonymity of the study participants' related data were maintained by avoiding possible identifiers such as participants' names. Identification code numbers only were used as a reference. During data collection, any under-five children with acute diarrhea in the previous two weeks who had not been treated were provided with oral rehydration salts (ORS) and linked to a health facility for further treatment.

Consent for publication
Not applicable.

Competing interests
The authors declare that they have no competing interest.