Consultation-Liaison Case Conference: Suicidal Ideation in a Patient at the End-of-Life

Abstract

Psychiatric comorbidities are common among patients approaching the end of life, often necessitating the involvement of consultation-liaison psychiatrists. We present the case of a patient with advanced metastatic prostate cancer and a complicated hospital course who made suicidal remarks and requested a hastened death. This common and challenging clinical scenario requires a multidisciplinary approach. In this article, experts in consultation psychiatry, palliative care, and psychooncology describe helpful diagnostic and therapeutic strategies for such cases. The key learning points are the differential diagnoses in end-of-life patients endorsing suicidal ideation, the psychiatric management of oncological and palliative care patients, the implementation of a safe discharge plan, and the role of the consultation-liaison psychiatrist in hospice care.

CASE PRESENTATION (DR. EDUARDO ANDRES CALAGUA-BEDOYA)

A 69-year-old male with no past psychiatric history and a medical history of hypertension, congestive heart failure, and prostate cancer with metastases to bone and liver was admitted to the hospital due to facial cellulitis and a nasal abscess, which were treated with vancomycin plus abscess drainage. One day later, he developed hypoxic respiratory failure and a drop in his hemoglobin and platelet count. Abdominal and pelvic imaging did not reveal bleeding. The anemia and thrombocytopenia were thought to be secondary to his chemotherapeutic agents, olaparib and leuprolide, and the blood dyscrasias and respiratory distress improved with iron, blood transfusions, and supplemental oxygen. Three days later, the patient developed new respiratory problems secondary to volume overload. He was transferred to a step-down unit and started on intravenous diuretics and cefepime for possible aspiration pneumonia with adequate response.

His primary team thought his prognosis was grim given his advanced oncological disease. On his fifth day of hospitalization, after a conversation with his physicians and relatives about his deteriorating clinical status and the possibility of needing intubation in case of respiratory failure, he requested his code status to be changed to do not resuscitate/do not intubate. The next day, the patient had an episode of epistaxis, which was controlled by applying pressure with a gauze. His hemoglobin started to decrease again, but he declined a new transfusion and reported being tired of the medical complications he had experienced since his admission. He refused to pursue any further treatments. The primary team consulted palliative care in order to help clarify the patient’s goals of care, as he was thought to be terminally ill. Palliative care determined that the patient had capacity to participate in this conversation. During the evaluation, he mentioned uncontrollable oncological pain as one of his major concerns. Of note, he was receiving hydromorphone for pain management. Although he reported it was inefficacious, he declined other treatments. The patient eventually became tearful, stated he no longer wanted to fight, and requested assistance from the palliative care team to terminate his life. He further noted that “it would be easy to find some way to do it at home” but did not elaborate on details. After his statements, the psychiatry team was consulted to assess for suicidal ideation and depression.

During our initial interview, the patient described feeling “tired and frustrated” due to his medical ailments. He felt that he had suffered enough, that he no longer had an acceptable quality of life (QOL), and that future treatments would be futile. When probed for details, he stated “I cannot even walk anymore, everything went to hell.” He expressed the wish to die peacefully and requested an injectable medication to hasten his death. The patient reiterated his desire to harm himself at home, but not at the hospital, if he were not provided with a physician-assisted death—an intervention that he could have been eligible for in jurisdictions where it is available but illegal in the state in which the patient was receiving care. He did not verbalize a specific suicide plan and said “I have not thought about that yet, I will worry about it later.” A review of systems and mental status exam were notable for decreased appetite and sleep, irritability, and guardedness but no evidence of anhedonia, anxiety, psychosis, agitation, disorientation, or inattention. He did not seem overtly delirious as he was attentive to us and all clinicians who entered the room and never demonstrated evidence of waxing and waning consciousness or of disorientation. However, a comprehensive evaluation for this condition was not possible given his lack of cooperation. He denied prior psychiatric history, including diagnoses, treatments, admissions, or suicide attempts. The patient did have a 30-year history of cannabis use, but he had stopped after his cancer diagnosis. A review of ancillary tests at the time of the psychiatric evaluation showed anemia, thrombocytopenia, uremia, and hypocalcemia. His vital signs showed a blood pressure of 151/81 mm Hg, pulse of 103, respiratory rate of 15, temperature of 36.4°C (97.6°F), and oxygen saturation of 95% while using a nasal cannula. A computerized tomography brain scan without contrast from admission day showed age-appropriate parenchymal volume loss and microvascular ischemic changes, but no acute hemorrhages or ischemia. We gathered collateral from his relatives at bedside, who reported that the patient had been a very independent man his whole life and described him as the “my way or the highway” type of person. He had never made remarks in the past about suicide. He did seem upset due to his medical issues, mainly because he was losing autonomy, one of his core values. The family also mentioned that the patient lived with his wife and with several relatives, including his adult children and their spouses, his grandchildren, and his nieces. We wondered whether his resolve to go home was strengthened by limitations on visitation of some of these important family members in the context of the COVID-19 pandemic.

