From governance scores to governance impact: Closing the know–do gap in AMR policy evaluation

Abstract

As the 2026 revision of the Global Action Plan (GAP) on antimicrobial resistance (AMR) approaches, attention is shifting from policy adoption to the more complex challenge of evaluating implementation. Although most countries now have national action plans, reductions in resistance remain limited, exposing a persistent know–do gap between formal commitments and real-world impact. This commentary argues that prevailing governance assessments prioritise institutional presence over functional performance, overlook political economy constraints, and insufficiently integrate environmental and agricultural domains central to One Health. We highlight how surveillance asymmetries and static scoring distort comparisons and misdirect investment. We propose an impact-oriented evaluation agenda that incorporates temporal dynamics, incentive structures, equity considerations, and causal inference approaches to distinguish genuine progress from reporting artefacts. Closing the know–do gap is essential if AMR governance is to move beyond preparedness and deliver measurable, sustained change.

As preparations accelerate for the 2026 revision of the Global Action Plan (GAP) on antimicrobial resistance (AMR), attention is shifting from policy formulation to the more demanding task of evaluating implementation [1]. Since 2015, more than 180 countries have developed national action plans (NAPs), signalling broad political acknowledgement of the AMR threat. Yet these strategic commitments have not translated into a proportional reversal of rising AMR trends or a measurable decrease in its health impact. Instead, rising AMR burdens continue to challenge universal health coverage, food system sustainability, and environmental integrity. This divergence underscores a persistent know–do gap. Although consensus exists on what must be done, the conversion of policy commitments into measurable and sustained impact remains uneven and insufficiently understood [2].

Governance scores capture aspirations, not system functioning

Governance indices have improved visibility and accountability in global AMR coordination, but they overwhelmingly assess the presence of policies rather than their performance [3]. Countries may achieve high scores for publishing guidelines, establishing multisectoral committees, or issuing stewardship frameworks, yet these instruments frequently lack operational authority, durable financing, and cross-sectoral enforcement. AMR governance is inherently a long-cycle enterprise: strengthening laboratories, reforming antimicrobial sales regulation, or realigning agricultural incentives requires years of institutional adaptation [4]. Static evaluations cannot capture these trajectories, and thus risk privileging formal compliance over functional capacity.

Despite widespread endorsement of One Health, AMR monitoring remains dominated by human health systems, whereas critical arenas for resistance evolution such as environmental and agricultural sectors, are underrepresented in both data systems and governance assessments [5]. Wastewater, aquaculture effluents, pharmaceutical contamination, and agricultural run-off contribute substantially to AMR emergence, yet they remain peripheral in most NAP evaluations. This sectoral imbalance produces an incomplete risk picture and obscures ecological drivers. Without integrating environmental and agricultural evidence as core governance domains, One Health will remain conceptually embraced but operationally incomplete.

Measurement distortions and capacity asymmetries

Comparisons of AMR governance performance are frequently skewed by structural differences in surveillance capacity. Countries with robust detection systems inevitably identify more resistance and may appear to perform worse, while settings with limited data infrastructure may present an inconsistently measured or artificially low reported burden. This measurement distortion risks penalising data-poor countries and misdirecting investment and technical support. Mitigating these biases requires systematic adjustment for data completeness, triangulation with non-routine sources, and causal inference approaches capable of disentangling true epidemiological trends from surveillance artefact.

Yet capacity asymmetries also operate within countries: subnational gaps in surveillance, regulatory enforcement, and diagnostic access create concentrated pockets of AMR risk that national averages obscure. Without explicit attention to these internal divergences, governance evaluations will misclassify progress and overlook the areas where strengthened capacity would yield the greatest marginal returns.

Political economy constraints in AMR implementation

Many AMR governance failures arise less from missing policies than from the political economy of antimicrobial use. Economic pressures, diagnostic uncertainty, and permissive market structures drive predictable patterns of overuse: farmers rely on prophylactic antibiotics to stabilise yields, clinicians prescribe defensively when uncertainty raises the perceived cost of restraint, and distributors profit from weakly regulated sales [6]. Such structural incentives, rather than formal institutional gaps, shape antimicrobial behaviour in ways that conventional governance indicators largely overlook.

Effective governance evaluation must therefore assess whether interventions shift these underlying incentive structures, altering economic signals and reducing behavioural dependence on antimicrobials across prescribers, producers, and distributors. Embedding equity as a core criterion prioritising high-burden, low-capacity regions, ensures that governance reforms translate into tangible reductions in risk where they matter most.

Conclusion: Toward impact-oriented governance for GAP 2026

Closing the know–do gap requires evaluation frameworks that capture how institutions evolve and how policies reshape the incentives driving antimicrobial use. Governance assessments must incorporate temporal dynamics tracking multi-year implementation trajectories and milestones, rather than rely on static annual scores. Methodological approaches also need recalibration to correct for surveillance bias, integrate environmental and agricultural evidence, and apply causal inference tools capable of distinguishing real changes in antimicrobial use and resistance from reporting artefacts.

As AMR governance enters its second decade, the credibility of global efforts hinges on shifting evaluation from documenting preparedness to assessing real-world impact. An impact-oriented approach—rooted in political economy, equity, and causal reasoning—is essential to identifying which policies genuinely alter antimicrobial practices and resistance trajectories. With the 2026 GAP revision approaching, the central question is no longer whether national plans exist, but whether they generate measurable and sustained change. Bridging this gap will determine whether the global response to AMR remains aspirational or becomes a durable source of progress.

CRediT authorship contribution statement

Xiao-Xi Zhang: Writing – review & editing, Writing – original draft, Funding acquisition, Formal analysis, Data curation, Conceptualization. Yong-Zhang Zhu: Writing – original draft, Formal analysis, Data curation, Conceptualization. Xiao-Nong Zhou: Writing – review & editing, Writing – original draft, Supervision, Formal analysis, Data curation, Conceptualization.

Funding

The project was supported by National Key Research and Development Program of China (No. 2024YFE0199000) and National Natural Science Foundation of China (No. 72204160).

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

Data will be made available on request.