In October 2025, the World Health Organization published its most comprehensive look yet at drug-resistant infections. Drawing on more than 23 million laboratory-confirmed infections reported by 104 countries, it found that one in six common bacterial infections in 2023 was resistant to standard antibiotic treatment. In WHO’s South-East Asia and Eastern Mediterranean regions, the figure was one in three. Between 2018 and 2023, resistance rose in more than 40% of the drug and bug combinations WHO monitors.
Antimicrobial resistance is often called the slow pandemic, and the name fits. There is no single outbreak, no daily case count on the evening news. Just a steady narrowing of the drugs that work.
The standard response is a stewardship campaign: prescribe less, finish your course, stop giving antibiotics to animals. All of that matters. But it tells only half the story, and in much of the world, the less important half. For millions of people with resistant infections, the problem is not too many antibiotics. It is not getting the right one at all.
Who is actually dying
The most detailed estimate of the burden comes from the Global Research on Antimicrobial Resistance project, published in The Lancet in September 2024. It estimated that bacterial resistance has been directly responsible for at least one million deaths a year since 1990, and forecast 39 million deaths directly attributable to resistance between 2025 and 2050.
Buried in that study is a striking shift. Between 1990 and 2021, deaths from resistant infections among children under five fell by 50%, which the authors linked to vaccination, including pneumococcal vaccines, and better water and sanitation. Over the same period, deaths among people aged 70 and older rose by more than 80%, driven by ageing populations and older people’s vulnerability to infection.
So resistance is increasingly a disease of ageing. It is concentrated in hospital wards, intensive care units and nursing homes. It is closely tied to surgery, cancer treatment and chronic illness, the very parts of modern medicine that depend on effective antibiotics.
And the most effective interventions against child deaths were not antibiotic policies at all. They were vaccines and clean water, which prevent the infections before anyone reaches for a prescription.
The access gap nobody campaigns about
In 2025, a study in The Lancet Infectious Diseases led by the Global Antibiotic Research and Development Partnership (GARDP) looked at nearly 1.5 million carbapenem-resistant Gram-negative infections across eight low- and middle-income countries, including India, Pakistan, Brazil and South Africa. Using sales data for the antibiotics that work against these infections, it estimated that appropriate treatment was potentially available for just 6.9% of cases.
Read that again. For more than nine in ten patients with some of the most dangerous resistant infections in those countries, the drugs that could cure them were not reaching them.
This is the uncomfortable part of the AMR story. Much of the rich-world messaging treats antibiotics as something to be rationed. In many settings, the same drugs are simultaneously overused in some places and absent in others. A pharmacy may sell broad-spectrum antibiotics without a prescription for a cold, while the hospital down the road cannot get the last-line drug a septic patient needs.
A stewardship campaign that only says “use less” risks making the access problem worse, especially for new antibiotics that are already hard to find.
The market that punishes success
Behind both problems sits a broken market. A new antibiotic that works against resistant bacteria is exactly the drug doctors should hold in reserve. That is good medicine and terrible economics. A company that spends years developing one is told, in effect, to sell as little as possible.
The United Kingdom has tried to fix this. Its antimicrobial products subscription model, sometimes called the “Netflix model,” pays manufacturers a fixed annual fee for access to selected antibiotics, regardless of how much is used. After a pilot with Pfizer and Shionogi, the NHS adopted the model permanently in 2024, with annual fees in England ranging from £5 million to £20 million per product depending on its assessed value.
It is a small pool of money by pharmaceutical standards. But it is the right idea: pay for having the drug available, not for how many boxes are sold. Few other countries have followed at meaningful scale, and a single national scheme cannot sustain a global pipeline on its own.
What would actually change the curve
If resistance is growing among older, sicker patients in hospitals, and if the drugs that treat it are both undersupplied and under-rewarded, then the priorities look different from the usual stewardship poster:
- Prevent infections first. Vaccination, infection control in hospitals and nursing homes, and water and sanitation cut antibiotic use and resistance at the same time.
- Treat access as part of stewardship. The goal is the right drug for the right patient, not fewer drugs everywhere.
- Pay for availability. More countries need pull incentives that reward new antibiotics for existing, not for selling.
- Invest in diagnosis. Much inappropriate prescribing happens because a clinician cannot tell what the patient has. A test that answers that question saves drugs.
A scenario from both sides of the gap
Consider two patients on the same day.
The first is a young man in a large city in South Asia with a sore throat and a mild fever. He goes to a private pharmacy, describes his symptoms and leaves with a course of broad-spectrum antibiotics. He takes three days’ worth, feels better and stops. The infection was probably viral. The antibiotics did nothing for him, but they did apply selective pressure to the bacteria living in his gut.
The second is an older woman in a hospital a few kilometres away, recovering from surgery. She develops a bloodstream infection. The laboratory, when it eventually reports, finds a carbapenem-resistant organism. The drugs that might work are expensive, not on the hospital formulary, or simply not available in the country. Her doctors try what they have.
