Skip to main content
NCBI home page
Bulletin of the World Health Organization logoLink to Bulletin of the World Health Organization
. 2025 May 3;103(6):392–402. doi: 10.2471/BLT.24.292505

Testing indoor residual spraying coverage targets for malaria control, Bioko, Equatorial Guinea

Test des objectifs de couverture de pulvérisation intradomiciliaire d’insecticide à effet rémanent dans la lutte contre le paludisme à Bioko, Guinée équatoriale

Evaluación de los objetivos de cobertura del rociamiento intradomiciliario para el control del paludismo en Bioko, Guinea Ecuatorial

اختبار أهداف تغطية بقايا الرش في الأماكن المغلقة لمكافحة الملاريا، بيوكو، غينيا الاستوائية

对赤道几内亚比奥科岛疟疾防治中室内滞留喷洒覆盖率指标的检测

Оценка целевых показателей охвата распыления инсектицидов остаточного действия в помещениях для борьбы с малярией – о. Биоко, Экваториальная Гвинея

Guillermo A García a,✉, Dianna E B Hergott b, David S Galick d, Olivier Tresor Donfack d, Liberato Motobe Vaz d, Lucas O Nze Nchama d, Jeremías N Mba Eyono d, Restituto M Nguema Avue d, Matilde Riloha Rivas e, Marcos M Iyanga d, Faustino E Ebang Bikie d, Teresa A Ondo Mifumu d, Wonder P Phiri d, Michael E von Fricken f, Robert C Reiner Jr c, David L Smith c, Carlos A Guerra a
PMCID: PMC12161160  PMID: 40511394

Abstract

Objective

To test 50% indoor residual spraying coverage (percentage of households sprayed) for non-inferiority against the recommended 80% coverage for malaria control.

Methods

Indoor residual spraying was done in 2021 and 2022 on Bioko, Equatorial Guinea, in a control arm (80% coverage) and intervention arm (50% coverage) with 37 clusters each. We assessed malaria infection in a representative sample of the population during annual surveys using rapid diagnostic tests. We compared the change in the odds of Plasmodium falciparum infection between baseline and post-intervention using difference-in-differences analysis within a survey-weighted binomial generalized linear model. Given differences between the arms at baseline, we adjusted the model for indoor residual spraying coverage at baseline.

Findings

Relative to baseline, the odds of malaria infection post-intervention were 1.11 (95% confidence interval, CI: 0.81–1.52) in the 80% arm and 0.97 (95% CI: 0.72–1.29) in the 50% arm. In the adjusted model, the change in the odds of P. falciparum infection was no greater in the intervention arm than in the control arm (odds ratio: 0.89; 95% CI: 0.58–1.36), with the upper CI being lower than the non-inferiority margin of 1.43.

Conclusion

There was no evidence that 50% coverage was inferior in preventing malaria, which supports the use of this target in settings where this level makes indoor residual spraying feasible by increasing the cost–effectiveness and equity of the intervention.

Introduction

Indoor residual spraying has been widely used as a key malaria intervention. Early indoor residual spraying campaigns with dichlorodiphenyltrichloroethane contributed to the rapid elimination of malaria from the last areas in southern United States of America.1 This spraying was also credited for substantial reductions in malaria in other countries before and during the Global Malaria Eradication Programme (1955–1969).2–4 Although large-scale indoor residual spraying at the time was limited across most of Africa,5 pilot projects showed significant reductions in malaria transmission4,6 and provided evidence for scaling up the intervention in the 21st century.7

Indoor residual spraying is most beneficial when applied to multiple neighbouring households. Coverage, that is the percentage of households sprayed in a given area, is key to achieve community-wide protection. Universal coverage with indoor residual spraying, where all households in a target population are sprayed, was the original aim for malaria control programmes,8 but this target is unrealistic.

Indoor residual spraying is expensive and costs often prevent universal coverage. The implementation costs are due to the logistics of indoor residual spraying,9,10 including substantial investment in personnel and equipment.11 Additionally, insecticides are expensive, representing up to 50% of the overall cost.11,12 This cost has increased since 2012, when, to preserve the efficacy of pyrethroid-based insecticidal nets, the World Health Organization (WHO) called for phasing out the relatively inexpensive pyrethroid insecticides.13 The use of new formulations has led to an almost tenfold increase in cost of indoor residual spraying,12 which many countries cannot afford.7 Thus, there is a need to re-examine recommendations that can improve their feasibility.

One approach is to optimize coverage targets that use spillover effects, where people who do not receive the intervention still benefit from its effects.14 Given spillover, less than 100% coverage would still protect people whose houses were not sprayed, improving the cost–effectiveness of the intervention.14 Under this premise, WHO recommends that the indoor residual spraying coverage target be at least 80%.15 However, the basis of this recommendation is unclear.