Later that day, the family visited the patient, and he again manifested his desire to go home and stop his treatment. His relatives wanted to respect his wishes and inquired about home hospice. They were committed to monitoring him closely at his house given his recent comments about self-harm. The primary team requested psychiatric clearance to send the patient home, but he refused to engage further with our psychiatry team to clarify his diagnosis and prior suicidal statements and ensure a safe discharge plan. Our differential diagnosis was broad and included a primary mood or adjustment disorder, delirium, desire for hastened death, or psychological factors such as frustration or demoralization. Additionally, we did not know if the patient had a clear plan or intention to harm himself upon discharge. In subsequent days, we made multiple attempts to engage the patient with assistance of his relatives, all of which were unsuccessful due to the lack of collaboration and lethargy. Unfortunately, the patient’s clinical condition quickly deteriorated, and 3 days after our first encounter, he passed away in the hospital surrounded by his family.

EVALUATION AND DIAGNOSIS (DR. CARRIE ERNST)

The above case demonstrates the complexity of psychiatric comorbidity among individuals at the end-of-life (EOL).

This patient presented with a “frustrated” mood, irritable affect, hopelessness about the potential for any symptomatic improvement, and a strong desire to die. The desire for death included both a request for assistance from his health care providers to end his life and then a statement that he would find a way to end his own life upon discharge to home. Additional symptoms included poor sleep, decreased appetite, low energy, and anxious ruminations about his situation. The presence of this constellation of symptoms in a medically complex, terminally ill patient should first prompt consideration of the most likely diagnosis. More than 60% of patients with cancer report experiencing psychological distress.¹ Distress associated with normal grieving/coping or high somatic symptom burden must be differentiated from distress associated with psychiatric disorders (Table 1). The differential diagnosis in this case is broad and includes delirium, adjustment disorder with depressed mood, major depressive episode, mood disorder secondary to general medical condition, demoralization, grief/normal coping, medication-induced mood disorder, and subsyndromal depressive symptoms.

TABLE 1.

