New findings from UCLA research indicate that initial prescriptions for pharmaceuticals influencing cognitive function, such as antipsychotic agents, originate with significantly greater frequency from acute and post-acute care environments – encompassing emergency departments, hospitals, and skilled nursing facilities – in contrast to conventional physician offices.
A substantial proportion of these pharmaceutical agents are administered to individuals experiencing cognitive decline, for instance, those with dementia, who are inherently more vulnerable to adverse outcomes associated with these medications, including falls, disorientation, and subsequent hospitalizations.
Pharmaceuticals like antipsychotics and benzodiazepines are recognized for their impact on cognitive capabilities, exacerbating delirium, and elevating the risk of falls among the elderly population. Furthermore, the issuance of prescriptions for these types of medications tends to be enduring and difficult to discontinue; for example, among older individuals diagnosed with dementia who are prescribed a cognition-altering agent, over half continue this regimen a year later.
Dr. Dan Ly, an assistant professor of medicine within 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. Dr. Ly also holds the position of lead author for this research letter.
The outcomes of this investigation are slated for publication on April 28 in the esteemed peer-reviewed periodical, JAMA Network Open.
Dr. Ly highlighted a recognized imperative to decrease the prescribing of these cognition-modifying pharmaceuticals. “Our findings suggest that endeavors to curtail these prescriptions could yield the most significant results if our focus is specifically directed toward acute and post-acute care settings,” he commented.
The research team undertook an analysis of data sourced from the Health and Retirement Study (HRS), encompassing adults aged 66 and older, cross-referenced with Medicare fee-for-service claims spanning from January 1, 2008, through December 31, 2021. The inquiry examined four distinct medication classes: benzodiazepines, nonbenzodiazepine hypnotics, antipsychotics, and anticholinergics. A particular emphasis was placed on identifying the clinical environment where the initial prescription was dispensed, comparing initiations from acute and post-acute settings (which also encompassed skilled nursing facilities) against those originating from physicians’ offices.
Participants were categorized into one of three distinct groups based on their cognitive status: individuals with no discernible cognitive impairment; those with cognitive impairment but not dementia (CIND); or individuals diagnosed with dementia.
Across the studied population, 14% of individuals without cognitive impairment, 17% of those classified as CIND, and 22% of individuals with dementia received a new prescription for a cognition-affecting medication during an acute or post-acute care encounter. Moreover, the persistence of use was notable, with 38% of those without cognitive impairment, 44% of CIND patients, and 51% of individuals with dementia continuing to take these medications one year post-prescription.
When contrasted with the overall patient visits to these types of facilities, a disproportionately large segment of these prescriptions were initiated in acute and post-acute environments. For illustration, 43% of antipsychotic prescriptions were first issued in acute and post-acute settings for patients with dementia, despite only 22% of all patient visits by individuals with dementia occurring in such locations.
“We were taken aback by the substantially higher incidence of cognition-influencing medications being prescribed outside of traditional office settings,” stated Dr. Ly. “Our work illuminates the specific loci of prescription initiation for these agents, thereby enabling more precise targeting of intervention strategies aimed at reducing their overall administration.”
The present findings are subject to a limitation stemming from the researchers’ presumption that the last observed setting for patients corresponds to the location where prescriptions were initiated. The subsequent phase of this research will involve an examination of the characteristics of healthcare providers who exhibit a higher propensity for prescribing these medications, according to Dr. Ly.
“This research builds upon our previous investigations, which uncovered that as many as 70% of prescriptions for Central Nervous System-active medications lacked an adequately documented clinical rationale. This underscored a critical necessity to address and curtail excessive and potentially hazardous prescribing practices,” remarked Dr. John N. Mafi, an associate professor of medicine at the Geffen School and the senior author of the study. “Our most recent investigation now equips policymakers and clinicians with a strategic blueprint, precisely indicating the initial areas where their intervention efforts should be concentrated: namely, acute and post-acute care facilities.”
Ly, D. P, et al. (2026) Initiation Setting and Persistence of Medications Affecting Cognition in Older Adults. JAMA Network Open. DOI: 10.1001/jamanetworkopen.2026.10234. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2848234