Evidence shows the effect of indoor residual spraying on the risk of acquiring malaria, although few high-quality trials allow measuring effect sizes,16,17 and these trials have only evaluated spraying at high coverage (≥ 80%).17 Observational data provide mixed results of the effect of different levels of coverage. Data from Equatorial Guinea showed a decrease in the risk of malaria infection in individuals inhabiting an area with indoor residual spraying coverage ≥ 80% compared with people living in areas with poor coverage (< 20%), but no effect for people living in areas with medium coverage (50–79%).18 However, in Malawi, the same study showed a decreased risk of infection in people in areas with 50–79% coverage relative to people in low-coverage areas (< 20%), but no significant protection was observed with higher coverage (≥ 80%).18 In Madagascar, in areas where malaria prevalence in children younger than 5 years of age ranged between 5% and 14%, a significant reduction in incidence was seen in areas with very high spraying coverage (86–90%) relative to areas with ≤ 85% coverage, although the observed effect was not robust to sensitivity analyses and disappeared at higher coverage (> 90%).19 In a high malaria transmission area of Zambia, a reduction of 4–5% was observed with each 10% increase in spraying coverage, depending on the season, but it is not clear if increasing coverage had diminishing returns on this effect.20

These studies mostly support the need for high coverage, but cannot recommend the optimal coverage to trigger community protection from spillover and eventually suppress transmission.21 Such information can help malaria programmes weigh the benefits and costs of indoor residual spraying and optimize resources while maximizing the effect of spraying.10

We examined indoor residual spraying coverage targets on Bioko, Equatorial Guinea, from 2020 to 2022. Between 2004 and 2014, the Bioko Island Malaria Elimination Project supported the National Malaria Control Programme of Equatorial Guinea in delivering yearly indoor residual spraying island-wide. From 2015 to 2020, only the highest transmission areas were sprayed, leaving large fractions of the population uncovered. By 2019, the gains in reducing malaria prevalence had stalled, suggesting that spraying more areas may again be needed. However, budget constraints made spraying the whole island at ≥ 80% coverage unfeasible. Our recent dose–response analyses of observational data from Bioko showed the indoor residual spraying spillover effect could occur at 30% coverage with no significant gains thereafter.22 This finding prompted us to investigate a lower coverage that better balanced impact and intervention equity.

Methods

Study design

We conducted an operational study using two arms to test the non-inferiority of 50% indoor residual spraying coverage (intervention) against the WHO-recommended 80% coverage (control). The outcome was change in the odds of Plasmodium falciparum infection in the intervention arm relative to the control, expressed as odds ratios (OR). We followed the WHO reporting guidelines for operational research.23

Setting

Bioko has a population of about 270 000, around 80% of whom live in Malabo, the country capital.24 Malaria transmission is perennial, although higher between June and December. Before malaria control, transmission rates were among the highest for any malaria-endemic setting.25 Anopheles funestus and An. gambiae sensu stricto were the main vectors, but were eliminated after indoor residual spraying was introduced in 2004.26–29 Currently, the main vector is An. coluzzii with a small contribution to transmission by An. melas. Despite considerable gains in the past 20 years,26,27,30–32 transmission remains stable across the island with occasional outbreaks.32

In 2021, we assigned almost all inhabited areas of Bioko to 74 clusters, randomly allocated (1:1) to control or intervention coverage (Fig. 1 and Fig. 2). Clusters were based on primary sampling units used in annual malaria indicator surveys, which were assigned to either an urban (34 clusters, 16 in control and 18 in intervention areas) or rural (40 clusters, 21 in control and 19 in intervention areas) stratum. We defined urban as having both high population density and relatively low residual transmission.33 We calculated the number of clusters based on a median number of 108 individuals per cluster at baseline (in 2020) and an estimated intracluster correlation coefficient of 0.09 with 80% power and α = 0.05. The number of clusters was influenced by the feasibility of spraying primary sampling units. Based on the above-mentioned calculations, 37 clusters per arm was enough to detect a change of 6–7% in P. falciparum parasite rate in the intervention group. We randomly assigned clusters and checked the number of households in each arm after each iteration. We chose the randomization that showed the best balance of spraying workloads to optimize deployment and facilitate implementation. We adopted a so-called fried-egg design, with the so-called yolk of each cluster located at least 300 m away from its border.34,35 We based analyses on data within these segregated areas to reduce the likelihood of contamination between control and intervention arms.

Fig. 1.

Distribution of clusters, arms and yolks, Bioko, Equatorial Guinea, 2021–2022

Notes: Clusters were based on primary sampling units of areas about 1 x 1 km. Populated, 100 x 100 m sectors (small squares in the right panel) received either 80% (purple) or 50% (yellow) spraying coverage. Cluster yolks, based on a distance of at least three sectors (300 m) from the closest sprayed sector in the neighbouring cluster, are shown in dark grey. The red box roughly corresponds to the city of Malabo.

Fig. 1

Open in a new tab

Fig. 2.

Households sprayed and malaria indicator surveys sample size by year, Bioko, Equatorial Guinea, 2021–2022

Notes: Percentages represent indoor residual spraying coverage. The number of female participants sampled by arm and within yolk is included for reference.

Fig. 2

Open in a new tab

We derived baseline data for indoor residual spraying coverage (percentage of households sprayed) and P. falciparum infection (as a binary variable) from the indoor residual spraying round and malaria indicator survey in 2020 (Table 1). We calculated the parasite rate as the proportion of P. falciparum-positive individuals. We balanced risk factors for malaria infection, including age, sex, bednet use and history of off-island travel,36 between arms. Variables were balanced at baseline, except the parasite rate (higher in the intervention arm in rural clusters; Fig. 3) and median spraying coverage (higher in the control arm; Fig. 4).

