Deprescribing considerations for central nervous system-active polypharmacy in patients with dementia

Abstract

Older adults with dementia are much more likely than those without dementia to experience polypharmacy, defined as taking at least five medications. Approximately 72% of older adults with dementia, versus only 44% of those without dementia, experience polypharmacy.¹ Although multiple medications may be prescribed to treat multiple chronic conditions, polypharmacy in older adults is associated with increased risks of adverse drug events,² cognitive and physical impairment,³ frailty, falls, and mortality.⁴ For older adults with dementia, the most common contributors to polypharmacy include cardiovascular medications and medications acting on the central nervous system.¹ An estimated 73% of adults aged 65 and over with dementia use at least one cardiovascular medication, and an estimated 85% use at least one medication acting on the central nervous system.^{1, 5} Yet another risk beyond polypharmacy alone is the concomitant use of three or more medications all acting on the central nervous system, termed central nervous system-active polypharmacy. These medications typically include: antiepileptics, antidepressants, antipsychotics, benzodiazepines, nonbenzodiazepine benzodiazepine receptor agonist hypnotics (i.e., z-drugs), opioids, and skeletal muscle relaxants.⁶ The concomitant use of these medications is associated with increased risks of falls,⁷ cognitive decline,⁸ emergency room visits, and hospitalizations.^{9, 10} The 2023 Beers Criteria recommend against central nervous system-active polypharmacy.⁶

Older adults with dementia are more likely to experience central nervous system-active polypharmacy because many of the medications can be used to manage neuropsychiatric symptoms, such as agitation, aggression, sleep disorders, mood disorders, and psychotic symptoms, related to the underlying dementia. This is concerning because antipsychotics, benzodiazepines, and z-drugs are specifically advised against in persons with dementia.⁶ In 2005, the Food and Drug Administration added a black box warning for atypical antipsychotics for persons with dementia due to increased mortality risks, and in 2008, the black box warning was expanded to all antipsychotics (including typical antipsychotics). Nonetheless, in community-dwelling older adults living with dementia in the United States, 14% in 2018 concomitantly used at least three medications acting on the central nervous system for at least 30 overlapping days.¹¹

In this month's issue, Dr. Vordenberg and colleagues sought to understand how the 14% came to be, by analyzing 2019 prescription claims data from a cohort of community-dwelling Medicare beneficiaries aged 65 and above with Alzheimer's disease and related dementias. Specifically, the study authors determined how many prescribers contributed to the concomitant use of at least three central nervous system-active medications. This is particularly relevant because one might expect that multiple medications could be prescribed at different times by multiple prescribers and that prescribers may not be aware that all three medications are being used at the same time. Aligned with this expectation, the authors find that 75% of community-dwelling older adults with dementia who experienced central nervous system-active polypharmacy were prescribed the medications by at least two prescribers.¹² These findings suggest that communication among prescribers is essential for reducing polypharmacy. Critically, prescribers need to know their patient's up-to-date and full list of medications. Patients and caregivers also play an important role by bringing a comprehensive and up-to-date medication list (i.e., a “living medication list”) to their provider appointments. This is particularly helpful when patients are cared for by multiple healthcare providers in different health systems that do not share electronic health records.

Another important finding from this study is revealed when study authors take a population-level view and consider total exposure to central nervous system-active polypharmacy (i.e., days exposed to central nervous system-active polypharmacy across all patients, as opposed to number of patients exposed). Using a denominator of total person-days exposed to central nervous system-active polypharmacy, study authors find that 45% of these exposure days are attributed to a single provider.¹² This is contrasted with the 25% of individuals who experience central nervous system-active polypharmacy attributed to a single provider.¹² Thus, those who experience central nervous system-active polypharmacy due to multiple prescriptions from a single provider are more likely to stay on these medications, or be exposed to concomitant use for longer. This potentially suggests that these prescribers intended for the medications to be used concomitantly. Deprescribing efforts should focus on better understanding why individual prescribers prescribe multiple medications acting on the central nervous system and whether it is possible to substitute with other medications or nonpharmacological alternatives. In the study, the main drug classes contributing to central nervous system-active polypharmacy were antidepressants (93% of exposure days), antiepileptics (62%), and antipsychotics (54%).¹² Benzodiazepines were involved in of 37% of exposure days, opioids were involved in 26%, and z-drugs and skeletal muscle relaxants were rarely involved (5% and 2%, respectively).¹² The use of antipsychotics in 54% of days alongside additional medications acting on the central nervous system is particularly disconcerting, given that these medications already hold black box warnings for their use by themselves in persons with dementia. However, managing the behavioral and psychological symptoms of dementia can be extremely difficult. At a minimum, 2023 Beers Criteria recommend “periodic deprescribing attempts” for antipsychotics to assess ongoing need and the lowest effective dose.⁶

