Effectiveness of Long-lasting Insecticide-Impregnated Bed Nets in Controlling Phlebotomus argentipes in Sri Lanka: A Cluster Randomized Controlled Trial

Abstract

Background

Sri Lanka is endemic for cutaneous leishmaniasis (CL) caused by Leishmania donovani. Long-lasting insecticide-impregnated bed nets (LLINs) have been used for vector control, but their impact on interruption of sand fly bites remains untested in Sri Lanka. This study assessed the efficacy of LLINs as compared with untreated bed nets in reducing vector-human contact and density of Phlebotomus argentipes, the vector of CL in Sri Lanka.

Methods

A cluster-randomized trial was conducted with an intervention group (treated nets) and a control group (untreated nets) in CL-endemic sites. Blood was collected at baseline and 6 and 12 months to measure antibodies against a P argentipes salivary marker. Sand flies were collected over 32 months with light traps (Centers for Disease Control and Prevention) and cattle-baited traps. Bioassays assessed LLINs’ impact on P argentipes mortality, and new CL cases were monitored postintervention.

Results

Postintervention, antibodies against P argentipes salivary marker significantly decreased by 28.8% at 6 months (P = .00002) and 44.9% at 12 months (P = .00001) in the intervention group as compared with the control group. Although indoor and outdoor monthly captures were significantly reduced in the intervention group vs controls, LLINs’ impact on reducing P argentipes density was more pronounced indoors, decreasing by 77.3% after 2 months (P = .00001), 89.3% after 10 months (P = .00001), and 63.1% after 24 months (P = .00003). During follow-up, knockdown and Phlebotomus mortality rates ranged from 100% to 81.54% and 100% to 90.43%, respectively. Six CL cases vs 1 case occurred in the control and intervention groups.

Conclusions

LLINs effectively reduced vector exposure and indoor P argentipes density and is a promising intervention for control of CL in Sri Lanka.

Leishmaniasis, a neglected tropical disease, affects >1 billion people in endemic regions, with limited treatment access causing substantial morbidity and mortality [1]. It presents in 3 major forms: cutaneous leishmaniasis (CL), visceral leishmaniasis (VL), and mucocutaneous leishmaniasis [1]. Annually, there are approximately 30 000 new VL cases and >1 million CL cases worldwide [1].

CL is the most common form globally, particularly in the Middle East, South America, and Central Asia [1, 2]. VL, though less prevalent, is the most severe and is fatal if untreated with >40 000 deaths annually, affecting mainly the Indian subcontinent (ISC), East Africa, and Brazil [1, 3]. These geographic differences reflect the complex distribution of Leishmania species, sand fly vectors, and varying environmental and socioeconomic conditions [1, 2, 4]. Core components of leishmaniasis control, such as vector control, case detection, and health education, are broadly similar in principle, but the specific strategies and implementation approaches vary significantly by region, type of leishmaniasis, and local epidemiology [1, 2].

The World Health Organization’s (WHO's) 2030 road map for neglected tropical disease elimination prioritizes eradicating VL caused by Leishmania donovani, with a focus on the ISC [5–7]. Sri Lanka primarily reports CL and was the first ISC nation to report CL cases caused by L donovani [8, 9]. While VL incidence has decreased through elimination programs, new CL cases linked to L donovani have surfaced in previously nonendemic areas of Sri Lanka, India, Bangladesh, Nepal, and Bhutan [10–14].

Control programs emphasize comprehensive surveillance, timely case detection and treatment, and vector control to reduce leishmaniasis transmission [15, 16]. In the ISC, efforts focus on controlling P argentipes, the primary leishmaniasis vector [17, 18]. India, Nepal, and Bangladesh have tested various chemical and integrated methods to control vectors, including long-lasting insecticide-treated nets (LLINs), indoor residual spraying (IRS), insecticide-treated dog collars, durable wall linings, and insecticidal wall painting [19–24]. Given the high cost of IRS, cost-effective alternatives such as LLINs, durable wall linings, and insecticidal wall paint are crucial for sustainable vector control since economic feasibility is essential for long-term use [23–26].

