Dementia Patients Are Still Getting Drugs That Make Their Symptoms Worse

There is something quietly alarming about a medical system that recognizes a problem, issues guidelines to fix it, and then continues the same practice anyway. That is roughly the situation described by a new study on anticholinergic and sedative drug prescribing among people living with dementia, a population already navigating one of the most cognitively and physically precarious conditions a person can face. Despite decades of warnings from geriatricians and pharmacologists, millions of dementia patients are still receiving medications known to cloud thinking, impair balance, and raise the risk of falls, often without any documented clinical reason for doing so.

The study found that while overall prescribing of these drugs has declined, people with cognitive impairment remain disproportionately likely to receive them compared to the broader population. That gap is not a minor statistical footnote. It points to a structural failure in how dementia care is coordinated, how prescribing decisions get reviewed, and how the medical record system either captures or ignores the reasoning behind a prescription. In a meaningful share of cases, there was simply no documented justification at all.

The drugs in question belong to a broad category that includes certain antihistamines, bladder medications, antidepressants, and antipsychotics. Their anticholinergic properties, meaning they block a key neurotransmitter called acetylcholine, are precisely what makes them problematic for people whose brains are already losing cholinergic function through the disease process itself. Prescribing them to a dementia patient is, in a neurochemical sense, pushing in the same direction the disease is already pulling. The result can be accelerated confusion, increased agitation, and a higher likelihood of falls that lead to fractures, hospitalizations, and a sharp decline in functional independence.

The persistence of this prescribing pattern despite clear clinical guidance raises a systems-level question that goes beyond individual physician behavior. Dementia care is rarely managed by a single provider. A patient might see a neurologist for their cognitive symptoms, a urologist for bladder issues, a cardiologist for heart disease, and a primary care physician who tries to hold all of it together. Each specialist operates within their own prescribing logic, and none of them necessarily has a complete view of the patient's full medication list. The result is what pharmacologists call polypharmacy, a compounding of drug interactions and side effects that no single prescription, reviewed in isolation, would predict.

Electronic health records were supposed to help solve this. Automated alerts can flag potentially inappropriate medications for older adults, and tools like the Beers Criteria, published by the American Geriatrics Society, give clinicians a concrete reference for which drugs to avoid in elderly populations. But alert fatigue is real. When a system flags dozens of interactions per shift, clinicians learn to click through warnings the way most people click through cookie consent banners. The signal gets lost in the noise.

There is also a subtler incentive problem at work. Managing behavioral symptoms in dementia, such as agitation, sleep disturbance, and anxiety, is genuinely difficult. Non-pharmacological interventions like structured activity, sensory therapies, and caregiver training take time, coordination, and resources that many care settings simply do not have. A prescription, by contrast, takes about ninety seconds. The system, in other words, is not just failing to prevent inappropriate prescribing. In some respects, it is actively incentivizing it.

The second-order consequences of this pattern are where the real systemic cost accumulates. A dementia patient who falls and fractures a hip faces a hospitalization that is, for someone with cognitive impairment, its own medical hazard. Hospital environments are disorienting, sleep is disrupted, new medications are introduced, and delirium, a acute and often underrecognized condition, can develop and leave lasting cognitive damage even after the underlying cause is treated. What began as a prescription for a bladder problem can, through this cascade, accelerate the very decline it was never meant to touch.

For family caregivers, the downstream effects are equally significant. A fall that leads to a hospitalization that leads to a faster transition into residential care compresses the timeline of a disease that was already rewriting every plan the family had made. The emotional and financial costs are enormous, and they are largely invisible to the prescriber who signed the original order.

The study's findings suggest that the solution is not simply better guidelines, since those already exist. What is needed is a redesign of the care coordination infrastructure itself, one that makes it harder to prescribe inappropriately than to prescribe carefully. Until the system's default settings change, the guidelines will remain exactly what they are now: correct, well-intentioned, and widely ignored.