Summary: New research identifies acute and post-acute care settings—such as emergency departments, hospitals, and skilled nursing facilities—as the primary places where older adults are first prescribed medications that can impair cognition. The study finds these medications are often started during high-stress care transitions and frequently remain in use long after discharge, especially among people with dementia.
Key Facts
- Disproportionate initiation: Although only about 22% of medical visits for people with dementia occur in acute or post-acute settings, roughly 43% of antipsychotic prescriptions for this group are initiated there.
- Persistent use: Prescriptions begun in these settings tend to persist. Approximately 51% of dementia patients who were started on cognition-affecting medications in a hospital continued taking them one year later.
- Highest-risk patients: Individuals with dementia or cognitive impairment not dementia (CIND) are most likely to receive these medications despite being especially vulnerable to side effects such as falls, delirium, and rehospitalization.
- Documentation gaps: This study builds on prior findings that up to 70% of prescriptions for central nervous system–active drugs lacked a recorded clinical indication.
- Target for intervention: Researchers recommend focusing deprescribing efforts on care transitions—when patients leave hospitals or nursing facilities—to reduce inappropriate initiation and long-term use.
Source: UCLA
Initial prescriptions for medications that can impair cognition—such as antipsychotics, benzodiazepines, certain hypnotics, and anticholinergics—are far more likely to be started in emergency rooms, hospitals, or skilled nursing facilities than in outpatient doctors’ offices, new UCLA analysis shows.
The largest share of these medications is prescribed for older adults with cognitive impairment, particularly those with dementia. This group faces the greatest risk of harm from these drugs, including worsened confusion, increased fall risk, and a higher chance of hospital readmission.
“Medications such as antipsychotics and benzodiazepines can impair thinking, raise the risk of delirium, and increase fall risk for older adults,” said Dr. Dan Ly, assistant professor of medicine in the division of general internal medicine and health services research at the David Geffen School of Medicine at UCLA and the VA Greater Los Angeles, who led the research. “These prescriptions also tend to be sticky: many patients who start these medications in the hospital remain on them months to a year later.”
The study will appear April 28 in the peer-reviewed journal JAMA Network Open.
To determine where these drugs are initiated and how long patients remain on them, researchers analyzed data from the Health and Retirement Study (HRS) linked to Medicare fee-for-service claims for adults aged 66 and older between Jan. 1, 2008 and Dec. 31, 2021. The analysis focused on four medication classes: benzodiazepines, nonbenzodiazepine hypnotics, antipsychotics, and anticholinergics. The team compared initiation events that occurred in acute and post-acute settings (including skilled nursing facilities) with those started in outpatient doctors’ offices.
Participants were categorized as having no cognitive impairment, cognitive impairment not dementia (CIND), or dementia. Overall, initiation in acute and post-acute settings occurred for 14% of patients without cognitive impairment, 17% of those with CIND, and 22% of people with dementia. Persistence at one year was substantial: 38% of people without cognitive impairment, 44% with CIND, and 51% with dementia were still taking the medication a year after initiation.
The data show a clear mismatch between where patients receive care and where these medications are started. For example, although only about 22% of visits by people with dementia occur in acute and post-acute settings, 43% of antipsychotic starts for that group took place there. That suggests these high-intensity settings are disproportionately responsible for initiating medications that can harm cognition.
“We were surprised by how frequently cognition-affecting medications were started outside of the usual office setting,” Dr. Ly said. “Identifying the clinical locations where these prescriptions originate gives clinicians and policymakers a clearer roadmap for targeted interventions to reduce risky prescribing.”
The study has limitations. The authors assumed the last clinical setting recorded in claims was the place of prescription initiation, which may not always be accurate. Future research will examine provider characteristics and decision-making patterns to better understand who prescribes these drugs and why.
Key Questions Answered:
A: Emergency departments and hospitals frequently confront acute behavioral crises or severe agitation in patients with dementia. Clinicians may use antipsychotics or sedatives to manage immediate safety concerns, combat severe agitation, or address “sundowning.” However, what are intended as short-term measures can become long-term prescriptions when they are not explicitly reassessed at discharge.
A: Beyond cognitive slowing and increased confusion, these medications raise the risk of physical injury. Antipsychotics and benzodiazepines impair balance and reaction time, increasing the likelihood of falls, hip fractures, head injuries, and hospital readmissions.
A: The study suggests a clear approach: insist on thorough medication reconciliation at hospital discharge. Families should ask which medications are new, why they were started, what benefits and risks were considered, and when or how these medications will be reviewed and stopped if no longer needed.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- The full journal paper was reviewed.
- Additional editorial context was added by staff.
About this neurology and neuropharmacology research news
Author: Enrique Rivero
Source: UCLA
Contact: Enrique Rivero – UCLA
Image: Image credited to Neuroscience News
Original Research: Open access. “Initiation Setting and Persistence of Medications Affecting Cognition in Older Adults” by Dan P. Ly, Annie W. Yang, Mei Leng, Catherine Sarkisian, Cheryl L. Damberg, and John N. Mafi. JAMA Network Open. DOI: 10.1001/jamanetworkopen.2026.10234
Abstract
Initiation Setting and Persistence of Medications Affecting Cognition in Older Adults
Medications that adversely affect cognition—such as antipsychotics—remain commonly prescribed for older adults despite evidence linking them to increased delirium and fall risk. This study evaluated the clinical settings where these medications are first started and measured how often patients remain on them one year later, with a focus on differences by cognitive status.
Funding: This work was supported by awards from the National Institute on Aging, the National Institutes of Health, the US Department of Veterans Affairs Health System Research, the UCLA Clinical and Translational Science Institute, and the US Deprescribing Research Network, among other sources acknowledged by the authors.