When Antibiotics Stop Working: Why Antimicrobial Resistance Is a Public Health Emergency

Antimicrobial resistance is still often spoken of as a crisis waiting in the wings — something dramatic that will arrive later, once “the antibiotics stop working.” That language is already behind the evidence. According to the World Health Organization, antimicrobial resistance occurs when bacteria, viruses, fungi and parasites no longer respond to the medicines used to treat them, making infections harder — and sometimes impossible — to treat. In clinics and hospitals today, that process is already narrowing options for urinary tract infections, pneumonia, wound infection and bloodstream infection. It also sits quietly underneath surgery, cancer care, intensive care and neonatal medicine: those pathways assume bacterial complications remain treatable. When that assumption weakens, the loss is not only clinical. It is a public-health emergency measured in preventable deaths, longer illness, and health systems that lose margin for ordinary care.

What antimicrobial resistance actually means

“Antibiotics stop working” is a public shorthand for a biological and clinical process. Microbes can acquire or develop traits that allow them to survive medicines that once killed or slowed them. In practice, that may mean a first-line antibiotic no longer clears an infection; that a second- or third-line drug is needed; or that the remaining options are slower, more toxic, harder to obtain, or simply unavailable. Resistance is not the same as a person becoming “immune” to antibiotics. It is a property of the microbe — and once resistant strains circulate, they can spread between patients, facilities and countries.

Not every resistant infection is immediately untreatable. Many remain manageable with alternative regimens. The public-health problem is the narrowing corridor: each lost antibiotic class reduces the margin of safety for everyday infections and for sepsis. Elective surgery, chemotherapy, transplantation and intensive care all depend on treatable bacterial complications. When that assumption weakens, the risk reaches ordinary care pathways — which is why antimicrobial resistance belongs in prevention and health-system planning, not only in specialist footnotes.

A global burden already measured in millions of deaths

The strongest global death estimates for bacterial antimicrobial resistance come from the Global Research on Antimicrobial Resistance (GRAM) Project, published in The Lancet and cited in WHO’s antimicrobial-resistance fact sheet. For 2021, the analysis estimated about 4.71 million deaths associated with bacterial antimicrobial resistance and about 1.14 million deaths attributable to it. Those two figures are not interchangeable. Attributable deaths answer a causal counterfactual — deaths judged unlikely to have occurred in the absence of resistance. Associated deaths describe a wider burden in which resistant infection was present among those who died. Confusing them either inflates or understates the problem; both matter for public health because one measures a stricter causal claim and the other the scale of severe resistant infection in fatal pathways.

Looking ahead, the same Lancet modelling forecasts roughly 1.91 million attributable and 8.22 million associated deaths in 2050 if trajectories continue without stronger control. Those numbers are modelled forecasts — scenarios under stated assumptions, not destiny. Even so, they underline a hard point: bacterial antimicrobial resistance is already among the world’s major infectious threats, and delay does not freeze the curve. Waiting for a single “breakthrough antibiotic” is not a strategy. The burden is present tense.

In 2021, about 4.71 million deaths were associated with bacterial antimicrobial resistance and about 1.14 million were attributable to it.

4.71M

Associated

1.14M

Attributable

Sources: Lancet GBD 2021 Antimicrobial Resistance Collaborators (2024); WHO antimicrobial resistance fact sheet.

Laboratory surveillance sharpens the same message. WHO’s Global Antimicrobial Resistance and Use Surveillance System (GLASS) picture for 2023 indicates that about one in six laboratory-confirmed bacterial infections causing common infections worldwide were resistant to antibiotics. Between 2018 and 2023, resistance rose in more than 40% of the pathogen–antibiotic combinations monitored, with average annual increases of about 5–15%. At the same time, WHO notes that Access-group antibiotics — first-choice options for many common infections — accounted for only about 53% of global human antibiotic use in 2022, against a target of at least 70% by 2030. Stewardship is not only about using fewer antibiotics overall, but about using the right ones when they are needed — and preserving last-line agents for when they are truly required.

