What does the latest data reveal about the global burden of antibiotic resistance?
A Lancet modelling study of 204 countries and territories estimates that bacterial antimicrobial resistance was associated with 4.71 million deaths in 2021, including 1.14 million deaths directly attributable to it, and forecasts 1.91 million attributable deaths by 2050 without stronger action. This is a statistical modelling analysis rather than a trial: the authors pooled cause of death records, hospital discharge data, microbiology results, insurance claims, pharmaceutical sales and published literature covering 520 million individual records or isolates, then modelled burden for every location, including those with no data.
The analysis covers 22 pathogens, 84 pathogen-drug combinations and 11 infectious syndromes from 1990 to 2021, with forecasts to 2050. It is the first assessment to put three decades of trend data behind the resistance problem rather than a single year’s snapshot. The earlier cross-sectional estimate for 2019 found 1.27 million deaths attributable to bacterial resistance and 4.95 million deaths associated with it.
That history matters for the penicillin story. The drugs that made modern surgery and cancer care possible are losing ground, and this study is the first attempt to measure how fast, and in whom.
What the data show:
- Current death toll: 4.71 million deaths associated with bacterial resistance in 2021, of which 1.14 million were directly attributable to it
- Opposing age trends: deaths fell by more than 50% in children under 5 between 1990 and 2021, while rising by over 80% in adults 70 and older
- MRSA rose the most: meticillin-resistant Staphylococcus aureus went from 261,000 associated deaths in 1990 to 550,000 in 2021
- Forecast to 2050: 1.91 million attributable and 8.22 million associated deaths, a 69.6% rise in resistance deaths from 2022 to 2050
The authors also model two alternative futures. Better care of severe infection and improved access to appropriate antibiotics could cumulatively avert 92.0 million deaths between 2025 and 2050. A drug pipeline aimed at gram-negative pathogens could avert 11.1 million resistance deaths.
Dr. Kumar’s Take
The headline finding here is not the total. It is the split. Resistance deaths in children under 5 fell by more than half over three decades, and they fell in every super-region. Over the same period, deaths in adults 70 and older rose by more than 80%. Those are two different epidemics running in opposite directions, and averaging them into one number hides the whole story.
The paediatric decline is proof that this problem responds to intervention. Vaccination, clean water, sanitation and basic infection control moved that curve. The trend in older adults is the one that concerns me clinically. By 2050 the model puts 65.9% of all attributable resistance deaths in people 70 and older, in a world that is ageing quickly. Those are the patients I see with prior hospitalisations, indwelling hardware, immunosuppression and repeated antibiotic courses behind them.
There is one more figure worth holding onto: deaths are forecast to rise 69.6% by 2050 while disability-adjusted life-years rise only 9.4%. That gap is the arithmetic of a burden shifting into late life. It does not make the deaths less real.
Historical Context
In 2014 the Review on Antimicrobial Resistance projected that 10 million deaths a year could be caused by resistance by 2050. That figure drew scientific criticism, but it put resistance on the global agenda. The WHO committed to a global action plan on resistance in 2015, the UN General Assembly held a high-level meeting on it in 2016, and a resistance indicator was written into the Sustainable Development Goals.
Progress since then has been uneven. Funding and implementation of national action plans have varied widely, and until now the available global estimates were a single cross-section for 2019, useful for setting priorities such as the 2024 WHO bacterial priority pathogens list but unable to show whether anything was improving. A 2024 Lancet series recommended a 10% reduction in resistance mortality from the 2019 baseline by 2030. Judging progress against that target requires exactly the kind of time trend this study provides.
What the Research Shows
Direct and Associated Mortality In 2021, 1.14 million deaths were attributable to bacterial resistance, meaning deaths that would not have occurred had the infections been drug-susceptible. A further count of 4.71 million deaths was associated with resistance, comparing against a scenario of no infection at all.
