The push to reduce antibiotic prescriptions is working — U.S. outpatient prescribing dropped 19.2% between 2011 and 2022, falling from 877 to 709 prescriptions per 1,000 people, according to the CDC’s 2025 Stewardship Report. That sounds like unqualified good news, especially given that antimicrobial resistance now kills an estimated 1.17 million people globally each year. But the downside is real and rarely discussed: undertreated infections can escalate to sepsis, certain populations are being cut off from antibiotics they genuinely need, and no one has actually established how low prescribing can safely go before patients start dying from treatable infections.
For families dealing with dementia care, this matters more than most people realize. Older adults are already the population most vulnerable to both antibiotic-resistant infections and the consequences of delayed treatment. A 2024 Lancet forecast projects that by 2050, 65.9% of all deaths directly attributable to antimicrobial resistance will occur in people aged 70 and older. When your loved one can’t clearly articulate symptoms — when a urinary tract infection or pneumonia might present as confusion or agitation rather than the textbook signs — the margin between appropriate caution and dangerous undertreatment gets razor thin. This article breaks down the prescribing trends, the real risks of going too far in either direction, who falls through the cracks, and what families and caregivers should watch for.
Table of Contents
- Why Are Doctors Prescribing Fewer Antibiotics — and What’s the Real Downside?
- The Sepsis Risk That Lurks Behind Reduced Antibiotic Use
- Older Adults Bear the Greatest Burden of Antimicrobial Resistance
- How Families Can Navigate the Prescribing Tightrope
- The Health Equity Problem Hidden in the Prescribing Data
- What the Global Forecasts Mean for Your Family
- Where Antibiotic Stewardship Goes From Here
- Conclusion
- Frequently Asked Questions
Why Are Doctors Prescribing Fewer Antibiotics — and What’s the Real Downside?
The decline in antibiotic prescribing is largely driven by antimicrobial stewardship programs, which emerged in response to decades of overprescription. For years, doctors handed out antibiotics for viral infections, sinus congestion, and sore throats that would have resolved on their own. That practice bred resistant bacteria at an alarming rate. So health systems pushed back. Intravenous antibiotics in hospitals decreased roughly 7% per year from 2012 to 2020, then roughly 9% annually through 2021. Long-term care facilities — the places where many dementia patients live — saw an 8% decrease in prescribing rates from 2013 to 2021, with the steepest drop occurring between 2019 and 2021 during the COVID-19 pandemic. The downside is that this reduction isn’t surgical.
It’s a broad cultural shift, and broad shifts create collateral damage. The CDC itself warns that undertreated infections can progress to sepsis, which can quickly lead to tissue damage, organ failure, and death without fast antibiotic treatment. In prescriber surveys, doctors identified sepsis as the consequence they fear most when withholding antibiotics. A 2024 study published in the Journal of Infection found limited evidence on where the safe lower threshold for antibiotic prescribing actually sits — meaning we’re reducing prescriptions without a clear map of when we’ve gone too far. The problem is compounded in populations that can’t advocate for themselves. A person with moderate-to-advanced Alzheimer’s disease who develops a bacterial infection may not be able to describe worsening symptoms. If the default clinical posture is now “wait and see,” that wait can cost critical hours. This isn’t an argument against stewardship — it’s an argument for nuance that the numbers alone don’t capture.
The Sepsis Risk That Lurks Behind Reduced Antibiotic Use
Sepsis is where the prescribing debate turns life-or-death. When a bacterial infection isn’t adequately treated, the body’s immune response can spiral out of control, damaging its own tissues and organs. Research published in JAMA Network Open found that inadequate antibiotic coverage — meaning the prescribed antibiotic didn’t actually target the infecting pathogen — is associated with higher mortality in sepsis patients. This isn’t a theoretical risk. It’s a documented pattern in hospital data. Here’s where it gets particularly concerning for older adults: patients treated with certain “high-risk” broad-spectrum antibiotics had a nearly 80% higher risk of subsequent sepsis compared to no antibiotic therapy at all, according to research highlighted by CIDRAP. So giving the wrong antibiotic can be worse than giving none. But giving none when one is needed can be fatal.
Unnecessary broad-spectrum antibiotic use was separately linked to a 22% increase in mortality and a 26% increased risk of C. difficile infection, per JAMA Network Open. C. difficile is itself a major killer in long-term care settings, where dementia patients are disproportionately housed. However, if a patient has clear signs of a bacterial infection — fever, elevated white blood cell count, localized symptoms — the calculus shifts sharply. The danger isn’t just overprescription anymore. It’s also the growing reluctance to prescribe when the situation genuinely calls for it. Families and caregivers should understand that stewardship programs are designed to reduce inappropriate prescribing, not to delay necessary treatment. If you sense that a doctor is hesitant to prescribe for a loved one with dementia who shows clear infection signs, it’s worth pushing back and asking directly what would happen if this infection progresses untreated.