Differential Diagnosis of Psychological Distress in Terminally Ill Patients

Assessment	Adjustment disorder*	Major depressive disorder*	Delirium*	Grief	Demoralization
Diagnostic criteria (*Adapted from DSM-5)	The development of emotional or behavioral symptoms in response to an identifiable stressor(s) occurring within 3 mo of the onset of the stressor(s). Marked distress, i.e., out of proportion to the severity or intensity of the stressor and/or significant impairment in functioning. Low mood, tearfulness, or feelings of hopelessness are predominant in depressed mood subtype.	Five or more of the following: (including depressed mood or anhedonia) Depressed mood Anhedonia Weight or appetite changes Sleep disturbance Psychomotor changes Fatigue/low energy Worthlessness or guilt Impaired concentration Recurrent thoughts of death/SI Clinically significant distress or functional impairment. Symptoms not due to medical/substance cause or other psychiatric disorder.	Disturbance in attention and awareness. Additional disturbance in cognition. Develops over a short period of time. Change from baseline. Tends to fluctuate. Evidence of direct physiological cause.	Emotions and behaviors result from a loss. May experience somatic symptoms. May have passive wish to hasten death.	Helplessness, hopelessness, meaningless, subjective feelings of lack of competence and sense of failure, inability to cope and problem-solve. Closely related to loss of autonomy.
Distinguishing features	Not enough to meet the criteria for major depression or a full anxiety disorder but can still reduce QOL. Does not represent normal bereavement. Distress focused on real issues related to the illness.	Symptoms tend to be constant, unremitting. Anhedonia, hopelessness/despair, helplessness, worthlessness, guilt, social withdrawal, and SI may help to distinguish from other etiologies. May also present with intractable somatic symptoms, poor cooperation with care, and disproportionate disability. Attentional/cognitive disturbances less likely. Treatment usually necessary.	Prominent disturbances in attention, cognition, and awareness/arousal. Often have perceptual disturbances. Can be hyperactive or hypoactive. Hypoactive subtype most often confused withdepression.	Often future-oriented and able to experience pleasure. Less likely to see sustained functional impairment. May be able to cope without treatment but may also benefit from psychological therapies and support.	Feelings of isolation and entrapment. Existential distress usually present. Apprehension and panic-like symptoms may develop. Somatization may occur. Hedonic capacity is preserved. Responds to supportive psychotherapy.
Onset/timing	Onset within 3 mo of a stressful event. Can become chronic but resolves within 6 mo once the stressor has ended.	Two-week period; often has a gradual onset. Change from functioning.	Develops over short period (hours to days). Often resolves with treatment of precipitating etiology but may represent terminal delirium at the EOL.	May come in waves and increase as the disease progresses.	Usually develops after a markedly stressful life event, such as physical illness. May abate if stressor is resolved.
Risk factors	Higher prevalence in women, patients with a metastatic disease.	Higher prevalence in those with an advanced disease or pain.	Multiple predisposing and precipitating factors including medications, poor nutritional status, medical diagnoses, older age.	Associated with disease progression.	Acute debilitating illness, chronic medical conditions, terminal diagnoses. Poor family and social support. Decreased QOL.
Epidemiology	Most often reported psychiatric morbidity in cancer patients.	Occurs in many terminally ill patients, but not inevitably.	Very high prevalence at EOL.	Common in terminally ill patients and their families.	Highly prevalent in medical settings, very common in oncological patients. May represent a prodromal state to MDD and suicidality.

^*Adapted from DSM-5.

DSM-5 = Diagnostic and Statistical Manual of Mental Disorders, Fifth Version; EOL = end of life; MDD = major depressive disorder; QOL = quality of life; SI = suicidal ideation.

Depression or other forms of psychological distress in patients approaching EOL can impair QOL, amplify pain, interfere with the grief process, and increase the risk of suicide, and thus, accurate identification and treatment are critically important.² This patient had several of the neurovegetative symptoms of depression, including poor appetite, sleep disturbance, and fatigue. However, these are also typical symptoms of advanced cancer, chemotherapy, or radiation treatment. Additionally, adverse effects of other medications used in cancer treatment such as opioids, glucocorticoids, and dopamine-blocking antiemetics can mimic mood or anxiety disorders. One systematic review found that 5 symptoms (fatigue, pain, lack of energy, weakness, and appetite loss) were reported in more than 50% of patients with an incurable cancer.³ The Hospital Anxiety and Depression Scale in particular has been designed to eliminate the influence of somatic symptoms in screening for depression and has been validated as a screening tool for psychiatric disorders in the cancer and palliative care setting.^4–6 Psychological symptoms such as anhedonia, hopelessness/despair, helplessness, worthlessness, guilt, social withdrawal, and suicidal ideation may help to distinguish a major depressive episode from other somatic etiologies, especially in terminally ill patients. The American College of Physicians-American Society of Internal Medicine Endof-Life Care Consensus Panel suggests that intractable pain/somatic symptoms, disproportionate disability, and poor cooperation with care may be additional indicators of depression in terminally ill patients.²

It is important to note that depressive symptoms are not an inevitable or “normal” response to the dying process, nor is the desire for hastened death. It is estimated that 15%–50% of individuals with cancer experience depressive symptoms and just 5%–20% meet the diagnostic criteria for a major depressive disorder.^7,8 Accurate diagnosis and treatment of depression may help alleviate suffering and improve QOL in terminally ill patients, so the diagnosis should not be missed.

