Doctors across the country are pulling dementia patients off five classes of commonly prescribed medications — antipsychotics, benzodiazepines, barbiturates, certain antidepressants, and sleeping pills — because mounting evidence shows these drugs accelerate cognitive decline, increase fall risk, and raise the likelihood of death in people already struggling with memory loss. A landmark January 2026 study published in JAMA found that one in four Medicare beneficiaries with dementia were prescribed at least one of these risky CNS-active medications, compared to just 17 percent of older adults with normal cognition. Even more troubling, over two-thirds of those prescriptions lacked a documented clinical reason for being written.
Consider a 78-year-old woman with moderate Alzheimer’s disease who has been taking Xanax for anxiety and Ambien to sleep for the past six years. Her neurologist discovers that both drugs are worsening her confusion, contributing to two recent falls, and may have accelerated her cognitive decline in the first place. This scenario is playing out in geriatric clinics nationwide as the medical community reckons with decades of overprescribing to its most vulnerable patients. This article breaks down which medications are being removed and why, what the research actually says about the risks, how the deprescribing process works in practice, and what families and caregivers need to know about navigating these medication changes safely.
Table of Contents
- Which Common Medications Are Doctors Removing From Dementia Patients and Why?
- How Anticholinergic Drugs Contribute to Cognitive Decline
- The Benzodiazepine Problem in Dementia Care
- What Happens When Dementia Medications Themselves Are Deprescribed
- The Cultural Shift Required for Safe Deprescribing
- What Congress and the FDA Are Doing About Antipsychotic Use
- Where Dementia Medication Management Is Heading
- Conclusion
- Frequently Asked Questions
Which Common Medications Are Doctors Removing From Dementia Patients and Why?
The five drug classes now under scrutiny all share a common trait: they act on the central nervous system in ways that compound the cognitive and physical vulnerabilities already present in dementia. The JAMA study, led by Dr. John N. Mafi of UCLA, analyzed nationwide Medicare claims data from 2013 to 2021 for adults ages 65 and older and identified these categories as posing the greatest risk. Antipsychotics have carried an fda black box warning since 2005 for atypical versions, extended to all antipsychotics in 2008, due to increased death risk in elderly dementia patients. Benzodiazepines like Valium and Xanax are linked to falls, hip fractures, cognitive worsening, and pneumonia hospitalizations.
Barbiturates, tricyclic antidepressants with strong anticholinergic properties, and non-benzodiazepine sleeping pills like Ambien round out the list. The good news is that prescribing rates for these medications have been declining — dropping from 20 percent to 16 percent among all Medicare beneficiaries over the nine-year study period. But the bad news is that dementia patients still receive these drugs at far higher rates than the general older population. That gap suggests a systemic problem: the people most susceptible to harm from these medications are the ones most likely to receive them, often because behavioral symptoms of dementia are treated reflexively with sedation rather than through non-pharmacological approaches. To put the comparison in stark terms, an older adult without cognitive impairment has roughly a one-in-six chance of being prescribed one of these drugs. An older adult with dementia has a one-in-four chance. That difference represents hundreds of thousands of patients nationwide receiving medications that may be making their condition worse.
How Anticholinergic Drugs Contribute to Cognitive Decline
Many of the medications being deprescribed share anticholinergic properties, meaning they block the neurotransmitter acetylcholine — the same chemical that dementia itself depletes. A meta-analysis of 14 studies involving more than 1.5 million subjects found that anticholinergic drug use is associated with increased risk of both all-cause dementia and Alzheimer’s disease specifically. Researchers observed a dose-response relationship over a 10-year cumulative exposure window: the longer and higher the use, the greater the dementia risk. What makes this particularly insidious is that anticholinergic medications are not limited to the five drug classes in the JAMA study. They include over-the-counter options that millions of older adults take routinely — first-generation antihistamines like Benadryl for allergies, bladder antimuscarinics for incontinence, and tricyclic antidepressants.
A person with early-stage dementia who takes diphenhydramine every night for sleep and oxybutynin for bladder control is effectively doubling down on a mechanism that worsens the very disease they are fighting. However, it is important to note that not all anticholinergic burden is equal. A patient taking a low dose of a mildly anticholinergic drug for a short period faces far less risk than someone on multiple strongly anticholinergic medications for years. The clinical challenge lies in calculating cumulative anticholinergic load across a patient’s entire medication list, which is why geriatricians increasingly use scoring tools like the Anticholinergic Cognitive Burden Scale during medication reviews. If a patient has been on these drugs for decades and has stable symptoms, the decision to stop is not automatic — abrupt withdrawal carries its own dangers, including rebound insomnia, anxiety spikes, and seizure risk with benzodiazepines and barbiturates.
