Study design and description of study settings

A community-based cross-sectional study was conducted in May 2017 in rural Dembiya. Dembiya is one of the districts in North Gondar Zone, the Amhara National Regional State, Ethiopia. In Dembiya district, hygiene and sanitation-related communicable diseases were highly prevalent. The district health office report indicated that intestinal parasitic infections and diarrheal diseases were the top 4 and 5 prevalent diseases, which accounted 5161 (9.97%) and 4981 (9.62%), respectively, during 2017. The population in the area had also poor access to water, sanitation, and hygiene (WASH). According to the report of the district health department, clean water and latrine coverage in the district was 26.60% and 55%, respectively, in 2017.¹³

Sample size determination and sampling procedures

This study is part of the baseline survey for Dembiya water, sanitation, and hygiene-neglected tropical diseases (Dembiya WASH-NTD) project. The project was implemented to prevent intestinal parasitic infections through improved WASH. Sample size was, therefore, determined based on the prevalence of intestinal parasitic infections using single population proportion formula with the following assumptions: P = 85.1% (prevalence of intestinal parasitic infections among children aged 6–59 months in Shesha Kebkele, Wondo Genet, Southern Ethiopia during 2010),¹⁴ 95% confidence interval, 5% margin of error (d), and 15% nonresponse rate. A total of 225 children aged 6 to 59 months were selected from 5 rural kebeles (the lowest administrative units in Ethiopia). The study subjects were selected by systematic random sampling technique.

Data collection tools

We used pretested and structured questionnaire to collect sociodemographic, hygiene- and sanitation-related, and common GI symptom data. We also used observation checklists for food safety, environmental sanitation, and hygiene condition of children. Direct stool examination technique was used to identify parasitic infections in children. Children were provided a plastic stool container and asked to bring approximately 15 g of their own stool. A drop or drops of saline were placed on a slide. Approximately 0.05 g of stool specimen was placed using an applicator stick and mixed with a drop of saline and covered by cover slide. Finally, the specimen was examined under the microscope at low power (×10 objective) and high power (×40 objective) magnifications for the identification of intestinal parasites. Stool specimen was analyzed immediately after collection.¹⁵ Water samples were taken from individual households at point of use using sterilized sampling bottles, and the samples were transported to the central laboratory within 4 h with cold chain. Escherichia coli was used as a biological indicator for drinking water quality.

Measurement of variables

GI symptoms among children, the primary outcome variable of this study, were defined as the presence of one or more of these symptoms: abdominal discomfort, abdominal cramp, diarrhea, nausea, vomiting, bloating, and perianal itching. We primarily used mothers’ report to assess these symptoms. Health professionals also diagnosed for some symptoms. Different predictor variables like presence of intestinal parasitic infections, hand washing practice, drinking water quality, food safety, households’ environmental sanitation, and cleanliness of living houses were clearly explained elsewhere.¹⁶

Quality management

To minimize potential biases, we did the following: adequate or representative sample was calculated and study subjects were selected randomly from a well-defined source or target population; pretested the tools and amended as needed; used well-defined checklists for observation; we trained data collectors and field supervisors in making correct assessments; monitored each stage of research, including close follow-up during data collection; encouraged all the study subjects to participant in the study and to provide complete responses; double checked the data and cleaned it of errors; identify possible confounding factors; analyzed the data with proper statistical methods; and interpreted the results in an objective manner based on evidence.

Data analysis

Statistical Package for Social Sciences (SPSS) version 20 was used to analyze data. Frequencies and percentages were used for most variables. Univariable binary logistic regression analysis was used to choose variables for the multivariable binary logistic regression analysis on the basis of P value less than .2. Multivariable binary logistic regression was employed to identify factors associated with GI symptoms on the basis of odds ratio (OR) with 95% confidence interval (CI) and P < .05 and to control the possible effect of confounders. Hosmer and Lemeshow test was used to check model fitness.

Hygiene and sanitation conditions

A total of 225 mother-child pairs were participated in this study with 100% response rate. One hundred nineteen (52.9%) of the children were female, and 166 (73.8%) of them were aged between 24 and 59 months. The median age of children was 42 months, and the interquartile range was 24 to 48 months. One hundred thirty-four (59.6%) of the mothers were 30 and below years old. Majority, 180 (80.0%), of mothers did not attend formal education and 50 (22.2%) of the households had at least 1 member whose education level is secondary and above.

