Quantitatively Evaluating the Cross-Sectoral and One Health Impact of Interventions: A Scoping Review and Application to Antibiotic Resistance [article]

Nichola R Naylor, Jo Lines, Jeff Waage, Barbara Wieland, Gwenan M Knight
2020 medRxiv   pre-print
Current published guidance on how to evaluate antibiotic resistance (ABR) from a One Health perspective has focussed on the evaluation of intervention design and of the implementation process. For efficient resource allocation, it is also important to consider quantitative measures of intervention impact. In particular, there has been little discussion of how to practically evaluate ABR-related agri- and aquaculture interventions from a public health perspective. Lessons can be learned from
more » ... r One Health and cross-sectoral intervention impact evaluations. WebofScience, EconLit, PubMed and grey literature were searched for literature quantitatively evaluating interventions across humans, animals and/or the environment. The review included 90 studies: 73 individual evaluations (from 72 papers) and 18 reviews, all including some measure of human impact, but only 29 papers covered all three One Health perspectives (human, animal and environmental). To provide decision makers with expected outcome estimates that are related to their objective, evaluations should provide outcome estimates from multiple different perspectives; individual, microeconomic and/or macroeconomic perspectives across the One Health system should be taken into account. Based on the methods found in this review, a multi-level compartmental modelling approach for ABR-related intervention evaluation is proposed. The outcomes of such models can then feed into multi-criteria-decision analyses that weigh outcomes alongside other chosen outcome estimates (for example equity or uncertainty). It is key that future quantitative evaluation models on ABR-related interventions are shared (for example through open source code sharing websites) to avoid duplication of effort and to enable more comprehensive estimates of intervention impact to be modelled in the future.
doi:10.1101/2020.01.30.20019703 fatcat:l4yw4iseofgjba6vq2fsfcy26q