What constitutes high-quality evidence for malaria vector control?

The 21st century has seen remarkable progress against malaria, particularly from 2005 to 2015. Although wide implementation of insecticide-treated mosquito nets and indoor residual spraying of insecticide, in African countries with a high burden, has resulted in an 80% reduction in malaria infection and illness,1Bhatt S Weiss DJ Cameron E et al.The effect of malaria control on Plasmodium falciparum in Africa between 2000 and 2015.Nature. 2015; 526: 207-211Crossref PubMed Scopus (1689) Google Scholar these interventions have limits due to suboptimal coverage or use, emergence of insecticide resistance, and nuances in mosquito behaviour. The WHO's 2021 World Malaria Report2WHOWorld malaria report 2021. World Health Organization, Geneva2021Google Scholar documented stagnation of progress, with malaria mortality rising from 14·8 deaths per 100 000 population in 2015 to 15·3 per 100 000 population in 2020.2WHOWorld malaria report 2021. World Health Organization, Geneva2021Google Scholar The impact of our standard vector-control interventions has plateaued: mosquitoes evolve, and so must our interventions. Each year since 2015, more than 550 000 people—mostly children—have died of malaria.2WHOWorld malaria report 2021. World Health Organization, Geneva2021Google Scholar Countries and funding agencies rely on WHO for science-based guidance on the safety, quality, and efficacy of vector-control tools. We contend that the slow adoption and roll-out of new vector-control interventions is associated with an overly stringent definition of high-quality evidence of public health value. When WHO guidance lags, so too does implementation of new or improved interventions. The pathway of nets treated with pyrethroids plus piperonyl butoxide is illustrative. While currently recommended for areas with monooxygenase-based pyrethroid resistance, these nets were introduced in 2007, but it took 10 years for an interim policy recommendation to be released, despite solid evidence of effectiveness from multiple studies with entomological outcomes.3Tungu P Magesa S Maxwell C et al.Evaluation of PermaNet 3.0 a deltamethrin-PBO combination net against Anopheles gambiae and pyrethroid resistant Culex quinquefasciatus mosquitoes: an experimental hut trial in Tanzania.Malar J. 2010; 9: 21Crossref PubMed Scopus (84) Google Scholar By that time, pyrethroid resistance had spread throughout Africa. When developing vector-control guidance, WHO places a higher value on evidence from randomised controlled trials (RCTs) with epidemiological outcomes—a method developed for assessment of medical interventions for individual patients. Over the past decade various vector-control interventions have been assessed by Cochrane Reviews, but each places the highest value on RCTs, thereby downgrading other evidence. This weighting leads to surprising conclusions in WHO's vector-control guidelines; eg, that the evidence for the effectiveness of indoor residual spraying of insecticide is of low certainty, despite this intervention having been a pillar of WHO's malaria eradication effort in the 1960s, which resulted in markedly reduced malaria in many regions.4Nájera JA González-Silva M Alonso PL Some lessons for the future from the Global Malaria Eradication Programme (1955–1969).PLoS Med. 2011; 8e1000412Crossref PubMed Scopus (254) Google Scholar The guidelines also state that for larval habitat modification, “No recommendation can be made because the evidence…was deemed to be insufficient”, despite decades of experience and documentation of its impact dating back to Ronald Ross in the early 20th century.5Ross R The logical basis for the sanitary policy of mosquito reduction.Science. 1905; 22: 689-699Crossref PubMed Scopus (17) Google Scholar This minimisation of evidence from multiple continents spanning more than a century reflects a remarkably narrow view of what constitutes high-quality evidence. In areas of high transmission, a background of high coverage of insecticide-treated mosquito nets and improved case management means that RCTs are increasingly complex and costly; furthermore, the heterogeneity of malaria vector systems and environments makes generalisation across trials tenuous. In low-transmission settings, the sample size requirements for RCTs with epidemiological outcomes are prohibitive. The RCT was recognised as the gold standard for epidemiological research in 1982,6Jones DS Podolsky SH The history and fate of the gold standard.Lancet. 2015; 385: 1502-1503Summary Full Text Full Text PDF PubMed Google Scholar initiating the so-called RCT era. As varieties of methods for inference have proliferated, some have argued for a more flexible approach in assessment of evidence in support, or not, of new medicines, vaccines, and surgical procedures. Arguments for greater flexibility in evaluation of such interventions have greater force for those aimed at vectors of disease, but do not diminish the importance of well designed RCTs for evaluation of interventions, where appropriate. Hackett,7Hackett LW Malaria in Europe: an ecological study. Oxford University Press, London1937Google Scholar a pioneer of malaria elimination, noted that “perfection can be the enemy of the good” for malaria control. Making WHO guidance more inclusive of entomological outcomes8Sherrard-Smith E Ngufor C Sanou A et al.Inferring the epidemiological benefit of indoor vector control interventions against malaria from mosquito data.Nat Commun. 2022; 133862Crossref PubMed Scopus (9) Google Scholar and study designs other than RCTs—including historical evidence or implementation of novel test-negative designs—would allow countries to more easily access a range of innovative interventions, resulting in improved locally appropriate vector control, thereby accelerating malaria control and elimination globally and saving lives now. We declare no competing interests.