Delirium, often of the hypoactive subtype, is common in individuals with a terminal illness, occurring in up to 88% of patients during the last few weeks of life.⁹ The Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, defines delirium as a syndrome with symptoms that include disturbance of consciousness, attention, and cognition.¹⁰ The onset is abrupt, and the course fluctuates. The disturbance must be due to a medical cause. Hypoactive delirium in particular is often misdiagnosed as depression, due to overlapping symptoms such as psychomotor abnormalities and sleep and energy disturbances. However, attention, cognitive disturbances, and perceptual disturbances are less common in depression, and the onset is more gradual.¹¹ Although our patient did not display symptoms of delirium on initial evaluation, it is important to continuously evaluate terminally ill patients for potential medical contributors to neuropsychiatric symptoms.

In addition to the conditions discussed above, our patient was likely experiencing grief or an adjustment disorder as he faced death and coped with the loss of QOL and autonomy. His high somatic symptom burden also undoubtedly contributed to his distress and his perception of suffering. Assessing and treating the symptom burden, especially pain, is an important aspect of managing the psychological distress. In patients with cancer, uncontrolled pain is both a common symptom as well as a major risk factor for depression and suicide.

Lastly, demoralization is a syndrome that needs to be considered in chronically ill patients. Loss of independence (which was one of the main problems verbalized by this patient) can lead to lack of hope, powerlessness, self-esteem issues, and a decreased sense of meaning.¹² At times called as the “giving up-given up” complex, it is fairly common in medical settings and can mimic neurovegetative symptoms, and it is distinguished from major depression due to the absence of anhedonia. Demoralization responds to psychotherapy, but if left untreated, it can progress to clinical depression.¹³

In addition to establishing the most likely diagnosis, the consultation-liaison (C-L) psychiatrist was faced in this case with the difficult question of whether the patient had true suicidality (i.e., specific thought of ending his life) and presented an imminent safety concern or whether the patient’s statement was an expression of psychological distress or more of a passive desire for hastened death. Our inability to answer this question was one of the factors that prevented the patient from being discharged to hospice or home hospice.

In the case of patients with cancer, it is known that the suicide rate is at least twice that of the general population, with particularly increased risk in the advanced stages of the illness.^14,15 Other risk factors for suicide include a preexisting psychiatric illness, male sex, older age, depression, hopelessness, isolation, unemployment, demoralization, delirium, pain, debility, a sense of being a burden to others, and a family or personal history of suicide/suicide attempts.^2,7,16 The patient presented had a number of these risk factors. While suicidal thoughts are often temporary and associated with psychological factors such as loss of control, a wish to be free of suffering, anxiety about future pain, and an acceptance of death, a smaller percentage of terminally ill cancer patients do report a sustained and serious desire for death. The desire for death in terminally ill cancer patients has been found to be closely associated with clinical depression, pain, and low family support. At least 2 of these factors are potentially modifiable.¹⁷ In the case of our patient, a protective factor against suicide was his family. They were extremely supportive and actively involved during the whole process. His relatives were concerned about his statements and tried to work as mediators when he became uncooperative with the psychiatrists, but they were also respectful of his wishes to withhold treatments as they thought they aligned with his value system.

It is important to note that the expression of a desire for hastened death generally does not represent suicidal ideation and may in fact signify a request for assistance in living with less distress. Regardless of the actual intent, a thorough clinical assessment is warranted, and the clinician has the responsibility to explore, clarify, and interpret such requests. Block and Billings reviewed the clinical approach to further exploration and emphasized consideration of the following: inadequately controlled physical symptoms; difficulties in relationships with loved ones; psychiatric disorders such as depression, anxiety, delirium, dementia, personality, and substance use disorders; and the presence of psychological factors and personal values such as the need for control or the fear of loss dignity.¹⁸ Character disorders and certain personality traits should also be explored as they can correlate with depression and worse QOL.¹⁹ An approach to the assessment of suicidal thoughts is outlined in Table 2. Further helpful lessons about assessing death-hastening requests may be learned from the experience with patients requesting dialysis discontinuation. Of particular relevance, Cohen and colleagues found that most patients referred for requests to terminate life support were neither depressed nor suicidal. The decisions to hasten death were generally from patients with a lifelong strong need for independence and control and prompted more acutely by dissatisfaction with a deteriorating illness course and the wish for a dignified EOL.²⁰

TABLE 2.

