Strongly disagree with the conclusion of the analysis
Mark Rowland, London School of Hygiene & Tropical Medicine
13 December 2014
While the majority of trials failed to show a protective effect of topical repellents, two RCT trials did. It is incorrect to conclude in the abstract that 'that topical repellents are unlikely to provide effective protection against malaria'. Trials which fail do not mean that the intervention does not work. They can fail for many reasons. Key to the success of topical repellent protection is adherence. The trial will fail to show an effect if the recipients are culturally unwilling to adhere to the (new) intervention, if there is inadequate health information messaging, if the monitoring of the effect is inadequate. Trials that require behavioural change are very vulnerable to failure. Because it is difficult to monitor daily application and usage per protocol analysis is usually not possible.
In the case of the Afghan refugee study, the trial population were very receptive to the new intervention particularly among mothers and children. who did not have other effective forms of protection (e.g. ITN) at the time. In a additional case-control study on a another population (not included in the metaanalysis) there was also good evidence for protection against clinical malaria among repellent users and the strongest effect was in repellent+ITN users.
Trials of topical repellents depend more strongly on daily routine behavioural compliance (self implementation) than trials of ITNs or IRS and if the messaging is wrong the intervention fails. We know repellects are less likely to succeed than ITN and IRS which do not such strong behavioural change. But we also know that repellents do protect against biting when used properly and should therefore protect against malaria.
Just because the majority of trials failed to show an effect, means that repellents do not protect against malaria. The point is that in some populations, some cultures, and in some epidemiological situation, repellents were shown to protect against malaria. That cannot be denied, and therefore the overall conclusion is wrong and misleading.
Strongly disagree with the conclusion of the analysis
13 December 2014
While the majority of trials failed to show a protective effect of topical repellents, two RCT trials did. It is incorrect to conclude in the abstract that 'that topical repellents are unlikely to provide effective protection against malaria'. Trials which fail do not mean that the intervention does not work. They can fail for many reasons. Key to the success of topical repellent protection is adherence. The trial will fail to show an effect if the recipients are culturally unwilling to adhere to the (new) intervention, if there is inadequate health information messaging, if the monitoring of the effect is inadequate. Trials that require behavioural change are very vulnerable to failure. Because it is difficult to monitor daily application and usage per protocol analysis is usually not possible.
In the case of the Afghan refugee study, the trial population were very receptive to the new intervention particularly among mothers and children. who did not have other effective forms of protection (e.g. ITN) at the time. In a additional case-control study on a another population (not included in the metaanalysis) there was also good evidence for protection against clinical malaria among repellent users and the strongest effect was in repellent+ITN users.
Trials of topical repellents depend more strongly on daily routine behavioural compliance (self implementation) than trials of ITNs or IRS and if the messaging is wrong the intervention fails. We know repellects are less likely to succeed than ITN and IRS which do not such strong behavioural change. But we also know that repellents do protect against biting when used properly and should therefore protect against malaria.
Just because the majority of trials failed to show an effect, means that repellents do not protect against malaria. The point is that in some populations, some cultures, and in some epidemiological situation, repellents were shown to protect against malaria. That cannot be denied, and therefore the overall conclusion is wrong and misleading.
Competing interests
None