When examining the types of prescribers involved in central nervous system-active polypharmacy, the study authors found that primary care physicians contributed to 63% of exposure days, so deprescribing efforts may focus on this prescriber group.¹² In the study, primary care physicians included those with training in internal medicine, family medicine, and geriatric medicine. The next most common contributor (32%) included primary care advanced practice providers, including nurse practitioners, clinical nurse specialists, and physician assistants in fields related to primary care, adult health, family health, gerontology, chronic care, and home health. Meanwhile, psychiatrists were only involved in 22% of exposure days, other prescribers in 14%, neurologists in 13%, and other advanced practice providers in 7%.¹² Primary care providers face myriad barriers to achieving successful deprescribing in clinical practice, ranging from fragmented care systems to lack of access to nonpharmacological treatment options to inadequate time for careful medication reconciliation.^{13, 14} Thus, potential solutions targeting these groups are more likely to be successful if they extend beyond educational efforts and encompass comprehensive changes to clinic structures (e.g. through support for embedded clinical pharmacists or nonpharmacological treatment strategies for behavioral and psychological symptoms of dementia) and avoid unnecessary increases in primary care provider workloads.

Solutions for how to deprescribe central nervous system-active polypharmacy among older adults with dementia are urgently needed, and Dr. Vordenberg and colleagues shed new light to inform the careful design of future deprescribing interventions. It should be acknowledged that given the high medication burden faced by older adults with dementia and their caregivers, 87% of them in 2016 reported being willing to stop at least one of their medications, 50% reported being uncomfortable with taking as many as five or more medications, and 22% believed they could be taking medications that they no longer needed.¹⁵ Yet, polypharmacy and central nervous system-active polypharmacy remain vexing problems for older adults with dementia. Efforts to encourage deprescribing while managing the distressing behavioral and psychological symptoms of dementia can be challenging. This difficult balance must be front and center when designing future deprescribing interventions. Innovative comprehensive dementia care programs incorporating medication management and caregiver support strategies overseen by an interdisciplinary team, such as those supported by the ongoing Centers for Medicare and Medicaid Services' nationwide “Guiding an Improved Dementia Experience” (GUIDE) Model, represent a particularly promising venue in which these deprescribing interventions may thrive.¹⁶ Given patient willingness and discomfort with polypharmacy, an important component to include in any deprescribing intervention is engagement of the patient and caregiver, in addition to the primary care provider. It is critical that these key individuals understand the medication risks and are involved in potentially repeated attempts to decrease dose and discontinue medications whenever possible.

All authors meet the International Committee of Medical Journal Editors criteria for authorship. AH was responsible for drafting the article. AH and MEG made substantial contributions to conception and design of the manuscript and interpretation of data, revised critically for important intellectual content, and gave approval of the final version.

AH discloses research funding from Abbott, AstraZeneca, VA, and NIH; consulting fees from Genentech; and honoraria from Academy of Managed Care Pharmacy. AH serves as an Assistant Editor for the Journal of Managed Care & Specialty Pharmacy and on the Editorial Advisory Board for JAGS.

The funders had no role in the design, analysis, and preparation of the article. The contents do not represent the views of the U.S. Department of Veterans Affairs or the United States Government.

AH was supported by VA HSR&D funding (IK2 HX003359), Durham Center of Innovation to Accelerate Discovery and Practice Transformation (CIN 13–410) at the Durham VA Health Care System, and by the Health Care Systems Research Network (HCSRN)-Older Americans Independence Centers (OAICs) AGING Initiative (R33AG057806). MEG was supported by NIA (K76AG088411 and R03AG078804). Both authors were supported by the US Deprescribing Research Network (R24AG064025).