“Pathogens do not follow borders,” says Afreenish Amir, Ph.D., a medical microbiologist whose work has helped strengthen antimicrobial resistance (AMR) surveillance, laboratory capacity, and stewardship in Pakistan.

She has worked with hospitals and microbiology laboratories in Pakistan participating in the World Health Organization’s Global Antimicrobial Resistance and Use Surveillance System. Her work subsequently expanded across infection prevention and control, antimicrobial stewardship, fungal and bacterial genomics, wastewater and environmental surveillance, and One Health initiatives. She has supported fungal diagnostics and surveillance in country.

As a disAMR Ambassador, she advocates for stronger regional collaboration and greater exchange of practical knowledge among countries confronting similar challenges.

What first drew you to AMR, and when did it become clear that this was the problem you wanted to work on?

I am a medical doctor by training, and I specialized in microbiology. I was supporting work on AMR surveillance through national and international platforms. At the same time, I became involved in antimicrobial stewardship and infection prevention and control initiatives.

That work opened the door to many related areas, including fungal genomics, wastewater environmental surveillance, cholera genomics, and metagenomics. AMR is not a single project or the responsibility of one discipline. Once you begin working in surveillance, it becomes clear how closely laboratory capacity, clinical practice, public health, policy, and multiple sectors are interconnected. Addressing AMR therefore requires a coordinated, multidisciplinary and One Health approach, bringing together evidence, expertise, and action across these different areas.

What do people outside Pakistan need to understand about the country’s AMR challenges?

Pakistan is the fifth-most populous country in the world, and it carries a high burden of infectious diseases. Through national surveillance, we are seeing resistant bacteria and other pathogens that require greater attention.

These pathogens do not follow borders. An outbreak or emerging threat in one country can rapidly have implications for the wider region, underscoring the importance of cross-border surveillance, timely information sharing, and coordinated regional preparedness and response.

Fungal infections are increasingly recognized as an important public health concern, with emerging antifungal resistance adding to the challenge. Strengthening diagnostic capacity for timely and accurate detection, improving access to essential antifungal medicines, and increasing clinical awareness and engagement are critical for improving the diagnosis, treatment, and surveillance of fungal diseases.

Once a surveillance system begins generating data, how do you make sure those data lead to action?

The first approach always focuses on surveillance, to generate evidence and build the capacity to produce high-quality data. The next question is how to use those datasets. This requires engaging more directly with clinicians about antimicrobial and antifungal stewardship. The information generated by the laboratory has to support patient safety and more responsible use of antimicrobials.

The same principle applies to broader AMR surveillance. Data generated by hospitals and laboratories should inform antimicrobial stewardship, prescribing practices, and the development of antibiograms that clinicians can use when selecting treatment.

Surveillance cannot operate alone. Laboratories, clinicians, stewardship teams, infection prevention specialists, and policymakers all need to work together. The goal is to build a coherent system in which evidence reaches the people who can act on it.

How do you balance antimicrobial stewardship with the need to improve access to essential medicines and diagnostics?

This remains an important concern. Evidence generated by hospitals, laboratories, and public health institutions ultimately has to reach the community.

The Essential Medicines List must reflect the evidence emerging from surveillance and identify which antibiotics should be available through primary health centers and government hospitals. The Essential Diagnostics List must also evolve so that health facilities have the tools needed to determine which treatments patients require.

Access also means access to quality-assured medicines. We continue to discuss the need to regulate nonprescription antimicrobial sales, maintain appropriate standards for products sold through pharmacies, and prevent falsified or substandard medicines from reaching patients.

We cannot look at stewardship, diagnostics, medicine quality, and access as separate issues. The decisions developed from hospitals, pharmacies, laboratories, and regulatory agencies experiences ultimately have to improve care for the patient.

How can low-income countries organize laboratory capacity when advanced resources cannot be placed in every facility?

Low-income countries can adopt a tiered laboratory network, with basic diagnostic services at peripheral facilities and advanced testing concentrated in regional and national reference laboratories. Strong specimen referral systems, standardized protocols, quality assurance, workforce training, and digital data-sharing platforms can ensure equitable access to diagnostics. The priority should be building an integrated and sustainable laboratory network rather than duplicating expensive technologies at every facility.

Why is a One Health approach essential to understanding and controlling AMR?

AMR cannot be addressed through human health alone. Human health, animal health, agriculture, livestock, and the environment are all connected. Microorganisms move among the environment, animals, and people.

If we invest only in human health while neglecting antimicrobial use and resistance in livestock, poultry, agriculture, and the environment, we will not adequately reduce the burden of resistant infections in people.

Integrated AMR surveillance projects provide an important foundation, but these efforts must be institutionalized as part of routine surveillance systems. Sustained, longitudinal data collection is essential to understand how antimicrobial resistance emerges and evolves within individual sectors, and how resistance patterns and transmission dynamics may influence other sectors.

Antibiotics reach patients at the end of a manufacturing chain that produces its own waste. In some places where antibiotics are made, waste and wastewater contain high concentrations of active pharmaceutical ingredients (APIs). When those compounds enter waterways or soils, bacteria encounter antibiotics released during production, independent of the medicines’ use by patients. That exposure raises a question for antimicrobial resistance (AMR) efforts: can the production of antibiotics create conditions that favor resistance to them?

Research near manufacturing sites gives a qualified answer: yes, it can. Investigators have measured antimicrobial contamination and resistance in environments affected by production waste. At one wastewater treatment plant, researchers found that a resistant bacterium became more abundant as wastewater moved through treatment. These findings establish a reason to examine what leaves production sites and how well controls address the risk, although they cannot establish manufacturing’s share of global AMR or trace a particular patient’s infection to a particular discharge.

