Improving the efficiency of reactive case detection for malaria elimination in southern Zambia: a cross-sectional study

• Main Authors: Fiona R. P. Bhondoekhan, Kelly M. Searle, Harry Hamapumbu, Mukuma Lubinda, Japhet Matoba, Michael Musonda, Ben Katowa, Timothy M. Shields, Tamaki Kobayashi, Douglas E. Norris, Frank C. Curriero, Jennifer C. Stevenson, Philip E. Thuma, William J. Moss, for the Southern and Central Africa International Centers of Excellence for Malaria Research
• Format: Article
• Language: English
• Published: BMC 2020-05-01
• Series: Malaria Journal
• Subjects: Malaria; Zambia; Elimination; Screening; Reactive case detection; Environment
• Online Access: http://link.springer.com/article/10.1186/s12936-020-03245-1

Abstract Background Reactive case detection (RCD) seeks to enhance malaria surveillance and control by identifying and treating parasitaemic individuals residing near index cases. In Zambia, this strategy starts with passive detection of symptomatic incident malaria cases at local health facilities or by community health workers, with subsequent home visits to screen-and-treat residents in the index case and neighbouring (secondary) households within a 140-m radius using rapid diagnostic tests (RDTs). However, a small circular radius may not be the most efficient strategy to identify parasitaemic individuals in low-endemic areas with hotspots of malaria transmission. To evaluate if RCD efficiency could be improved by increasing the probability of identifying parasitaemic residents, environmental risk factors and a larger screening radius (250 m) were assessed in a region of low malaria endemicity. Methods Between January 12, 2015 and July 26, 2017, 4170 individuals residing in 158 index and 531 secondary households were enrolled and completed a baseline questionnaire in the catchment area of Macha Hospital in Choma District, Southern Province, Zambia. Plasmodium falciparum prevalence was measured using PfHRP2 RDTs and quantitative PCR (qPCR). A Quickbird™ high-resolution satellite image of the catchment area was used to create environmental risk factors in ArcGIS, and generalized estimating equations were used to evaluate associations between risk factors and secondary households with parasitaemic individuals. Results The parasite prevalence in secondary (non-index case) households was 0.7% by RDT and 1.8% by qPCR. Overall, 8.5% (n = 45) of secondary households had at least one resident with parasitaemia by qPCR or RDT. The risk of a secondary household having a parasitaemic resident was significantly increased in proximity to higher order streams and marginally with increasing distance from index households. The adjusted OR for proximity to third- and fifth-order streams were 2.97 (95% CI 1.04–8.42) and 2.30 (95% CI 1.04–5.09), respectively, and that for distance to index households for each 50 m was 1.24 (95% CI 0.98–1.58). Conclusion Applying proximity to streams as a screening tool, 16% (n = 3) more malaria-positive secondary households were identified compared to using a 140-m circular screening radius. This analysis highlights the potential use of environmental risk factors as a screening strategy to increase RCD efficiency.