While vector control strategies have been tested in India, Nepal, and Bangladesh within the ISC region, Sri Lanka is yet to implement a robust control program for eliminating CL caused by L donovani. This study seeks to fill this critical gap by evaluating the effectiveness of LLINs in reducing human-vector contact to control CL in Sri Lanka. Utilizing a cluster-randomized controlled intervention trial (CRT), our study established the effectiveness of LLINs as compared with untreated nets by assessing the percentage reduction of human-vector exposure through measuring antibodies against rPagSP02 + rPagSP06, a composite recombinant salivary marker specific to P argentipes bites [27, 28] (herein, P argentipes salivary marker [PASM]).

METHODS

Study Area and Duration

The study was conducted in the Ambalantota Ministry of Health area in Sri Lanka's Southern Province, an endemic region for CL located approximately 190 km from Colombo and 15 km from Hambantota. Spanning 213 km² (6.08–6.24°N, 80.89–81.03°E), the Ministry of Health area consists of 16 administrative divisions. Among these, Walawewatta and Elegoda were selected as intervention and control sites, respectively, based on comparable CL incidence (83 cases from 2018 to 2020; Supplementary Table 1). The sites are 4.5 km apart and were assigned to receive either insecticide-treated nets (intervention) or untreated nets (control). The 32-month study (January 2022–August 2024) included a 6-month preintervention survey, 1 month for intervention implementation, and 24 months of follow-up.

Study Design

This study employed a CRT design, dividing the area into 10 clusters per group, with each cluster comprising 30 households approximately 50 m apart (Figure 1). The number of clusters and households was determined to achieve 80% power and 95% confidence (Supplementary Formula 1), incorporating a design effect of 1.86 based on previous studies from Bangladesh, India, and Nepal [20–23]. Households within each cluster were selected consecutively by spatial proximity. At least 1 individual from each household was included for blood sampling and net distribution. Inclusion criteria were healthy individuals aged ≥18 years, and those with a confirmed or suspected history of CL were excluded. Sociodemographic data were collected from the same individual, and all comparisons were based on participants who provided blood samples and received nets. Study areas were first grouped by geographic location to ensure separation between intervention (LLINs) and control (untreated nets) clusters and minimize potential contamination, while ensuring similar environmental and epidemiologic conditions across the broader study area. Within these geographic strata, clusters were then randomly assigned to intervention or control via a computer-generated sequence. Within each cluster, households were selected by identifying a random starting point and enrolling spatially adjacent households consecutively until the target number was reached. Household-level random effects were included in the analysis to account for spatial variability. Further details on household selection and follow-up are provided in Figure 2.

Figure 1.

The map illustrates the location of the 2 study sites designated as intervention (Walawewatta) and control (Elegoda) within the Ambalantota Ministry of Health area in the Hambantota district, Southern Province, Sri Lanka. Locations of clusters within each study site are indicated by symbols. Each group consists of 10 clusters and each cluster consists of 30 households. Abbreviations: DS, Divisional Secretary; GN, Grama Niladhari.

Figure 2.

Consortium diagram of the study. The diagram illustrates the flow of the study, including the screening process, enrollment, and assignment to the intervention group (LLINs) and control group (untreated nets). Participants who met the inclusion criteria were selected for the study. Participants <18 years of age and those with a medical history of cutaneous leishmaniasis were excluded from the study. HH, household; LLIN, long-lasting insecticide-treated net.

