Economic Evaluation of Interventions to Reduce Antimicrobial Resistance: A Systematic Literature Review of Methods.

Abstract

Background and objective: Economic evaluation of antimicrobial resistance (AMR) interventions is complicated by the multisectoral, inter-temporal and international aspects of the problem, further hindered by a lack of available data and theoretical understanding of the emergence and transmission of AMR. Despite the substantial global focus on the problem, there is a lack of comprehensive economic evaluation literature on AMR policies. The goal of this work is to review the available literature on the economic evaluation of AMR interventions focusing on methods used to quantify the effects on AMR and the associated health consequences and costs.

Methods: The studies included in the review were identified by a previous study by Painter et al. that included all full economic evaluations of AMR policies in the peer-reviewed and grey literature published between 2000 and 2020. The current review extracted additional information to (1) summarise the types and the key features of the AMR intervention economic evaluation literature available; (2) systemise the types of intervention effects on AMR quantified and describe these across the dimensions of AMR burden: time, space, wider pathogen pool and different sectors (One Health framework); and (3) categorise the methods used to derive these outcomes and how were these linked to health consequences and costs.

Results: Thirty-one studies were included within this review, of which 18 evaluated interventions that aimed to reduce infection rates and 11 evaluated interventions that aimed to optimise antimicrobial use. Almost all were conducted with a high-income and/or upper-middle income country perspective and focused on human health. Thirteen of 31 studies were cost-utility analyses. Fifteen of 31 and 7/31 studies estimated the AMR effects through decision tree and/or Markov models and transmission models, respectively. Transmission models and linkage of AMR outcomes to quality-adjusted life-years and costs were more common in evaluations of interventions aimed at reducing infection rates. Most of the included studies restricted the scope of evaluation to a short time horizon and a narrow geographical scope and did not consider the wider impact on other pathogens and other settings, potentially resulting in an incomplete capture of the effects of interventions.

Conclusions: This review found limited available literature that mainly focused on high-income countries and infection prevention/reduction strategies. Most evaluations used a narrow study scope, which might have prevented the full capture of the costs and outcomes associated with interventions. Finally, despite the known complexities associated with quantifying AMR effects, and the corresponding methodological challenges, the implications of these choices were rarely discussed explicitly.