The Antibiotics Are Failing: Inside the CDC's Warning on a Post-Antibiotic Future

In September 2025, the Centers for Disease Control and Prevention published findings in the Annals of Internal Medicine that confirmed what infectious disease specialists had feared: infections caused by NDM-producing carbapenem-resistant Enterobacterales — bacteria resistant to nearly every antibiotic in the clinical arsenal — had surged more than 460% in the United States between 2019 and 2023 [1]. Early 2024 data suggested those levels had not declined [2].

The report landed in a landscape already shaped by years of worsening trends. Each year, more than 2.8 million antibiotic-resistant infections occur in the U.S., killing more than 35,000 people [3]. When Clostridioides difficile, a gut infection often triggered by antibiotic use, is included, the toll rises above 3 million infections and 48,000 deaths [3]. The pandemic made things worse: during 2020, more than 29,400 people died from resistant infections commonly linked to healthcare settings, with nearly 40% of those infections acquired in hospitals [4].

These are not abstract projections. They represent a measurable erosion of one of modern medicine's foundational capabilities.

The Strains Driving the Crisis

NDM-CRE — short for New Delhi metallo-β-lactamase-producing carbapenem-resistant Enterobacterales — sits at the center of the CDC's current concern. The NDM enzyme destroys carbapenems, a class of antibiotics typically reserved as a last line of defense for serious infections. Bacteria carrying NDM genes can cause pneumonia, bloodstream infections, urinary tract infections, and wound infections, with very few treatment options remaining [1].

Source: CDC Annals of Internal Medicine Report, 2025

Data as of Sep 26, 2025CSV

The five-fold increase documented across 29 states with mandatory reporting represents a rapid expansion for a pathogen class that was, until recently, uncommon in the U.S. [2]. Most NDM-1 producers remain susceptible only to colistin, fosfomycin, and tigecycline — a narrow set of drugs, each with significant toxicity or efficacy limitations [5]. The emergence of strains carrying both carbapenem and colistin resistance genes in a single isolate has been documented, raising the prospect of infections with no reliable antibiotic treatment [6].

Beyond NDM-CRE, the broader picture includes several resistant organisms responsible for significant mortality. In 2020, approximately 12,700 CRE infections caused 1,100 deaths in the U.S. [1]. MRSA (methicillin-resistant Staphylococcus aureus) remains one of the highest-burden resistant pathogens, with tens of thousands of invasive infections annually [3]. Candida auris, a drug-resistant fungal pathogen, saw cases increase nearly five-fold from 2019 to 2022 [4]. The CDC's 2021-2022 threat update found that six categories of hospital-onset resistant bacterial infections increased by a combined 20% during the pandemic compared to pre-pandemic levels, peaking in 2021 and remaining elevated through 2022 [4].

The Economic Toll

The financial burden of antibiotic resistance extends well beyond individual hospital bills. The CDC and University of Utah researchers estimated that treating just six of the most alarming resistant threats costs more than $4.6 billion in direct healthcare expenditures annually — and that figure excludes downstream costs after initial hospitalization, lost productivity, and long-term disability [7].

A separate analysis found that antibiotic resistance added an average of $1,383 to the cost of treating each patient with a bacterial infection, amounting to roughly $2.2 billion in additional national costs per year [8]. Among older Americans alone, multidrug-resistant infections led to $1.9 billion in healthcare costs, more than 400,000 hospital days, and over 10,000 deaths in 2017 [9].

These figures have grown substantially since the CDC's first major antimicrobial resistance (AMR) threat report in 2013, which estimated $20 billion in excess direct healthcare costs and $35 billion in lost productivity annually [3]. The upward trajectory reflects both rising infection rates and the increasing complexity of treating resistant organisms that require longer hospital stays, more expensive drugs, and more intensive monitoring.

A Pharmaceutical Pipeline in Collapse

The antibiotics that exist today are the product of a research infrastructure that has largely been dismantled. Over the past four decades, the share of antibiotics among all FDA drug approvals has dropped from 20% in 1980 to roughly 6% [10]. More than 82% of all antibiotic approvals occurred before 2000 [10].

The industry exodus is well documented. The number of multinational pharmaceutical companies with active anti-infective programs fell from 18 to just 6 by 2020 — Merck, Shionogi, GSK, Pfizer, Otsuka, and Johnson & Johnson [10]. Large-company antibiotic investigational new drug (IND) filings declined from over 75% of the total in the 1980s to under 20% in the 2010s [10].