Table 1. Baseline characteristics of the study population, by study arm, Bioko, Equatorial Guinea, 2020.

Characteristic Control arm (80% coverage)
(n = 6047)		 Intervention arm (50% coverage)
(n = 6157)
Median age (range), years 18 (1–92) 18 (1–104)
Female participants, no. (%) 3242 (53.6) 3267 (53.1)
P. falciparum parasite rate, % (95% CI)
Total 14.0 (11.3–16.6) 15.8 (11.0–20.6)
Females 12.6 (9.3–16.0) 14.6 (10.3–19.0)
Urban stratum 12.4 (9.7–15.1) 13.5 (8.7–18.3)
Rural stratum 18.5 (13.3–23.7) 28.4 (22.3–34.5)
Slept under a net the previous night, no. (%) 2386 (39.5) 2413 (39.2)
Travelled to mainland, no. (%) 111 (1.8) 110 (1.8)
Inhabited houses, no. 34 853 36 951
Median indoor residual spraying coverage, % households sprayed 57.8 31.2
Open in a new tab

CI confidence interval; P.: Plasmodium.

Notes: Data are based on the 2020 malaria indicator survey and indoor residual spraying round. Estimates of the parasite rate are weighted by malaria indicator survey design. The low rate of travel in 2020, relative to previous years,33 was due to coronavirus disease 2019 travel restrictions. The number of inhabited houses by arm varies year on year as the population tends to move houses frequently, or houses are built or destroyed.

Fig. 3.

Fig. 3

Open in a new tab

Mean Plasmodium falciparum parasite rate, by year and study arm, Bioko, Equatorial Guinea, 2021–2022

CI: confidence interval.

Fig. 4.

Cluster-level indoor residual spraying coverage, by year and study arm, Bioko, Equatorial Guinea, 2021–2022

Notes: The thick black line within each box is the median coverage and the lower and upper box bounds is the interquartile range. The range is represented by the horizontal lines of the whiskers and the empty circles show outliers or data points at least 1.5 times under or over the first and third quartiles.

Fig. 4

Open in a new tab

Implementation

We did two consecutive indoor residual spraying rounds (February–July 2021 and February–June 2022). We used the Bioko Island Malaria Elimination Project’s spatial decision support system10,37 for implementation of the study. We used 100 × 100 m sectors as the spatial units for deployment and coverage monitoring.10

We used a mix of non-pyrethroid insecticides according to available stock. In 2021, 43 314 houses were sprayed: 40 522 (93.6%) with clothianidin (SumiShield® 50WG, Sumitomo Chemical Co. Ltd, Japan); 1941 with pirimiphos-methyl (4.5%; Actellic® 300CS, Syngenta, Switzerland); and 851 (2.0%) with Fludora® Fusion (Bayer S.A.S., France), a combination of clothianidin and deltamethrin. In 2022, 41 452 houses were sprayed using exclusively Fludora® Fusion. Local insecticide resistance monitoring data showed mosquitoes remained highly susceptible to all insecticides used.

Variables and data sources

We measured the primary outcome (P. falciparum infection) during malaria indicator surveys conducted in the rainy season (August and September) of the same years. In each survey, we used stratified cluster sampling of 5% of inhabited households in urban clusters and 25% in rural clusters to obtain an overall sample size about 5% of the island population (Fig. 2). We tested all consenting individuals in selected households for malaria infection (CareStart™ Pf/PAN (HRP2/pLDH) Ag Combo RDT, Access Bio, Somerset, USA). Individuals positive for malaria received a course of antimalarial medicines as per national guidelines.38 The parasite rate estimates presented here are survey-weighted estimates of individuals tested within the yolk of each cluster.

To determine spraying coverage, we calculated community coverage around each household (household-level coverage) as the fraction of homes sprayed within a 300 m radius of each household.34,35 We averaged household-level coverage for each cluster to provide cluster-level coverage and standardized for use in the models. We defined per protocol coverage as a range of 70–100% in the control and of 45–60% in the intervention arm, allowing for expected variability in coverage during implementation.

Analyses

We used a difference-in-differences analysis to compare changes in odds of plasmodium infection post-intervention and between arms. We used a survey weighted binomial generalized linear model. The model contained three basic terms. First, there was an arm term representing the difference in odds of infection at baseline between the control and intervention arms. To investigate non-inferiority in a setting where there is perfect balance at baseline, we would expect this term to be close to 0. Second, we had a time term, reflecting the difference between the odds of infection in the control arm before (2020) and after (2022) the intervention. Third, we had a difference-in-differences term for the interaction between arm and time, thus providing an estimate of the difference in odds of infection over time between the intervention and control arms. We used the difference-in-differences term to estimate the effect size of the intervention relative to the control arm and its uncertainty to assess non-inferiority. With no difference between the intervention and control arms, the odds ratio of change in infection would be 1 (i.e. difference-in-differences term = 0). We used a non-inferiority margin of 1/0.7 = 1.43,39 meaning if the upper confidence interval (CI) was lower than this margin, we would conclude non-inferiority. Given differences in pre-intervention spraying coverage at baseline, we adjusted the model using a baseline coverage term corresponding to coverage in 2020 (Table 1).

