Sorry, you need to enable JavaScript to visit this website.
Skip to main content

I Just Want to Go Home: Understanding Delirium’s Impact on Treatment Preferences

Karl Steinberg, MD, CMD, HMDC, HEC-C | March 27, 2024
View more articles from the same authors.

The Case

A 76-year-old man with a history of heart failure, chronic lung disease, and chronic urinary retention of unknown etiology was readmitted for altered mental status after a recent hospitalization. He was intermittently combative and refused to accept bilevel positive airway pressure (BiPAP), nasal cannula, or mask for his low oxygen saturation. Over the first three days of his hospitalization, he repeatedly requested to go home, declined most medical interventions, and made angry comments about the care he was receiving. His family, who lived a few hours away, acted as his medical decision-makers. They were consistently updated and started working on finding a housing solution closer to them, based on his reported wishes, but were not able to travel to the hospital to see him.

On hospital day 3, the patient had a discussion alone with a hospitalist, in which the patient was documented as agreeing to receive care at home or at a board and care through hospice. The family was alerted, and the palliative medicine service was consulted to assist with transition. The palliative medicine consultant visited the patient and determined that he did not have capacity to make decisions due to delirium, nor was he clearly eligible for hospice. In discussion with the family, it became clear that they had only agreed to hospice care for their father after their discussion with the hospitalist, but the family had not been present for the discussion between the patient and hospitalist. Accordingly, everyone agreed to continue to treat the patient’s underlying medical issues and to wait to see if his delirium improved. In subsequent days, the patient's mental status cleared, and he had an in-depth conversation with his family in which he stated that he vehemently opposed hospice care, wished to receive the full scope of medical treatments (including cardiopulmonary resuscitation [CPR]), and intended to return to the hospital if needed in the future. He was then transferred to a skilled nursing facility near his family for rehabilitation.

The Commentary

By Karl Steinberg, MD, CMD, HMDC, HEC-C

In this case, a seriously ill, encephalopathic patient with active cardiac, respiratory, urological and neurological conditions was rehospitalized, experiencing hypoxia requiring supplemental oxygen. Probably at least in part due to delirium, the patient refused several recommended treatments and focused on his desire to go home. The case study does not discuss what strategies were undertaken to address the patient’s care refusals, which may have been clinically significant. A treating clinician documented a discussion where the patient agreed to hospice enrollment, presumably because the patient was informed that this was a way his request to go home would be promptly honored. The case study does not describe the patient’s mental status, especially his cognitive functioning, at the time of this encounter. The clinician and care team apparently assumed that the patient had sufficient decisional capacity to express a preference for hospice admission and to assent to this plan—which was then presented to his family, who believed this was what he wanted based on their communication with the team, but not corroborated through their own conversations with the patient.

Because the details of the hospitalist’s conversation with the patient were not documented, it is impossible to know definitively whether the patient had decisional capacity during that conversation. However, subsequent conversations in which the patient voiced a desire to receive life-prolonging treatments, after he had clearly regained capacity, make it improbable that he had the capacity to understand during the earlier conversation. More likely, he assented to the notion of hospice because he was so focused on his immediate goal to “go home.” It is also unclear whether the patient understood what hospice entails—while healthcare professionals are generally readily familiar with hospice, many nonmedical members of the public are not.

For a patient to be eligible for the Medicare Hospice Benefit,1 which comes through Medicare Part A, they must have a six-month (or less) life expectancy if their illness runs its ordinary course (defined as a terminal illness or combination of illnesses), and this prognosis must be certified by a physician. They must also agree to forgo treatments directed at curing the nominal terminal condition, “waiv[ing] all rights to Medicare payments for the terminal illness and related conditions” and allowing hospice to manage the overall medical plan of care with its interdisciplinary team, consisting of physicians, nurses, social workers, spiritual care counselors, volunteers and bereavement providers. Hospice usually provides wraparound services including medications, durable medical equipment, some custodial services such as home health aide assistance with activities of daily living (ADLs), and psychosocial and spiritual assistance for the patient and their family. The services can be provided in almost any care setting, but with limited short-term exceptions, hospice does not cover 24-hour care.

Fortunately, the palliative care consultant carefully assessed the patient’s decisional capacity and determined that in fact, the patient lacked the capacity to request or consent to hospice and was unable to make an informed decision in that regard. Moreover, the consultant had doubts as to whether the patient would be eligible for hospice. Although the case presentation does not clarify whether this doubt related to prognostic versus philosophical criteria, it appears that the patient expressed a desire not to forgo life-prolonging medical treatments—hence rendering him ineligible for hospice for philosophical reasons. Moreover, if the patient was indeed willing to accept medical treatments (e.g., oxygen supplementation, intravenous fluids and antibiotics), that would render him ineligible for hospice based on prognosis, since it is assumed that with appropriate treatments, he would recover from this treatable acute medical condition—which in fact he did.

It is not clear whether this patient had any form of advance health care directive or designated surrogate (agent, proxy, decision-maker). Different states have different mechanisms for identifying a surrogate when the patient has not designated one, and it is presumed from the narrative that in the state where this occurred, members of the patient’s family would have been permitted to consent to hospice election.2,3 Clinicians should be familiar with their state laws governing surrogate medical decision-making for incapacitated patients, which vary across the United States.4 Whenever possible, the healthcare team should strive to identify a single preferred point person through whom to communicate with a patient’s family. In this case, the fact that the patient did not get to speak with his family directly about the purported hospice request created an additional potential safety issue. If the family had been at the bedside, they probably would have recognized that he was not cognitively intact or at his baseline, which would have precluded further efforts at hospice enrollment.

