INTRODUCTION
Antipsychotics and benzodiazepines are common components of the hospice toolkit prescribed for both physical and behavioral symptom management. 1 Despite their routine use in hospice, there are currently limited guidelines or consensus on best prescribing practices. 2 Antipsychotics and benzodiazepines are often utilized to treat distressing symptoms at end of life, however, these medication classes are not without risk when prescribed to older adults including side effects such as sedation, falls, and increased mortality. 2 , 3 Hospice represents an increasingly heterogenous population, including many with non‐cancer diagnoses, some of whom will outlive and be discharged from the hospice benefit. 4 For those patients who are not imminently dying, patients and their care partners may prefer avoiding such medications in order to maintain function and minimize sedation.
Given a clear lack of clear evidence to inform antipsychotic and benzodiazepine prescribing best practice for patients enrolled in hospice, there may be variation or inconsistent utilization of these medications. The goal of this study is to describe the variation in the prevalence of antipsychotic and benzodiazepine prescribing among beneficiaries enrolled in hospice across states. Understanding variation in practice patterns may help to inform appropriate prescribing in hospice and to clarify the current standard of care.
METHODS
We identified all Medicare beneficiaries ≥65 years old enrolled in hospice for ≥1 day in 2017 using the 100% Medicare hospice file and identified those that had any antipsychotic or benzodiazepine prescription fill during hospice enrollment. 5 We examined state‐level prescribing using a logistic regression model controlling for beneficiary age, sex, hospice qualifying diagnosis, and state. The state was included as a fixed effect and beneficiaries were assigned to the state of their hospice agency. Results are presented as predictive margins, which represent the average predicted prescribing rate if the entire sample had been in a given group. The average adjusted prescribing rate was computed for each state by setting all beneficiaries in the cohort to a given state (e.g., Alaska) and then computing the predictive margins. This was repeated for each state, resulting in state‐level average adjusted prescribing rates. Lastly, we determined Spearman's correlation to examine the relationship between antipsychotic and benzodiazepine prescribing. Analyses were conducted using SAS 9.4. This study was approved by the Michigan Medicine IRB; informed consent was waived.
RESULTS
Of 1,393,622 beneficiaries enrolled in 4219 hospice agencies in 2017, 59.4% (standard deviation [SD]: 7.4) were female with a mean age of 83.8 years (SD: 2.0). 84.6% were non‐Hispanic white, 8.9% non‐Hispanic Black, and 2.8% Hispanic. The most common hospice qualifying diagnoses were cancer (22.9%), heart disease (21.8%), and dementia (21.3%).
Figure 1 shows adjusted state‐level prescribing rates for antipsychotics and benzodiazepines. Rates of antipsychotic prescribing varied widely from 61.8% of hospice beneficiaries in Oregon to 15.4% in Oklahoma (Figure 1, SD: 10.9). Benzodiazepine prescribing rates also varied across states, with a high of 76.9% in Oregon and low of 41.6% in New York (SD: 6.7). All states had higher prescribing rates of benzodiazepines than antipsychotics. There was a weak correlation between adjusted state‐level antipsychotic and benzodiazepine prescribing rates (Spearman's rho, 0.38; p‐value = 0.008), but five states were in the top quartile of both antipsychotic and benzodiazepine prescribing (Oregon, New Mexico, Washington, New Hampshire, and Florida).
FIGURE 1.
(A–C) State‐level average adjusted antipsychotic and benzodiazepine prescribing rates among hospice beneficiaries 65 years and older, 2017. The interpretation of an adjusted prescribing rate for a given state is the average predicted probability of being prescribed an antipsychotic or benzodiazepine in hospice had all beneficiaries lived in the given state. This was repeated for each state, resulting in state‐level average adjusted prescribing rates.