We did secondary analyses of the impact of actual intervention coverage to account for deviations from the per protocol coverage range. Additionally, we explored the differences in impact by stratum and investigated the effect size at different levels of baseline coverage (online repository).40

We used R version 4.4.0 (R Foundation, Vienna, Austria) for analyses, and the survey package41 to generate survey weights according to the sampling strategy of malaria indicator surveys.

Ethical considerations

This study was part of routine malaria control activities and operational research on Bioko. All interventions within the Bioko Island Malaria Elimination Project and National Malaria Control Programme of Equatorial Guinea, including vector-control activities, are approved by the Ministry of Health and Social Welfare of Equatorial Guinea through the presentation of annual work plans, which include ethical considerations.

The indoor residual spraying implementation strategy was approved within these routine work plans, with regular progress updates presented to the health ministry, donors and other stakeholders. The health ministry considered that the strategy was non-experimental as Bioko routinely undergoes changes in indoor residual spraying target areas as part of normal operations. No households were denied the intervention. As our study constituted operational research within routine malaria control activities rather than an intervention or cluster-randomized trial as defined by WHO and the scientific community, it was not registered in a clinical trials registry.

The malaria indicator surveys have ethical approval from the health ministry and the London School of Hygiene and Tropical Medicine granted at the start of the Bioko Island Malaria Elimination Project. As part of the protocol, all individuals found positive for malaria receive antimalarial treatment according to national guidelines.

While malaria control on Bioko is funded by a public–private partnership between the government and a consortium of oil companies, the funders had no influence on our study design and analyses.

Results

Fig. 2 summarizes the numbers of individuals sampled and the overall indoor residual spraying coverage by arm and year. Age and sex of the population sampled in 2021 and 2022 were similar to baseline (Table 1).

Table 2 and Fig. 3 give the mean survey weighted P. falciparum parasite rate measured before (2020), during (2021) and after (2022) the intervention. The rate was higher in the rural stratum and significantly so in the intervention arm, peaking in 2021. In the urban stratum, mean prevalence was not statistically different between arms and relative to the baseline.

Table 2. Plasmodium falciparum parasite rate and indoor residual spraying coverage in control and intervention arms, by stratum and year, Bioko, Equatorial Guinea, 2021–2022 .

Stratum 2021
 2022
Control Intervention Control Intervention
Rural
Median indoor residual spraying coverage, % (IQR) 81.6 (79.1–83.9) 67.9 (56.4–78.9) 82.1 (75.0–86.1) 56.3 (52.0–63.5)
Mean parasite rate, % (95% CI) 18.8 (13.5–24.1) 31.7 (23.2–40.1) 18.5 (14.5–22.4) 29.2 (22.5–35.9)
Urban
Median indoor residual spraying coverage, % (IQR) 78.0 (72.1–79.9) 54.8 (51.5–58.8) 73.0 (64.9–77.7) 51.3 (50.6–52.5)
Mean parasite rate, % (95% CI) 13.0 (10.0–16.0) 13.3 (8.8–17.8) 14.3 (10.8–17.8) 12.8 (8.8–16.8)
Open in a new tab

CI: confidence interval; IQR: interquartile range.

Note: Data were determined by the 2021 and 2022 malaria indicator surveys and indoor residual spraying rounds.

Fig. 4 shows the highly heterogeneous spraying coverage at baseline as a result of the targeting strategy; many clusters received no spraying, particularly in the intervention arm. During the study, only two clusters received < 40% coverage (33.7% and 37.9%, both in the intervention arm and the rural stratum in 2022), with coverage narrowly distributed, especially in the urban stratum (Fig. 4 and Fig. 5). Per-protocol coverage was achieved in many clusters, although it proved operationally challenging in others (Table 3). Coverage in the control arm was more easily attained in rural clusters, with little variation around the 80% mark, particularly in 2021. This coverage was harder to achieve in urban clusters, particularly in 2022, when only 11 of 16 clusters (68.8%) were within the per protocol coverage range. In contrast, intervention-arm coverage was easier to realize in urban clusters (Fig. 5), notably so in 2022 when all 18 clusters were within range (Table 3). In the rural stratum, many intervention clusters received more spraying than per protocol, especially in 2021, when only six of 19 clusters (31.6%) were within range and median coverage was 67.9% (Table 3 and Fig. 4).

Fig. 5.

Indoor residual spraying coverage in urban Malabo, by year, Bioko, Equatorial Guinea, 2021–2022

Notes: The inset in the 2020 map shows the location of Malabo on Bioko. The lined areas are the clusters. Coverage is expressed as the average household-level coverage by sector using a 300 m buffer.

Fig. 5

Open in a new tab

Table 3. Clusters within range of per protocol coverage, by arm and year, Bioko, Equatorial Guinea, 2021–2022.

Year No. (%)
Control arm Intervention arm
2021
Rural 20/21 (95.2) 6/19 (31.6)
Urban 13/16 (81.3) 14/18 (77.8)
2022
Rural 19/21 (90.5) 8/19 (42.1)
Urban 11/16 (68.8) 18/18 (100.0)
Open in a new tab

Note: We defined an acceptable range as between 70% and 100% for the control arm and between 45% and 60% for the intervention arm.