Age-Specific Trends Move in Opposite Directions Between 1990 and 2021, resistance deaths in children under 5 fell by more than 50%, and that decline held in every super-region. In people 5 and older, resistance mortality rose in every super-region, with deaths in adults 70 and older increasing by over 80%.
MRSA Is the Single Largest Riser Meticillin-resistant Staphylococcus aureus increased more than any other pathogen-drug combination globally, from 261,000 associated and 57,200 attributable deaths in 1990 to 550,000 associated and 130,000 attributable deaths in 2021.
Carbapenem Resistance Is Growing Fastest Among Gram-Negatives Resistance to carbapenems rose more than any other antibiotic class in gram-negative bacteria, from 619,000 associated deaths in 1990 to 1.03 million in 2021, and from 127,000 attributable deaths to 216,000 over the same period. Carbapenems are among the last-line agents for serious gram-negative infection.
Where the Burden Is Concentrated Earlier estimates place the highest burden attributable to and associated with resistance in sub-Saharan Africa and south Asia. By 2050 the super-regions forecast to carry the highest all-age resistance mortality rates are south Asia and Latin America and the Caribbean.
Practical Takeaways
- Prevention works: the more than 50% fall in resistance deaths among children under 5 tracks vaccination, water and sanitation investment, and infection control
- Older adults are the growing risk group: by 2050, 65.9% of all attributable resistance deaths are forecast to occur in people 70 and older
- Better care has the largest modelled payoff: improved treatment of severe infection and access to appropriate antibiotics could avert 92.0 million deaths cumulatively between 2025 and 2050
- New drugs matter, but do less alone: a gram-negative drug pipeline is modelled to avert 11.1 million resistance deaths
- Stewardship is not optional: the authors call for combining infection prevention, vaccination, reduced inappropriate antibiotic use in both farming and human medicine, and research into new antibiotics
Related Studies and Research
- Penicillin: The Accidental Discovery That Changed Medicine and Won a War
- Abraham & Chain (1940): First Discovery of Penicillinase
- 2019 Antibiotic Resistance Threats Report
- Penicillin’s Discovery and Antibiotic Resistance: Lessons for the Future
FAQs
How many deaths does antimicrobial resistance cause?
In 2021, an estimated 1.14 million deaths were directly attributable to bacterial resistance and 4.71 million deaths were associated with it. The earlier 2019 estimate was 1.27 million attributable and 4.95 million associated deaths.
Which age groups are most affected, and is that changing?
The trends run in opposite directions. From 1990 to 2021, resistance deaths in children under 5 dropped by more than 50%, while deaths in adults 70 and older rose by over 80%. By 2050, adults 70 and older are forecast to account for 65.9% of all attributable resistance deaths.
What are the projections for future AMR burden?
The reference forecast is 1.91 million deaths attributable to resistance and 8.22 million deaths associated with it in 2050, representing a 69.6% increase in resistance deaths from 2022 to 2050. Disability-adjusted life-years rise far less over that period, by 9.4% to 46.5 million.
Which bacterial threats are growing fastest?
MRSA increased more than any other pathogen-drug combination globally between 1990 and 2021. Among gram-negative bacteria, carbapenem resistance grew more than resistance to any other antibiotic class, reaching 1.03 million associated deaths in 2021.
Bottom Line
This analysis estimates 4.71 million deaths associated with bacterial antimicrobial resistance in 2021, including 1.14 million directly attributable to it, and forecasts 1.91 million attributable deaths by 2050. The trend is not uniform. Resistance deaths in young children have more than halved since 1990, which shows prevention works, while deaths in adults 70 and older have climbed by over 80% and are set to dominate the burden. The modelled interventions point the same way the trend data do: better care of severe infection and access to appropriate antibiotics could avert 92.0 million deaths between 2025 and 2050, with a gram-negative drug pipeline averting a further 11.1 million. Prevention, vaccination, stewardship in both medicine and agriculture, and new antibiotics all have to move together.
