Top 3 Takeaways
- Mosquito modification had the strongest evidence. Wolbachia wMel replacement was associated with approximately a 77% reduction in dengue in both the AWED cluster-randomized trial and pooled nonrandomized studies.
- Evidence quality varied substantially. Over half of the review’s entries had serious-to-critical risk of bias, and certainty was very low for most outcomes.
- Uncertainty is not proof of ineffectiveness. The researchers called attention to geographical gaps and methodological limitations, with literature updates planned every 6 months.
Assessing Disease Outcomes Across Intervention Types
The review examined vector control measures targeting mosquito-human transmission among people at risk of Aedes-borne diseases. There were no restrictions on participants’ age, sex, or location. Vaccines and antivirals were excluded because the investigation focused on vector control.
Eligible studies measured epidemiological outcomes, such as disease incidence or prevalence. Comparators included no intervention, standard care, routine vector control, or alternative interventions. The researchers considered randomized controlled trials and cluster-randomized trials, as well as quasi-experimental and observational studies with comparators.
The search covered Embase, PubMed, Web of Science, and Scopus. Reviewers independently screened studies and extracted information. The team assessed the risk of bias and used GRADE to evaluate the certainty of evidence. The meta-analysis excluded studies at high risk of bias.
The living-review approach includes planned 6-month updates and an online dashboard. This structure allows the researchers to revisit their findings as additional evidence becomes available.
Evidence Was Unevenly Distributed
The included studies were clustered in the dengue-endemic Americas, Southeast Asia, and Australia. Few eligible studies came from South Asia or the African continent, identifying geographical gaps in the available research.
Study designs also varied across intervention classes. Nonrandomized designs dominated most categories, while cluster-randomized trials were concentrated in mosquito modification, insecticide-treated materials, and mass trapping. This variation was important to the review’s assessment of the robustness of the reported effects.
Over half of the entries were judged to have serious-to-critical risk of bias, and certainty was very low for most outcomes. Mosquito modification stood out in the GRADE (Grading of Recommendations, Assessment, Development, and Evaluation) assessment: Of 13 outcomes in that category, 9 had high-certainty evidence, 1 had moderate-certainty evidence, 2 had low-certainty evidence, and 1 had very-low-certainty evidence.
By comparison, all 6 outcomes assessed for integrated programs had very-low-certainty evidence. For insecticide-treated materials, 3 outcomes had low-certainty evidence, and 6 had very-low-certainty evidence.
Mosquito Modification Produced Consistent Findings
The meta-analysis reported an approximately 77% reduction in dengue with Wolbachia wMel replacement. That estimate was similar whether derived from the AWED cluster-randomized trial or pooled nonrandomized studies.
For the mosquito modification approach labeled IIT–SIT in the poster, the reported reduction was approximately 57% in nonrandomized studies and 77% in a cluster-randomized trial. The authors characterized the mosquito modification findings as “Consistent, substantial protection against dengue.”
Results for other intervention classes were less consistent after studies at high risk of bias were excluded. The poster reported that the effects of autodissemination were not statistically significant. It also reported no effect in the cluster-randomized trials of insecticide-treated materials and indoor residual spraying, whereas nonrandomized evidence for indoor residual spraying showed an increased incidence.
These findings were presented alongside the review’s broader concerns about sparse evidence and study quality. The researchers did not conclude that all approaches outside mosquito modification were ineffective.
Research Gaps Remain
The authors identified short follow-up as a priority gap, alongside the limited number of studies from Africa and South Asia. They also highlighted unaddressed interference or spillover, as well as differences in outcomes and in how those outcomes were ascertained.
Their interpretation separates the absence of convincing evidence from evidence that an intervention has no benefit. Although mosquito modification currently has the most robust, replicated support in the review, the findings leave important questions about other intervention classes unresolved. Continued updates are intended to maintain the evidence assessment as new studies emerge.
Poster authors: Jo Yi Chow; Thaddaeus Wu I-Khai; Min Xian Wang; Luis Ponce; Esther Choo Li Wen; Liang Yao; and Jue Tao Lim.
Affiliations: Lee Kong Chian School of Medicine, Nanyang Technological University, Singapore; National Centre for Infectious Diseases, Singapore; and SingHealth Duke-NUS Academic Medical Centre, Singapore.
Source: Poster P-861, living review version 1.0, searches through June 2026. PROSPERO registration: CRD420251060653.