Intervention

The intervention was carried out in August 2022 with WHO-recommended LLINs (Olyset Net; treated with 2% w/w permethrin) [29, 30]. Prior to implementation, a pilot study assessed the susceptibility of wild-caught P argentipes to permethrin via a cone bioassay, as detailed under the LLIN bioefficacy methods. The results confirmed full susceptibility, supporting the use of permethrin-treated nets. LLINs measuring 180 × 160 × 150 cm, with a mesh size of 56 holes/sq in and hole dimensions of 6.7 × 11.1 mm [29, 30], were distributed in the intervention group, 1 net per bedroom. The control group received untreated nets in 2 sizes (180 × 100 × 150 cm and 180 × 160 × 150 cm), with a finer mesh size of 100 holes/sq in. All households were instructed to use the nets nightly, and they received information sheets on correct use and washing practices, emphasizing that repeated washing may reduce insecticidal efficacy over time.

Pre- and Postintervention Blood Collection

Preintervention activities included collecting 3-mL baseline blood samples from at least 1 consenting adult (≥18 years) per household in the control and intervention groups. Blood was collected from groups at 6 and 12 months postintervention. The serum was separated and stored at −20 °C until further use. The number of samples collected at each time point is detailed in Table 1.

Table 1.

Percentage Reduction of Anti-PASM Antibody Levels Indicative of a Decreased Exposure to Phlebotomus argentipes bites

Time: Group	No. of Samples	GM of ELISA OD	SD	No. Positive	Anti-PASM IgG Reduction, %	P Valuea
Baseline
Intervention	300	0.118	0.078	199
Control	246	0.142	0.101	174
6 mo
Intervention	259	0.073	0.063	108	28.81	.00002
Control	208	0.131	0.075	145
12 mo
Intervention	233	0.061	0.054	28	44.91	.00001
Control	165	0.138	0.066	131

Abbreviations: ELISA, enzyme-linked immunosorbent assay; GM, geometric mean; OD, optical density; PASM, P argentipes salivary marker.

^a P values were determined via a linear mixed effect model.

Measurement of Antibody Levels to PASM by Enzyme-Linked Immunosorbent Assay

Exposure to P argentipes bites was measured by antibodies against the composite recombinant salivary biomarker rPagSP02 + rPagSP06 (PASM) with an optimized indirect enzyme-linked immunosorbent assay protocol [27, 28]. The cutoff value for the assay was determined by a receiver operating characteristic curve, with optical density values from nonendemic (n = 15) and CL-positive (n = 30) controls, achieving 96.7% sensitivity and 100% specificity.

Reporting of CL Case Incidence During the Study

New CL cases were recorded in the intervention and control groups at 2-month intervals throughout the 24-month follow-up period. CL diagnosis followed WHO guidelines [31], with confirmed cases recorded in official registry books issued to patients receiving treatment. Regular monitoring enabled comparison of CL incidence between the groups over time.

Entomologic Assessment

Sand fly density was monitored monthly during a 6-month preintervention period (January–July 2022) and then bimonthly (October 2022–August 2024), following the intervention. Two trapping methods were used: Centers for Disease Control and Prevention light traps (CDC-LTs) and cattle-baited net traps (CBNTs), with trap setup and collection timings detailed in Supplementary Figure 1. All collected sand flies were morphologically identified according to taxonomic keys by Lewis [32].

Collection of Meteorologic Data

Meteorologic data from the Hambantota national station were collected to complement entomologic findings. Variables included monthly average, maximum, and minimum temperatures, as well as cumulative precipitation and relative humidity, from January 2022 to August 2024. Although no direct site comparison was done, the sites are geographically close with similar seasonal conditions based on meteorologic data. Additionally, the sand fly's limited flight range (<2000 m) [33, 34] and reliance on the immediate microenvironment minimize environmental differences between the groups.

Bioefficacy of LLINs During the Intervention

Cone bioassay tests were conducted following the WHO Pesticide Evaluation Scheme guidelines [35] on the intervention and control groups at baseline and 2, 4, 6, 8, 10, 12, 14, 16, 18, and 24 months postintervention. Sand flies were collected with CBNTs by trained personnel. Gender identification was performed immediately after collection where males were identified by the presence of 3-lobed external genitalia, which are absent in females. Only female P argentipes sand flies from randomly selected households were used for testing. Flies were exposed to insecticide-treated netting surfaces for 3 minutes; knockdown was recorded at 1 hour and mortality at 24 hours. Tests were done in quadruplicate and mortality percentages calculated.