What Leaves an Antibiotic Manufacturing Site?

The issue extends across the production chain. Antibiotic active ingredients are made, incorporated into finished products, and packaged through processes that can generate liquid and solid waste. The World Health Organization’s (WHO’s) guidance addresses these stages, from active ingredient production through formulation and primary packaging. It sets out targets intended to reduce the risk of resistance emergence and spread, as well as harm to aquatic life.1,2

This broad scope matters when several facilities contribute to one antibiotic. A company that sells the finished medicine may rely on other manufacturers to produce its active ingredient or formulate the product. Examining the final facility alone would leave earlier production stages outside the assessment. The WHO therefore identifies manufacturers, regulators, antibiotic purchasers, auditors, and waste-management services among the audiences for its guidance.

The available research offers evidence of contamination, but not a complete inventory of manufacturing emissions. In a systematic review, researchers at the Chinese Center for Disease Control and Prevention assembled published measurements of antibiotics and resistance genes in wastewater and solid waste associated with pharmaceutical production.3 They also identified a shortage of publicly available data about manufacturing processes and waste. That gap makes it difficult to judge how consistently controls are applied across production sites or to compare releases across the wider industry.

What Studies Near Production Sites Show

An investigation in Hyderabad, India, sampled water in and around a major bulk-drug production area, including locations near manufacturing facilities and sewage treatment plants, the Musi River, and nearby communities. The researchers detected antimicrobials at all 28 sampling sites and found resistant Gram-negative bacteria in more than 95% of their samples. They described the contamination as apparently associated with the selection and dissemination of resistant pathogens. The study measured contamination and resistance in the same environments; it could not identify a facility as the source of a particular resistant infection.4

Another study examined effluents from an azithromycin-synthesis facility and a veterinary-drug formulation facility, along with sediments at their discharge sites and upstream and downstream. Sediments exposed to the discharges contained a greater proportion of resistant bacteria than upstream sediments. The researchers also measured increases in the relative abundance of resistance genes and genetic elements associated with their movement. In laboratory experiments, resistance carried by material recovered from exposed sediments transferred to a recipient Escherichia coli strain. These results provide evidence of changes in the receiving environments at the sites studied.5

The upstream comparisons show differences associated with the discharges. They do not by themselves reveal what happened to resistance after it left the sampling area. The WHO’s explanation of its manufacturing guidance emphasizes concern about the emergence of new resistance and says manufacturing emissions are unlikely to contribute significantly to transmission of resistance already widespread in society.5,6

When Treatment Changes the Wastewater

Removing antibiotic residues is one aim of wastewater treatment. A study of a full-scale plant serving a manufacturer of primarily beta-lactam antibiotics shows why conventional measures of chemical removal may not tell the entire story. Researchers examined bacterial communities and resistance genes across 11 treatment units.7 The plant substantially reduced chemical oxygen demand and biochemical oxygen demand, meeting the local discharge limits the researchers reported for those measures. Yet beta-lactam residues persisted, and some exceeded the predicted no-effect concentration thresholds used in the study. Antibiotic-resistant Aeromonas veronii also became more abundant as wastewater moved from influent to effluent, while some resistance genes persisted through treatment.

The researchers attributed the enrichment of A. veronii probably to prolonged beta-lactam selection pressure and the organism’s adaptation to aquatic environments. Their findings concern one plant. What makes the result consequential is the difference between the measures: satisfactory performance against conventional chemical indicators did not mean the absence of antibiotic residues or enrichment of a resistant organism.

The investigators also compared their wastewater-derived A. veronii with publicly available genomes. In that analysis, its closest evolutionary relative was an isolate from the stool of a patient with diarrhea in the same province. The comparison did not establish that the plant was the source of the patient’s isolate. The authors called for further epidemiological work to determine whether a transmission pathway exists.

For facilities assessing their wastewater, the study makes a case for looking beyond conventional chemical indicators and antibiotic concentrations. Measurements of resistant organisms and resistance genes can reveal changes as water passes through treatment. The study does not specify how often this happens at other plants, but it identifies a question that chemical measurements alone cannot answer.

From Concern to Discharge Targets

The WHO published guidance on wastewater and solid waste management for antibiotic manufacturing in 2024.1,2 It provides targets addressing resistance-related human health risks and risks to aquatic life. WHO describes the guidance as an independent scientific basis that regulators and other decision-makers can use when incorporating targets into binding requirements. Publication of the guidance did not itself create one legal discharge limit that applies in every manufacturing country.

The AMR Industry Alliance’s Antibiotic Manufacturing Standard supplies another framework. It requires manufacturers to have an environmental management system and meet predicted no-effect concentration targets for antibiotics. A predicted no-effect concentration is a risk-assessment target: a concentration intended to be low enough to avoid a specified adverse effect. It gives manufacturers a basis for assessing emissions, rather than a guarantee that every possible resistance risk has disappeared. The Alliance updated its standard in 2025 to align more closely with WHO guidance, including provisions on waste management, active ingredient discharge limits, and transparency about suppliers.8,9

Applying a target requires evidence about performance. The relevant production sites and waste streams must be identified, and controls must be assessed and maintained. The BSI Kitemark for Minimized Risk of AMR provides independent assessment against the Alliance’s standard.10 BSI describes an initial document review and on-site assessment that can validate adherence to concentration targets through approaches including mass balance or sampling and analysis. Certification is followed by annual surveillance and periodic recertification.