Europe is not exempt

High-income European health systems sometimes treat antimicrobial resistance as a problem “elsewhere.” The European evidence does not support that comfort. ECDC estimates that more than 35,000 people die each year in the EU/EEA as a direct consequence of antimicrobial resistance, based on estimates using 2020 EARS-Net data. That is already a large, recurring annual toll — not a theoretical future cost. Through EARS-Net, ECDC also tracks resistance in invasive isolates across the EU/EEA. The Annual Epidemiological Report for 2024 shows both success and warning signs. Estimated EU incidence of methicillin-resistant Staphylococcus aureus (MRSA) bloodstream infection was 4.48 per 100,000 population — down 20.4% compared with 2019 and described as meeting the 2030 target trajectory. By contrast, third-generation cephalosporin-resistant Escherichia coli bloodstream infection incidence was 11.03 per 100,000 (+5.9% vs 2019), and carbapenem-resistant Klebsiella pneumoniae was 3.51 per 100,000 — a 61% increase compared with 2019, against a 2030 target of 2.07 per 100,000.

Estimated EU incidence of carbapenem-resistant Klebsiella pneumoniae bloodstream infection: 2.18 per 100,000 in 2019, 3.51 in 2024, and a 2030 target of 2.07.

2.18

2019

3.51

2024

2.07

2030 target

Source: ECDC, Antimicrobial resistance in the EU/EEA (EARS-Net), Annual Epidemiological Report for 2024. 2019 baseline calculated from the reported +61% rise to 3.51 in 2024; 2030 target 2.07 as cited by ECDC.

That mixed pattern resists a comforting story of uniform European improvement. Progress against MRSA shows coordinated infection prevention and stewardship can move population indicators. The rise in carbapenem-resistant K. pneumoniae shows other threats can advance at the same time — especially where last-line agents are already under pressure. For EU/EEA countries, the useful question is whether national action plans are closing the gaps EARS-Net keeps illuminating: laboratory capacity, hospital infection prevention, outpatient prescribing, and containment of highly resistant organisms before they become endemic.

What happens when antibiotics stop working

When treatment options shrink, the first consequences are clinical: longer illness, more complications, more intensive care, higher mortality risk for some infections, and greater reliance on toxic or poorly tolerated regimens. The second consequences are systemic: delayed surgery, harder cancer pathways, outbreak-control costs, and bed occupancy that crowds out other care. In other words, antimicrobial resistance does not only kill through “untreatable infection” headlines. It raises the baseline risk of care that modern medicine treats as routine.

CDC’s United States estimates illustrate the scale in a high-income setting. The agency reports more than 2.8 million antibiotic-resistant infections and more than 35,000 deaths in the United States each year, drawing on the 2019 antimicrobial-resistance threats assessment. Those figures are not global totals, and US surveillance definitions differ from European and global systems — but they show that even well-resourced systems already absorb a large annual toll. Read alongside ECDC’s EU/EEA estimate and the Lancet global burden, the pattern is consistent: antimicrobial resistance is already rewriting the risk profile of ordinary infection and of medicine that depends on reliable antibiotics.

Why resistance keeps spreading

Resistance spreads for reasons that are biological and organisational at once. Misuse and overuse of antimicrobials — in human medicine and beyond — create selection pressure that favours surviving resistant strains. Gaps in infection prevention and control, and in water, sanitation and hygiene (WASH), then do the rest: through hands, devices, contaminated surfaces and environments that allow transmission in hospitals, long-term care and communities. Limited access to vaccines, diagnostics and quality-assured medicines pushes systems toward broader empirical therapy, delayed confirmation, or unsafe substitutes. A weak research and development pipeline means that when last-line options fail, replacements arrive slowly. And because resistant organisms move with people, animals, food and environmental reservoirs, a One Health frame — linking human, animal and environmental health — is not optional rhetoric. It is how the ecology of resistance actually works.