Statistical Analysis

All statistical analyses and visualizations were performed in Prism version 8.0 (GraphPad) and R version 4.2.0. Descriptive statistics summarized baseline characteristics. To assess the intervention effect on antibody responses (optical density values to PASM), a linear mixed effects model was used with fixed effects for group (intervention/control), time (baseline, 6 and 12 months), and their interaction and a random effect for participant. Log transformation was applied to meet normality assumptions. Estimated marginal means with 95% CIs supported post hoc comparisons.

The impact on indoor P argentipes density was assessed via a Poisson generalized linear model with a log link and household-level random effects. Seasonal variables (temperature, humidity, rainfall) were excluded due to multicollinearity. Model diagnostics and percentage reduction calculations are detailed in the supplementary methods, Table 1, and Supplementary Table 5.

RESULTS

Study Population

The study included participants of varying ages. The mean (SD) age in the intervention and control groups was 49 (15) and 48 (14) years, respectively. Females, mostly housewives, constituted the majority in both groups (66.3% in the intervention and 54.0% in the control), probably as the survey was conducted during the day when most males were away working. Males were primarily engaged in outdoor work, especially paddy and banana cultivation. Farmers and laborers, who are at higher risk of sand fly bites, made up 16.8% of the intervention group and 22.7% of the control group. Travel outside the local area within the past 6 months was reported by 29.0% of the intervention group and 19.3% of the control group. Most houses had only nonplastered brick walls (69.3% in intervention, 74.0% in control), and tiled roofs were common in the intervention group (68.0%) and asbestos roofs in the control group (55.7%). Nearly all participants slept on beds (95.7% in intervention, 93.0% in control) and used mosquito nets (74.3% in intervention, 68.0% in control). Insect repellent use was moderately higher in the intervention group (46.0% vs 39.3%). A proportion of participants (27.0% in intervention, 34.3% in control) engaged in outdoor activities for 1 to 3 hours during the day, and a few were active during dawn or dusk, particularly those involved in banana cultivation. The presence of pet cats, dogs, or both was reported by 54.5% of the intervention group and 57.0% of the control group. Outdoor latrines, storage rooms, and cattle sheds were common in both groups (61.6% in intervention, 58.0% in control). None of the sociodemographic factors or living conditions differed significantly (P > .05) between the intervention and control groups (Supplementary Table 2).

Use of Bed Nets by Study Participants

A total of 956 LLINs and 874 untreated bed nets were distributed to intervention and control households, respectively. Allergic reactions to LLINs were minimal (0.67% of households). Net usage remained high (99.9% intervention, 100% control). LLIN washing increased from 0.016% (12 months) to 60.7% (24 months), while 64.2% of control households washed nets within 6 months. LLIN damage rose from 2.7% to 6.0% (12–24 months) as compared with 8.5% to 21.2% in control nets, with no significant difference between groups (P > .05).

Effect of Intervention on Reducing Exposure to Bites of P argentipes

A significant percentage reduction in antibody levels to PASM, indicative of P argentipes exposure intensity, was observed among intervention group participants at 6 and 12 months postintervention as compared with baseline and the control group (Figure 3). In the intervention group, the estimated marginal mean of antibody levels to PASM decreased by 28.81% at 6 months postintervention (P = .00002) and by 44.91% at 12 months (P = .00001; Table 1). In contrast, no significant differences (P > .05) in anti-PASM levels were observed among control group participants.

Figure 3.