WHO’s guidance, the Alliance’s standard, and BSI certification have different roles. The guidance supplies independently developed targets and risk-management recommendations. The standard states requirements a manufacturer can adopt, and certification assesses conformity with those requirements. Regulators can draw on guidance when developing enforceable rules. Keeping those roles clear makes it easier to ask what a manufacturer is required to do, who verifies it, and what evidence is available to others.

When Purchasing Standards Reach the Factory

The National Health Service (NHS) antimicrobial products subscription model offers a concrete example of a purchaser requiring evidence about manufacturing.11 Under the model, the NHS pays a fixed annual fee for access to selected antibiotics, rather than paying according to the volume used. Companies bidding for contracts under this program need the BSI Kitemark for Minimized Risk of AMR. This environmental requirement applies to the defined subscription program, not every antibiotic purchase by the NHS.

BSI says certification must cover all manufacturing sites involved in producing the relevant antibiotic. That includes active ingredient facilities, finished-product facilities, and sites operated by contract manufacturers. The requirement reaches beyond the company submitting a bid to the production steps on which its medicine depends. It gives the purchaser a way to ask whether controls have been assessed throughout that chain, rather than relying on a statement about the final product’s manufacturing site alone.

The route to certification allows for implementation over time. According to BSI’s description of the NHS tender process, an applicant can provide evidence of certification or submit a delivery plan acceptable to the NHS before a contract is awarded. BSI says certification must then be achieved within 12 months of the contract’s start. Its assessments and continuing surveillance provide a means of checking specified controls as production continues.

The NHS example gives practical force to the WHO’s inclusion of antibiotic purchasers among the audiences for its guidance. A purchaser can make independently assessed manufacturing performance part of a supply decision, while manufacturers must account for the facilities that contribute to their product. The example also illustrates why implementation needs careful design: the WHO calls for stepwise improvement that protects antibiotic supply and affordable access alongside stronger environmental controls.

Making Performance Visible

Manufacturers can identify where antibiotic residues enter waste streams, assess those streams against relevant targets, and verify controls across the production sites involved. The treatment-plant study adds a reason to consider resistant bacteria and resistance genes alongside chemical pollutants when evaluating wastewater. Independent assessment offers one way to check whether specified controls are in place and maintained.

Regulators and purchasers have different means of acting on that evidence. Regulators can use the WHO’s guidance as a basis for requirements, inspections, and monitoring. Purchasers can specify what evidence suppliers must provide, as the NHS subscription model demonstrates. Both approaches depend on information about actual production sites and waste-management practices. The WHO calls for audits and public transparency, while the systematic review of manufacturing waste identifies limited data availability as a barrier to evaluating mitigation.

The evidence does not yield a global percentage of AMR attributable to manufacturing, and the studies considered here do not establish a path from a particular discharge to a particular patient’s infection. Manufacturers, regulators, and purchasers can still act on the risks the evidence does identify. Discharge measurements, coverage of each production site, and independent verification would give them a firmer basis for judging whether antibiotic manufacturing is controlling its contribution to environmental resistance risk.

One Sample, Many Contributors

A single wastewater sample can contain resistance signals contributed across an entire sewer catchment. It may capture organisms shed by patients with diagnosed infections, asymptomatic carriers, and people who never interact with the health care system. What it cannot reveal is which of those contributors produced a particular signal or what that signal means clinically.

This broad population coverage is the principal attraction of wastewater surveillance. Clinical surveillance depends on people entering the health care system, undergoing testing, and having their results reported. Wastewater collects microbial material across a connected population, including from people who may never appear in clinical surveillance data.1

Programs may look for resistant organisms that can be cultured from wastewater, specific antimicrobial resistance genes (ARGs), or the broader collection of resistance genes present in a sample. A composite sample collected at a wastewater treatment plant can represent a large municipal catchment, while samples drawn from a hospital or long-term care facility can provide a more localized view of the population using that building. Repeated collection can show whether selected organisms or genes are present and how their abundance changes over time.2

Population coverage should not be confused with individual resolution. A wastewater sample is not equivalent to testing every person in its catchment. It combines material from many contributors, along with organisms and genes originating elsewhere in the wastewater system. The result is an aggregate signal whose meaning depends on what was measured, where the sample was collected, and what entered the sewer before it reached that point.

This distinction defines both the value and the limits of wastewater surveillance for antimicrobial resistance (AMR). It can reveal presence, distribution, and change across a population. It is considerably less capable of identifying the clinical consequences, microbial sources, or transmission pathways behind the signal.

What the Signal Can Reveal

At its most straightforward, wastewater surveillance can determine whether a selected resistant organism or ARG is detectable within a sampled catchment. With consistent sampling, investigators can establish a baseline and look for changes in abundance, persistence, and geographic distribution. This makes wastewater potentially useful for following resistance patterns across communities that would be difficult to sample individually.

The approach can also support comparisons across large geographic areas. A global study used metagenomic sequencing to analyze ARGs in 757 sewage samples collected from 243 cities in 101 countries between 2016 and 2019.3 The study demonstrated that sewage can provide a common matrix for comparing the diversity and abundance of resistance genes across locations, including countries with limited existing microbiome data.