The Benzodiazepine Problem in Dementia Care
Benzodiazepines deserve special attention because they remain among the most commonly prescribed drugs for older adults despite decades of warnings. Research shows that benzodiazepine users have an odds ratio of 1.39 for dementia risk, with the danger climbing higher for long-acting formulations with half-lives exceeding 20 hours and for use lasting more than three years. In patients who already have dementia, these drugs increase the risk of falls, hip fractures, further cognitive impairment, and drug-associated hospital admissions. The clinical trap is straightforward. A dementia patient becomes agitated or anxious — common behavioral symptoms of the disease — and a prescriber reaches for lorazepam or diazepam because the effect is immediate and visible. The patient calms down. Everyone in the room feels better.
But what is not visible is the cascade of harm: the drug lingers in an aging body with slower metabolism, the sedation increases fall risk overnight, and the anticholinergic effects chip away at whatever cognitive reserve remains. A single hip fracture in an 80-year-old with dementia carries roughly a 30 percent one-year mortality rate, turning a well-intentioned prescription into a potentially lethal chain of events. Tapering off benzodiazepines in dementia patients requires particular caution. Long-term users can experience withdrawal seizures if the drugs are stopped abruptly. The standard approach involves gradual dose reduction over weeks or months, with close monitoring for withdrawal symptoms. For some patients, a crossover to a longer-acting benzodiazepine followed by slow tapering provides a smoother transition. The key point for families: if your loved one has been on a benzodiazepine for years, do not stop it suddenly on your own. Work with their physician to create a structured tapering plan.
What Happens When Dementia Medications Themselves Are Deprescribed
The deprescribing conversation extends beyond risky CNS-active drugs to the very medications prescribed to treat dementia. Cholinesterase inhibitors like donepezil and rivastigmine, along with memantine, are the standard pharmacological treatments for Alzheimer’s disease. But their benefits are modest, their side effects are real, and a growing body of evidence suggests that stopping them may not be as harmful as once feared. A study of more than 16,000 Medicare beneficiaries found that discontinuing cholinesterase inhibitors when starting memantine was not associated with increased institutionalization risk and was actually associated with lower risk of fall-related injury. The tradeoff is emotionally charged.
Families often view Alzheimer’s medications as the last line of defense, and the suggestion to stop them can feel like giving up. But the numbers tell a different story: 56 percent of patients switched or discontinued their initial Alzheimer’s treatment within six months, and only 15 percent remained on their initial treatment at 18 months. These medications were never designed to be taken indefinitely, and in advanced stages of dementia, their marginal cognitive benefit may be outweighed by side effects including nausea, diarrhea, dizziness, and bradycardia. Canadian clinical guidelines from the CCCDTD5 recommend that deprescribing Alzheimer’s medications should happen through shared decision-making with patients and caregivers, including honest discussion of benefits and harms from the very start of therapy. This approach — setting expectations early that no medication is forever — is fundamentally different from the traditional model where drugs are prescribed and never revisited.
The Cultural Shift Required for Safe Deprescribing
The biggest barrier to deprescribing is not clinical — it is cultural. Families fear that stopping a medication means abandoning treatment. Prescribers worry about liability if a patient worsens after discontinuation. Nursing homes face regulatory pressure to manage behavioral symptoms, and medications remain the fastest visible intervention. A 2025 roadmap published in Drugs & Aging lays out a framework for deprescribing in dementia patients that emphasizes patient-centered collaborative approaches involving care partners as essential participants in the decision. Key clinical trials, including OPTIMIZE and DPRESCRIBE-AD, are testing structured deprescribing interventions and reinforcing a message that the dementia care community needs to hear: no medication should be prescribed indefinitely without regular reassessment.
This means building deprescribing conversations into every care visit, not waiting until a crisis — a fall, a hospitalization, a sudden behavioral change — forces the issue. The practical limitation here is time. A thorough medication review and deprescribing conversation takes 30 to 45 minutes, and most primary care visits last 15. Until the healthcare system creates space and reimbursement for these conversations, many dementia patients will continue receiving medications they no longer need. Families should also be aware that deprescribing does not mean doing nothing. Non-pharmacological interventions for dementia-related behavioral symptoms — structured activities, environmental modifications, music therapy, caregiver training in communication techniques — have evidence behind them and carry none of the drug risks. The challenge is that these approaches require more effort, more time, and more coordination than writing a prescription.
