From malaria control to sustainable elimination: Cluster randomised trial comparing targeted versus generalised vector control in South Africa

Countries that have reduced malaria incidence to low levels face major challenges when trying to eliminate the disease altogether. In trying to reduce transmission further, considerable resources are required for disease prevention through mosquito vector control, for example by indoor residual spraying (IRS) of all houses. Such mass prevention efforts can lead to reduced compliance in communities and control programs who no longer perceive a risk of malaria, and waning political and donor commitment when the disease burden is low, thereby endangering the sustainability of the elimination effort. Evidence based methods of scaling back blanket IRS have to be developed which ensure that populations are not put at risk when IRS is no longer routinely applied.
In this study, targeted focal IRS in response to new cases being reported will be compared with generalised annual IRS of all houses, to determine whether it is as effective, less costly, more acceptable, results in higher coverage and compliance and increased malaria prevention seeking behaviour. A pre-condition for this approach, is a reliable rapid malaria case surveillance system, based on definitive diagnosis of suspected cases. This trial will be carried out in South Africa, which has practised blanket IRS for many decades, where case incidence in many districts is now low enough to be considered pre-elimination, and where a high quality case reporting system is well established. Spray localities will be grouped into clusters of 5,000 to 10,000 persons which will be randomly allocated to either targeted IRS, or to blanket routine IRS. Targeted neighbourhood IRS will be triggered in response to two or more local cases occurring within 4 weeks of each other and residing within 0.5km from one another. Spraying and community awareness activities will be carried out in a radius of 0.5km from each case house. It is postulated that focal spraying will lead to higher quality of IRS application because it can be better supervised and will be seen as protection against real risk of infection due to the occurrence of recent local cases; that it will be more acceptable to householders and hence lead to better co-operation with access to premises and hence higher spray coverage; and that householders will exercise better compliance with not repainting, washing or re-plastering of walls after spraying. As a result we expect that incidence of cases will be no higher in targeted IRS areas than in those receiving mass IRS and that targeted spraying will be more economical and hence more sustainable. The study will measure malaria case incidence, householder acceptability and compliance, spray coverage, and economic costs of the interventions as outcome indicators.
There is some evidence from other countries that in very low transmission settings, incident malaria cases occur in hotspots that are stable over time. If such hotspots can be accurately located, they can be singled out for focal interventions such as targeted IRS at the beginning of each season. To investigate whether such hotspots of local transmission exist in South Africa, it is proposed that filter paper blood spots are collected in communities where targeted IRS is carried out, to be tested for the presence of antibody sero-positivity to malarial antigens. The sero-prevalence of 'outbreak' communities will be compared with sero-prevalence of randomly selected communities in which no recent cases have occurred. This comparison will determine whether neighbourhoods with recent cases have historically been exposed to malaria parasites, and are therefore likely to be hotspots of transmission. The existence of such hotspots would strengthen the case for targeted control efforts.
For countries that have set elimination of malaria as a policy objective this study will provide evidence upon which sustainable policy decisions about mass vector control can be based during the pre-elimination period.

To achieve malaria elimination, interventions have to be sustainable in low transmission settings. Annual mass treatment of houses with indoor residual spraying (IRS) is costly per case averted when the malaria burden is low and can lead to poor compliance when the perceived risk of malaria is small. An open label two arm cluster randomised non-inferiority trial will compare effectiveness, cost, IRS coverage and householder acceptability between targeted focal IRS, initiated in response to two or more new local cases being reported in an area, no more than 4 weeks apart and within no more than 0.5 km from one another, with generalised annual IRS of all structures. Aggregations of spray localities will be used to form study clusters of between 5,000 and 10,000 inhabitants. Assuming an average of 3 cases/1000 person-years, and a coefficient of variation of 0.4 the trial will require at least 21 clusters per arm (42 in total) for 80% power and 5% significance if cluster populations consist of 6000 persons on average, incidence is measured over a 2 years period and if 1 case/1000 is regarded as the largest difference in the two methods that can be regarded as unimportant. The existence of hotspots of transmission from which cases arise will strengthen the case for focal vector control. Hotspots will be investigated through sero-prevalence surveys in the neighbourhoods of incident cases, compared to control neighbourhoods. Study endpoints will be malaria incidence, IRS coverage, intervention costs, compliance and antibody sero-prevalence. If targeted IRS is shown to be as effective, results in higher coverage, is more acceptable and more economical than mass IRS it will enable countries that have set elimination as a policy objective to make evidence based decisions about mass vector control in the pre-elimination phase.

The following will benefit from this research:
1. The general public, policy makers and national and provincial malaria control programs (NMCP and PMCP) in countries that have set malaria elimination as a policy objective and who are currently using similar interventions;
2. Communities who are being protected from malaria;
3. National and international donor and development agencies who support malaria control and elimination;

If the trial is successful, the general public in the elimination countries will benefit economically from the ultimate elimination of malaria from their countries, provided the elimination effort is achievable by sustainable means.

Policy makers will benefit because it will help them make policy decisions about possible scaling back of blanket IRS and replacing it with rapid response targeted IRS. This will make elimination efforts more sustainable.

NMCPs and PMCPs have asked for this trial since they want to have evidence upon which to base decisions about withdrawing generalised IRS from areas that have seen very few local cases.

More sustainable malaria elimination will benefit the communities at risk of malaria by protecting them against clinical malaria morbidity, reducing household costs associated with malaria episodes, and enhancing quality of life through less disruption resulting from having their houses sprayed.

National and international donor and development agencies will benefit from the results of this trial since it will enable them to make rational decisions about programme support, and provide exit strategies from supporting IRS programs indefinitely, once a pre-elimination level of transmission has been attained.