The information obtained depends heavily on the analytical method. Targeted quantitative polymerase chain reaction (qPCR) can sensitively measure selected genes associated with clinically important resistance. Metagenomic sequencing examines the collection of genetic material in a sample and can provide a broader profile of the resistome. A national-scale comparison found metagenomics more comprehensive for resistome profiling, while high-throughput qPCR permitted sensitive quantification of selected genes of clinical significance.4

Wastewater results can also correspond with patterns observed in human sampling. A 2022 systematic review found reasonably high overall concordance between wastewater and human AMR prevalence estimates despite substantial variation in study designs, organisms, resistance targets, and methods.1 It found no statistically significant relationship between particular methodological features and high-agreement studies, although the analysis was limited by heterogeneous approaches, inconsistent reporting, and a small evidence base. In a descriptive analysis, longer, time- and location-matched studies using composite influent samples generally showed better agreement.

The defensible output is therefore a community resistance profile rather than a count of resistant infections. Wastewater may show that a target is present, becoming more common, declining, or distributed differently between locations. Those findings can identify where closer investigation is warranted, even when the contributors to the signal remain unknown.

What Detection Does Not Establish

The presence of an ARG is only one component of clinically meaningful resistance. In a mixed wastewater sample, detecting a gene does not establish that it came from a human. Resistant organisms and genes may enter wastewater from human or animal excreta, health care facilities, industrial sources, runoff, or environmental microorganisms. Some may also persist in sewer systems or wastewater treatment infrastructure.

However, gene detection does not necessarily identify the organism carrying it. Genetic material from many microorganisms is mixed within the same sample. Unless an analytical method links an ARG to a particular genome, cell, or cultured isolate, investigators cannot assume that the gene sits within the pathogen of greatest clinical concern. Multiple resistance genes detected together may also reside in different organisms rather than in a single multidrug-resistant bacterium.

Even a confirmed gene–host association does not automatically demonstrate phenotypic resistance. A gene may not be expressed at a sufficient level, and other mechanisms may be required for the organism to withstand an antimicrobial. Permeability, efflux, antimicrobial modification, and additional genetic or physiological factors can affect the phenotype. Detection also does not show that the host is pathogenic or possesses the virulence characteristics required to cause human disease.

A hospital study of carbapenemase-producing Enterobacterales illustrates the problem. Researchers collected 20 composite sewage samples over two years and compared molecular and culture-based findings with organisms identified in hospitalized patients.5 Levels of three carbapenemase genes corresponded with the abundance of relevant carbapenemase-producing Enterobacterales in sewage. Two other genes did not show the same correspondence, most likely because they were carried by non-Enterobacterales organisms within the wastewater. A gene-level result was therefore informative for some targets but misleading if interpreted as evidence of a specific resistant pathogen for others.

Wastewater is similarly limited in what it can establish about human disease. A signal does not identify which individuals are infected or colonized, how many resistant infections have occurred, or whether any of the people contributing to it are ill. Consequently, wastewater evidence cannot be translated directly into clinical disease burden or deaths. Clinical specimens remain necessary to connect resistance with a diagnosed infection and its treatment implications.

A wastewater detection alone also cannot reconstruct transmission, although cultured isolates, genomic comparisons, and epidemiological data can help determine whether organisms found in wastewater overlap with those circulating in clinical settings. If the same resistance mechanism appears in wastewater and patient isolates, that overlap does not prove that organisms moved between individual patients, from a facility into the community, or from the community into a facility.

A 2026 Swiss study shows what a more resolved investigation entails.6 Researchers analyzed 762 extended-spectrum beta-lactamase-producing Escherichia coli genomes collected from wastewater, clinical settings, cattle, and wildlife between 2021 and 2023. Wastewater captured lineages that also appeared in clinical settings, but exact clonal matches between wastewater and corresponding clinics were rare. The findings supported wastewater as an indicator of community circulation while showing that overlap between compartments is easier to detect than the direction or frequency of specific transmission events.

Conditions within the sewer add another layer of uncertainty. Transport time, storage conditions, freezing and thawing, biological activity, and interactions with sewer biofilms can affect the microbial or resistance profile eventually measured in the laboratory. The sampled signal may therefore reflect both what entered the sewer and what happened before analysis.

When Wastewater Could Provide Useful Early Warning

A surveillance result does not qualify as early warning merely because it precedes a clinical report. An operational early-warning system needs a defined target, a stable baseline, recurring sampling, adequate analytical sensitivity, an interpretable change threshold, and a planned response. Without those elements, detection may be scientifically interesting but difficult for a health system to act upon.

In practice, a potential alert could take several forms: the first detection of a previously absent target, a sustained increase above an established baseline, or a difference between facility- and community-level measurements. The most plausible targets are low-prevalence resistance mechanisms with high clinical importance, particularly those not already well characterized through routine clinical surveillance. Facility-level sampling may be especially useful in hospitals and long-term care settings, where a rare signal could become too diluted to detect reliably at a municipal treatment plant.

Evidence from health care facilities shows both the potential and the interpretive difficulty. In the two-year hospital study, sewage monitoring detected some clinically important carbapenemase-producing organisms and gene patterns that corresponded with findings in patients.5 The mismatches for other genes demonstrated why an alert would still require culture, clinical testing, and genomic investigation before infection-control conclusions could be drawn.

A separate 16-month study examined wastewater surveillance for five carbapenemase genes at long-term care facilities.7 Several genes were routinely detected in composite wastewater, passive samples, and sewer biofilms. However, the facilities recorded few clinical infections during the study period, and contributions from sewer biofilms complicated attempts to relate wastewater measurements to clinical incidence. Although the abundance of two genes increased after resistant infections were identified in patients, another increase late in the study could not be attributed to a patient infection.