Potential harm to the patient was averted by the careful capacity assessment and action of the palliative care consultant, as discharging the patient home on hospice with comfort-focused treatment before all his active acute medical problems had been addressed likely would have resulted in his premature death.

Approaches to Improving Safety

Delirium is common in hospitalized older patients, and it carries significant negative prognostic implications. Delirium may manifest as nonadherence with care, including behavior that negatively impacts safety, such as refusing medications, pulling lines and tubes, trying to get out of bed unsafely, or even violent behavior. Strategies to screen for delirium, prevent delirium, avoid management that may contribute to delirium when possible, and alleviate it promptly when it occurs, are highly advisable.5,6,7 If this patient had not developed delirium, or if it had been promptly identified and steps taken to ameliorate it, it is improbable that he would have assented to the plan for hospice enrollment.

It is essential for clinicians to be able to assess decisional capacity, which is not always black-and-white and should be evaluated considering the specific clinical scenario. A patient (or surrogate) ideally should be able to understand the choices being considered, assess the likely consequences of each choice, use reasoning to decide among the options, and be able to express their reasoning and the decision itself.8 Documentation of these abilities is advisable when capacity (or incapacity) is not obvious. In especially challenging situations, formal mental health evaluations and/or ethics committee input should be requested. In this case, if the clinician who placed the hospice referral had performed such a diligent capacity assessment, they probably would not have proceeded with the hospice referral or told the family that this was what the patient wanted. It is important when discussing goals of care with patients (and/or their surrogates) to provide objective information to guide their decisions, to engage in shared decision-making, and to avoid overtly or subtly injecting our own beliefs and values into the conversation—or recommending a particular decision for reasons of expediency (e.g., discharging a ‘difficult’ patient).9,10

Take Home Points

  • Clinicians should be mindful that delirium can be dangerous and carries significant negative short-term and long-term prognostic implications. Health care institutions should implement policies and procedures that promote consistent screening, prevention, prompt identification, and strategies to attempt to resolve delirium as quickly as possible.
  • Decisional capacity can be fluid and is situation-specific, so clinicians should have the knowledge and skills to perform capacity assessments in patients with a variety of deficits and should request assistance from ethics and/or mental health consultants when presence or absence of capacity for a particular decision is unclear.
  • Whenever possible, shared decision-making that involves robust communication between the healthcare team and the patient, and when possible, involved family members, should be sought.
  • When considering hospice referrals, clinicians should feel confident that patients (and/or their decision-makers) have at least a basic understanding of what hospice entails, the requirements for hospice eligibility, and how hospice election can limit future treatment options—although it is reasonable to expect hospice personnel to explain these nuances in detail before a patient is enrolled.
  • Determining who has legal standing to make decisions for an incapacitated patient is an important responsibility of the healthcare team and may vary in different jurisdictions.
  • Clinicians engaging in goals of care conversations and determining patients’ treatment preferences should strive to be objective information sources and should avoid pushing the patient and/or their surrogates toward a decision based on expediency or the clinician’s own personal values and beliefs.

Karl Steinberg, MD, CMD, HMDC, HEC-C
California State University Shiley Haynes Institute for Palliative Care
President, National POLST Collaborative
Past President, AMDA – The Society for Post-Acute and Long-Term Care Medicine, Coalition for Compassionate Care of California, California Association of Long Term Care Medicine
Medical Director, Hospice by the Sea, Solana Beach, CA


  1. Hospice. Centers for Medicare and Medicaid Services (CMS). Accessed March 14, 2024. [Available at]
  2. Comer AR, Gaffney M, Stone CL, Torke A. Physician understanding and application of surrogate decision-making laws in clinical practice. AJOB Empir Bioeth. 2017;8(3):198-204. [Available at]
  3. Patel J, Ackermann RJ. Issues in Geriatric Care: Medical Decision-Making. FP Essent. 2018;468:35-38.
  4. Arias, M. Recent updates to default surrogate statutes. Bifocal. 2023;44(3):54-57. [Free full text]
  5. Hshieh TT, Inouye SK, Oh ES. Delirium in the Elderly. Clin Geriatr Med. 2020;36(2):183-199. [Available at]
  6. Burton JK, Craig L, Yong SQ, et al. Non-pharmacological interventions for preventing delirium in hospitalised non-ICU patients. Cochrane Database Syst Rev. 2021;11(11):CD013307. [Free full text]
  7. Jackson TA, Wilson D, Richardson S, Lord JM. Predicting outcome in older hospital patients with delirium: a systematic literature review. Int J Geriatr Psychiatry. 2016;31(4):392-399. [Available at]
  8. Hamilton RKB, Phelan CH, Chin NA, et al. The U-ARE Protocol: A Pragmatic Approach to Decisional Capacity Assessment for Clinical Research. J Alzheimers Dis. 2020;73(2):431-442. [Free full text]
  9. VitalTalk. Accessed March 14, 2024. [Available at]
  10. Jain N, Bernacki RE. Goals of Care Conversations in Serious Illness: A Practical Guide. Med Clin North Am. 2020;104(3):375-389. [Free full text]
This project was funded under contract number 75Q80119C00004 from the Agency for Healthcare Research and Quality (AHRQ), U.S. Department of Health and Human Services. The authors are solely responsible for this report’s contents, findings, and conclusions, which do not necessarily represent the views of AHRQ. Readers should not interpret any statement in this report as an official position of AHRQ or of the U.S. Department of Health and Human Services. None of the authors has any affiliation or financial involvement that conflicts with the material presented in this report. View AHRQ Disclaimers
Related Resources