Superbugs Could Kill 39 Million People by 2050

The 39 Million Question

Here’s a number that should stop you cold: 39 million.

That’s how many people the latest global forecast says could die directly from drug resistant infections between now and 2050 (Naghavi et al., 2024). Another 169 million deaths will be associated with superbugs, meaning the infection was present and contributed to their death even if it wasn’t the sole cause.

But here’s the part that makes this more than just another terrifying headline. The same study shows we have already cut superbug deaths in half for children under five since 1990. We know what works. We just haven’t done it for everyone.

The new paper, published in The Lancet by a team led by Mohsen Naghavi, Stein Emil Vollset, Kevin Ikuta, and Lucien Swetschinski, is the first comprehensive look at bacterial antimicrobial resistance (AMR) from 1990 all the way to 2050. It covers 22 pathogens, 84 drug pathogen combinations, and 204 countries. The researchers analyzed 520 million individual records. This is not a model built on thin air. This is epidemiology at its most ambitious.

And what it reveals is a story of two worlds. One world is getting better. The other is getting worse. Fast.

The Child Who Got Saved and the Grandparent Who Did Not

The most hopeful line in the entire paper is buried in the abstract: from 1990 to 2021, deaths from AMR decreased by more than 50% among children younger than 5 years (Naghavi et al., 2024).

That is extraordinary. In three decades, we halved the number of young children dying because antibiotics stopped working. Better sanitation, vaccines, improved infection control in hospitals, and wider access to basic antibiotics all played a role. The study shows this trend held across every global super region. It wasn't a fluke of one wealthy country.

Now the other side. Among adults 70 years and older, AMR deaths increased by more than 80% over the same period (Naghavi et al., 2024). Not a small uptick. An explosion.

Why? Older people have more health problems. They spend more time in hospitals and nursing homes, where superbugs thrive. Their immune systems are weaker. And as the global population ages, the number of people in this vulnerable group is growing fast. The authors forecast that by 2050, people 70 and older will account for 65.9% of all AMR attributable deaths (Naghavi et al., 2024).

So the same interventions that saved children worked. But we did not apply them with the same urgency to older adults. The result is a demographic split in the AMR crisis that will define the next 25 years.

The Rise of MRSA and the Carbapenem Crisis

Not all superbugs are created equal. The study tracks which specific pathogens and drug resistance combinations are driving the numbers.

Methicillin resistant Staphylococcus aureus, or MRSA, is the single biggest mover. In 1990, MRSA was associated with 261,000 deaths globally. By 2021, that number had more than doubled to 550,000 associated deaths, and 130,000 deaths directly attributable to the infection (Naghavi et al., 2024). MRSA is the poster child for why we need new antibiotics. It used to be a hospital problem. Now it is everywhere.

But the more alarming trend involves Gram negative bacteria. These are the hard shelled bugs that cause pneumonia, bloodstream infections, and urinary tract infections. They are also the ones that have learned to defeat carbapenems, our strongest class of last resort antibiotics.

Resistance to carbapenems rose from 619,000 associated deaths in 1990 to 1.03 million in 2021 (Naghavi et al., 2024). That is a 66% increase in three decades. And the pipeline for new drugs that can kill Gram negative bacteria is nearly empty. Developing a new antibiotic takes 10 to 15 years and costs over a billion dollars. The market incentives are broken. Pharmaceutical companies have largely abandoned the field.

The study makes this explicit. Under a scenario where new drugs targeting Gram negative pathogens are developed, the authors estimate 11.1 million AMR deaths could be averted between 2025 and 2050 (Naghavi et al., 2024). That is a lot of lives. But it is also a massive if.

How They Built the Map

To understand why this paper matters more than previous estimates, you need to know what they actually did.

The team collected data from multiple cause of death records, hospital discharge databases, microbiology lab reports, published literature, pharmaceutical sales data, antibiotic use surveys, mortality surveillance systems, insurance claims, and more. In total, 520 million individual records or bacterial isolates, covering 19,513 study location years (Naghavi et al., 2024).

They built a statistical model with five components: the number of deaths involving sepsis, the proportion of infectious deaths linked to a specific syndrome (like pneumonia or bloodstream infection), the proportion of those syndrome deaths caused by a specific pathogen, the percentage of that pathogen resistant to a given antibiotic, and the excess risk of death or longer infection from that resistance.

Then they defined two counterfactuals. Attributable deaths assume the drug resistant infection is replaced by a drug sensitive version of the same bug. Associated deaths assume the drug resistant infection is replaced by no infection at all. Attributable is the direct cost of resistance. Associated is the full burden of the infection, including the fact that the patient was sick enough to get infected in the first place.

Both numbers matter. Attributable tells you how many deaths are directly caused by the failure of antibiotics. Associated tells you the total scale of the problem, including the underlying illness that made the patient vulnerable.

The Forecast: Better Care Beats New Drugs

Here is where the paper gets genuinely useful for policymakers. The authors ran three scenarios out to 2050.

The reference scenario, a probabilistic forecast of the most likely future, projects 1.91 million attributable AMR deaths and 8.22 million associated AMR deaths in 2050 (Naghavi et al., 2024). That is the baseline. That is where we are headed if nothing changes.

The Gram negative drug scenario assumes new antibiotics come online specifically for the hardest to treat bugs. Under that scenario, 11.1 million AMR deaths could be averted cumulatively (Naghavi et al., 2024). That is a meaningful number. But it is not the biggest lever.

