Menu
Menu
Menu
Menu
Introduction: Handwashing is fundamentally an inexpensive means of reducing the spread of communicable diseases. In developing countries, many people die due to infectious diseases that could be prevented by proper hand hygiene. The recent coronavirus (COVID-19) pandemic is a threat to people who are living in resource-limited countries including sub-Saharan Africa (SSA). Effective hand hygiene requires sufficient water from reliable sources, preferably accessible on premises, and access to handwashing facility (water and or soap) that enable hygiene behaviors. Therefore, this study aims to determine the prevalence of limited handwashing facility and its associated factors in sub-Saharan Africa. Methods: Data from the Demographic and Health Surveys (DHS) were used, which have been conducted in 29 sub-Saharan African countries since January 1, 2010. A two-stage stratified random cluster sampling strategy was used to collect the data. This study comprised a total of 237,983 weighted samples. The mixed effect logistic regression model with a cluster-level random intercept was fitted. Meta-analysis and sub-group analysis were performed to establish the pooled prevalence. Results: The pooled prevalence of limited handwashing facility was found to be 66.16% (95% CI; 59.67%—72.65%). Based on the final model, household head with age group between 35 and 60 [AOR = 0.89, 95% CI; 0.86—0.91], households with mobile type of hand washing facility [AOR = 1.73, 95% CI; 1.70—1.77], unimproved sanitation facility [AOR = 1.58, 95% CI; 1.55—1.62], water access more than 30 min round trip [AOR = 1.16, 95% CI; 1.13—1.19], urban residential area [AOR = 2.08, 95% CI; 2.04—2.13], low media exposure [AOR = 1.47, 95% CI; 1.31—1.66], low educational level [AOR = 1.30, 95% CI; 1.14—1.48], low income level [AOR = 2.41, 95% CI; 2.33—2.49] as well as lower middle-income level [AOR = 2.10, 95% CI; 2.14—2.17] and households who had more than three children [AOR = 1.25, 95% CI; 1.20—1.31] were associated with having limited handwashing facility. Conclusion and recommendation: The pooled coverage of limited handwashing facility was high in sub-Saharan Africa. Raising awareness of the community and promoting access to handwashing materials particularly in poorer and rural areas will reduce its coverage.
The Demographic and Health Surveys (DHS) program began in 1984 [20]. It is a nationally representative cross-sectional household survey conducted in low- and middle-income countries. We have used Demographic and Health Surveys (DHS) data which was conducted from 1st January 2010 to 31st December 2016 in 29 sub-Saharan African countries [21]. The survey is designed to collect information about maternal and child health, nutrition, household characteristics and other health issues. For comparison, the DHS survey adheres to the same basic protocols throughout the country. Households in DHS are selected using a two-stage cluster sampling methodology. In the first stage, cluster enumeration areas (EAs) (typically villages in rural areas or blocks in urban areas) were sampled using a probability proportional to the population size technique. In the second stage, all households in the selected area were listed, and then 25–30 households were chosen at random for interviews. This sampling strategy was utilized to get a representative sample of households. A total of 237,983 weighted samples were included in the study. All households located in sub-Saharan African countries were the source of population, while households found in 29 sub-Saharan African countries at the time of the DHS survey were the study population. The outcome of this study was a limited handwashing facility. DHS collected information on handwashing facility, such as the place where handwashing facility found, whether fixed (such as Sink with tap and Tube with outlets) or mobile (such as Tippy tap, Raised bucket with tap/ outlet, Two buckets suspended, Suspended bottle or bag with outlet/hole/ pop-up plug and Foot pump sink). Furthermore, data on the presence of water, soap, and any detergents (ash, mud, or sand) on the premises were gathered through face-to-face interviews and observation. Based on this information, households with both fixed and mobile places and household members washing their hands without water and or soap (confirmed by observation) at the time of the interview, were considered as “having limited handwashing facility” [22]. Individual and household level variables for limited handwashing facility extraction included the following; age of household head, sex of household head, marital status of the household head, household size, household wealth index, educational status of household head, floor material type (standard vs. substandard) [11], place of handwashing facility (fixed vs. mobile), water sources, sanitation facility and the number of under-five children in the household. Community-level factors that affected the availability of limited handwashing facility were place of residence, region, community-level education, income level, and community media exposure. Some of the individual variables were taken directly from DHS such as the sex of the household head. Other variables were computed and categorized further. The operational definition and coding of variables are summarized in the supplementary table (S 1& S 2). We have used STATA version 14.0 software to extract and analyze the data. After the samples were weighted, descriptive statistics were performed. Because of the hierarchical and clustering nature of the DHS data, a mixed effect multilevel logistic model was used. A cluster-level random intercept was utilized to determine the difference in limited handwashing facility between clusters. Meta-analysis was conducted to determine the pooled prevalence of limited handwashing facility in sub-Saharan Africa, as well as sub-group analysis by region, income level, and year of the survey were also employed. Four models were fitted in the multilevel analysis. The first was a null model (Model 1) that was designed to check the variability in limited handwashing facility and which only contains the outcome variable. Model 2 and Model 3 were for individual/household and community-level variables, respectively. In the fourth model (Model 4), both the community and individual/ household variables were fitted simultaneously. Model comparison was done using deviance and the model with the lowest deviance was chosen as the best-fitting model. Permission for data was obtained from the DHS program (https://dhsprogram.com/data/available-datasets.cfm). On the website, a request was made. The researchers had no ethical concerns since the DHS program handled ethical issues both before and throughout the survey.