The odds of malaria infection at the end of the study (2022) were not significantly different from baseline (2020) in both arms: 1.11 (95% CI: 0.81–1.52) in the control and 0.97 (95% CI: 0.72–1.29) in the intervention arms. The difference-in-differences term in both models was not significant: base model OR: 0.87 (95% CI: 0.57–1.31); adjusted model OR: 0.89 (95% CI: 0.58–1.36), with the 95% CI overlapping the no difference line and not exceeding the 1.43 non-inferiority margin (Fig. 6). We did additional model iterations and found no evidence of inferiority (online repository).40.

Fig. 6.

Difference-in-differences analysis of the effect of 50% versus 80% indoor residual spraying coverage on malaria infection odds, Bioko, Equatorial Guinea, 2020–2022

CI: confidence interval; OR: odds ratio.

Notes: The OR of the difference-in-differences term is illustrated with uncertainty.

Fig. 6

Open in a new tab

Discussion

We compared the spillover effect of high (≥ 80%) versus medium (50%) indoor residual spraying coverage and found no significant difference in P. falciparum infection odds between the two arms, indicating no evidence of 50% coverage being inferior. In fact, the mean model estimates and lower CI were inclined towards superiority of 50% coverage. This finding can be explained by the lower baseline coverage in some intervention clusters compared with control clusters, particularly in urban areas, where receiving a medium coverage had a higher relative effect (online repository).40

The data in our study were collected and recorded at the household level and linked to a robust, error-controlled, geographically accurate spatial decision support system.10,37 As such, data were unambiguously assigned to a location, which eliminated recall bias when reporting house-spraying status. This system also allowed monitoring and guiding of spray teams, thus giving a precise representation of coverage. Furthermore, we conducted the study under real-world conditions in an operational context.

We originally aimed to use a non-inferiority, cluster randomized trial design. However, operational constraints resulted in two important limitations. First, differences existed in baseline spraying coverage as described earlier. To account for the differences, we adjusted the model for baseline coverage, although we observed no significant change in the impact of the intervention. The difference in baseline coverage could also partly explain the difference in baseline P. falciparum infection particularly noticeable in rural clusters. When running the base model separately in low baseline coverage communities (< 45%), the observed impact of the intervention leant even further towards superiority. Conversely, running the model in high baseline coverage communities (≥ 70%) inclined the effect towards inferiority, although the overall result was inconclusive (online repository).40 Second, while we aimed to achieve per protocol coverage, it would have been unethical not to spray in the intervention arm if someone wanted it. This situation resulted in many rural clusters in the intervention arm receiving higher than 50% coverage (Table 3). At the same time, spraying at 80% coverage in urban areas had challenges, including large numbers of houses to spray and higher refusal rates, resulting in suboptimal coverage in the control arm. In a sensitivity analysis, we adjusted the model with actual coverage achieved during the study as an intention-to-treat covariate but observed no difference in the modelled outputs (online repository).40 These limitations led to the use of the adjusted difference-in-differences binomial models described rather than standard t-tests used in cluster randomized trials.

Notably, on Bioko, An. coluzzii exhibits a mix of endo- and exophaghic activity. Nevertheless, recent analyses pairing local vector and human behaviours showed indoor biting is still driving most transmission, suggesting indoor residual spraying and long-lasting insecticidal nets are reducing transmission but not interrupting it.42

Our results were not likely confounded by environmental factors, notably rainfall, because the annual rounds of spraying, including the rounds reported here, are timed before the start of the main rains on Bioko. However, the impact of lower targets for indoor residual spraying coverage may differ depending on other local factors. Transmission intensity may influence the effect size of lower indoor residual spraying coverage. Similarly, insecticide resistance of local vectors could affect the optimal spraying coverage required. Another potential confounder is that Bioko is an island, although previous studies showed its high connectivity to mainland Africa through the movement of people.33,43 Investigation of lower targets for indoor residual spraying coverage is needed in other malaria-endemic settings, particularly where mosquito ecology, vector composition, malaria transmission and insecticide resistance differ.

Despite limitations for the generalizability of our findings, they have important implications for vector control. Achieving high indoor residual spraying coverage can be particularly challenging where the population is likely to refuse the intervention or where houses are difficult to access.10 High population density areas are especially hard to spray at high coverage due to the volume of houses. In Bioko, spraying budgeting and planning is based on four households sprayed per sprayer per day.10 Therefore, coverage of 80% in an urban sector with 78 houses requires 15.6 worker days, in contrast to 2.2 worker days in a rural sector with 11 households. Reducing the coverage target to 50% would equate to 9.8 worker days in urban and 1.4 worker days in rural sectors. Spraying urban Malabo is further complicated by high refusal and absenteeism, which slow productivity and require multiple visits. A lower coverage target in Malabo would reduce the problem of refusal. In contrast, spraying productivity in rural sectors is facilitated by easier access, lower refusal and higher demand from the population, so a lower target would have a lower impact. In absolute terms, lower spraying coverage translates into substantial cost and logistics savings in urban Malabo but less so in the rural periphery. This situation is probably true in other malaria-endemic areas and is especially important where financial support for malaria vector control is insufficient.

