What is the value of reactive case detection in malaria control? A case-study in India and a systematic review

Abstract

Background: Reactive case detection (RCD) for malaria is a strategy to identify additional malaria infections in areas of low malaria transmission and can complement passive surveillance. This study describes experiences with RCD in two Indian sites, and aimed to synthesize experiences with RCD across endemic countries.

Methods: RCD programmes were piloted in two urban areas of India with a low prevalence of mainly Plasmodium vivax malaria in 2014. Cases were identified in a clinic by microscopy and contacts were screened within 2 weeks; PCR, in addition to microscopy, was used to detect Plasmodium parasites. A systematic review was conducted to identify RCD experiences in the literature.

Results: In Chennai, 868 contacts were enrolled for 18 index cases of clinical malaria; in Nadiad, 131 contacts were enrolled for 20 index cases. No new malaria infections were detected in Nadiad among contacts, and four new infections were detected in Chennai (three P. vivax and one Plasmodium falciparum), of which two were among household members of index cases. An additional five studies describing results from an RCD strategy were identified in the literature: four in Africa and one in Thailand. Including the results from India, the average number of contacts screened per index case in a total of seven studies ranged from four to 50, and 126 in a case study in Thailand with one index case. Malaria was detected in 0-45 % of the contacted persons. The average number of index cases needed to be traced to find one new case of malaria ranged from one to five, and could not be assessed in one study in India (no contacts positive for 20 cases). Sharing the household with an index case was associated with a five-fold increased risk of malaria compared to contacts from households without an index case (pooled risk ratio 5.29, 95 % CI 3.31-8.47, I(2) 0 %, four studies).

Conclusions: RCD in areas of low malaria transmission is a labour-intensive strategy, and its benefit is not clear. Studies are needed to assess how RCD can be optimized or into alternatives where interventions are targeted to family members or hotspots.

Fig. 1

Steps in RCD. Sources: WHO 2012 Disease surveillance for malaria elimination: an operational manual [4]; Zanzibar malaria control programme 2009 Malaria elimination in Zanzibar: a feasibility assessment [26]; Smith Gueye et al. [10]

Fig. 2

Mapping of RCD clusters. Representative blocks (~3.6 sq km) of the RCD areas of a Chennai and b Nadiad are shown. Large spheres index case households; small spheres proximal and distal RCD households; red malaria-positive reactive cases, blue negative

Fig. 3

Malaria cases over time in the clinics where the index cases were recruited. a Chennai, Tamil Nadu, India, b Nadiad, Gujarat, India

Fig. 4

Flow diagram for systematic review

Fig. 5

Comparison of malaria detected during contact tracing among members of index case households versus among members of other (more distal) households, studies in Africa and India, 2009–2015. This is an analysis where the raw numbers have been used, and no adjustment was done for clustering at the household level or by index case or other factors. In the study by Stresman et al., the non-index households were randomly selected from the same locality [19]. CI confidence interval, HH household, PCR polymerase chain reaction, RDT rapid diagnostic malaria test, RR risk ratio