Small biotech firms have tried to fill the gap, but the economics are punishing. Of 12 antibiotic companies that went public in the past decade, only 5 remain active today [11]. Achaogen filed for bankruptcy within a year of receiving FDA approval for plazomicin. Tetraphase, which brought eravacycline to market in 2018, saw its peak market capitalization of $1.8 billion collapse to near zero within five years [10].

The current global pipeline contains just 77 antibacterial agents in clinical development, of which only 44 non-tuberculosis, non-C. difficile candidates target bacterial infections — compared to more than 1,300 cancer drugs in development [11]. Venture capital tells the same story: oncology companies raised approximately $7 billion in 2020, a 900% increase from 2011; antibiotic companies raised $160 million, less than they attracted in 2011 [11].

Source: OpenAlex

Data as of Jan 1, 2026CSV

The root cause is structural. Developing a new antibiotic through clinical trials costs well over $1 billion [11]. But unlike cancer drugs or biologics, a successful new antibiotic is deliberately kept in reserve — prescribed sparingly to slow resistance — which means low sales volume. The result is a market where the drugs society needs most are the ones least profitable to make.

The PASTEUR Act and the Subscription Fix

Congress has attempted to address this market failure through the PASTEUR Act (Pioneering Antimicrobial Subscriptions To End Upsurging Resistance), reintroduced in the 119th Congress in 2026 [12]. The bill proposes a subscription model: the federal government would pay pharmaceutical companies fixed annual fees — between $750 million and $3 billion over 5 to 10 years — for access to critically needed antibiotics, regardless of how many doses are actually used [13].

The mechanism is designed to delink revenue from sales volume, providing stable returns that make antibiotic development financially viable. Payments would be tied to a drug's contributions to patient care, clinical innovation, and public health benefit [13]. The bill's total authorization has been scaled back from $11 billion to $6 billion to improve its chances of passage [11].

The UK has adopted a similar approach. The National Health Service agreed to pay up to £10 million per year for up to 10 years for access to cefiderocol and ceftazidime-avibactam, two antibiotics effective against some resistant gram-negative bacteria [13].

Critics of the PASTEUR Act, including Médecins Sans Frontières, have argued that combining pull incentives with existing push funding (from programs like CARB-X and BARDA) could produce a surplus of marginally improved antibiotics rather than genuinely novel ones [14]. The debate centers on whether subscription payments are calibrated to reward true innovation or simply subsidize incremental modifications.

The U.S. vs. Northern Europe: A Policy Gap

International comparisons highlight the role of policy in shaping resistance rates. In the EU and European Economic Area, community antibiotic consumption ranges from 10.4 defined daily doses (DDD) per 1,000 inhabitants per day in the Netherlands to 36.3 in Greece [15]. The Netherlands also records the lowest hospital antibiotic consumption in the EU/EEA [15]. Carbapenem resistance rates are lowest in Denmark, the Netherlands, Norway, Finland, and Sweden [16].

These outcomes are not accidental. The Netherlands and Nordic countries have implemented national antibiotic prescribing surveillance, facility-specific guidelines, and coordinated stewardship networks across healthcare institutions, including nursing homes [17]. Research across European countries found that ambulatory antibiotic consumption and per-capita health expenditure together explain 74% of the variation in AMR rates — suggesting that prescribing practices, not just pathogen biology, drive resistance levels [18].

The U.S. lacks a comparable national framework for outpatient antibiotic stewardship. While hospital stewardship programs have expanded — the CDC now recommends core elements for both hospital and outpatient settings [19] — enforcement and adoption remain uneven, particularly in ambulatory care, where most antibiotic prescriptions originate.

Agriculture's Role: Evidence and Counterarguments

Approximately 70% of medically important antibiotics sold in the U.S. are used in food-producing animals, primarily for growth promotion and disease prevention rather than treatment of sick animals [20]. The question of whether agricultural antibiotic use drives resistance in human clinical infections has been debated for decades, but the weight of evidence points in one direction.

A review of 92 published studies found that 59% openly stated or contained evidence linking agricultural antibiotic use to increased resistant infections in humans [21]. Among 139 academic studies specifically, 72% found evidence of a link, while only 5% argued there was no connection [21]. Transmission pathways include direct animal contact, exposure to manure, consumption of undercooked meat, and environmental contamination of water and soil [22].

