Prevalence and associated factors of acute diarrhea among under ve children living in Hargeisa IDPs, Somaliland

Background: In developing countries, diarrhea is the major cause for child death when children are less than ve years’ old. Dehydration, malnutrition, delayed physical development and early childhood mortality are the major consequences of diarrheal diseases. In Somaliland, diarrheal diseases have been endemic and a major problem since 1994, with epidemics occurring annually. Objective: The aim of this study was to assess prevalence and risk factors of acute diarrhea among under ve children living in Hargeisa IDPs, Somaliland Methods: A community based cross sectional study was conducted among mothers of under ve children from August to September, 2020, in Hargeisa IDPs. 383 mothers were selected by using single population proportional formula. Data was entered, cleaned and analyzed using SPSS version 22. To explore association between variables, bivariate logistic regression was done for each independent variable with the dependent variable and those variables with a p-value of < 0.05 were adjusted into multivariate logistic regression. Finally, those variables having a P value < 0.05 were recognized as determinants of acute diarrheal disease. Results: The prevalence of diarrhea among under ve children ling in Hargeisa IDPs was 51%. The children whose age older than one year [AOR= 3.586, 95% CI, 1.054, 12.196], didn’t breastfed exclusively [AOR= 4.006, 95% CI, 3.274, 4.596], didn’t met colostrum milk [AOR= 36.41, 95% CI, 24.312, 40.489], drinks water stored in Jerry-cans [AOR = 4.901 95% CI, 1.306, 8.387] and poorly practices hand washing [A0R = 5.739, 95% CI, 1.382, 7.822] were more likely to develop diarrhea than their respective contraparts. Conclusion: From this study we concluded that the prevalence of was very high (51%). Lack of awareness of exclusive breastfeeding and colostrum feeding, storing drinking water in unprotected container and poor hand washing practice were identied to be signicance predictors for childhood diarrhea


Background
According to the WHO, diarrhea is de ned as "the passage of three or more loose or liquid stools per day" (1). It's usually a symptom of an infection in the intestinal tract, which have multiple etiologic agents including bacteria, viruses and parasites (2)(3)(4). Diarrheal disease is the second leading cause of death in children under ve years old. Globally, it is estimated that there are 2.5 billion cases and 1.5 million deaths annually in children under ve years (5,6). In developing countries, diarrhea is the major cause for child death when children are less than ve years old (6). The morbidity and mortality of diarrhea is attributed by poor socio-economic status, lack of access to safe water, lack of hand washing, poor housing conditions, unsanitary disposal of human waste, improper feeding practices, and lack of access to adequate and affordable health care (7).
Dehydration, malnutrition, delayed physical development and early childhood mortality are the major consequences of diarrheal diseases (8). Loading [MathJax]/jax/output/CommonHTML/jax.js In Somaliland, diarrheal diseases have been endemic and a major problem since 1994, with epidemics occurring annually. Very poor sanitation conditions and the human consumption of unsafe water are the main causes of this health situation. IDP camps which have even worse living conditions provide foci for endemic cholera outbreaks (9).
The two studies conducted in Somalia and Somaliland revealed that the prevalence rates of acute diarrhea were 11% and 47% respectively. Based on ndings of this studies, lack of mother's knowledge towards hand washing at four critical times, drinking unsafe water and failure of exclusive breastfeeding increases the odds of acute diarrheal diseases (10,11).
According the neighbor countries, the prevalence of acute diarrhea in under ve-children ranged between 14.5% and 36.5% (12)(13)(14)(15) in Ethiopia whereas prevalence rates ranging between 10.6% and 19.6% reported in Kenya (16) which are lower rates compared to Ethiopia.
This study aimed to assess (1); prevalence of acute diarrhea among under ve children (2); to determine socio-demographic and economic of mother and child (3); explore child characteristics; and (4)  After substitution the formula, the nal sample size was 383 mothers/caregivers.

Sampling procedure
First, four IDPs (Stadium, Ayaxa 1, Digale and Ayaxa 2) was selected from total 16 IDPs in the city by using simple random sampling method followed by allocation of each IDP to probability sample proportional to its size. Household numbers having under-ve children were taken from health extension workers registration books in the MCHs. Systemic random sampling was used to select households. The youngest child was selected for a household having two or more under 5 years of children. (Fig. 1)

Study variables
The dependent (outcome) variable of this study was the status of acute diarrheal disease in under ve children as reported by the mother/caregiver of the child in two-week period prior to the survey.
The independent variables were socio-demographic characteristics of mother (maternal age, occupation, educational level, household income and household size); child characteristics (child's age, child's sex, vaccination status, EBF practice and colostrum feeding); environmental and hygienic factors (water storage container, hand washing practice at critical times, household oor type and frequency of latrine utilization).

Data collection procedure
Structured questionnaire was rst prepared in English and then translated to Somali language. Retranslation was also being made to see the consistency of both questionnaires. After this, the questionnaire was used in interviewing the study participants through face to face interview method. Data collection was facilitated by health professionals using structured and pretested questionnaire.

