The outcome variable in this study was diarrheal illness. The passing of loose, watery stool more than three times in a day for an individual is referred to as diarrhea (World Health Organization, 2017). This study sought to assess the predictors of diarrheal disease among children under the age of five in Ghana based on data from the 2014 GDHS. The association between a wide range of crucial elements including various socio-demographic, economic and environmental factors and diarrhea was assessed among young children in Ghana.
The total prevalence of diarrhea from this study was about 11.7% which is relatively lower than the reported prevalence of diarrhea among children under five years in Ghana from the 2008 GDHS which was 19.8% (Ghana Statistical Service (GSS) et al., 2015). Similarly, a study conducted on the trends and determinants of diarrheal disease among children under five in Ethiopia also reported a similar decrease in the prevalence of diarrhea from 26% in 2000 to 12% in 2016 (Negesse et al., 2021). Other studies in Plateau state, Nigeria however established an increasing trend in the prevalence of childhood diarrhea from 13% in 2013 to 24% in 2017 (Jiwok et al., 2021). The decrease in the prevalence of diarrhea in Ghana may be attributed to the improvements in sanitation over time, the rise in the number of health facilities and the various public health interventions that have been put in place over the years.
Children aged 12 to 23 months recorded a higher prevalence of diarrhea when compared to other age groups. This study's results contradicts those from a study conducted in Nairobi, Kenya, which established that most diarrhea cases were seen in children aged 6–11 months(Mutama et al., 2019) and aligns with findings from a study conducted in Western Ethiopia that saw the largest peak of diarrhea amongst infants who were 12–23 months (Alemayehu et al., 2021). Results from our study further established there was a significant association between children aged below 35 months of age and diarrhea. Children in this age group also had the greatest likelihood of experiencing diarrhea. This finding corroborates results from a study conducted among children below the age of five years in Sudan where participants aged 6–35 months had a higher likelihood of experiencing diarrhea than those who were either younger or older (Siziya et al., 2013).
According to the study conducted in Sudan, several mechanisms, including maternal antibodies against enteric pathogens and ongoing breastfeeding, provided some protection against diarrhea in children under 6 months old, thus resulting in a lower likelihood of experiencing diarrhea in this age group. Additionally, the lower prevalence of diarrhea among children aged older than 35 months could be attributed to the intrinsic immunity of children within this age group that may have developed. On the other hand, it was determined that the introduction of supplemental foods and modifications in dietary practices were to blame for the high prevalence of diarrhea in children aged 6–35 months.
Our study found no significant association between improved drinking water sources and childhood diarrhea. These findings were consistent with findings from a study conducted in Ethiopia and India (Soboksa, 2021) (Bawankule et al., 2017). Even though ensuring that people have access to improved sources of water is crucial for lowering the prevalence of diarrheal disease, multiple studies have shown that simply collecting water from these sources does not necessarily lessen the risk of developing the condition. Water may be contaminated during its distribution to homes even after it has been treated (Clasen et al., 2007).
Similarly, our study did not find significant association between type of toilet facility and childhood diarrhea. Even though improved toilet facility has the tendency to reduce diarrheal infections, most households have toilets that may serve as potential risk factors for diarrhea and other fecal-oral disease transmission. Some toilets smell due to the fact that these facilities are not always flushed or washed immediately after use; thus, attracting houseflies and suggesting poor hygiene practices (Nyambe et al., 2020).
According to our study, there was significant association between regions in the northern parts of Ghana (transition and savannah zones) and diarrhea. These findings were consistent with findings from a study conducted in Ghana on the spatiotemporal patterns of diarrhea (Asare et al., 2022). One possible explanation to this finding is the varied performance of the rotavirus vaccine in Ghana, where the northern regions of the country has seen a reduced impact from the vaccine (Asare et al., 2020). Rotavirus is known to be the leading cause of diarrhea among children under the age of five years in Ghana (Enweronu-Laryea et al., 2011). Secondly, the sparse distribution of WASH infrastructure, particularly in the transition and savannah zones, could be a contributing factor.
Our study revealed the sex of children was significantly associated with diarrhea. The odds of developing diarrhea among female children was less likely than male children. This finding corroborates results from a study conducted in Yemen, which identified that more male children were affected by diarrhea than female children (Bahartha & Alezzi, 2015). Findings from this study were attributed to a possible variation in sampling techniques and size that might have been used in the study.
The odds ratio for the educational level of mothers indicates that children born to mothers with some form of education were less likely to have diarrhea than their counterparts who had mothers with no education. Evidence which exists to support the impact of maternal education on children's overall living conditions, feeding practices, health amongst others might have accounted for this finding (Desmennu et al., 2017). The wealth index of familes showed a significant association with the occurrence of diarrhea in our bivariate analyses. Rather unexpectedly, the association became insignificant when other variables were included in the multivariate analysis. We found that children living in families with the highest wealth index were less likely to have diarrhea as compared to their poorer counterparts. This finding is supported by studies conducted in India (Pinzón-Rondón et al., 2015) (Bawankule et al., 2017). This finding might be due to the fact that children living in impoverished settings may not get access to improved environmental conditions and health care.
There was no significant association between childhood diarrhea and type of place of residence of children from our study. However, the odds of developing diarrhea were lower for children living in rural areas as compared to their counterparts living in urban areas. This finding contradicts findings from a study carried out by Alemeyahu et al. (2021) that concluded that diarrhea was more common in remote areas 24.8% than in metropolitan areas 21.7%. On the other hand, this study also goes in line with another study by Bahartha et al. in 2013, which suggested children living in urban areas were more susceptible to developing diarrhea than their rural peers.
Lastly, although stool disposal showed a significant association with diarrhea in the bivariate analysis, it was not significant when other variables were included in the analysis. In the multivariate analysis conducted, we found that there was an increased risk of childhood diarrhea among children who lived in households where stool was disposed improperly. This is in line with a study conducted in Nairobi, Kenya, which concluded, improper stool disposal was more likely to cause diarrhea (AOR = 2.05, CI = 1.36, 3.10) (Guillaume et al., 2020). This finding is also consistent with findings from a study conducted in Bangladesh and some sub-Saharan countries (Nigeria, Niger, and Burkina Faso) (Messelu & Dumga, 2016) (Bado et al., 2016) (Tambe et al., 2015). A likely explanation to this observation is that burying stool in the home and poor disposal of refuse and used diapers could potentially serve as a reservoir for various micro-organisms and pathogens, which may eventually potentiate the spread of diarrheal diseases, especially when insect vectors carry these pathogens from refuse to food items in the home (Cairncross et al., 2010).
Limitations of this study
The definition of the dependent and independent variables for this study was limited to the information collected by the Ghana Demographic and Health Survey Program; hence other variables of interest could not be explored. Another limitation of this study was that the classification for the prevalence of diarrhea was according to the symptoms and signs reported by the mothers of children surveyed, and therefore was not validated by any medical personnel. Also, there was a possibility for recall bias since the mothers were asked about past events they might have forgotten about, leading to the acquisition of incorrect responses. Lastly, due to the cross-sectional nature of the study, all of the data used in the regression analysis were collected at the time of the survey, therefore they can only indicate statistical relationships between the predictor and dependent variables and not a cause-and-effect link.