One of the strongest natural experiments comes from Denmark. After the country banned avoparcin — an antibiotic similar to vancomycin used exclusively in livestock — levels of vancomycin-resistant enterococci in both Danish animals and humans dropped within two years [21]. The European Medicines Agency and the World Health Organization both recognize that antibiotic overuse in farming contributes to resistance in human infections [21].

Agricultural industry researchers and trade groups have pushed back, arguing that direct transmission from farm animals to clinical infections in humans is difficult to prove on a case-by-case basis, that resistance can arise independently in human and animal populations, and that modern farming practices have already reduced antibiotic use significantly since the FDA's 2017 guidance restricting growth-promotion uses [20]. These counterarguments have some merit in narrow terms — tracing a specific resistant gene from a specific farm to a specific patient remains methodologically challenging — but the epidemiological and ecological evidence for population-level effects is strong and consistent across multiple countries and study designs.

Who Bears the Burden

Antibiotic-resistant infections do not fall equally across the population. Black patients have higher rates of both hospital-onset and community-acquired MRSA infections than white patients [23]. Individuals living in areas of low socioeconomic status are more likely to carry multidrug-resistant infections and to exhibit resistance to penicillin-class antibiotics [24].

The structural factors driving these disparities are well characterized: less access to affordable primary care, higher rates of chronic medical conditions, crowded living conditions, and longer hospital stays — all of which increase both exposure to resistant organisms and the likelihood that infections become severe before treatment begins [23]. Nursing homes and long-term care facilities face particular challenges. Staff in these settings have reported lack of training, insufficient infection control resources, and limited access to continuing education as barriers to implementing effective stewardship [25].

Community social vulnerability — a composite measure of poverty, housing instability, and limited healthcare access — correlates with both higher antibiotic prescribing rates and weaker infection prevention practices in nursing homes [26]. The result is a feedback loop: the populations with the fewest resources face the greatest burden of resistant infections and have the least capacity to respond.

The Stewardship Debate: Restriction vs. Access

Antibiotic stewardship — the practice of ensuring antibiotics are prescribed only when necessary and in the right dose and duration — is the primary tool for slowing resistance at the clinical level. But it carries a tension that is not always acknowledged in public health messaging.

Critics of aggressive restriction policies argue that pre-authorization requirements and formulary limits can delay treatment for critically ill patients, shifting the risk of mortality from a population-level problem onto individual patients [27]. In settings where diagnostic uncertainty is high — emergency departments, rural hospitals, facilities without rapid microbiology testing — the cost of withholding antibiotics while awaiting culture results can be severe.

The clinical outcomes data, however, largely favors well-designed stewardship. Studies of implemented programs report reductions in C. difficile infection, lower resistance rates, improved cure rates, decreased mortality, and hospital cost savings — without net increases in adverse outcomes [28]. The CDC's own guidance acknowledges the risk of undertreatment and emphasizes that stewardship should include expedited pathways for genuinely necessary prescriptions, not blanket restriction [19].

The patient groups at greatest risk of harm from under-treatment tend to be those with sepsis, immunocompromised patients, and individuals in settings with limited diagnostic capacity. For these populations, the clinical calculus differs from that of a patient with an uncomplicated urinary tract infection, and stewardship protocols need to account for that difference.

What Comes Next

The CDC plans to release updated estimates for at least 19 antimicrobial resistance threats in 2026, with new data in an electronic format and a commitment to biennial updates going forward [4]. The PASTEUR Act's fate in Congress will determine whether the U.S. adopts a subscription-based model for antibiotic procurement. Meanwhile, academic research on antimicrobial resistance continues to expand — publications on the topic have grown from roughly 12,000 per year in 2011 to over 97,000 in 2025 [29].

Source: OpenAlex

Data as of Jan 1, 2026CSV

The fundamental challenge remains: the bacteria are evolving faster than the institutions designed to contain them. The 460% surge in NDM-CRE infections is a data point, not a destination. Whether it becomes the inflection point that triggers sustained policy action or another warning absorbed and forgotten will depend on decisions made in the next few years — in Congress, in pharmaceutical boardrooms, in hospital infection control committees, and in the veterinary and agricultural sectors that account for the majority of antibiotic consumption.

The antibiotics are not gone yet. But the margin is narrowing.

Source: CDC Annals of Internal Medicine Report, 2025

Data as of Sep 26, 2025CSV