Older Adults Bear the Greatest Burden of Antimicrobial Resistance
The Lancet’s GRAM Project, published in September 2024, produced forecasts that should alarm anyone involved in elder care. By 2050, annual deaths directly attributable to antimicrobial resistance are projected to reach 1.91 million, with another 8.22 million deaths associated with resistant infections — a roughly 68% increase from 2022 levels. The starkest number: 65.9% of all AMR-attributable deaths in 2050 will be in people aged 70 and older. This isn’t just because older adults get more infections. It’s because their immune systems are weaker, they’re more likely to live in congregate settings like nursing homes where resistant bacteria circulate, and they’re more likely to have had prior antibiotic exposure that selects for resistant organisms. For someone with dementia, add another layer of vulnerability: difficulty communicating symptoms, potential inability to comply with treatment regimens, and the behavioral symptoms that can mask or mimic infection.
Consider a specific scenario. A 78-year-old woman with moderate Alzheimer’s in a memory care facility develops a urinary tract infection. She can’t describe burning or urgency — instead, she becomes more agitated and confused. Staff may attribute the change to dementia progression. If the facility’s stewardship program has made clinicians more cautious about prescribing, the window for simple oral antibiotics may close. By the time the infection is caught, she may need hospitalization and IV antibiotics — which themselves carry risks of C. difficile, further cognitive decline from the hospital environment, and exposure to additional resistant organisms.
How Families Can Navigate the Prescribing Tightrope
The tension between overprescription and underprescription isn’t one that families can resolve — but they can position themselves on the right side of it. The first step is understanding that not all antibiotic reductions are equal. The 19.2% decline in outpatient prescribing between 2011 and 2022 largely reflects appropriate cuts: fewer antibiotics for viral colds, fewer “just in case” prescriptions. That’s genuinely beneficial. The problem is when appropriate caution bleeds into inappropriate delay. For dementia caregivers specifically, the tradeoff looks like this: broad-spectrum antibiotics carry real risks (C. difficile, resistance, the nearly 80% higher subsequent sepsis risk documented for certain high-risk antibiotics), but untreated bacterial infections carry risks that are just as real and often more immediate.
The practical move is to insist on diagnostic testing — urine cultures, blood cultures, chest X-rays — rather than accepting a “let’s wait” approach when symptoms are present. A targeted narrow-spectrum antibiotic based on culture results is the best of both worlds: it treats the infection while minimizing resistance pressure. That said, cultures take 24 to 48 hours, and a critically ill patient may need empiric broad-spectrum coverage in the interim, with de-escalation once results come back. Document behavioral changes meticulously. If your loved one with dementia suddenly becomes more confused, agitated, lethargic, or stops eating, record the timeline. This documentation can be the difference between a doctor ordering cultures and a doctor attributing the change to disease progression. You are, in many cases, the only person who can signal that something has changed.
The Health Equity Problem Hidden in the Prescribing Data
The prescribing reductions are not distributed evenly, and the disparities raise serious ethical questions. Research published in Oxford Academic found that Black children were 25% less likely to receive antibiotics overall and 28% less likely to receive them even when they were clinically indicated for acute respiratory infections. This suggests that the forces behind prescribing reductions — whether conscious policy or unconscious bias — hit some populations harder than others. In the dementia world, equity concerns manifest differently but no less acutely. Black and Hispanic Americans develop dementia at higher rates than white Americans, are diagnosed later, and are more likely to receive care in under-resourced facilities. If those same facilities are also more aggressive in cutting antibiotic prescriptions — whether due to stewardship mandates, liability fears, or resource constraints — the result is a compounding of disadvantage.
A person who already faces barriers to timely dementia diagnosis may also face barriers to timely infection treatment. The limitation here is that most stewardship research doesn’t stratify outcomes by race, socioeconomic status, or cognitive impairment. We know prescribing is going down. We know disparities exist. We don’t yet have a clear picture of whether stewardship programs are actively widening or narrowing those gaps. Families from historically underserved communities should be especially vigilant about ensuring that their loved ones receive appropriate — not just reduced — antibiotic care.