Reduction in IgG levels measured against the Phlebotomus argentipes salivary marker after the LLIN intervention. Participants in the intervention and control groups, using LLINs or untreated bed nets, respectively, were tested against P argentipes salivary marker by indirect ELISA. Serum samples were collected at baseline and 6 and 12 months postintervention. The cutoff value for the indirect ELISA was 0.082, as determined by a receiver operating characteristic curve analysis. Data are presented as marginal mean (SD) OD value for each group at each time point. A 2-tailed paired t test was performed. *P < .05. Abbreviations: ELISA, enzyme-linked immunosorbent assay; LLIN, long-lasting insecticide-treated net; OD, optical density.

CL Case Incidence in the Intervention and Control Sites During the Postintervention Study Period

The number of CL cases occurring from October 2022 to August 2024 were documented. Interestingly, after the LLIN intervention, 6 CL cases were reported and confirmed in the control group, as compared with only 1 case in the intervention group.

LLINs Reduced P argentipes Sand Fly Density in the Postintervention Phase

P argentipes counts fluctuated seasonally for CDC-LT and CBNT (Figure 4A and 4B). There were no consistent differences between intervention and control clusters for either indoor or outdoor CDC-LT captures preintervention, with intervention clusters showing significantly higher captures in March and April 2022 (Supplementary Table 3, Figure 4A). Postintervention, a significant reduction was observed in indoor and outdoor P argentipes counts as compared with their control groups. The CDC-LT captures remained stable in the control groups and then declined sharply in August 2023. To assess the overall impact of LLINs, we performed a simple linear regression analysis comparing the slopes of intervention vs control groups for indoor and outdoor captures (Supplementary Figure 2A and 2B). Over the intervention study period, the slopes were significant for indoor captures (P = .0025) but not outdoor (P = .5533).

Figure 4.

LLINs reduced Phlebotomus argentipes indoor sand fly density after the LLIN intervention. A, Indoor and outdoor CDC light trap collections of P argentipes sand flies. B, Cattle-baited net trap collections of P argentipes sand flies. C, Climatic changes during the study period from January 2022 to August 2024 show average monthly temperature, rainfall, and relative humidity. Data are presented as mean (SD). A and B, Two-way analysis of variance with Tukey multiple-comparison test was performed. *P < .05. **P < .01. ****P < .0001. Abbreviations: CBNT, cattle-baited net trap; CDC, Centers for Disease Control and Prevention; LLIN, long-lasting insecticide-treated net.

The overall preintervention CBNT collections of P argentipes flies (January–July 2022) were higher than the postintervention collections (October 2022–August 2024; Supplementary Table 4), and density peaked in July 2022 with the highest counts of blood-fed and gravid females, although no significant differences in CBNT counts were noted between intervention and control clusters. Similar to CDC-LT captures, a marked decline in P argentipes was observed in August 2023 (Figure 4B).

Of note, high vector counts were recorded during periods with ambient temperatures around 28 °C and low rainfall (Figure 4C). However, a marked decrease in overall sand fly collections was observed after August 2023, coinciding with increases in average rainfall and humidity despite a stable temperature. A prolonged and unseasonably high rainfall from August 2023 to February 2024 may account for the low number of sand flies collected throughout our study during 2024 (Figure 4A and 4B).

The generalized linear model assessed intervention efficacy by measuring sand fly density reduction via CBNT and CDC-LT collections over 24 months. No significant reduction (P > .05) was found in CBNT or outdoor CDC-LT collections. However, indoor CDC-LT collections in intervention clusters showed a sustained P argentipes reduction, starting at 77.31% (2 months, P = .00001), peaking at 89.34% (10 months, P = .00001), and declining to 63.11% (24 months, P = .00003). No significant changes (P > .05) were observed in control clusters (Supplementary Table 5).

Effect of LLIN on P argentipes Mortality With the Cone Bioassay

Baseline cone bioassays showed 100% knockdown at 1 hour and 100% Abbott-corrected mortality at 24 hours (Figure 5). Over the 24-month follow-up, both rates declined, reaching 81.54% and 90.43%, respectively, at 24 months postintervention.

Figure 5.