DISCUSSION
Among hospice enrollees in 2017, we found marked variation in prescribing of antipsychotics and benzodiazepines between states. This variation in prescribing may reflect the limited evidence base to guide end of life psychotropic prescribing, 2 and is consistent with other work evaluating differences in practice patterns when there is not consensus regarding treatment appropriateness. 6 These regional differences may reflect prescribing choices informed by provider beliefs about the effectiveness of medications or organizational norms influencing order‐set defaults rather than individualized assessment of patient needs. 7 However, the limited correlation between antipsychotic and benzodiazepine prescribing suggests these medications are not being uniformly prescribed to all hospice enrollees through comfort kits.
Among our study's limitations, medications prescribed may not be taken by the patient and we are not able to ascertain the appropriateness of prescribing because prescription claims do not include indication for use. Our adjusted prescribing models do not account for all state‐level variables that may influence prescribing such as access to inpatient hospice facilities.
Future work is needed to understand the sources of state variation and the associated patient outcomes. A better understanding of the factors that influence hospice prescribing culture can help inform guidelines and best practices for medication use to balance relief of distressing symptoms while avoiding medication‐related harms.
AUTHOR CONTRIBUTIONS
Dr. Gerlach had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Gerlach, Teno, Bynum, Maust. Acquisition, analysis, or interpretation of data: all authors. Drafting of the manuscript: Gerlach, Strominger, Maust. Critical revision of the manuscript for important intellectual content: all authors. Statistical analysis: Zhang, Strominger. Obtaining funding: Gerlach. Administrative, technical, or material support: Gerlach. Supervision: Gerlach, Maust.
CONFLICT OF INTEREST
The authors have none to disclose.
SPONSOR'S ROLE
The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Gerlach LB, Zhang L, Strominger J, Teno J, Bynum JPW, Maust DT. State variation in antipsychotic and benzodiazepine prescribing among hospice beneficiaries in the United States. J Am Geriatr Soc. 2022;70(12):3637‐3640. doi: 10.1111/jgs.17992
No related papers have been published or submitted from this study.
Funding information National Institute on Aging, Grant/Award Numbers: K23AG066864, R01AG056407
REFERENCES
- 1. Gerlach LB, Kales HC, Kim HM, et al. Prevalence of psychotropic and opioid prescribing among hospice beneficiaries in the United States, 2014–2016. J Am Geriatr Soc. 2021;69(6):1479‐1489. doi: 10.1111/jgs.17085 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Finucane AM, Jones L, Leurent B, et al. Drug therapy for delirium in terminally ill adults. The . Cochrane Database Syst Rev. 2020;1(1):CD004770. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Agar MR, Lawlor PG, Quinn S, et al. Efficacy of Oral risperidone, haloperidol, or placebo for symptoms of delirium among patients in palliative care: a randomized clinical trial. JAMA Intern Med. 2017;177(1):34‐42. [DOI] [PubMed] [Google Scholar]
- 4. Teno JM, Plotzke M, Gozalo P, Mor V. A national study of live discharges from hospice. J Palliat Med. 2014;17(10):1121‐1127. [DOI] [PubMed] [Google Scholar]
- 5. Center for Medicare Chronic Care Policy Group . Medicare hospice payment reform: analysis of how the Medicare hospice benefit is used. Published December 3, 2015. Available at: https://www.cms.gov/Medicare/Medicare‐Fee‐for‐Service‐Payment/Hospice/Downloads/December‐2015‐Technical‐Report.pdf. Accessed September 15, 2017.
- 6. Mitchell SL, Teno JM, Roy J, Kabumoto G, Mor V. Clinical and organizational factors associated with feeding tube use among nursing home residents with advanced cognitive impairment. JAMA. 2003;290(1):73‐80. [DOI] [PubMed] [Google Scholar]
- 7. Gerlach LB, Zhang L, Strominger J, et al. Variation in benzodiazepine and antipsychotic prescribing among hospice agencies. J Gen Intern Med. 2022. doi: 10.1038/gim.2015.30. [DOI] [PMC free article] [PubMed] [Google Scholar]