These studies clarify the respective roles of wastewater and clinical surveillance. Wastewater can flag a change that deserves attention. The health system must then determine whether it represents a clinically significant organism, human carriage, active infection, environmental persistence, or another source. Follow-up could include prioritized clinical testing, more localized wastewater sampling, recovery and sequencing of cultured isolates, or review by infection-prevention personnel. Wastewater initiates the investigation; it does not supply the complete explanation.

Building Surveillance Around Decisions

Program design should begin with the public-health decision the data are expected to support. Assay selection follows from that decision. A hospital seeking early evidence of an emerging carbapenemase has different requirements from a national system comparing long-term resistome patterns. The sampling location, collection frequency, analytical method, and response threshold should reflect the objective.

Sampling scale is particularly important. Facility-level monitoring may improve sensitivity to organisms concentrated in a hospital or long-term care population. Municipal influent can provide broader community coverage but may dilute localized signals and combine contributions from residential, commercial, health care, industrial, animal, and environmental sources. Sampling at both levels may help investigators distinguish a facility-associated signal from wider community circulation, although complementary evidence would still be required to determine its precise origin.

Methodological consistency is equally important for trend interpretation. Choices involving sampling, transport, concentration, extraction, sequencing, and bioinformatics can materially alter the resistance profile reported from a sample. Variation among workflows limits comparison and reproducibility, particularly when programs attempt to distinguish a meaningful change from analytical noise.8

Wastewater data therefore need to be integrated with clinical microbiology, antimicrobial susceptibility testing, infection-prevention information, and genomic epidemiology. Antimicrobial-use data may add context, while clinical isolates can establish whether a detected resistance mechanism occurs in an organism causing human infection.2,9 The surveillance design should also specify what finding would trigger confirmatory investigation and what additional evidence would be required before initiating a clinical or infection-prevention response.

This operational framework remains under development. The World Health Organization’s 2025 assessment characterized most AMR wastewater studies as pilots or proofs of concept.2 No standard methods had emerged, and routine programs had not yet demonstrated integration with health-impactful actions across multiple independent settings. Possible applications include prioritizing diagnostic testing, strengthening infection-control measures, and informing longer-term decisions about preferred antimicrobials, but these uses remain prospective rather than operationally established.

The need for interpretation standards is becoming more immediate. The recast European Union Urban Wastewater Treatment Directive introduces a requirement to monitor AMR in wastewater, even as WHO reports that standardized methods, validated triggers, and routinely demonstrated health-impactful responses remain undeveloped.2,10

An Alert, Not a Diagnosis

Wastewater can show that a resistance signal is present, where it is appearing, and whether it is changing. It cannot determine by itself who carries it, whether it is causing infections, or how it moved through the population. Its most credible role is to identify where clinical and epidemiological investigation should begin.

An AMR wastewater program becomes useful when every meaningful signal has a method of confirmation and a defined health-system response.

Antimicrobial resistance (AMR) policy has long had an intuitive objective: reduce unnecessary exposure to antibiotics so that they remain effective when they are needed. That objective addresses only one of two failures that AMR policy must manage. In many health systems, inappropriate antibiotic use coexists with inadequate access to effective treatment. Limited access to quality-assured antimicrobials can lead to inadequate treatment, substandard medicines, or incorrect dosing, while inappropriate use can persist within the same broader health system.

New guidance from the World Health Organization (the WHO) makes that tension much harder for national AMR programs to treat as a secondary concern. The September 2026 guidance provides 20 recommendations for integrating equity across the 13 core interventions in the WHO’s people-centred approach to AMR.1 Its evidence synthesis concentrates on three areas where inequity can influence AMR or undermine interventions intended to control it: water, sanitation, and hygiene; access to health services and essential antibiotics; and surveillance of AMR and antimicrobial use.

The guidance matters less for documenting unequal AMR risk than for asking how AMR interventions themselves distribute access, exposure, and benefit. The WHO asks countries to examine whether stewardship policies, diagnostic systems, prescribing practices, procurement mechanisms, and restrictions on antibiotic sales produce different effects across populations. Equity, in this formulation, becomes something that national action plans (NAPs) need to measure and manage rather than simply acknowledge.

A health system must therefore reduce unnecessary antimicrobial exposure while making effective treatment reliably available to patients who need it. The difficult policy question is how to accomplish both at once.

Why Lower Use Is Not Always Better Use

Global antibiotic-use data illustrate why AMR policy cannot be reduced to a single goal of lowering consumption. The WHO’s Global Antimicrobial Resistance and Use Surveillance System reported a tenfold difference between the highest- and lowest-using countries and territories in its 2022 antibiotic-use data set. The WHO cautioned that these differences require further investigation, but said they point toward both antibiotic overuse and low access. Some low- and middle-income reporting settings also recorded little or no use of reserve antibiotics, which are intended for infections caused by the most drug-resistant bacteria.2

Low antibiotic consumption may reflect successful stewardship, but it can also signal inadequate access. Limited access to health care, weak diagnostic capacity, unaffordable medicines, fragile supply chains, or lack of access to antibiotics needed for particular resistant infections can all suppress use without producing better care.

The 2026 equity guidance places this tension within a wider critique of approaches that concentrate heavily on individual behaviour. The WHO notes that AMR interventions have often prioritized changing what patients and health workers do while paying less attention to structural conditions that shape those decisions. Self-medication, for example, may be linked to lack of health coverage or health insurance rather than simply inadequate awareness of appropriate antibiotic use. The guidance also identifies inadequate living and working conditions and insufficient health-promoting infrastructure among factors that can influence antimicrobial use.

