RR:C19 Evidence Scale rating by reviewer:
Reliable. The main study claims are generally justified by its methods and data. The results and conclusions are likely to be similar to the hypothetical ideal study. There are some minor caveats or limitations, but they would/do not change the major claims of the study. The study provides sufficient strength of evidence on its own that its main claims should be considered actionable, with some room for future revision.
***************************************
Review: The study examined very important questions on malaria elimination especially for the near-elimination regions: 1) what is the population-level impact of test-and-treat border posts on preventing between-country transmission and its subsequent impact on malaria prevalence in the involved regions, and 2) how certain parameters governing malaria transmission and border post implementation affect the results.
The study used an open-source individual-based model to simulate malaria dynamics and interventions in sub-Saharan Africa, combined with a metapopulation model to simulate interactions and human movements between regions. They conducted a main analysis in which they selected one cluster (one seed region and its seven adjacent regions) at a time, from the 401 administrative units touching an international boarder, and conducted 50 iterations of simulations for each cluster, and analyzed number of cases averted per year and changes in malaria prevalence for each cluster. They also performed two generic case studies in which they manually altered malaria prevalence and border post coverage to examine the impact of these two parameters on the intervention effectiveness.
The study is well designed, and the model and data used are proper. However, due to lack of data on some key model assumptions and parameters, I’m concerned that the quantitative results (e.g., number of cases averted, relative reduction in malaria prevalence) can be inaccurate. For example, the probability of influence of one region on its neighboring unit is drawn from a user-specified mixing matrix, which is somewhat subjective. In addition, due to lack of human movement data, a gravity model was used to estimate travels between regions based on destination population size and travel time between the origin and destination regions. The authors used the best approximation to overcome the data gap, but these can still be substantially different from real-world data, and population movement is a key parameter in this analysis that can substantially impact the results. Another question is, can the travelers be healthier (having less infection) than the average as depicted in healthy immigrant effect? The assumption of 80% of border post coverage is another example of a key parameter that was subjectively selected. And the uncertainty in these model inputs can also affect the quantitative results in the case studies, although I expect less impact on the qualitative results.
The authors are transparent on the methods and limitations, and the discussion was well written. It inspires future research in this area to fill the data gap. The qualitative results are reliable, more data and economic evaluation are needed to enhance the analysis for health policymaking, especially in specific countries/regions.
Creative Commons Attribution 4.0 International License (CC-BY 4.0)