Data quality control
Continuous guidance and supervision from principle investigator was done to keep data quality. The data collectors were trained for one day on the study instrument, consent form, how to interview and data collection procedures. Data was checked and rechecked for reliability and the questionnaire was pretested by 10% of sample size.

Data processing and analysis
The completeness of questionnaire was checked visually and coded with a whiteboard marker. Then was entered, cleaned and analyzed using SPSS version 22. Frequency tables and graphs were used to summarize the socio-demographic characteristics of the study participants and magnitude of diarrhea.
To explore association between status and risk factors, bivariate logistic regression was done for each independent variable with the dependent variable and those variables with a p-value of < 0.05 were adjusted into multivariate logistic regression. Finally, those variables having a P value < 0.05 were recognized as determinants of acute diarrheal disease.   Table 2). 178 (50.1%) children had diarrhea in two weeks' prior the data collection. Majority of them (82%) had watery type of diarrhea (Fig. 2).

Environmental and hygienic characteristics
Majority 274 (78.7%) of households were used metal tank for water storage container. One hundred (28.7%) of children were rarely practiced hand washing at critical times where as 40.2% and 31% of them were sometimes and often practiced handwashing respectively. Considering with the livestock ownership, fourth-three percent of households had at least one kind of livestock mainly goats. Only fty-eight (16.7%) of households had mud oor. All households had toilets however less than half (46.8%) of children were rarely used that latrines (Table 3).  The variables which were signi cant (p-value ≤ 0.05) during bivariate analysis were further considered in multivariate regression analysis in order to control potential confounders (Table 5). Accordingly, being a child age is older than one year; being a child who had poor hand washing practice; being from mothers not breastfed their children exclusively; being a child didn't meet colostrum milk and storing drinking water by jerry cans were found to be independent predictors of diarrheal disease in children.
Children whose age ranged between one and two years were 3.586 times more likely to have diarrhea than those aged less than one year [AOR = 3.586, 95% CI, 2.054, 5.196]. Exclusive breastfeeding practice was also signi cantly associated with childhood diarrhea.  * Signi cant association as p-value < 0.05, AOR = adjusted odds ratio, CI = con dence interval

Discussion
The two weeks' prevalence of diarrhea among under ve children in Hargeisa IDPs was 51%. The ndings of this study was extremely very high compared to SLHDS 2020 report which was only 4%. However closely related results of 47% were reported from Ceel-afwayn distric in Sanag region (11). Current studies conducted in Somalia, 11% (10), Ethiopia, 26% (17), Kenya, 19.6% (16) and Tanzania, 12.1% (18) shows lower results compared to the ndings of this study. This variation may be attributed to; the current study setting consider only urban dwellers; differences in socio-demographic characteristics, environmental factors (climate and geographical differences) and behavioral factors (availability of water, presence and usage of latrine, availability of hand washing facilities and ways of waste disposal in the study area compared to other studies. Loading [MathJax]/jax/output/CommonHTML/jax.js Children whose age ranged between one and two years were 3.586 times more likely to have diarrhea than those aged less than one year [AOR = 3.586, 95% CI, 1.054, 12.196]. The nding is supported with that of other similar studies conducted in other areas (19)(20)(21)(22)(23). This may be justi ed so because children aged more than one-year start crawling or walking which increases their exposure to infectious agents. Moreover, such children start complementary feeding, and this may increase their exposure to different types of infections through contaminated food and water.
Exclusive breast feeding was found to be a signi cant predictor of diarrheal morbidity in children as breastfeeding is an effective means of protecting children from diarrheal disease (24,25). For those children who didn't breastfed exclusively from their mothers were 4 times more likely to The colostrum is de ned as a sticky white or yellow uid secreted by the breasts during the second half of pregnancy and for a few days after birth, before breast milk comes in. Mothers colostrum provides effective passive immunity to the newborn against a wide range of enteric pathogens in the form of antibodies. They are central to the immunological link that occurs when the mother transfers passive immunity to the offspring (30 washing practice were identi ed to be signi cance predictors for childhood diarrhea (P < 0.05). Its highly recommended to pay special attention in EBF promotion: health care providers and decision makers should comprehensively address issues to improve EBF practices in the community. Moreover, implementation of health education programs towards importance of colostrum milk and hand washing practice could reduce the morbidity of childhood diarrhea.

Limitations
Not using enough literatures from the surrounding towns on the study area in speci c due to the unavailability of published journals on the topic of interest could be mentioned as a limitation. In addition to that, this study was cross-sectional study design, which made it di cult to establish causal effect relationship.

Ethical Considerations
The study protocol was reviewed and approved by Ethical review committee of University of Hargeisa.
Before data collection, written or oral agreement was obtained from the respondents such as parents and caregivers. informed consent was obtained from all a parent and/or legal guardian for participants under 18 and illiterates. Acceptance Permission of the study was obtained from every relevant authority in the Hargeisa city. Data collection was conducted con dentially. All methods were carried out in accordance with relevant institutional guidelines and regulations.