That perspective is consistent with a broader approach to sustainable antibiotic access proposed in the stewardship literature. Many patients in low- and middle-income countries lack effective and affordable treatments, while conventional stewardship programs may be difficult to implement where financial resources, health-system capacity, and specialized personnel are limited. A context-adapted approach still seeks the right antibiotic at the right time and for the right duration, but it adds affordability and practical access to the equation.3

The implication is not that stewardship should become less restrictive wherever access is poor. It is that stewardship has to distinguish inappropriate exposure from necessary treatment rather than treating lower consumption as an adequate proxy for success.

When Restricting Antibiotics Changes Access

One of the most consequential passages in the new WHO guidance concerns a familiar AMR intervention: restricting non-prescription sales of antimicrobials.

Such regulations have a clear stewardship rationale. The WHO’s people-centred framework includes regulation of non-prescription antimicrobial sales among the interventions intended to support appropriate, quality-assured treatment.4

The new equity guidance adds another criterion for judging whether such policies work. Recommendation 20 asks countries to assess the unintended consequences of policies or regulations that reduce non-prescription antimicrobial sales, specifically examining their effects on access to essential antimicrobials among different population groups. It also calls for consideration of fair and equitable access to new antimicrobials.

That recommendation exposes the access paradox in a particularly concrete form. A rule can reduce inappropriate antibiotic purchases and still create a problem if people who previously depended on that channel have no viable route to medical evaluation, diagnosis, prescription, and affordable treatment. The relevant question therefore extends beyond whether fewer antibiotics are sold without prescriptions. Policymakers also need to know whether patients who require those drugs can obtain them through an appropriate alternative pathway.

This does not argue against regulation. It changes what regulators have to evaluate. A restriction designed to improve antibiotic use operates within a health system whose clinics, laboratories, pharmacies, financing mechanisms, and supply chains may be distributed unequally. Applying the same rule to everyone does not guarantee the same effect for everyone.

For AMR policy, unintended access consequences therefore form part of the evaluation of whether an intervention succeeds.

Access Depends on the Health System

Treating antibiotic access as a simple question of whether a product is technically available obscures much of the problem. A drug can be registered in a country and still remain practically inaccessible because diagnosis, financing, procurement, distribution, or local availability fails somewhere along the path to treatment.

The WHO consequently treats access as a health-system issue extending from insurance coverage and diagnostics to procurement and supply continuity. Its recommendations call for health insurance or health-benefit packages to cover AMR-responsive services, including diagnostics and antimicrobials for drug-resistant infections, with particular attention to groups facing the greatest risk of inequity. They also ask countries to identify barriers to quality-assured medicines, including obstacles that push populations away from formal health-care channels and increase the risk of purchasing substandard or falsified antimicrobials.

The guidance extends the same logic into supply planning. Medicine forecasting and procurement should consider local infection epidemiology by sex, gender, age, and at least one health-equity indicator, in part to ensure that relevant antimicrobials are included and stockouts are reduced. Health-care facilities are also encouraged to develop contingency plans that preserve access to essential antibiotics during supply disruptions, including in low-resource, rural, and remote settings.

Recent regional evidence illustrates how varied these failures can be. A 2026 WHO survey of antibiotic access in the African Region was sent to all 47 Member States, with 25 responding. Countries reported barriers spanning stockouts, affordability, supply chains, registration, and public-sector availability. These barriers affected not only newer reserve antibiotics but also more widely used agents, indicating that access problems can occur across the antibiotic portfolio for different reasons.5

Similar problems appear in Latin America and the Caribbean. A 2026 review of reserve antibiotic access identified interconnected barriers involving regulatory approval, manufacturer participation, formulary inclusion, procurement, and health-technology-assessment capacity. It also found greater constraints in public hospitals than in private facilities, illustrating how substantial access inequalities can exist within a single national market rather than only between richer and poorer countries.6

Medicine availability alone therefore provides an incomplete measure of access. Effective treatment depends on a functioning pathway from care and diagnosis through prescribing, financing, procurement, and supply. Weakness at any point can turn nominal availability into practical absence.

The Patients Missing From AMR Surveillance

There is another difficulty with designing equitable AMR policy: some of the populations experiencing the greatest barriers to care may also be the least visible in surveillance systems.

Routine clinical microbiology provides much of the information used to understand resistance patterns. When aggregated, laboratory data can inform empiric treatment guidance, antibiotic procurement, national surveillance, and NAPs. However, those data depend on which patients reach a facility, which patients are tested, and what diagnostic capacity exists once they arrive.7

Those conditions differ substantially across settings. In some low-income countries, microbiological diagnostics are not routinely available. Where cultures depend on willingness or ability to pay out of pocket, laboratories are under-resourced, or patients do not use formal health-care services, the infections captured in routine datasets may be incomplete or unrepresentative. The same analysis notes that groups outside formal health care may be systematically underrepresented in hospital data even when marginalized populations carry substantial burdens of drug resistance.

The consequence extends beyond statistical completeness. If some populations are less likely to enter the diagnostic system, their infections are also less likely to become surveillance observations. Policies built from those observations may therefore describe resistance accurately for the population being tested while offering a less complete picture of the population actually experiencing infection.

The WHO’s equity recommendations respond directly to this limitation. Countries are asked to disaggregate relevant AMR program data, at minimum, by sex, age, and location, with other equity indicators included where feasible. The WHO also recommends defining equity variables within existing AMR and antimicrobial-use surveillance systems and strengthening the capacity of surveillance professionals to analyze inequalities and inequities.