Abbott-corrected knockdown rate at 1 hour and mortality rate at 24 hours for Phlebotomus argentipes exposed to insecticide-treated netting surfaces. The y-axis represents the count of P argentipes that were knocked down or found dead at each time point, expressed as a percentage of the total sand flies exposed to the insecticide. Error bars indicate the standard deviation (SD) from the three independent replicates. The threshold mortality rate (80%) is indicated by the dashed line.

DISCUSSION

This study represents Sri Lanka's pioneering effort to assess the efficacy of LLINs in controlling leishmaniasis vectors, as assessed by a PASM of vector exposure. In this study, we observed that LLINs significantly reduced exposure to P argentipes bites in participants from the intervention group as compared with controls, who used untreated normal bed nets. Additionally, there was a notable and sustained reduction in the indoor density of P argentipes in intervention clusters using LLINs as compared with control clusters using untreated nets over the intervention period. A weaker but significant reduction was also noted for outdoor captures. This significant percentage reduction underscores the efficacy of LLINs in providing enhanced protection against leishmaniasis vectors.

PASM was used to evaluate LLIN effectiveness, as it was previously validated to assess P argentipes exposure in endemic populations in India [27] and Sri Lanka [28]. Its performance supports PASM as a reliable tool to monitor human-vector contact in the ISC region. A minor nonsignificant antibody reduction in the control group likely reflects natural seasonal variation. In comparison, a study based on salivary gland homogenate in India and Nepal reported modest LLIN-induced reductions (12% at 12 months, 9% at 24 months) [21]. PASM, targeting 2 immunogenic proteins, offers improved specificity and sensitivity and lower background interference [27, 28]. Sustained low antibody levels in the intervention group over 12 months indicate effective reduction in P argentipes exposure.

In addition to measuring direct vector-human contact, our study provides a more comprehensive evaluation of the efficacy of LLINs by incorporating sand fly density and mortality rates and CL incidence, establishing their impact on a broader scope. The sand fly density results reported in this study align with regional studies conducted to assess the efficacy of LLINs in Bangladesh, where reductions in P argentipes density of 60% to 70% were observed up to 18 and 22 months after LLIN interventions [20, 36]. Other studies conducted in Bangladesh observed reductions in sand fly density between 62% and 82% but for periods of 3 to 12 months postintervention [22, 37]. A multicountry trial in Bangladesh, India, and Nepal showed LLIN effectiveness only up to 9 months postintervention [38]. A paired CRT in India and Nepal reported a 25% reduction in P argentipes density over 24 months, with no significant differences in seroconversion or new L donovani infections between groups [39]. These studies show varied outcomes regarding LLIN efficacy in reducing P argentipes density in the ISC, although climatic or seasonal influences were not considered [39].

The efficacy of LLINs in reducing sand fly density is influenced by factors such as study design, cohort size, seasonal and geographic changes, and community participation. Factors such as personal use, washing practices, insecticide type, and mesh size also play key roles. For example, a study in Nepal compared LLINs with mesh sizes of 156 and 625 holes/sq in, showing a mortality rate of 61% for the former and 78% for the latter [40]. A study in India found a 74.29% reduction in indoor P argentipes density with a mesh size of 156 holes/sq in, 91.90% with 196 holes/sq in, and 93.67% with 196 holes/sq in and a 75-cm border [41]. Our LLINs (56 holes/sq in) had a coarser mesh than typical LLINs and control nets but still significantly reduced indoor P argentipes density over 24 months while enhancing airflow and comfort in hot climates.

Due to variations in LLIN efficacy, countries in the ISC have started using a combination of control methods. In Bangladesh, environmental management reduced sand fly density by 72.4%, LLINs (PermaNet 2.0) by 42%, and IRS by 43.7% [42]. A study combining IRS and LLINs in Bangladesh showed a 70%–80% reduction at 4 to 5 months, decreasing to 60% at 11 months [43], and the efficacy of combined methods is further supported by a CRT that evaluated the same [20]. In India, combining LLINs (PermaNet 3.0) and IRS reduced P argentipes density by 93.59% and halted sand fly reemergence for up to 12 months [44].

