By Shobha Shukla & Bobby Ramakant
A 20-hospital, 3-year ICMR study of over 150,000 patients finds that carbapenem drug-resistant infections are driving up both deaths and treatment costs
A large multi-centre study of over 150,000 patients in India’s 20 tertiary-care hospitals has put a stark number on one of the top ten global health threats today: Antimicrobial Resistance (AMR). AMR happens when infections easier to treat earlier become difficult (or impossible) to treat due to disease-causing microbes (such as bacteria, virus, fungi, parasites) becoming drug resistant.
The findings offer one of the clearest multi-centre pictures yet of the clinical and economic burden when common Gram-negative bacteria become resistant to carbapenem antibiotics.
And the message is blunt: AMR is not simply making infections harder to treat. It is translating into more deaths and substantially higher treatment costs.
Researchers compared outcomes among patients infected with four major Gram-negative pathogens – Escherichia coli (E. coli), Klebsiella pneumoniae (K. pneumoniae), Acinetobacter baumannii (A. baumannii) and Pseudomonas aeruginosa (P. aeruginosa) – looking specifically at infections that were drug-resistant or susceptible to carbapenem antibiotics. Carbapenem antibiotics are a powerful group of broad-spectrum antibiotics, upon which clinicians have traditionally relied on when infections caused by Gram-negative bacteria become resistant to other drugs. Resistance to these medicines therefore leaves doctors with fewer therapeutic options.
E.coli is a common cause of urinary and bloodstream infections; K. pneumoniae can cause pneumonia, urinary infections and sepsis; A. baumannii is particularly associated with critically ill hospital patients; and P. aeruginosa can cause severe infections in people with prolonged hospital exposure or underlying illness.
The study authored by Dr Kamini Walia, Dr Sonam Vijay, Dr Nitin Bansal, et al, published in The Lancet Regional Health – Southeast Asia (Volume 53, October 2026), analysed data from 20 tertiary-care hospitals across India between April 2022 and April 2025. Dr Kamini Walia serves as a senior scientist at Indian Council of Medical Research (ICMR), heads its Descriptive Research Division and leads ICMR on Antimicrobial Resistance (AMR).
Each of the 20 participating hospitals was required to designate a minimum of 150 inpatient beds for Antimicrobial Resistance (AMR) surveillance activities, including at least 50 intensive care unit (ICU) beds and 100 ward beds.
“All patients were followed up until their final clinical outcome (discharge from the hospital or death),” notes the study.
“Carbapenem drug-resistant infections caused substantially higher mortality and treatment costs than susceptible infections. Mortality was particularly high in bloodstream infections, reaching 46–51% for carbapenem drug-resistant Acinetobacter, Pseudomonas and Klebsiella,” posted Dr Kamini Walia on LinkedIn. “For India, the study highlights the need to move beyond drug resistance surveillance to integrated surveillance linking AMR with patient outcomes, while strengthening infection prevention, right and timely diagnostics, antimicrobial stewardship and access to effective newer antibiotics.”
A nationwide snapshot of a growing threat
The scale of the surveillance effort is significant.
Researchers examined records from 159,336 hospitalised patients in the 20 tertiary-care hospitals, representing more than 1.75 million patient-days of hospital care. “Patient-days” measures the total number of days all admitted patients spend in a hospital or healthcare facility over a specific period.
Of these patients, 27,283 had infections confirmed microbiologically, while 26,213 (16.4% of the entire hospitalised population studied) had Gram-negative bacterial infections.
Among those with Gram-negative infections, more than six in ten were resistant to carbapenem antibiotics, according to the study’s analysis.
Mortality rose across all four bacteria
For every major pathogen examined, patients with carbapenem drug-resistant infections faced a higher risk of death than those infected with carbapenem susceptible strains.
The difference was particularly striking for E. coli and P. aeruginosa. Patients with carbapenem-resistant E. coli had a 41% higher relative risk of death than those with susceptible infections. For P. aeruginosa, the corresponding increase was 43%.
For K. pneumoniae, the relative risk was 33% higher, while carbapenem-resistant A. baumannii was associated with a 16% higher relative risk.
Absolute mortality data underline the severity
According to the study’s reported results, mortality was 24.4% for carbapenem-resistant E. coli compared with 17.3% for susceptible infections. For K. pneumoniae, mortality was 31.2% versus 23.5%. A. baumannii had particularly high mortality overall: 37.9% among resistant infections compared with 32.8% among susceptible infections. For P. aeruginosa, the corresponding figures were 28.9% and 20.2%.
It is important to note that bloodstream infections can rapidly progress to sepsis and multi-organ failure, leaving little room for delays or ineffective treatment. The most alarming data appeared in bloodstream infections caused by carbapenem drug-resistant non-fermenting bacteria, principally A. baumannii and P. aeruginosa. Mortality in these cases reached 46% to 51%.
AMR also comes with a financial penalty
The study did not stop at mortality.
Researchers also examined length of hospitalisation and antibiotic treatment costs. Drug-resistant infections required more expensive therapy, with antibiotic costs 1.1 to 2 times higher than those associated with susceptible infections.
Once first-line or commonly used antibiotics no longer work, clinicians may need newer, more expensive drugs or combinations of antibiotics. Patients may also require longer hospital stays, intensive monitoring and more complex (and may be more expensive) care.
For hospitals already operating under pressure, those costs accumulate quickly.
