Geographical weighted regression analysis of hotspots of acute respiratory infection and its associated factors among under- five children in Ethiopia, spatial and multilevel analysis

Abstract

Abstract Background Acute respiratory infections are among the most common childhood infections and continue to be a major public health concern worldwide. The prevalence of acute respiratory infection among children under the age of 5 years varies by geographic area, and previous studies have failed to indicate geographical variation in Ethiopia. Therefore, the aim of this study was to examine the geographic variation in Ethiopia using spatial analysis. Methods Using the 2016 EDHS, a total sample of 10,417 women were included in this study. Analysis was performed using STATA-17, Arc-GIS-10.8, and SaTScan-9.6. Multilevel analyses were employed because of the hierarchical nature of DHS data. Variables with a p-value < 0.25 at the bi-variable analysis were entered into the final model analysis and p-values < 0.05 were used to declare statistical significance. Ordinary least squares and geographic weighted regression were employed to explore the spatial relationship between the outcome and determinant variables. The model with the lowest corrected Akaike Information Criteria (AICc) value was considered as the best-fit model for the data. Results The prevalence of acute respiratory infection among children < 5 years in Ethiopia was 12.29% (95% CI = 11.68–12.94%). Most of the hotspot areas were located in Tigray, central Oromia, eastern SNNPR, and southern Amhara. The proportion of rural women, children with diarrhea, Muslims, women with no education, low media exposure, and the poorest women were significant predictors of hotspot areas in the spatial analysis. In the multilevel analysis, secondary education (AOR = 0.60, 95% CI = 0.40–0.90), child age 24–35 months (AOR = 0.74, 95% CI = 0.58–0.94), child age 48–59 months (AOR = 0.61, 95% CI = 0.46–0.79), recent diarrhea (AOR = 4.78, 95% CI = 3.99–5.70), Afar (AOR = 0.46, 95% CI = 0.27–0.79), Amhara (AOR = 0.57, 95% CI = 0.38–0.86), Benishangul (AOR = 0.13, 95% CI = 0.07–0.24), and SNNP (AOR = 0.54, 95% CI = 0.34–0.87) were significantly associated with acute respiratory infection. Conclusion and recommendation: Acute respiratory infections remain a major public health problem in Ethiopia. Age of the child, recent diarrhea, maternal education, and region were significant variables associated with acute respiratory infection. The government of Ethiopia must assess and evaluate the current acute respiratory infection prevention strategies and develop geographically targeted strategies.