Active case detection using bi-valent malaria rapid diagnostic test in nontransmission season impacts upcoming malaria transmission

Despite a considerable global effort to eradicate malaria in the last few years, the disease burden in 2020 was 241 million, with 627 000 deaths. India accounted for 83% of cases reported in the World Health Organization (WHO) South-East Asia Region (WHO malaria report 2021). Nevertheless, India has shown its commitment to eliminating malaria from the country by framing the National Strategic Plan (NSP), according to which the districts have been stratified into four categories based on the malaria transmission intensity and the indicator used for categorization of the districts was annual parasite incidence (API). In North-East India, many highly endemic districts (API>2) are in the “Intensified Control Phase” and fall under category III[1]. A few pockets in such districts have disproportionate (API>10) due to forest and tribal dominance in those regions. The previous reports demonstrated a high malaria burden (API>10) in the district East Garo Hills[2]; however, a few pockets of the district Udalguri are malaria hotspots with the API>5[3]. Evidence suggests that malaria-endemic areas harbor mainly subclinical infections in the dry season due to the presence of low-density asymptomatic Plasmodium infections[4], and the infectious reservoir peaks between the end of the dry season and the start of the wet season[5]. Such low-density asymptomatic infections sustain malaria transmission throughout the dry season[6], requiring treatment to alleviate clinical malaria episodes' risk[7]. In the present study, we have highlighted the importance of bi-valent Plasmodium falciparum/Plasmodium vivax (Pf/Pv) rapid diagnostic test (RDT) during mass blood surveys in the dry season that helped reduce malaria case incidence in high API districts. A mass blood survey was conducted as an intervention during February-April 2017, considered a dry malaria season in the district East Garo Hills (high API district) of Meghalaya state and the district Udalguri (low API district) of Assam district. We used bi-valent Pf/ Pv RDT kit to screen for malaria. Malaria-positive cases were given treatment according to the National Antimalarial Drug Policy of India[8]. In parallel, finger-pricked blood was used to prepare thin and thick smears on glass slides for performing light microscopy and applied on Whatman filter paper for Plasmodium species-specific polymerase chain reaction (PCR). A total of 9 118 participants were screened for malaria, 3165 from high and 5953 from low API districts. The villages for malaria screening were selected using the probability proportional to size sampling (PPS) method. Briefly, two Primary Health Centres (PHCs) from each district were selected at random (one from a high malaria-endemic PHC and one from a low-endemic PHC). Further, 10 villages from each PHC were selected using PPS sampling technique. In a situation where the required sample from the selected village was not available, the shortfall was covered from adjacent villages. The detailed methodology of the sampling technique and methods performed during the mass survey are described previously[4]. The epidemiological data (2010-2019) related to malaria cases, slide positivity rate (SPR), and API of the districts under intervention were obtained from the respective district units of the National Center for Vector Borne Diseases Control (NCVBDC), erstwhile known as the National Vector Borne Disease Control Programme (NVBDCP). In addition, malaria epidemiological data of adjoining districts as a control (non-intervention district) was also obtained from the NVBDCP to quantify the impact of mass blood surveys in reducing malaria burden in the districts under intervention (Supplementary Tables 1T2T3-4). We estimated the significance of the differences observed in each year (by taking the year of intervention as a reference year, i.e., 2017) to determine the importance of the change. The same exercise was performed for the adjoining districts (district South Garo Hills adjoining with a high API district and district Darrang with a low API district), where no such intervention was carried out. It was done to compare the change in the malaria burden between the nonintervention and intervention districts. The median age of participants was 12 years (IQR 7 to 28), 56.8% (5 197/9 118) were females, and 6.1% (554/9 118) had a history of fever. Overall malaria prevalence during the mass survey by RDT was 6.5% (596/9 118), out of which 84.4% (503/596) were asymptomatic Plasmodium infections with a higher prevalence in young children than in adults (Table 1).Table 1: Malaria prevalence during mass blood survey, 2017 conducted in the districts East Garo Hills (high API district) and Udalguri (low API district).Malaria prevalence during the mass blood survey in 2017 was 16.6% (524/3 165) and 1.2% (72/5 953) using RDT in high and low API districts, whereas PCR reported 14.7% (284/1 936) and 6.6% (100/1521) Plasmodium infections in the respective districts. It reflects that RDT in high API districts could detect Plasmodium infections near the diagnostic ability of PCR but lost