AMR was linked to 0.84 million under-five deaths in 2021, GBD analysis finds

Antimicrobial resistance was associated with 0.84 million deaths among children under five worldwide in 2021 and was the direct attributable cause of 0.19 million, according to a systematic analysis of Global Burden of Disease data published in the International Journal of Antimicrobial Agents on 1 September [s1]. Both figures represent a substantial decline since 1990 — but the burden remains deeply unequal [s1].

The wider picture is grim. Bacterial antimicrobial resistance was estimated to be associated with more than 4.7 million deaths globally in 2021 across all ages, the World Health Organization says, and approximately 1 in 6 laboratory-confirmed bacterial infections worldwide were resistant to antibiotics in 2023 [s2]. The new analysis narrows the lens to the youngest children, where resistant infections strike hardest and treatment options are thinnest.

What the analysis did

The authors conducted a secondary analysis of the Global Burden of Disease 2021 estimates, extracting AMR-associated and AMR-attributable mortality for children under five from 1990 to 2021, with projections to 2050 [s1]. The two measures answer different questions: AMR-associated deaths count all deaths involving a drug-resistant infection, while AMR-attributable deaths estimate how many would not have occurred if the infections had been drug-susceptible. The 0.84 million and 0.19 million figures for 2021 are the associated and attributable totals, respectively [s1].

The associated estimate carried a 95% uncertainty interval of 0.64 million to 1.04 million, and the attributable estimate an interval of 0.14 million to 0.24 million [s1].

Where the burden falls

Substantial regional disparities persisted, with sub-Saharan Africa bearing the highest burden [s1]. The pathogens behind the deaths split by measure: Streptococcus pneumoniae accounted for the largest number of AMR-associated deaths, while Klebsiella pneumoniae accounted for the largest number of AMR-attributable deaths [s1]. Bloodstream infections and lower respiratory infections accounted for the majority of AMR-related mortality [s1].

Among the specific pathogen-drug combinations driving attributable deaths, the analysis singled out beta-lactam-resistant S. pneumoniae, methicillin-resistant Staphylococcus aureus and carbapenem-resistant Acinetobacter baumannii [s1] — a list dominated by the hard-to-treat Gram-negative and hospital-acquired organisms that have the fewest working drugs left against them.

Two futures

The projections to 2050 indicate a continued decline in the under-five burden under a reference scenario [s1]. But the analysis also modelled two policy-relevant alternatives, and the gap between them is the study's sharpest point.

From 2025 to 2050, a "better care" scenario — assuming future improvements in health-care quality and access to appropriate antimicrobials — was projected to avert an estimated 10.65 million deaths [s1]. A "Gram-negative drug" scenario, assuming future drug development targeting Gram-negative pathogens, was projected to avert 1.09 million [s1]. In other words, delivering the care and drugs that already exist to the children who cannot reach them was modelled to prevent roughly ten times as many deaths as inventing new antibiotics for the most stubborn bugs.

That does not mean new drugs are optional. The world faces an antimicrobial research and development crisis, with few new medicines in the pipeline, WHO warns [s2]. But for young children specifically, the model puts most of the achievable gain on the access-and-quality side of the ledger.

What drives it

Resistance is not only a laboratory phenomenon. The misuse and overuse of antimicrobials are driving the development and spread of drug-resistant pathogens, and a lack of adequate access to appropriate new and existing vaccines, diagnostics and medicines also contributes to the crisis, WHO says [s2]. Those two forces pull in opposite directions in the places the analysis flags: the same low-resource settings that struggle to get children the right antibiotic in time can also struggle to use antibiotics judiciously, and both failures feed the burden. The study's own conclusion points beyond antibiotic development, calling for strategies that prioritise improved health-care access, infection prevention and vaccination, especially in high-burden regions [s1] — the levers the "better care" scenario assumes are pulled.

The caveats

These are modelled estimates built on GBD inputs, and they inherit the uncertainty of that exercise — reflected in the wide intervals around the 2021 figures and, still more, in scenario projections that run a quarter-century into the future. Attributing a death to resistance rather than to the underlying infection is an inference, not a measurement, and country-level surveillance of resistant infections in children remains patchy, especially in the regions carrying the most burden.

The headline that the burden is falling is real and worth stating plainly. So is the one underneath it: the decline is uneven, it is concentrated where health systems are weakest, and the analysis suggests the fastest way to push it lower for children is less about new molecules than about getting existing care, diagnostics, vaccines and antibiotics to the places still going without.

Sources

• Global burden of bacterial antimicrobial resistance among children under 5 years 1990-2021: A systematic analysis with forecasts to 2050 — International Journal of Antimicrobial Agents, 1 September 2026
• Antimicrobial resistance (fact sheet) — World Health Organization