Uganda has been exceptionally successful in scaling-up coverage of LLINs. Following the mass distribution campaigns to deliver free LLINs in 2013–14 and 2017–18, 90 and 83% of households, respectively reported ownership of at least one LLIN7,14. However, despite this success, the burden of malaria remains high in much of the country. Uganda had the third highest number of malaria cases reported in 2019, with reported case incidence increasing since 20142. If Uganda is to achieve the goals established by the World Health Organization’s Global Technical Strategy for malaria including reducing malaria case incidence by at least 90% by 2030 as compared with 201515, additional malaria control measures will be needed. This report highlights the critical role of IRS in substantially reducing the burden of malaria in areas where transmission remains high despite deployment of LLINs. Withdrawing IRS after 5 years of sustained use in three districts in northern Uganda was associated with a more than fivefold increase in malaria cases within 10 months. Restarting IRS with a single round in nine districts in Northern Uganda ~3 years after IRS had been stopped was associated with a transient but important (more than a fivefold) decrease in malaria cases within 8–12 months, returning to pre-IRS levels after 23 months. Initiating and sustaining IRS in five districts in Eastern Uganda was associated with a gradual reduction in malaria cases reaching almost a sevenfold reduction after 4–5 years.
Robust evidence supports the widespread use of LLINs for malaria control. In a systematic review of clinical trials conducted between 1987 and 2001, insecticide treated nets reduced all cause child mortality by 17% and the incidence of uncomplicated P. falciparum malaria by almost half16. However, there is concern that the effectiveness of LLINs may be diminishing due to widespread resistance to pyrethroids which until recently were the only class of insecticides approved for LLINs. Similar to many other African countries, high-level resistance to pyrethroids among the principle Anopheles vectors has been reported recently throughout Uganda17,18,19. In addition, behavioral changes in vector biting activity following the introduction of LLINs have been reported which could present new challenges for malaria control20,21,22. Finally, the effectiveness of LLINs may be further compromised by poor adherence and waning coverage in the setting of free distribution campaigns done intermittently. In Uganda, less than 18% of households reported adequate coverage (defined as at least one LLIN per two residents) 3 years after the 2013–14 distribution campaign23 and adequate coverage decreased from 71% to 51% between 6 and 18 months following the 2017–18 distribution campaign24. Although the World Health Organization recommends mass distribution campaigns every 3 years, mounting evidence suggests that LLINs should be distributed more frequently to sustain high coverage25,26,27,28,29,30,31.
Given concerns about the current effectiveness of pyrethroid-based LLINs and the persistently high burden of malaria despite aggressive scale-up of LLINs in countries like Uganda, additional malaria control measures are needed. IRS is an attractive option. Historically, IRS programs were used to dramatically reduce and even eliminate malaria in many parts of the world. Thus, while there is some evidence for the impact of IRS in the absence of LLINs32, it is surprising that the evidence base from contemporary controlled trials on the impact of adding IRS to LLINs for vector control is limited. A recent systematic review of cluster randomized controlled trials conducted in sub-Saharan Africa since 2008, reported that adding IRS using a “pyrethroid-like” insecticide to LLINs did not provide any benefits, while adding IRS with a “non-pyrethroid-like” insecticide produced mixed results5. Among the four trials comparing IRS plus LLINs with LLINs alone, three evaluated IRS with a carbamate (bendiocarb) and one evaluated a long-lasting organophosphate, pirimiphos-methyl (Actellic 300CS®)33,34,35,36. Only two trials (both using bendiocarb) assessed malaria incidence; one from Sudan found a 35% reduction when adding IRS to LLINs34, while another from Benin found no benefit of adding IRS33. All four trials assessed parasite prevalence, with an overall non-significant trend towards a lower prevalence when adding IRS to LLINs (RR = 0.67, 95% CI 0.35–1.28)5. However, when the analyses were restricted to include only the two studies with LLIN usage over 50%, adding IRS reduced parasite prevalence by over 50% (RR = 0.47, 95% CI 0.33–0.67)5. Of note, none of the trials that evaluated the impact of adding IRS with a “non-pyrethroid-like” insecticide assessed outcomes beyond 2 years. More recently, a number of observational studies have reported benefits of using IRS with pirimiphos-methyl (Actellic 300CS®). In the Mopti Region of Mali, delivery of a single round of IRS with Actellic 300CS® was associated with a 42% decrease in the peak incidence of laboratory-confirmed malaria cases reported at public health facilities37. In the Koulikoro Region of Mali, villages that received a single round of IRS with Actellic 300CS® combined with LLINs observed a greater than 50% decrease in the incidence of malaria compared to villages that only received LLINs38. In the Northern Region of Ghana, districts that received IRS with Actellic 300CS® reported 26–58% fewer cases of laboratory-confirmed malaria cases reported at public health facilities over a 2-year period, compared to districts that did not receive IRS39. In Northern Zambia, implementation of IRS with Actellic 300CS® targeting only high burden areas over a 3 year period was associated with a 25% decline in parasite prevalence during the rainy season, but no decline during the dry season40. In Western Kenya, the introduction of a single round of IRS with Actellic 300CS® was associated with a 44–65% decrease in district level malaria case counts over a 10 month period compared to pre-IRS levels41. In addition, several recent reports have documented dramatic resurgences of malaria following the withdrawal of IRS with bendiocarb in Benin42, and the withdrawal of IRS with Actellic 300CS® in Mali and Ghana37,39.
