Quantitatively evaluating the cross-sectoral and One Health impact of interventions: A scoping review and case study of antimicrobial resistance

Abstract

Background: Current frameworks evaluating One Health (OH) interventions focus on intervention-design and -implementation. Cross-sectoral impact evaluations are needed to more effectively tackle OH-issues, such as antimicrobial resistance (AMR). We aimed to describe quantitative evaluation methods for interventions related to OH and cross-sectoral issues, to propose an explicit approach for evaluating such interventions, and to apply this approach to AMR.

Methods: A scoping review was performed using WebofScience, EconLit, PubMed and gray literature. Quantitative evaluations of interventions that had an impact across two or more of the human, animal and environment sectors were included. Information on the interventions, methods and outcome measures found was narratively summarised. The information from this review informed the construction of a new approach to OH-related intervention evaluation, which then was applied to the field of AMR.

Results: The review included 90 studies: 73 individual evaluations (from 72 papers) and 18 reviews, with a range of statistical modelling (n = 13 studies), mathematical modelling (n = 53) and index-creation/preference-ranking (n = 14) methods discussed. The literature highlighted the need to (I) establish stakeholder objectives, (II) establish quantifiable outcomes that feed into those objectives, (III) establish agents and compartments that affect these outcomes and (IV) select appropriate methods (described in this review) accordingly. Based on this, an evaluation model for AMR was conceptualised; a decision-tree of intervention options, a compartmental-microeconomic model across sectors and a general-equilibrium (macroeconomic) model are linked. The outcomes of this multi-level model (including cost-utility and Gross Domestic Product impact) can then feed into multi-criteria-decision analyses that weigh respective impact estimates alongside other chosen outcome estimates (for example equity or uncertainty).

Conclusion: In conclusion, stakeholder objectives are key in establishing which evaluation methods (and associated outcome measures) should be used for OH-related interventions. The stated multi-level approach also allows for sub-systems to be modelled in succession, where resources are constrained.

Keywords: AMR, Antimicrobial resistance; Antimicrobial resistance; DALY, Disability Adjusted Life Year; Economic evaluation; GDP, Gross Domestic Product; Impact evaluation; MCDA, Multi-criteria decision analysis; NEOH, Network for Evaluation of One Health; OH, One Health; One Health.

Fig. 1

The system under evaluation for cross-sectoral an antimicrobial resistance intervention: Adapted from Ruegg et al [79] (Fig. 2). Ovals represent sectors, boxes represent agents, hexagons represent resources and parallelograms represent actions related to antimicrobial stewardship. Connecting lines represent potential relationships related to the issue and intervention. ‘Ministry’ may be multiple ministries in reality (for example, food system may include commerce and additional governmental offices). AMR: antimicrobial resistance.

Fig. 2

A conceptual multi-level model for evaluating cross-sectoral antimicrobial resistance interventions. White boxes represent health states or sector states. Segments (A) to (D) represent the model method. Shaded boxes represent settings in (A) – (C) and respective model results in (D). Transitions can occur between white boxes within each segment (including across setting), such as from animal antimicrobial susceptible carrier to antimicrobial susceptible human carrier within (B), but these lines have not been added for visual simplicity. Inputs refer to those changed through the intervention and not all model inputs. Abbreviations: AMR – antimicrobial resistance, AMS – antimicrobial susceptible,