Other alternatives have been proposed to reduce the burden of indoor residual spraying, such as partial or selective spraying where only parts of the walls are sprayed.44 Much of the effort in spraying is house preparation, which is unaffected if walls are sprayed partially, unless it means restricting spraying to the upper parts of the walls and ceilings.44 More importantly, spray teams must access the house, so intervention refusal would still affect productivity and coverage. Therefore, this alternative does not fully address the main complexities of indoor residual spraying at scale, although it could be complementary in a scenario of lower spraying coverage targets for even higher intervention efficiency. Our study suggests the lower coverage approach is a plausible strategy to expand the intervention area and reach a higher fraction of the population at risk using the same resources, or to save resources that could be reallocated to other interventions. The National Malaria Control Programme of Equatorial Guinea, with evidence from this study, has already adjusted indoor residual spraying coverage targets, which is facilitated by the spatial decision support system.10 A similar approach could bring indoor residual spraying back as an option for malaria control in countries that have moved away from the intervention due to perceptions of ineffectiveness and challenging expectations set by WHO. While innovative interventions, such as vaccines and monoclonal antibodies, promise support for malaria control, they face efficacy and cost–effectiveness issues.45 Driving innovation in the delivery of proven interventions, such as indoor residual spraying, can help overcome current bottlenecks to reducing the malaria burden.

Our findings provide evidence for revising current indoor residual spraying coverage recommendations. While such spraying is an effective vector-control intervention, operational challenges can limit its feasibility. Given the evidence of non-inferiority, we urge stakeholders and decision-makers revisit current standards and coverage targets15 and promote cost–benefit analyses and further research in other settings to validate our findings.

Acknowledgements

We thank the sprayers and surveyors involved in indoor residual spraying and malaria indicator surveys before and during the study. DLS is also affiliated with the Institute for Health Metrics and Evaluation, University of Washington, Seattle, USA

Funding:

Malaria control on Bioko is funded by the Government of Equatorial Guinea in partnership with a consortium of private companies led by Marathon Oil Corporation (a ConocoPhillips company). DLS, DEBH and RCR were funded by a grant from the National Institute of Allergies and Infectious Diseases, Maryland, USA (R01 AI163398). DLS was also funded by a grant from the Bill and Melinda Gates Foundation (INV030600).

Competing interests:

None declared.