The results from this study provides additional support for the critical role IRS can play in reducing the burden of malaria in African countries with high LLINs coverage. A strength of the study was its use of a large, rigorously collected dataset. Data were collected over nearly 7 years through an enhanced health facility-based surveillance system covering 14 districts in Uganda where IRS was being withdrawn, re-started, and initiated. This enhanced surveillance system facilitated laboratory testing and provided prospectively collected, individual-level data, allowing for analyses of quantitative changes in laboratory-confirmed cases of malaria over time, controlling for temporal changes in rainfall, seasonal effects, diagnostic practices, and health seeking behavior. Previous work by our group documented a marked decrease in malaria TPRs after 4 years of sustained IRS with bendiocarb in one district of Northern Uganda followed by a rapid resurgence over an 18-month period after IRS was withdrawn11. In this study we expand on these findings by including data from three districts and covering a 31-month period following the withdrawal of IRS. We were able to quantify more than a fivefold increase in malaria cases which was sustained over the 10–31 months following the withdrawal of IRS. This marked resurgence occurred despite the fact the first universal LLIN distribution campaign was timed to occur right after IRS was withdrawn. Given the dramatic nature of the resurgence, the Ugandan government was able to procure funding for a single round of IRS with Actellic 300CS® ~3 years after IRS was withdrawn in 10 districts of Northern Uganda. In this study, we assessed the impact of this single round in nine of these districts. This single round was associated with over a fivefold decrease in malaria cases after 8–12 months, with malaria cases returning to pre-IRS levels after almost 2 years. These data suggest that IRS with longer-acting formulations such as Actellic 300CS® administered every 2 years could be considered as a strategy for mitigating the risk of resurgence following sustained IRS and/or enabling countries to expand coverage when resources are limited, but formal assessment and a cost-effectiveness analyses are needed. This study also evaluated the impact of 5 years of sustained IRS in five districts of Eastern Uganda, starting first with bendiocarb and then switching to Actellic 300CS® after 18 months. Rounds of IRS were initially associated with marked decreases in malaria cases followed by peaks before subsequent rounds until the fourth and fifth years after IRS was initiated when there was a sustained decrease of almost sevenfold compared to pre-IRS level. Given the before-and-after nature of our study design, it is not clear whether the maximum sustained benefits of IRS seen after 4–5 years were due to the cumulative effect of multiple rounds of IRS, the switch from bendiocarb to Actellic 300CS®, improvements in implementation (although campaigns occurred regularly and coverage was universally high across rounds, see Supplementary Table 4), the second universal LLIN distribution campaign which occurred in this area in 2017, and/or other factors.
This study had several limitations. First, we used an observational study design, with measures of impact based on comparisons made before-and-after key changes in IRS policy. Although cluster randomized controlled trials are the gold standard study design for estimating the impact of IRS, it could be argued that withholding IRS would be unethical, given what is known about its impact in Uganda. Second, our estimates of impact could have been confounded by secular trends in factors not accounted for in our analyses. However, we feel that our overall conclusions are robust given the large amount of data available from multiple sites over an extended period with multiple complementary objectives providing consistent findings. Third, we could not assess the impact of IRS independent of LLIN use and did not have access to measures of IRS or LLIN coverage from our study populations. It is possible that some of the impacts we observed were from LLIN distributions in combination with IRS campaigns. However, we were able to provide a “real world” assessment of IRS in a setting where LLIN use is strongly supported by repeated universal distribution campaigns that are becoming increasingly common in sub-Saharan Africa. Similarly, we cannot draw conclusions on the impact of different IRS compounds given all sites received the same formulations consecutively. The results from Objective 3 indicate that malaria incidence dropped substantially in the years that districts stopped receiving bendiocarb and began receiving Actellic 300CS®. However, we cannot conclude whether this reduction was a result of this change or rather the cumulative impact of sustained IRS campaigns, as it has been suggested that in very high transmission settings, several years of IRS may be needed to maximize impact on measures of morbidity.43,44 Finally, our study outcome was limited to case counts of laboratory-confirmed malaria captured at health facilities. Thus, we were unable to measure the impact of IRS on other important indicators such as measures of vector distribution, parasite prevalence, or mortality.
There is a growing body of evidence that combining LLINs with IRS using “non-pyrethroid-like” insecticides, especially the long acting organophosphate Actellic 300CS®, is highly effective at reducing the burden of malaria in Uganda, and elsewhere in Africa. Despite these encouraging findings, IRS coverage in Africa has been moving in the wrong direction. The proportion of those at risk protected by IRS in Africa peaked at just over 10% in 2010. However, the spread of pyrethroid resistance has led many control programs to switch to more expensive formulations resulting in a 53% decrease in the number of houses sprayed between years of peak coverage and 2015 across 18 countries supported by the US President’s Malaria Initiative45 and an overall reduction in the proportion protected by IRS in Africa to less than 2% in 20192. Given the lack of recent progress in reducing the global burden of malaria coupled with challenges in funding, renewed commitments are needed to address the “high burden to high impact” approach now being advocated by the World Health Organization2. IRS is a widely available tool that could be scaled up, however demands currently exceed the availability of resources. Additional work is needed to optimize the use of IRS, prevent further spread of insecticide resistance, and better evaluate the cost-effectiveness of IRS in the context of other control interventions.
Source: Ecology - nature.com