The better care scenario assumes improvements in health care quality and access to appropriate antimicrobials. Across all age groups, 92 million deaths could be cumulatively averted between 2025 and 2050 (Naghavi et al., 2024).

Ninety two million.

That is nearly 10 times more than what new drugs alone would achieve. The single most powerful intervention against superbugs is not a miracle antibiotic. It is making sure people get the right antibiotic at the right time, in a clean setting, with proper follow up care.

This is not a popular message. It is easier to rally around a scientific breakthrough than around better hospital hygiene and more nurses. But the data is unambiguous. The biggest gains come from the boring stuff.

The Geography of Resistance

The burden of AMR is not evenly distributed. It never is.

The super regions with the highest all age AMR mortality rate in 2050 are forecast to be South Asia and Latin America and the Caribbean (Naghavi et al., 2024). These are places where antibiotic use is high, regulation is weak, and health systems are already strained. They are also places where the population is aging fast.

Sub Saharan Africa, which has the highest rates of infectious disease overall, also faces a severe AMR problem. But the pattern there is different. The dominant pathogens and resistance profiles do not always match those in wealthier regions. A drug that works in a London ICU may not work in a clinic in rural Kenya, and vice versa.

The authors note that AMR mortality for children under five decreased in all super regions, but the absolute numbers remain highest in the poorest places. Progress is real. It is also incomplete.

The DALYs Paradox

One of the most interesting findings in the paper is a statistical oddity that tells a human story.

From 2022 to 2050, the number of deaths attributable to AMR is forecast to increase by 69.6%. But the number of disability adjusted life years, or DALYs, a measure that accounts for both death and years of healthy life lost, increases by only 9.4% (Naghavi et al., 2024).

That sounds contradictory. How can deaths go up so much while DALYs barely move?

The answer is age. DALYs penalize deaths in young people more heavily because they lose more future life. The huge reduction in child AMR deaths means fewer DALYs lost early in life. The huge increase in elderly AMR deaths adds deaths, but those deaths happen at an age where fewer years of life are lost on average. The two trends offset each other.

This is not a reason to be complacent. Every death is a tragedy. But it does mean that the overall disease burden from AMR is not exploding as fast as the death count. The crisis is shifting from the very young to the very old. That changes how we think about intervention. Saving a 5 year old from a resistant infection gains you decades of life. Saving an 80 year old gains you a few years. Both matter. But the math of DALYs means the total burden grows slowly even as the death toll climbs.

What This Research Does Not Prove

The study is massive and careful. But it has limits worth understanding.

The authors rely on statistical modeling to fill in gaps where data does not exist. For many low and middle income countries, microbiology data is sparse. The model imputes resistance rates based on neighboring countries, income level, and other variables. That is reasonable. But it is not the same as having actual lab results.

The counterfactual scenarios, better care and new drugs, are hypothetical. They assume certain levels of investment and implementation that may not happen. The 92 million averted deaths under the better care scenario assumes health systems improve dramatically and consistently across the globe. That is aspirational. It is not a guarantee.

The study also does not account for the possibility of a truly transformative new class of antibiotics that works differently from anything we have now. The Gram negative drug scenario assumes incremental improvements to the existing pipeline. A breakthrough could change the math entirely. So could a catastrophic failure, like a pathogen that develops resistance to every known drug.

Finally, the paper focuses on bacterial AMR. It does not cover fungal, viral, or parasitic resistance. Those are real problems. They are just not this study.

The Methicillin Resistant Staphylococcus Aureus That Changed Everything

There is one pathogen in the data that tells the whole story in miniature.

MRSA was responsible for 57,200 attributable deaths in 1990. By 2021, that number had risen to 130,000 (Naghavi et al., 2024). It is the single largest contributor to AMR deaths, and it is growing faster than most.

MRSA is not a rare exotic bug. It is a common bacterium that lives on skin and in noses. It used to be easily killed by methicillin and related drugs. Now it is not. The result is that a routine skin infection or a surgical wound can turn into a life threatening bloodstream infection.

The rise of MRSA is a warning. It shows what happens when a pathogen adapts faster than our drug development pipeline. It also shows what happens when we overuse antibiotics in both humans and agriculture. MRSA is a man made problem. We bred it through decades of careless prescribing.

And it is not going away.

What This Actually Means

• ▸The single most effective intervention against superbugs is not a new drug. It is better infection control, better access to existing antibiotics, and better quality of care. The study estimates 92 million deaths could be averted by 2050 through these measures. That is roughly the population of Germany. We already know how to do this. We just have not done it everywhere.

• ▸Children under five have seen a 50% reduction in AMR deaths since 1990. The interventions that worked, vaccines, sanitation, basic antibiotics, and infection control, should be scaled globally and applied to older adults, whose AMR death rate has increased by 80% over the same period.

• ▸MRSA is the biggest single killer among resistant pathogens, and carbapenem resistance in Gram negative bacteria is the fastest growing threat. Any serious AMR strategy must prioritize these two targets above all others.

• ▸The geographic distribution of AMR is shifting toward South Asia and Latin America. Global health funding and drug development incentives should follow the data, not political convenience.

• ▸The DALYs paradox means the total disease burden from AMR is growing more slowly than the death count, because the crisis is shifting from children to the elderly. That is not a reason to relax. It is a reason to rethink how we measure success. Saving a life at any age is worth doing. But the math of DALYs can hide the true scale of the problem if we are not careful.

The future is not written. The data shows that 39 million deaths are possible. It also shows that 92 million deaths are preventable. The difference is not a miracle. It is a choice.