References

  • 1.Williams LLJ Jr. Malaria eradication in the United States. Am J Public Health Nations Health. 1963. Jan;53(1):17–21. 10.2105/AJPH.53.1.17 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Expert Committee on Malaria – report of second session. Washington, DC: World Health Organization; 1948. Available from: https://iris.who.int/bitstream/handle/10665/64063/WHO_IC_MAL_25.pdf [cited 2025 Apr 15].
  • 3.Expert Committee on Malaria – report of the fourth session. Geneva: World Health Organization; 1951. Available from: https://iris.who.int/bitstream/handle/10665/40156/WHO_TRS_39.pdf [cited 2025 Apr 15].
  • 4.Mabaso MLH, Sharp B, Lengeler C. Historical review of malarial control in southern African with emphasis on the use of indoor residual house-spraying. Trop Med Int Health. 2004. Aug;9(8):846–56. 10.1111/j.1365-3156.2004.01263.x [DOI] [PubMed] [Google Scholar]
  • 5.Indoor residual spraying: use of indoor residual spraying for scaling up global malaria control and elimination. Geneva: World Health Organization; 2006. Available from: https://iris.who.int/handle/10665/69386 [cited 2025 Apr 15].
  • 6.Kouznetsov RL. Malaria control by application of indoor spraying of residual insecticides in tropical Africa and its impact on community health. Trop Doct. 1977. Apr;7(2):81–91. 10.1177/004947557700700216 [DOI] [PubMed] [Google Scholar]
  • 7.Tangena JAA, Hendriks CMJ, Devine M, Tammaro M, Trett AE, Williams I, et al. Indoor residual spraying for malaria control in sub-Saharan Africa 1997 to 2017: an adjusted retrospective analysis. Malar J. 2020. Apr 10;19(1):150. 10.1186/s12936-020-03216-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Expert Committee on Malaria. sixth report. Geneva: World Health Organization; 1957. Available from: https://iris.who.int/handle/10665/88530 [cited 2025 Apr 15]. [PubMed]
  • 9.DeBoer KR, Vaz LM, Ondo Mfumu TA, Nlang JAM, Ondo L, Riloha Rivas M, et al. Assessing IRS performance in a gender-integrated vector control programme on Bioko Island, Equatorial Guinea, 2010-2021. Malar J. 2023. Oct 25;22(1):323. 10.1186/s12936-023-04755-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.García GA, Atkinson B, Donfack OT, Hilton ER, Smith JM, Eyono JNM, et al. Real-time, spatial decision support to optimize malaria vector control: The case of indoor residual spraying on Bioko Island, Equatorial Guinea. PLOS Digit Health. 2022. May 12;1(5):e0000025. 10.1371/journal.pdig.0000025 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Alonso S, Chaccour CJ, Wagman J, Candrinho B, Muthoni R, Saifodine A, et al. Cost and cost–effectiveness of indoor residual spraying with pirimiphos-methyl in a high malaria transmission district of Mozambique with high access to standard insecticide-treated nets. Malar J. 2021. Mar 10;20(1):143. 10.1186/s12936-021-03687-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Johns B, Haile M. PMI IRS country programs: 2020 comparative cost analysis. Rockville: VectorLink Project, Abt Associates Inc.; 2021. [Google Scholar]
  • 13.Global plan for insecticide resistance management in malaria vectors. Geneva: World Health Organization; 2012. Available from: https://iris.who.int/handle/10665/44846 [cited 2025 Apr 15].
  • 14.Benjamin-Chung J, Arnold BF, Berger D, Luby SP, Miguel E, Colford JM Jr, et al. Spillover effects in epidemiology: parameters, study designs and methodological considerations. Int J Epidemiol. 2018. Feb 1;47(1):332–47. 10.1093/ije/dyx201 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Operational manual on indoor residual spraying: control of vectors of malaria, Aedes-borne diseases, Chagas disease, leishmaniases and lymphatic filariasis. Geneva: World Health Organization; 2023. Available from: https://iris.who.int/handle/10665/375978 [cited 2025 Apr 15].
  • 16.Pluess B, Tanser FC, Lengeler C, Sharp BL. Indoor residual spraying for preventing malaria. Cochrane Database Syst Rev. 2010. Apr 14;2010(4):CD006657. 10.1002/14651858.CD006657.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Pryce J, Medley N, Choi L. Indoor residual spraying for preventing malaria in communities using insecticide-treated nets. Cochrane Database Syst Rev. 2022. Jan 17;1(1):CD012688. 10.1002/14651858.CD012688.pub3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Rehman AM, Coleman M, Schwabe C, Baltazar G, Matias A, Gomes IR, et al. How much does malaria vector control quality matter: the epidemiological impact of holed nets and inadequate indoor residual spraying. PLoS One. 2011. Apr 29;6(4):e19205. 10.1371/journal.pone.0019205 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Hilton ER, Rabeherisoa S, Ramandimbiarijaona H, Rajaratnam J, Belemvire A, Kapesa L, et al. Using routine health data to evaluate the impact of indoor residual spraying on malaria transmission in Madagascar. BMJ Glob Health. 2023. Jul;8(7):e010818. 10.1136/bmjgh-2022-010818 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Ferriss E, Chaponda M, Muleba M, Kabuya JB, Lupiya JS, Riley C, et al. The impact of household and community indoor residual spray coverage with Fludora Fusion in a high malaria transmission setting in Northern Zambia. Am J Trop Med Hyg. 2023. Jun 26;109(2):248–57. 10.4269/ajtmh.22-0440 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Hay SI, Smith DL, Snow RW. Measuring malaria endemicity from intense to interrupted transmission. Lancet Infect Dis. 2008. Jun;8(6):369–78. 10.1016/S1473-3099(08)70069-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Galick DS, Vaz LM, Ondo L, Iyanga MM, Bikie FEE, Avue RMN, et al. Reconsidering indoor residual spraying coverage targets: A retrospective analysis of high-resolution programmatic malaria control data. Proc Natl Acad Sci USA. 2025. Apr 22;122(16):e2421531122. 10.1073/pnas.2421531122 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Hales S, Lesher-Trevino A, Ford N, Maher D, Ramsay A, Tran N. Reporting guidelines for implementation and operational research. Bull World Health Organ. 2016. Jan 1;94(1):58–64. 10.2471/BLT.15.167585 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Fries B, Guerra CA, García GA, Wu SL, Smith JM, Oyono JNM, et al. Measuring the accuracy of gridded human population density surfaces: A case study in Bioko Island, Equatorial Guinea. PLoS One. 2021. Sep 1;16(9):e0248646. 10.1371/journal.pone.0248646 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Cano J, Berzosa PJ, Roche J, Rubio JM, Moyano E, Guerra-Neira A, et al. Malaria vectors in the Bioko Island (Equatorial Guinea): estimation of vector dynamics and transmission intensities. J Med Entomol. 2004. Mar;41(2):158–61. 10.1603/0022-2585-41.2.158 [DOI] [PubMed] [Google Scholar]