Data disaggregation alone cannot recover patients who never enter the data set. The WHO therefore also recommends periodic reviews of diagnostic services to identify population groups experiencing unequal access to testing for drug-resistant infections.

Equity-responsive surveillance consequently requires two questions. What differences are visible within the data already being collected? And whose infections are missing because the underlying system never captured them?

From Equity Language to Equity Metrics

Many AMR policies already contain language about vulnerable populations, equitable health care, or leaving no one behind. The harder question is whether those commitments affect budgets, indicators, implementation decisions, and evaluation.

An analysis published in 2025 examined NAPs from 14 West African countries using an equity-focused framework. Most plans did not explicitly reference equity, none included equity-related indicators in their monitoring frameworks, and interventions such as hygiene promotion, public-awareness campaigns, and health-worker training were usually generic rather than adapted to marginalized populations. Although stakeholder engagement was often multisectoral, disadvantaged groups were seldom specifically included.8

Those findings cannot be generalized to every national AMR plan, but they highlight the difference between recognizing that inequity exists and building a system capable of acting on it.

The new WHO guidance makes that distinction explicit from its first recommendations. Countries are asked to incorporate equity into the AMR situation analysis used to develop or update a NAP and then carry equity-responsive actions into objectives, prioritized activities, targets, indicators, and budgets. The guidance also calls for the disaggregated data needed to determine whether those measures are reaching different populations as intended.

That is ultimately the document’s most consequential shift. An equity commitment written into a statement of principles may have little influence over implementation. An equity variable in surveillance, a funded activity in a budget, a target in a monitoring framework, or an identified access barrier in procurement creates something that can be acted upon and evaluated.

The same logic reaches clinical stewardship. The WHO calls for prescription audits, treatment guidelines, and antimicrobial-stewardship program documents to be reviewed for evidence that unconscious prescribing bias linked to social, economic, or other characteristics may reinforce inequity. The current evidence base does not establish how frequently such bias occurs across health systems or which populations are most affected. The importance of the recommendation is therefore as a mandate to measure prescribing inequity rather than as evidence of its prevalence.

Equity becomes operational when AMR programs can detect disparities, identify the mechanisms producing them, and determine whether interventions narrow or widen them.

How the Definition of Success Changes

If equity becomes a design and performance criterion, the questions used to evaluate AMR interventions also need to change.

A program designed to reduce inappropriate antibiotic sales would still ask whether those sales declined, but it would also need to determine whether patients who require treatment retained a workable route to appropriate care. A surveillance program would still measure resistance trends, but it would also examine whether the populations contributing specimens reflect the populations affected by infection. Procurement systems would still monitor medicine availability, but they would also ask whether stockouts or shortages disproportionately compromise treatment for particular populations.

The same principle applies to diagnostic coverage and prescribing. Expanding laboratory capacity has limited equity value if some groups remain unable to reach or afford testing. A treatment guideline may be evidence based in its clinical content while still requiring evaluation of whether implementation produces systematic differences in who receives appropriate therapy.

The new guidance does not replace established AMR objectives with a separate equity agenda. It changes how those objectives are judged. Consumption, stewardship adherence, surveillance completeness, medicine availability, and policy implementation remain important, but aggregate performance can conceal substantial differences between populations.

The sustainable-access approach offers a useful clinical endpoint: the right antibiotic, delivered when needed, for an appropriate duration, and at an affordable price.3 Reaching that endpoint may require reducing use in some parts of a health system while expanding effective access in others.

Appropriate use is therefore not necessarily synonymous with less use everywhere. It requires enough information about who is receiving treatment, who is not, and why to distinguish excess from unmet need.

A Better Understanding of Stewardship Success

Equity is not new to the AMR conversation. The WHO has previously addressed gender inequalities in AMR NAPs, and its 2023 people-centred approach already placed equitable access to health services within a package covering prevention, essential services, diagnosis, and appropriate treatment.4

What the 2026 guidance adds is a more operational proposition. Equity appears not only as a principle but in recommendations about surveillance variables, diagnostic access, insurance and health benefits, medicine procurement, supply-chain continuity, prescribing audits, treatment guidance, regulatory policy, targets, indicators, and budgets.

That matters because AMR policy has to solve two failures at the same time. Antibiotics are used when they should not be, accelerating the selection pressure that contributes to resistance. Elsewhere, or even within the same health systems, patients who need effective treatment cannot reliably obtain it.

A stewardship program capable of correcting only the first failure remains incomplete. The more meaningful measure of success is whether a health system can prevent unnecessary antimicrobial use while reliably delivering effective treatment to every patient who needs it.

The Biotechnology Innovation Organization’s annual conference is the industry’s mecca, a gathering of thousands where biotech leaders showcase top-line data on experimental drugs and diagnostics, thousands of attendees wander eye-catching company exhibits that stretch for the equivalent of city blocks, and hope of new deals springs eternal.

And last week’s gathering in San Diego did not disappoint. BIO 2026 featured numerous sessions focused on drugs and diagnostics for key clinical challenges, from Alzheimer’s to oncology, from gene therapy to regenerative medicine. But one clinical field was dramatically under-represented. Only five of the more than 130 panel discussions focused on infectious disease. One of those focused on how to remain prepared for outbreaks when huge cuts have been proposed for pandemic prevention and infectious disease research by the U.S. government, the largest funder of such research on the planet. And just a single session focused on one of our largest public-health problems: antimicrobial resistance.