What the Global Forecasts Mean for Your Family
The WHO estimates that antimicrobial resistance could add $1 trillion in additional healthcare costs by 2050. The Lancet projects 39.1 million deaths directly attributable to AMR between 2025 and 2050. These numbers are staggering, but they can also feel abstract. Here’s how to make them concrete: if resistance continues to rise at current rates, the antibiotics that currently treat your mother’s UTI or your father’s pneumonia may simply stop working within their lifetime.
The infections won’t change. The drugs just won’t kill the bacteria anymore. The encouraging counterpoint is that the same Lancet/GRAM Project analysis found that 92 million cumulative deaths could be averted between 2025 and 2050 through better care of severe infections and improved antibiotic access globally, with another 11.1 million AMR deaths preventable through development of new drugs targeting Gram-negative bacteria. The path forward exists. Whether the investment and political will materialize is another question entirely.
Where Antibiotic Stewardship Goes From Here
The current trajectory — fewer prescriptions, but without a clear evidence base for how low is too low — cannot hold indefinitely. The 2024 Journal of Infection study calling out the lack of a safe lower threshold for prescribing is likely to prompt new research aimed at defining that line. Future stewardship programs will need to be smarter, not just leaner: incorporating rapid diagnostic testing, AI-driven pathogen identification, and individualized risk assessment rather than blanket prescribing reductions.
For the dementia care community, the most important development to watch is whether stewardship guidelines begin accounting for cognitive impairment as a distinct risk factor. A patient who cannot self-report symptoms needs a different clinical protocol than one who can. Until that distinction is codified, the responsibility falls to caregivers and families to bridge the gap — to be the voice that says, “Something is wrong, and waiting may not be safe.”.
Conclusion
The reduction in antibiotic prescribing represents genuine progress against the resistance crisis that threatens to kill nearly 2 million people annually by 2050. But progress has a cost when it’s applied without precision. Undertreated infections, sepsis, health equity gaps, and the particular vulnerability of older adults with dementia all represent the downside of a policy shift that treats prescribing reduction as an unqualified good. The reality is messier: some patients need fewer antibiotics, some need different antibiotics, and some need antibiotics sooner than they’re currently getting them. For families caring for someone with dementia, the takeaway is vigilance without panic.
Know the signs of infection that can masquerade as cognitive decline — sudden confusion, agitation, lethargy, loss of appetite. Push for diagnostic testing rather than accepting delay. Ask whether a narrow-spectrum antibiotic might be appropriate before a broad-spectrum one. And understand that you are navigating a healthcare system that is, in real time, trying to recalibrate one of its most fundamental tools. Your awareness of the tradeoffs is itself a form of protection.
Frequently Asked Questions
Can untreated infections make dementia worse?
Yes. Infections — particularly urinary tract infections and pneumonia — are well-documented triggers for delirium in people with dementia. Delirium involves a sudden worsening of confusion and cognition. While delirium itself may resolve once the infection is treated, repeated episodes are associated with faster long-term cognitive decline. Delayed treatment extends the period of delirium and may worsen outcomes.
Why are doctors reluctant to prescribe antibiotics now?
Antimicrobial resistance is a genuine crisis, with an estimated 1.17 million global deaths per year attributed to resistant bacteria. Stewardship programs have made physicians more cautious, and rightfully so — unnecessary antibiotics cause real harm including C. difficile infections and increased resistance. However, in prescriber surveys, doctors acknowledge that the fear of sepsis from withholding antibiotics weighs heavily on their decisions.
Are nursing homes cutting back on antibiotics too aggressively?
Long-term care facilities reduced prescribing rates by 8% between 2013 and 2021, with the sharpest decline during the COVID pandemic. Whether this is too aggressive depends on the facility. Some were dramatically overprescribing before, and reductions were overdue. Others may have over-corrected. The key indicator is whether residents are being evaluated with proper diagnostic testing or simply being denied antibiotics by default.
What should I do if I think my loved one with dementia has an infection?
Document the behavioral changes you’re observing, including the timeline. Request that the care team order diagnostic tests — urine cultures, blood work, chest X-ray if respiratory symptoms are present. Ask specifically whether antibiotics are indicated and, if not, what the plan is if symptoms worsen. Make sure there is a clear follow-up window rather than an open-ended “wait and see.”
Are certain antibiotics more dangerous than others for older adults?
Yes. Broad-spectrum antibiotics carry higher risks: research found that certain high-risk broad-spectrum antibiotics were associated with a nearly 80% higher risk of subsequent sepsis, a 22% increase in mortality, and a 26% increased risk of C. difficile. Narrow-spectrum antibiotics targeted to the specific pathogen are generally safer and more effective, which is why diagnostic cultures matter so much.
