The Comparative Effectiveness of Vector Control Approaches for Dengue: A Systematic Review and Multilevel Network Meta-Analysis.

Abstract

Background: Dengue vector control plays an important role in reducing the burden of dengue infection. Conventional meta-analysis can only include interventions that can be directly compared. Hence, this study aimed to compare the effectiveness of all interventions for dengue vector control directly and indirectly through a multilevel Bayesian network meta-analysis.

Methods: Cluster randomised controlled trials (cRCTs) and randomised controlled trials (RCTs) were included. Screening and full-text review were performed independently by two reviewers, with resolution by a third reviewer if there was disagreement. Outcomes measured for the included studies considered were entomological indices and epidemiological measures. The Cochrane risk of bias tool was used to assess the quality of the studies by two independent reviewers, with a third reviewer to resolve any discrepancies.

Results: A total of 48 cRCTs and 13 RCTs were included in this study. The quality of cRCTs and RCTs was overall deemed at high risk of bias except for two studies with low risk of bias. The network meta-analyses and narrative synthesis indicate that most interventions are not significantly superior to one another. Despite the lack of statistically significant results in the network meta-analyses, a standalone intervention appears to be less effective than combined interventions.

Conclusions: Although most interventions were not found to be significantly superior to one another in the network meta-analyses, it is noteworthy that standalone interventions were shown to be less effective than combined interventions. These results reinforce the importance of promoting integrated, multi-faceted vector control strategies for dengue prevention.