The Prevalence and Factors Associated with Prompt Diagnosis and Treatment of Fever among Under-five Children in Zambia; Evidence from a Country-wide Cross-Sectional Survey

Abstract

Introduction: Fever is one of the signs of a suspected infection as the body mounts an inflammatory response to fight infections such as malaria, pneumonia and gastroenteritis. Prompt diagnosis of the aetiology of the fever and appropriate treatment can prevent severe disease and mortality. Delayed diagnosis and treatment of infections contribute significantly to the high under-five mortality ratio. This study assessed the prevalence and associated factors to prompt diagnosis and treatment of fever among under-five children in Zambia. Methods: The study used secondary data from a nationwide cross-sectional survey carried out during the rainy season, a peak transmission season for malaria and oral-faecal transmitted diseases. The survey used multistage computer-generated random sampling by statistical enumeration areas (SEAs) as clusters and households within the SEAs. Informed consent was obtained from the adult caregivers of the children. All children within the selected households below five years were included in the study, and the caregivers were interviewed on the occurrence of fever in the two weeks preceding the survey. In addition, the children were tested for malaria using rapid diagnostic tests and haemoglobin count. The data was analysed in STATA version 14 using weights to account for inter and intra-cluster correlations (ICC). Descriptive statistics using measures of frequencies, medians and interquartile ranges were done, and cross-tabulations and logistic regression were used to assess measures of association. The significance level was set at a P-value of 0.05 and a confidence level of 95%. Results: Out of the 3003 under-five children included, 728 had a fever in the two weeks preceding the survey, reflecting the prevalence of fever of 19.5%. The fever was more common in poverty-related indicators such as lower education levels of heads of households, rural areas, lower wealth status, not using ITNs, and malaria and anaemia. The prevalence of prompt health-seeking behaviour was 57.3% which raises concerns that 42.7% of the children with fever did not have prompt treatment within the same of the next day of fever onset. Cross-tabulations showed that prompt health-seeking behaviour was more common in children who were boys, children older than one year compared to those who were below one year, heads of households with secondary or tertiary education, Eastern, Muchinga and North-western provinces and those who did use indoor residual spraying. Further, the adjusted multivariable model found that the male sex of the children was associated with significantly increased odds of prompt health-seeking behaviour aOR 1.52 and the use of IRS was also associated with increased odds of prompt health-seeking behaviour aOR 1.8. While having secondary or tertiary education compared to no education was not statistically associated with prompt health-seeking behaviour (P-value = 0.085), the odds were also increased nonetheless aOR 1.30. Conclusion: This study has found that the prevalence of fever among under-five children in Zambia is comparable to other sub-Saharan African Countries. Fever was associated with age, education, area of residence, wealth status and use of malaria prevention interventions like insecticide-treated nets and indoor residual spraying. Further, fever was found to be associated with the presence of anaemia and malaria. Health-seeking behaviour among children with fever was sub-optimal, as a considerable proportion did not seek treatment. Factors associated with health-seeking behaviour among children with fever include sex, education levels of the head of the household and staying in a house with indoor residual spraying.