The study found, for example, that patients with carbapenem-resistant E. coli had an average hospital stay of about 23 days, compared with about 18 days among patients with susceptible E. coli infections. However, length of hospital stay was carbapenem-resistant A. baumannii and P. aeruginosa was shorter, which could be because of high mortality. “This paradox likely reflects the early mortality in carbapenem-resistant A. baumannii and P. aeruginosa infections, leading to reduced length of hospital stay owing to poor survival,” notes the study.
In other words, drug resistance can create a vicious cycle: the infection becomes harder to treat, treatment becomes more complicated, hospitalisation can become longer, and costs rise while the patient’s risk of dying increases.
The changing antibiotic arsenal
The researchers also looked at how clinicians were treating drug-resistant infections.
For carbapenem-resistant E. coli and K. pneumoniae, ceftazidime–avibactam was used more commonly than polymyxin B, reflecting the changing therapeutic landscape for difficult-to-treat Gram-negative infections.
But the availability of newer antibiotics does not eliminate the underlying problem.
Every new antibiotic introduced into clinical practice exists within an evolutionary contest. Bacteria that survive exposure can acquire or select for resistance mechanisms, potentially eroding the usefulness of drugs that were once considered reliable.
That makes antimicrobial stewardship – the effort to ensure antimicrobials (like antibiotics, antivirals, antifungals and anti-parasitics) are used only when appropriate, and that, after right and timely diagnosis, the right drug is given at the right dose and for the right duration. This is not merely a prescribing principle but becomes a central part of preserving the effectiveness of the remaining treatment options.
Why health surveillance is indispensable
Much of the available evidence on AMR comes from laboratory surveillance, which can show whether bacteria isolated from patients are resistant to particular antibiotics. But knowing that drug resistance exists is different from knowing what it does to patients in real-life hospital settings across a diverse country like India.
The new study links microbiological drug resistance to outcomes such as mortality, hospitalisation and treatment costs across a large multi-centre population.
That distinction is crucial in India, where researchers have pointed to fragmented surveillance, incomplete laboratory and clinical data and a lack of standardised estimates of mortality associated with resistant infections.
Was infection acquired in the healthcare facility?
The study found that 85% of E. coli infections were healthcare-associated infections (or hospital acquired), 91.6% of K. pneumoniae infections were healthcare-associated, 95.1% of A. baumannii infections were healthcare-associated, and 91.9% of P. aeruginosa infections were healthcare associated. “Among blood stream associated infections across all pathogens, again majority (more than 85%) were classified as healthcare-associated infections,” notes the study.
Those findings point beyond antibiotic prescribing alone.
“Microbiologically confirmed infections were classified as community-acquired if the culture was positive within 48 hours of hospitalisation and there was no history of hospitalisation within the preceding 90 days. Infections not meeting these criteria were classified as healthcare-associated infections,” study authors explain.
Infection prevention as important as new antibiotics
If drug-resistant infections are concentrated in healthcare settings, then preventing them in the first place becomes as important as finding better drugs to treat them.
The researchers ultimately call for stronger surveillance, better infection prevention and control, improved access to effective therapies and stronger antimicrobial stewardship.
The implications extend from intensive-care units to hospital laboratories and pharmacy departments.
Hand hygiene, environmental cleaning, appropriate use of invasive devices used in medical procedures, rapid identification of drug-resistant organisms (right and timely diagnosis and drug-susceptibility testing where possible), isolation or cohorting where appropriate, and careful antibiotic prescribing, all form part of the efforts to save us from drug-resistant infections.
So does health surveillance.
A hospital cannot effectively control a drug-resistant organism it cannot detect or track. Government of India’s prestigious Indian Council of Medical Research (ICMR)’s AMR Surveillance Programme led by Dr Kamini Walia has made pathbreaking contributions over the years to identify patterns and measure whether AMR interventions are working.
The new multi-year study was itself funded through an ICMR task-force project.
A warning, not a verdict
The findings are powerful, but they should not be read as proof that drug resistance alone caused every death.
Persons with drug-resistant infections may differ from those with susceptible infections in important ways, including underlying disease, severity of illness, prior antibiotic exposure and healthcare-associated risk factors. Those differences can influence outcomes independently of drug resistance itself.
The researchers’ comparison nevertheless provides strong evidence of an association between carbapenem drug resistance and worse outcomes across multiple major pathogens.
And the consistency is difficult to ignore.
The study authors observe: “Most of the hospitals, included in the study have very high standards of infection control and are accredited by national and international quality control boards, but they are tertiary care hospitals, which predominantly manage referred and critically ill patients; therefore, the rates observed should not be interpreted as indicators of hospital performance. We suspect that hospitals with poor laboratory support to facilitate timely identification and treatment may have far worse patient outcomes compared to what is documented in this study.”
For the health system, this study is another reminder that preventing drug-resistance – and preventing infections from spreading in the first place – may ultimately be far less costly than trying to outrun bacteria after drug-resistance has emerged.
“Finally, improving access to newer effective antimicrobials is crucial to reduce mortality associated with drug resistant infections in India,” recommend study authors.
ICMR researcher Dr Sonam Vijay rightly shared on LinkedIn that “Our findings among 1.5 lakh (150,000) hospitalised patients, including 16.4% with Gram-negative bacterial infections, highlight the need for timely diagnostics and access to effective, affordable novel antimicrobials to improve outcomes and reduce the economic burden of carbapenem-resistant infections.”
(The writers are part of CNS Editorial team as well as on the board of Global AMR Media Alliance)