nearly 6-fold cases in low API districts. The reason for the lower detection ability of RDT in the low API district was the high persistence of low-density infections in this district[4,9]. The reported SPR by the district health department in 2017 was 3.7% (3 090/82 739) in high and 1.3% (1 129/85 062) in low API districts. In the intervention year, the high API district reported a nearly 5-fold reduction in SPR (3.7%) compared to RDT prevalence (16.6%) observed during the mass survey. This reduction in SPR was observed due to the active malaria case detection in the dry season using RDT, thereby abridging malaria transmission for the upcoming rainy season. Contrary to that, the low API district did not show any impact of the mass survey due to the low diagnostic accuracy of RDT[10] in this region, which is why the district reported SPR (1.2%) and the prevalence reported in the mass survey (1.3%) found similar in 2017. Malaria data of preceding and following years from the year of intervention i.e. 2017 show that overall mean percent change in malaria cases (-27.4, 95% CI -132.9 to 78.1) and API (-104.1, 95% CI -341.3 to 133.1) was declined but SPR (14.2, 95% CI -54.9 to 83.3) increased over the period in high API intervention district. In 2018, there was a sudden drop in malaria cases (3.1 fold; -343.3% vs. -111.4%), SPR (4.5 fold; -261.4% vs. -57.6%), and API (3.0 fold; -330.0% vs. -111.4%) in high API district compared to the control district. This decline showed a reversed trend in the subsequent year, 2019 with a fold change decline remained 0.6 fold (-385.1% vs. -642.2%) in malaria cases, 0.2 fold (-130.5% vs. -612.6%) in SPR, and 1.8 fold (-1 195.3% vs. -671.3%) in API of high API district (Supplementary Figure 1). On the other hand, low API intervention district showed an overall increase in the mean percent change in malaria cases (46.0, 95% CI 2.0 to 89.9), API (46.0, 95% CI 1.7 to 90.1) and SPR (35.4, 95% CI -16.4 to 87.2). There was no impact of mass survey observed in 2018 on malaria cases (0.97 fold; 52.5% vs. 54.2%), SPR (0.96 fold; 61.3% vs. 63.6%), and API (0.97 fold; 52.5% vs. 54.2%) in low API district compared to the control district. The reason is that there was an outbreak in 2018 at a sub-center situated in the intervention district, which contributed to a sudden increase in the number of malaria cases and resulted in the dilution of our interpretations of the low API district[11]. Nevertheless, a sharp decline in malaria cases (17.2 fold; -172.0% vs. -10.0%), SPR (1.6 fold; -229.7% vs. -147.0%), and API (62.6 fold; -173.8% vs. -2.8%) was observed in 2019 (Supplementary Figure 2) which was attributed to the extra efforts laid-down by the district health department for the management and control of an outbreak. To conclude, a mass blood survey using RDT in the dry season detected a significant proportion of Plasmodium infections that persisted in asymptomatic individuals during the dry season. Such infections, i.e., gametocytes, can potentially transmit in the upcoming favourable monsoon season in high API areas. Mass surveys in the dry season resulted in the contraction of the malaria reservoir, which is responsible for the transmission in the next transmission season[12]. That is why the year of the mass survey and the following year had a significant reduction in malaria, which was not observed over the preceding years. However, this reduction started losing the trend in the succeeding year, i.e., 2019, which further supports the observations that a single mass survey could only derail the transmission in the intervention year or maximally by the following year but cannot have prolonged effects further. We advocate the strategic use of bi-valent RDTs by stratifying the districts based on API and conducting mass screening for malaria during the non-transmission season in high API districts. In contrast, low API districts should deploy advanced molecular tools to deal with low-density infections. Conflict of interest statement The authors declare that there is no conflict of interest. Ethics approval The study was approved by the Institutional Ethics Committee of the National Institute of Malaria Research (ECR/NIMR/ EC/2015/490). All subjects or their representatives gave written informed consent prior to enrolment. Funding This study was funded by Indian Council of Medical Research India vide grant No. NER/55/2015-ECD-I. Authors’ contributions SP was the principal investigator of the study. SP and NM conceived the idea and designed the study. HS contributed to the data collection, analysis and interpretation. HS, SP, KS and MPS executed the field operations. HS, SSAH and MPS performed the analytic calculations and performed the numerical simulations. HS performed laboratory experiments under the supervision of NM. SP, KS and NM coordinated the field operations and logistics arrangements. MPS and NM critically reviewed the manuscript for important intellectual content. HS drafted the first version of the manuscript. All authors provided intellectual inputs and reviewed and approved the final version. Publisher’s note The Publisher of the Journal remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Edited by Zhang Q, Lei Y, Pan Y