None of this requires a villain narrative. It requires recognising antimicrobial resistance as a systems problem: prescribing culture, hospital hygiene, laboratory capacity, pharmaceutical supply, agricultural and environmental reservoirs where they contribute, and willingness to fund prevention that rarely makes headlines. The factors interact. Strong stewardship with weak infection prevention still loses ground; strong laboratories without outpatient stewardship still feed selection in the community.

What a serious public-health response looks like

A serious response is often reduced to the search for new antibiotics. New agents matter — especially for carbapenem-resistant gram-negative infections — but they are not a substitute for the quieter work that keeps existing medicines useful. The practical package is familiar because it has to be: antimicrobial stewardship; infection prevention and control; vaccination; diagnostics that shorten the time to targeted therapy; surveillance; equitable access to quality-assured medicines; research and development; and national action plans funded beyond a PDF. WHO, ECDC and CDC all return, in different language, to versions of that list. The harder question is whether countries treat it as continuous infrastructure. That framing now has an updated global policy anchor: in May 2026 the World Health Assembly adopted the Global Action Plan on Antimicrobial Resistance 2026–2036, a One Health framework meant to guide coordinated action through 2036 and to support countries in developing, implementing and financing ambitious national action plans — including progress toward the 2024 United Nations target of a 10% reduction in bacterial AMR-associated deaths by 2030.

Europe’s mixed EARS-Net results and WHO’s laboratory and use indicators point to the same test. Progress is possible — falling MRSA incidence in the EU shows that — but rising carbapenem-resistant K. pneumoniae and an Access share near 53% show that partial success is not enough. Without surveillance, prevention, stewardship and workable pathways when first-line drugs fail, every new antibiotic enters a system primed to burn through it.

Closing

Antimicrobial resistance is not waiting for a dramatic future announcement. The Lancet GRAM estimates already place bacterial antimicrobial resistance among the major causes of infectious death worldwide; WHO’s laboratory and antibiotic-use indicators show resistance common in everyday pathogens and Access antibiotics still short of the 2030 target; ECDC’s European death estimate and EARS-Net indicators show both achievable progress and dangerous reverse trends; and CDC’s United States figures show how large the annual toll already is in a well-resourced system. The honest reading is that “antibiotics stop working” is already a present constraint on care — not a metaphor for a distant collapse. The 2026–2036 Global Action Plan makes the policy direction clearer; the remaining test is whether national systems fund and deliver the infrastructure that plan describes.

What follows is less spectacular than a single breakthrough, and more demanding. Keep surveillance strong enough to see the problem early. Fund infection prevention as core capacity. Use antibiotics as if the next patient will need them. Expand vaccination and diagnostics so fewer infections require broad empirical therapy. And judge readiness not by strategy documents, but by whether bloodstream-infection indicators, antibiotic-use quality, and outbreak control improve under ordinary pressure — before the next crisis makes the narrowing of options unmistakable.

Sources & further reading

1. WHO — Antimicrobial resistance fact sheet
2. Lancet GBD 2021 Antimicrobial Resistance Collaborators — Global burden of bacterial antimicrobial resistance 1990–2021 and forecasts to 2050 (The Lancet, 2024)
3. ECDC — Antimicrobial resistance in the EU/EEA (EARS-Net): Annual Epidemiological Report for 2024
4. ECDC — Time to act and not react: how can the European Union turn the tide on antimicrobial resistance?
5. CDC — Antimicrobial Resistance Threats (data and research)
6. CDC — Antibiotic Resistance Threats in the United States, 2019
7. WHO — World Health Assembly adopts the Global Action Plan on Antimicrobial Resistance 2026–2036 (25 May 2026)
8. WHO — Global Antimicrobial Resistance and Use Surveillance System (GLASS) report 2025