Increased Case-Load of Malaria in an Epidemic Prone Zone that had been Masked by Insufficient Surveillance Data Reveled during Yellow Fever Outbreak Investigation in Kenya, 2022

Abstract

Abstract Background: Early detection is key to the control of malaria infection. Over 70% of the population in Kenya is constantly at risk of infection. In March 2022, several samples were collected from yellow fever suspects, during a yellow fever outbreak investigation in Isiolo County, with malaria being detected in more than 92% of them. This prompted further investigations to, find more cases characterizing them, conduct data quality audits, assess health facilities malaria surveillance, and establish possible factors associated with malaria infection in Isiolo County. Methods: This was a retrospective review of data from sixteen purposively selected health facilities. Outpatient, inpatient, pharmacy and parasitology laboratory registers were reviewed from 1st October 2021 to 31st March 2022, line listed suspected malaria cases in Excel, abstracted sociodemographic, clinical, and laboratory information. Plotted epidemic curve to illustrate distribution of cases. We interviewed clinicians, pharmacists, and laboratory officers to assess malaria surveillance. For Data Quality Audit (DQA), we used a standardized tool and automatically score data quality at various service delivery points, summarised continuous variables into frequencies and proportions using STATA 15. Bivariate analysis was applied to generate odds ratios and logistic regression to identify factors associated with malaria. Data were presented in tables, map, and figures. Results: Of the 5527 records analyzed, 54.9% (3453/5527) were females. Median age was 15 years (IQR 5-30). Participants with laboratory results were 89.6% (4957/5527), of these 11.5% (572/4957) were malaria positive. Exposure to malaria infection was two times more likely among participants ≥5 years, OR=1.85(95% CI 1.46, 2.35; p=0.00), males OR=1.45(95% CI 1.22,1.73; p=0.00), and residents in Garbatula Sub County OR=1.50(95% CI 1.07,2.10; p=0.01). In adjusted multivariable logistic regression analysis, ≥5 years aOR=1.5(95% CI 1.16,2.06; p=0.000), Males aOR=1.45(95% CI 1.223,1.732; p=0.000), and residents in Garbatula Sub County aOR=1.51(95% CI 1.078,2.105; p=0.016) were significantly associated with malaria infection. Data completeness was 83%, and timeliness 71%. Stock-outs of malaria commodities in 25% (4/16) of the facilities. Conclusions: malaria case load surpassed the alert threshold not depicted by the routine surveillance data. Incompleteness of data, untimeliness reporting and poor inventory may have attributed to the observed under-reporting. Interventions aimed at improving malaria data and inventory management would improve malaria surveillance indicators in the County.