  • 26.Sharp BL, Ridl FC, Govender D, Kuklinski J, Kleinschmidt I. Malaria vector control by indoor residual insecticide spraying on the tropical island of Bioko, Equatorial Guinea. Malar J. 2007. May 2;6(1):52. 10.1186/1475-2875-6-52 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Overgaard HJ, Reddy VP, Abaga S, Matias A, Reddy MR, Kulkarni V, et al. Malaria transmission after five years of vector control on Bioko Island, Equatorial Guinea. Parasit Vectors. 2012. Nov 12;5(1):253. 10.1186/1756-3305-5-253 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Molina R, Benito A, Roche J, Blanca F, Amela C, Sanchez A, et al. Baseline entomological data for a pilot malaria control program in Equatorial Guinea. J Med Entomol. 1993. May;30(3):622–4. 10.1093/jmedent/30.3.622 [DOI] [PubMed] [Google Scholar]
  • 29.Molina R, Benito A, Blanca F, Roche J, Otunga B, Alvar J. The anophelines of Equatorial Guinea. Ethology and susceptibility studies. Res Rev Parasitol. 1996. Jan;56:105–10. [Google Scholar]
  • 30.Kleinschmidt I, Torrez M, Schwabe C, Benavente L, Seocharan I, Jituboh D, et al. Factors influencing the effectiveness of malaria control in Bioko Island, equatorial Guinea. Am J Trop Med Hyg. 2007. Jun;76(6):1027–32. 10.4269/ajtmh.2007.76.1027 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Kleinschmidt I, Schwabe C, Benavente L, Torrez M, Ridl FC, Segura JL, et al. Marked increase in child survival after four years of intensive malaria control. Am J Trop Med Hyg. 2009. Jun;80(6):882–8. 10.4269/ajtmh.2009.80.882 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Guerra CA, Fuseini G, Donfack OT, Smith JM, Ondo Mifumu TA, Akadiri G, et al. Malaria outbreak in Riaba district, Bioko Island: lessons learned. Malar J. 2020. Aug 3;19(1):277. 10.1186/s12936-020-03347-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Guerra CA, Kang SY, Citron DT, Hergott DEB, Perry M, Smith J, et al. Human mobility patterns and malaria importation on Bioko Island. Nat Commun. 2019. May 27;10(1):2332. 10.1038/s41467-019-10339-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Gimnig JE, Kolczak MS, Hightower AW, Vulule JM, Schoute E, Kamau L, et al. Effect of permethrin-treated bed nets on the spatial distribution of malaria vectors in western Kenya. Am J Trop Med Hyg. 2003. Apr;68(4) Suppl:115–20. 10.4269/ajtmh.2003.68.115 [DOI] [PubMed] [Google Scholar]
  • 35.Hawley WA, Phillips-Howard PA, ter Kuile FO, Terlouw DJ, Vulule JM, Ombok M, et al. Community-wide effects of permethrin-treated bed nets on child mortality and malaria morbidity in western Kenya. Am J Trop Med Hyg. 2003. Apr;68(4) Suppl:121–7. 10.4269/ajtmh.2003.68.121 [DOI] [PubMed] [Google Scholar]
  • 36.García GA, Janko M, Hergott DEB, Donfack OT, Smith JM, Mba Eyono JN, et al. Identifying individual, household and environmental risk factors for malaria infection on Bioko Island to inform interventions. Malar J. 2023. Mar 1;22(1):72. 10.1186/s12936-023-04504-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.García GA, Hergott DEB, Phiri WP, Perry M, Smith J, Osa Nfumu JO, et al. Mapping and enumerating houses and households to support malaria control interventions on Bioko Island. Malar J. 2019. Aug 22;18(1):283. 10.1186/s12936-019-2920-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Galick DS, Donfack OT, Mifumu TAO, Onvogo CNO, Dougan TB, Mikue MIAA, et al. Adapting malaria indicator surveys to investigate treatment adherence: a pilot study on Bioko Island, Equatorial Guinea. Malar J. 2024. Aug 13;23(1):244. 10.1186/s12936-024-05057-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Technical consultation on determining noninferiority of vector control products within an established class: report of a virtual meeting, 31 August–2 September 2021. Geneva: World Health Organization; 2021. Available from: https://iris.who.int/handle/10665/349446 [cited 2025 Apr 15].
  • 40.García GA, Hergott DEB, Galick DS, Donfack OT, Motobe Vaz L, Nze Nchama LO, et al. Rethinking indoor residual spraying coverage targets: operational research on Bioko Island testing the non-inferiority of 50% against 80% [online repository]. London: Figshare; 2025. 10.6084/m9.figshare.28836299 [DOI]
  • 41.Lumley T, Gao P, Schneider B. Analysis of complex survey samples. Package ‘survey. [internet]. Vienna: The Comprehensive R Archive Network; 2024. Available from: https://cran.r-project.org/web/packages/survey/survey.pdf [cited 2025 Apr 15]. [Google Scholar]
  • 42.Ooko M, Bela NR, Leonard M, Maye VON, Efiri PBE, Ekoko W, et al. Malaria burden and residual transmission: two thirds of mosquito bites may not be preventable with current vector control tools on Bioko Island, Equatorial Guinea. Int J Infect Dis. 2024. Oct;147:107197. 10.1016/j.ijid.2024.107197 [DOI] [PubMed] [Google Scholar]
  • 43.Guerra CA, Citron DT, García GA, Smith DL. Characterising malaria connectivity using malaria indicator survey data. Malar J. 2019. Dec 23;18(1):440. 10.1186/s12936-019-3078-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Coleman S, Yihdego Y, Sherrard-Smith E, Thomas CS, Dengela D, Oxborough RM, et al. Partial indoor residual spraying with pirimiphos-methyl as an effective and cost-saving measure for the control of Anopheles gambiae s.l. in northern Ghana. Sci Rep. 2021. Sep 10;11(1):18055. 10.1038/s41598-021-97138-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Miura K, Flores-Garcia Y, Long CA, Zavala F. Vaccines and monoclonal antibodies: new tools for malaria control. Clin Microbiol Rev. 2024. Jun 13;37(2):e0007123. 10.1128/cmr.00071-23 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Bulletin of the World Health Organization are provided here courtesy of World Health Organization

RESOURCES