Assessing Suicidal Thoughts in Terminally Ill Patients

Explore desire to die: Why now, what are the meanings behind the desire to die?

Assess pain and symptom control: undertreated symptoms, side effects or fears about the dying process.

Review social supports: recent loss or rejection, worsening interpersonal relationships, new financial burden, anyone with the knowledge of the patient’s plan, attitudes of family and health care providers about the plan.

Assess cognitive status: cognitive deficits, new neurological symptoms, understanding of medical condition and prognosis and implications of suicide.

Assess for psychiatric comorbidity: untreated mood or anxiety disorder, delirium, substance use disorder, demoralization; if depression is present, is it interfering with the patient’s decision-making?

Assess for psychological factors and coping mechanisms: coping with loss of control, dependency, neuroticism, inflexibility, suspiciousness, avoidance, insecure attachment style, help-rejecting attitude, grief, uncertainty, loss of dignity, ambivalence.

Explore religious, spiritual, and existential concerns: unresolved concerns or questions.

Assess for suicide risk factors: age, sex, personal or family history, pain, other psychiatric diagnoses, medical comorbidities, social isolation, financial stressors, unemployment.

Assess for protective factors: hope, supports, future-oriented, reasons for living, children, religious beliefs.

Assess strength of desire for death: range from no or transient thoughts to constant thoughts of death.

Assess for methods, means, and intent: specific plan, access to weapons/means, preparations, likelihood of acting on plan.

Assess for room for negotiation: Would a delay of the decision be acceptable and would the patient consider an antidepressant medication trial or other attempts at symptom management?

Adapted from Block SD: Assessing and managing depression in the terminally ill patient; Kolva E, Hoffecker L, Cox-Martin E: Suicidal ideation in patients with cancer: A systematic review of prevalence, risk factors, intervention and assessment; McFarland DC, Morita J, Alici Y: Personality Disorders in Patients with Cancer.

MANAGEMENT STRATEGIES (DR. DANIEL SHALEV)

The management of psychiatric comorbidities in individuals at the EOL may entail unique considerations, even once a thorough evaluation is made and diagnosis achieved as described above. In particular, there may be specific dispositional issues (as described below) and pharmacologic considerations (as described in the subsequent section) that need to be considered.

In this case of a 69-year-old man with a metastatic prostate cancer and heart failure approaching the EOL, the consulting psychiatric team faced the dilemma of balancing the patient’s suicidal statements with his desire to die at home. Psychiatric management of patients at the very EOL (prognosis in the range of weeks to hours) is challenging. Many interventions deployed by C-L psychiatrists may be incongruent with the timeline (e.g., antidepressant medications) or structural needs (e.g., inpatient psychiatric hospitalization) of patients who are approaching death. Furthermore, even psychiatrists with subspecialty training in geriatrics or C-L psychiatry may have limited exposure to the common structures of EOL care such as hospice.^21,22 Gaps in knowledge of such structures can make it difficult to effectively problem-solve around the where, how, and who of managing high-acuity psychiatric needs in patients at EOL.

In the United States, only a minority (~ 30%) of patients with life-limiting illnesses like cancer, heart failure, or chronic pulmonary disease die in the hospital.²³ Through models of home-based care including home hospice, many patients in the United States are able to die at home (30%), the preferred place of death for the majority of people.²⁴ Sizable minorities of patients also die in nursing facilities (21%) or specialized hospice facilities (8%). Among patients who die in the hospital setting, C-L psychiatrists can provide psychotherapeutic and psychopharmacologic interventions and collaborate to ensure safety in patients with emergent psychiatric needs. Unfortunately, mental health integration in nursing home settings is often limited²⁵ and may be entirely nonexistent in home-based care models including home hospice.²⁶ As such, management of urgent psychiatric issues in non–hospital-based EOL care is not operationalized and may be suboptimal.^27,28

Would (Home) Hospice Be Appropriate for the Patient in This Case?