This study found that 59 of 225 (26.2%) children kept their finger nails short and clean at the time of the survey. One hundred two (45.3%) of mothers/care givers reported that they frequently washed hands of their child after playing and defecation, and before eating. Nearly a quarter, 55(24.4%), of mothers/care givers had good hand washing practice. The bacteriological test indicated that the quality of drinking water among 158 (70.2%) households was not good. The study revealed that 118 of 225 (52.4%) households practiced good food safety measures at the time of the survey. Less than half, 97 of 225 (43.15), of households had access to adequate sanitation. The vast majority, 198 of 225 (88.0%), of households lived in unclean quarter or house. One hundred eighty-five (82.2%) of the households had livestock at the time of the survey (Table 1).

Table 1.

Personal hygiene of children aged 6 to 59 months in rural Dembiya, northwest Ethiopia, May 2017.

Common intestinal parasitic infections

In this study, 58 of 225 (25.8%) children were infected with one or more intestinal parasitic infections. Of the infected children, the highest proportion, 45 (77.6%), of them were infected by Ascaris lumbricoides (Figure 1).

Figure 1.

Prevalence and commonly identified intestinal parasitic infections (IPIs) among under-5 children in rural Dembiya, northwest Ethiopia, May 2017.

Common GI symptoms

The current study depicted that 139 of 225 (61.8%) children had GI symptoms. Abdominal discomfort (137 of 139 [98.7%]), abdominal cramp (125 of 139 [89.9%]), and diarrhea (118 of 139[84.9%]) were the highest symptoms reported (Figure 2). Of the children who reported GI symptoms, almost all of them had more than 1 symptom (Table 2).

Figure 2.

Prevalence and commonly identified GI symptoms among under-5 children in rural Dembiya, northwest Ethiopia, May 2017.

Table 2.

Number of gastrointestinal symptoms found in under-5 children in rural Dembiya, northwest Ethiopia, May 2017.

Factors associated with gastrointestinal symptoms

Table 3 shows factors significantly associated with GI symptoms among under-5 children. Age of children, educational status of mothers/care givers, hand hygiene of children, hand washing practice of mothers/care givers, drinking water quality, food safety, households’ sanitation performance, cleanliness of the house, availability of livestock, and intestinal parasitic infections were variables entered into the univariable analysis. All other variables except age of children, educational status of mothers, and hand washing practice of mothers were selected for the multivariable binary logistic regression analysis.

Table 3.

Factors with GI symptoms among under-5 children in rural Dembiya, northwest Ethiopia, May 2017.

Abbreviations: CI, confidence interval; GI, gastrointestinal; OR, odds ratio.

Hosmer and Lemeshow test = 0.335.

*

Statistically significant at P < .05. **Statistically significant at P < .001.

This study found that intestinal parasitic infections were statistically associated with GI symptoms among children. The probability of GI symptoms was 13.69 times to be higher among children who had intestinal parasitic infections (OR = 13.69, 95% CI = 3.31-56.59).

The current study identified that hygiene and sanitation conditions were significantly associated with GI symptoms. The likelihood of having GI symptoms was 2.28 times to be higher among children who did not keep their finger nails short and clean (OR = 2.28, 95% CI = 1.02-5.10). Similarly, children with no access to adequate sanitation had higher odds to have GI symptoms compared with their counterparts (OR = 2.37, 95% CI = 1.08-5.18). The odds of having GI symptoms was 9.06 times to be higher among children who lived in unclean houses (OR = 9.06, 95% CI = 2.60-31.54). The likelihood of GI symptoms among children was higher in households who had livestock (OR = 4.68, 95% CI = 1.82-12.03).

Implications of findings

The findings of this study may indicate high prevalence of enteric infections as the reported GI symptoms are hallmarks or common manifestations of different enteric infections, and provide initial evidence to the local health authorities or health practitioners about the burden of enteric infections in the area and warrants further researches on enteric infections.

Limitation of the study

We primarily used mothers’ report to assess GI symptoms. Health professionals also diagnosed for some symptoms. Due to this phenomenon, the current study might be affected by social desirability bias. Moreover, the 95% CI for some predictor variables, for instance “intestinal parasitic infections,” is wide due to small number of cases in one of the cells in a 2-by-2 table.