What Congress and the FDA Are Doing About Antipsychotic Use
In March 2024, Congress directed the FDA to hold a public workshop to re-evaluate the antipsychotic black box warning for older adults, with an update due within 12 months. This move reflects a complicated reality: while antipsychotics are genuinely dangerous for dementia patients, the black box warning has not stopped their widespread use, and some clinicians argue it has instead driven prescribing underground — with drugs given off-label without proper documentation or monitoring.
The workshop aims to determine whether the warning needs revision, not necessarily weakening but possibly making it more specific about which patients face the highest risk and under what circumstances limited antipsychotic use might be clinically appropriate. For families, the practical takeaway is this: if your loved one with dementia is currently taking an antipsychotic, ask the prescribing physician to document the specific clinical indication, review whether the drug is still necessary at every visit, and outline a plan for discontinuation.
Where Dementia Medication Management Is Heading
The next several years will likely see deprescribing become a standard component of dementia care rather than an afterthought. The 2025 Drugs & Aging roadmap, the ongoing OPTIMIZE and DPRESCRIBE-AD trials, and increasing awareness among both clinicians and families are converging toward a model where every medication prescribed to a dementia patient comes with a built-in reassessment timeline.
Technology may help — electronic health record systems are beginning to incorporate anticholinergic burden calculators and automated alerts when a dementia patient is prescribed a high-risk medication. But the most meaningful change will be philosophical. The dementia care field is moving away from a model that treats every symptom with a pill and toward one that asks a harder question: is this medication still serving this patient, at this stage of their disease, given the risks it carries? For the roughly 7 million Americans living with Alzheimer’s disease and related dementias, getting that question right could mean fewer falls, fewer hospitalizations, clearer thinking on their best days, and more time at home with the people they love.
Conclusion
The evidence is now clear that millions of dementia patients have been receiving medications that worsen cognition, increase fall risk, and may hasten death — often without documented clinical justification. The five drug classes identified in the January 2026 JAMA study (antipsychotics, benzodiazepines, barbiturates, anticholinergic antidepressants, and sleeping pills) represent the most urgent deprescribing targets, but the conversation extends to Alzheimer’s medications themselves when their benefits no longer outweigh their harms.
For families and caregivers, the action steps are concrete: request a comprehensive medication review with your loved one’s physician, ask about the anticholinergic burden of every drug on the list, ensure that any deprescribing happens gradually and with medical supervision, and advocate for non-pharmacological alternatives to manage behavioral symptoms. Deprescribing is not giving up on treatment — it is making treatment smarter, safer, and more aligned with what actually helps the person in front of you.
Frequently Asked Questions
Can I stop my loved one’s medications on my own if I think they are harmful?
Never stop any medication without medical supervision. Many of these drugs, particularly benzodiazepines and barbiturates, require gradual tapering to avoid dangerous withdrawal effects including seizures. Always work with the prescribing physician to create a structured discontinuation plan.
If antipsychotics have a black box warning for dementia patients, why are doctors still prescribing them?
The black box warning does not prohibit prescribing — it alerts clinicians and patients to serious risks. In some cases of severe agitation or psychosis that pose a danger to the patient or others, physicians may determine that short-term, closely monitored antipsychotic use is the least harmful option. The problem is when these drugs are prescribed reflexively or continued indefinitely without reassessment.
Will stopping Alzheimer’s medications like donepezil cause a rapid decline?
Research suggests that discontinuation is not associated with increased institutionalization and may reduce fall-related injuries. Some patients may experience a temporary worsening of symptoms, which is why guidelines recommend shared decision-making and careful monitoring. In advanced dementia, the modest cognitive benefits of these drugs often no longer outweigh the side effects.
How do I know if my loved one is taking an anticholinergic medication?
Ask the pharmacist to run an anticholinergic burden assessment on their full medication list, including over-the-counter drugs. Common culprits include first-generation antihistamines like diphenhydramine (Benadryl), bladder medications like oxybutynin, and tricyclic antidepressants like amitriptyline. Many patients are on multiple mildly anticholinergic drugs that add up to a significant cumulative burden.
What non-drug alternatives exist for managing agitation and sleep problems in dementia?
Evidence-supported approaches include structured daily activities, regular exercise appropriate to ability level, music therapy, bright light therapy for sleep-wake cycle disruption, environmental modifications to reduce overstimulation, and caregiver training in validation and redirection techniques. These approaches take more effort than a pill but carry no risk of falls, cognitive worsening, or drug interactions.
How often should dementia patients have their medications reviewed?
At minimum, every six months — and ideally at every clinical visit. Any time a dementia patient experiences a fall, increased confusion, a hospitalization, or a change in disease stage, a full medication review should be triggered. The goal is to ensure every remaining medication has a clear, current clinical justification.