Hospice is a model of care for individuals with a prognosis of 6 months of less who have decided to discontinue curative or disease-modifying interventions for their life-limiting diagnosis. Hospice is one of the most common models of EOL care in the United States; approximately half of Medicare decedents are enrolled in hospice at the time of death.²⁹ There are a number of unique aspects of the hospice payment and care delivery structure that optimize it to providing care for individuals in the last phase of life including a capitated payment model, utilization of community members as volunteers, and interdisciplinary team-based care. Hospice can be provided across settings including hospital settings (as in scatter-bed programs or hospice inpatient units), nursing home settings, or at home. However, most hospice care is provided in the home.

Understanding the services provided by home hospice care can be helpful to C-L psychiatrists navigating EOL psychiatric care. Home hospice care provides medications, durable medical equipment, interdisciplinary team visits (generally including nursing/medical visits, chaplaincy, and social work), access to a telehealth clinician for urgent symptom-management questions, and bereavement services for caregivers. Many referring clinicians have a limited understanding of the immense task informal caregivers undertake when providing home hospice care. While hospice may provide limited home health aide hours (,20 hours/weekly), care is generally contingent on family caregivers who are able to provide around-the-clock care to the patient including administering medications and cleaning, repositioning, and managing medical equipment such as feeding or drainage tubes. As such, the success of the home hospice model depends on caregiver(s) who are physically, emotionally, intellectually, and logistically capable of providing constant care through the dying process.

It is unusual that psychiatrists participate in home hospice care. Generally, hospice programs do not have dedicated mental health clinicians and are unlikely to have the ability to manage high-acuity psychiatric needs like suicidality.^26,30 Financially, insurers will not concurrently pay for hospice and disease-oriented treatment. However, patients can still obtain medical services for conditions unrelated to their terminal diagnosis; as such, patients can theoretically still access mental health services alongside their hospice care if they have access to such services. While there are no data on such concurrent care, anecdotally such arrangements are vanishingly rare. Rather, hospice nurses, physicians, social workers, and informal caregivers of home hospice patients often shoulder the burden of managing psychiatric needs at the EOL.

C-L psychiatrists can play a major role in evaluating and optimizing patients for home hospice care. Understanding the natural history of an individual patient’s psychiatric needs and likely trajectory can help C-L psychiatrists evaluate whether a given patient can obtain appropriate care in the home hospice setting. Some psychiatric issues, like anxiety, insomnia, depression, or hypoactive delirium may be manageable or even improved by home disposition. Others, like dangerous aggressive behaviors or active suicidality, may not lend themselves to safe management at home by an informal caregiver. C-L psychiatrists can also implement optimized treatment plans for symptoms that can be leveraged by hospice teams. For instance, titrating an appropriate benzodiazepine for anxiety or a stimulant for cancer-related fatigue prior to a patient going home with hospice can allow for more rapid symptom management than would be feasible in the home setting. Finally, C-L psychiatrists can collaborate with palliative care and medical clinicians to investigate the logistical and psychological factors that may influence the success of a home hospice disposition: Do the patient and family understand the role of hospice? Are family members freely available and willing to provide EOL care, or are they giving affirmative responses out of a sense of duty and guilt? Curious and careful inquiry can help transitions to home hospice proceed smoothly and allow patients and families to experience the final phase of life with dignity, comfort, and autonomy.

Returning to the case described here, the psychiatry team faced the challenge of a patient who clearly wished to go home to die and who may have been appropriate for home hospice except for repeatedly stated suicidal intent that he refused to clarify to the psychiatry team. While there are no operationalized guidelines about psychiatric contraindications to hospice care, I generally avoid referring patients who report suicidality to home hospice care. The burden on informal caregivers and hospice staff of ensuring safety and providing psychiatric stabilization is an immense one. While patients enrolling in hospice are at the end of their lives, a completed suicide is nonetheless a catastrophic outcome with impacts on families, communities, and clinicians that cannot be understated. As a note, interest in hastened death is often a normative question by patients and caregivers at EOL, and I do not equate questions or request for hastened death with the actual intent to harm oneself. However, because the patient in this case repeatedly articulated his intent to harm himself in his home, his ongoing care in the hospital was an appropriate-if-unfortunate means by which to deliver high-quality EOL care.