The P argentipes killing effect in the bioassay cone test in our study lasted up to 24 months with a satisfactory knockdown rate of 81.54% at 1 hour and a mortality rate of 90.43% at 24 hours, demonstrating the efficacy of the 2% permethrin in LLINs in combating sand flies. This exceeded PermaNet's efficacy with 85-mg/m² deltamethrin, where >80% mortality lasted only 12 to 18 months [20, 36, 38], highlighting the impact of insecticide type and concentration on LLIN longevity and effectiveness.

This study is the first CRT evaluating LLIN efficacy against P argentipes in Sri Lanka. A key strength is the use of a community-based design with geographically distinct intervention and control groups, minimizing contamination between them. Additionally, the integration of entomologic, serologic (PASM), and epidemiologic (CL incidence) data provides a multifaceted assessment of intervention impact.

One limitation is that, due to the low number of reported CL cases, our sample size may have been insufficient to detect modest differences in incidence. While the 24-month follow-up was adequate for entomologic and serologic monitoring, it may still be too short to assess long-term disease incidence. As inferred from literature, at $10 per net, LLINs are highly cost-effective when compared with alternatives such as durable wall linings ($50) and insecticidal wall paint ($30) [37, 38]. We note the absence of direct comparison between interventions and emphasize the need for future comparative studies. We acknowledge that the lack of geographic overlap between intervention and control units may introduce spatial confounding.

In conclusion, our study highlights the significant impact of LLINs in controlling P argentipes sand flies in Sri Lanka and underscores their advantage over untreated nets. Using PASM antigen to assess efficacy of intervention strengthens these conclusions and underscores the value of scalable, cost-effective solutions for regional and global leishmaniasis management. Moreover, PASM highlights the importance of developing reliable surveillance tools to effectively monitor the progress of interventions. Importantly, our findings are expected to not only inform Sri Lanka's vector control strategies but also provide valuable insights for the ISC region and globally where leishmaniasis remains a major public health concern.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online (http://jid.oxfordjournals.org/). Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author.

Notes

Acknowledgment. The authors express their gratitude to Mr D. Sunil Shantha, Mr M. P. Ariyapala, and Mr Anura Mahakumara for their essential field support and to the Department of Parasitology, Faculty of Medicine, University of Colombo, for providing logistical assistance.

Author contributions. S. B. P. conducted the experiments and field surveys, analyzed the data, and wrote the manuscript. N. S., S. S., E. I., J. G. V., S. K., and N. D. K. conceptualized the study, interpreted the data, and revised the manuscript. S. S. and N. S. led and oversaw the field study design. E. I., J. G. V., and S. K. supplied the recombinant composite antigen for the research. N. N. P. conducted the statistical and mathematical modeling. N. D. K. provided overall project supervision, managed collaborations, interpreted the data, and edited the manuscript. All authors reviewed the findings and approved the final manuscript.

Data availability statement. Data supporting the conclusions of this article are included within the article. The datasets used and/or analyzed during the present study are available from the corresponding author upon reasonable request.

Disclaimer. The views expressed in the content are those of the authors and do not necessarily reflect the official position of the National Institutes of Health.

Ethics statement. This study obtained ethics approval (EC-22-125) from the Ethics Review Committee of the Faculty of Medicine, University of Colombo, Sri Lanka.

Financial support. This study was funded by the National Institute of Allergy and Infectious Diseases (NIAID), National Institutes of Health (NIH), under Award Number: U01AI136033; This reseasrch was also supported in part by the Intramural Research Program of NIH, NIAID (ZIAAI00133-04). The contributions of the NIH authors were made as part of their official duties as NIH federal employees, are in compliance with agency policy requirements, and are considered works of the United States Government. However, the findings and conclusions presented in this paper are those of the authors and do not necessarily reflect the views of the NIH or the U.S. Department of Health and Human Services.