No wonder antimicrobial resistance is called the silent pandemic.

Over the next quarter century, drug-resistant pathogens are projected to kill 39 million people, which is more than five times the number who died from COVID from 2020 to 2024.

Many of the people most at risk are already medically vulnerable. They’re residents of low- and middle-income countries with poorly funded healthcare systems. They’re children. They’re poor. They’re elderly. And many are projected to succumb following medical procedures that today are relatively safe. These include lifesaving cardiovascular surgeries, hip transplants and cancer treatments of all sorts. Without effective antibiotics we can trust to squelch incipient infections, much of modern medicine is at risk.

It doesn’t have to be this way.

disAMR exists to counter the threat of antimicrobial resistance. As an organization, we’re working to empower stakeholders in academia, industry, NGOs, and public health by building infrastructure the AMR fight needs.

We’ll spotlight innovative solutions and the people developing them, we’ll explore paths forward, and we’ll connect the AMR community by convening digital events, and, eventually, live ones. That community includes our more than 700 AMR Ambassadors, who span the globe and deserve a special shout-out for their dedication to the cause.

Together, we’re building a space for those passionate about fighting antimicrobial resistance to learn about the challenge, explore solutions and meet the people developing them. It’s time to make some noise about the silent pandemic. Come join us, and if you are not already receiving our newsletter, disAMR Matters, sign up to receive it.

The WHO report Mainstreaming antimicrobial resistance into primary health care: International workshop report (2025) summarizes a multi-country workshop held in Bangkok that aimed to integrate antimicrobial resistance (AMR) interventions into primary health care (PHC)-oriented health systems as a core strategy for strengthening universal health coverage. Recognizing that 80–90% of antibiotics are prescribed in primary care, the report emphasizes embedding antimicrobial stewardship (AMS), appropriate antimicrobial use (AMU), infection prevention and control (IPC), surveillance, and community engagement within PHC policies and service delivery. Drawing on country experiences (including Thailand, Indonesia, Kazakhstan, and Sweden), it identifies key challenges such as weak policy integration, limited AMR data, insufficient funding, over-the-counter antibiotic sales, workforce capacity gaps, and low prioritization of AMR. The workshop introduced practical tools, notably the WHO AMR-PHC scoping tool, to help countries align AMR-sensitive actions with PHC reforms, strengthen governance and multisectoral collaboration, and promote sustained behavioural change at provider and community levels. The overall conclusion is that systematically integrating AMR into PHC-oriented systems—supported by political commitment, financing, data-driven stewardship, regulation, and community education—is essential to build resilient health systems capable of mitigating AMR sustainably.

The Fleming Fund highlights how countries in East and Southern Africa have strengthened antimicrobial resistance (AMR) strategies by building coordinated, data-driven surveillance systems across human and animal health sectors. Since 2016, support has helped countries such as Uganda, Tanzania, Kenya, Malawi, Zimbabwe, Zambia, Eswatini and Rwanda improve laboratory capacity, standardise data collection, and report to global platforms like WHO GLASS and WOAH ANIMUSE. Importantly, surveillance data is increasingly informing national policy and stewardship decisions—for example, guiding antibiotic use in poultry in Zimbabwe and revising carbapenem prescribing practices in Malawi. Tools such as WHONET enable more systematic reporting and early detection of resistance trends. While progress has been substantial, the report stresses that sustained financing, trained personnel, and long-term system integration remain essential to ensure that AMR data continues to translate into effective policy and public health action.

Researchers in Los Angeles identified a previously unreported genotype of extensively drug-resistant (XDR) Shigella sonnei harboring an AmpC-type β-lactamase gene blaDHA-1 in isolates from two unrelated patients with no clear epidemiologic link, suggesting broader community spread beyond traditional high-risk groups; whole-genome sequencing showed the isolates were closely related, and their resistance profiles compromised most commonly used oral antibiotics, underscoring an urgent public health concern and the need for enhanced genomic surveillance and tailored treatment strategies.

This multicentre stepped-wedge cluster-randomized trial across 10 hospitals evaluated whether a discharge-focused antimicrobial stewardship intervention—combining guideline dissemination with prospective audit and feedback for patients nearing discharge—could reduce antibiotic overuse at hospital discharge. Among more than 21,000 admissions, the intervention did not reduce the proportion of patients prescribed antibiotics after discharge, nor did it shorten postdischarge or inpatient antibiotic duration, length of stay, or 30-day readmission rates. However, detailed chart review showed a significant improvement in prescribing quality, with optimal antibiotic use at discharge more common during the intervention period. Most surveyed prescribers perceived the initiative as beneficial. Overall, while discharge-focused audit and feedback improved the appropriateness of antibiotic prescribing, it did not reduce overall antibiotic use, indicating that additional or alternative stewardship strategies are needed at this transition of care.

Since the 2017 global call to action, the Hub has strengthened global cooperation in antimicrobial resistance (AMR) research and development by reducing duplication and improving the efficient allocation of resources. It has built a comprehensive knowledge base across push and pull incentives, fostered collaboration through working groups, member engagement, and policy outputs, and advocated for AMR action at the highest political levels, including the G20 and G7.

Its flagship Dynamic Dashboard is the world’s most comprehensive database on public and philanthropic AMR R&D investments, covering more than 18,100 projects valued at USD 17.4 billion, alongside WHO pipeline data on antibacterials and bacterial vaccines and a global overview of incentive mechanisms. Additional impact is driven through dedicated working groups on One Health AMR and pull incentives, supported by active contributions from members, observers, and stakeholders.