Both of these patients are part of the AMR story, and a policy that only addresses one will fail the other. Tighter pharmacy controls would help the first patient’s case and do nothing for the second. Better access to new antibiotics would help the second and, if poorly managed, make the first problem worse. The GARDP estimate that appropriate treatment was potentially available for only 6.9% of carbapenem-resistant infections across eight countries is the second patient’s story told at scale.
Why the market fails, in plain terms
The economics are worth spelling out because they explain why the pipeline is thin.
A company that develops a new cancer drug expects to sell it widely at a high price for the length of its patent. A company that develops a new antibiotic against resistant bacteria faces the opposite. Doctors are told to hold it in reserve. Stewardship programmes restrict its use. Treatment courses are short. By the time resistance makes it essential, the patent may be close to expiry.
Research and development costs are similar. Revenues are much lower. Investors notice, and capital flows to therapeutic areas where success is rewarded with sales.
Subscription models like the UK’s attempt to break that link between volume and revenue. The logic is simple: pay a fixed amount for the drug being available, not for each dose sold. But the amounts must be large enough, and paid by enough countries, to make development worth the risk. One country’s subscription, however well designed, cannot sustain a global pipeline alone.
The strongest counterargument
A serious objection to emphasising access is that it could accelerate resistance.
If newer antibiotics become widely available in health systems with weak stewardship, weak laboratory capacity and unregulated pharmacy sales, they may be misused just as older drugs were. Resistance to them could emerge quickly, wasting the few new options we have. From this perspective, restricting access is a painful but necessary protection for everyone.
This argument has real merit. Many older antibiotics lost effectiveness precisely because they were widely available without controls. Any access programme that ignores stewardship will repeat that history.
But the argument goes wrong when it becomes a reason to accept that patients with resistant infections in poorer countries go untreated. The answer is to pair access with stewardship: getting reserve antibiotics into hospitals with diagnostic capacity and prescribing controls, rather than onto pharmacy shelves. WHO’s AWaRe classification, which groups antibiotics into Access, Watch and Reserve categories, is designed precisely to support this kind of targeting. The choice is not between access and stewardship. It is about where each belongs.
Second-order effects
Ageing will make AMR harder to see. As the GRAM findings show, deaths are shifting toward older adults. Deaths among frail, elderly patients with multiple conditions are often attributed to their underlying illness rather than to a resistant infection. The burden may grow while remaining statistically quiet.
Modern medicine is exposed. Surgery, chemotherapy, organ transplants and intensive care all depend on reliable antibiotics to prevent and treat infections. As resistance spreads, the risks of these procedures rise. That makes AMR a threat not just to infectious disease care but to the whole of hospital medicine.
Vaccines become AMR tools. The GRAM study linked falling child AMR deaths partly to vaccination. Vaccines that prevent bacterial infections, and even those that prevent viral infections that often lead to antibiotic prescriptions, reduce antibiotic use. That makes vaccine programmes part of the AMR response, although they are rarely funded or counted that way.
The measurement problem underneath
One more reason AMR stays slow and quiet: much of the world cannot measure it well. Surveillance depends on laboratories that culture samples and test them against antibiotics. Many hospitals in low-income countries have limited microbiology capacity, so clinicians treat on best guesses and resistant infections go unrecorded. The WHO’s GLASS report draws on millions of confirmed infections, but confirmed is the operative word. Where there is no laboratory, there is no data point.
That has a practical consequence. Countries with the weakest laboratories may look, statistically, as if they have less of a problem. Funding for diagnostic capacity is therefore not a technical side issue. It is what makes the problem visible enough to act on, and it is also what lets clinicians switch from broad-spectrum guesses to targeted treatment.
What this means for decision-makers
For ministries of health, it means treating AMR as a health system problem, not a single programme. The biggest gains are likely to come from infection prevention in hospitals, laboratory capacity so that clinicians know what they are treating, and regulated access to the right drugs.
For hospital leaders, it means investing in infection prevention and control, which reduces both infections and antibiotic use, and in stewardship teams that help clinicians choose the right drug rather than simply restricting options.
For funders, it means backing pull incentives and access programmes together, and counting vaccine and water investments as AMR interventions.
For clinicians, it means recognising that stewardship includes making sure the patient who needs a reserve antibiotic gets it, promptly.
What to watch next
Watch whether other high-income countries adopt subscription-style payments at a scale that matters. Watch whether the next rounds of WHO GLASS data show resistance continuing to climb across the monitored combinations. And watch whether access programmes for reserve antibiotics reach the countries where the GARDP study found the largest treatment gaps.
The takeaway
Antimicrobial resistance will not be solved by persuading people to take fewer pills. It will be solved, if it is, by preventing infections, getting the right drugs to the patients who need them, and paying companies to build drugs we hope never to use.
That is a harder message than “finish your course.” It is also the accurate one.
Where you work, is the bigger problem antibiotic overuse or lack of access to the right drug? We’d like to hear which.