TREATMENT OPTIONS (DR. PHILIP A. BIALER)

The details of the case, possible diagnoses, and management strategies have been discussed. Thus, it would be appropriate to explore the current available treatments for patients approaching the EOL.

Symptom management is the key. Ensuring maximum pain relief without oversedating the patient to allow them to continue to interact with their family is crucial.

While the differential diagnosis of the psychiatry evaluation includes a primary mood disorder such as major depressive disorder, the details of the case do not necessarily support this. One might consider a trial of an antidepressant, although given the patient’s lack of engagement with the psychiatry team, it is doubtful that he would agree to take one. Also, adding an antidepressant is unlikely to have any benefit given the limited time of survival for this patient. Delirium was also included in the differential diagnosis, so a low-dose of a neuroleptic such as olanzapine might have provided some relief.

The patient also expresses thoughts about suicide although what he may actually be asking for is help with hastening his death. Some therapies such as ketamine³¹ or esketamine³² have been shown to quickly reverse thoughts of suicide. Ketamine has been studied in palliative care settings for management of pain, depression, and suicidality,³³ and esketamine has been studied in depressed patients at the EOL in its subcutaneous form with fair results.³⁴ Nevertheless, these treatments would not be appropriate for the patient from this case. Both medications have been associated with hallucinations and dissociation,^33,35 which could preclude meaningful interactions between terminally ill patients and their families. Ketamine is also associated with respiratory depression and sedation.³⁵ The patient described in our case was already on supplemental oxygen due to respiratory distress, thereby making ketamine a potentially dangerous treatment option. Additionally, the patient was not sufficiently participatory to undergo ketamine treatment. Some structural challenges were in place as well, including the need for possible transfer to a highly monitored setting to receive ketamine given his frail and complex condition.

Given their serotoninergic properties, hallucinogens have been explored as new treatment options for depression and anxiety.³⁶ One study using psilocybin in a controlled outpatient environment demonstrated its effectiveness of alleviating depressive symptoms in patients with end-stage cancer, but again this would not be appropriate for this patient, as psilocybin is a schedule I controlled substance and is not available for clinical settings up to date.³⁶

Regarding nonpharmacological treatments, a number of psychotherapeutic modalities have been developed specifically for individuals at the EOL.³⁷ The focus of such modalities is existential distress amelioration, but they are not evidence-based interventions for suicidality. One type of psychotherapy that has been implemented in palliative care settings is dignity therapy (DT). This brief intervention aims to decrease psychological distress in EOL patients.³⁸ The protocol includes a questionnaire that collects information about the person’s life story, including the roles they have played, the lessons they have learned, their accomplishments, any unsaid things they would like to tell their loved ones before death, and even instructions for their families after they are gone. The information is transcribed into a “legacy document” that is given to the patient at the end of the intervention so it can be passed to their relatives. DT is a simple therapeutic approach that fosters hope, meaning, and dignity of life in patients with advanced diseases.³⁸ Evidence suggests that DT improves the EOL experience for the patients, and their relatives and clinicians also find DT helpful for navigating the EOL process.

A similar intervention is meaning-centered psychotherapy (MCP), which has demonstrated to improve spiritual well-being and to diminish psychological distress among patients with an advanced cancer.³⁹ Individual MCP is a manual-driven treatment usually comprised of 7 outpatient sessions (Table 3) with a trained clinician.⁴⁰ Noteworthy, the patient described in this case expressed a desire for hastened death. Individual MCP has only a modest effect in this variable.³³ However, group MCP does appear to reduce the desire for death hastening in oncological patients.⁴¹ In addition, some of the principles of MCP can be adapted into 1 or 2 sessions at the bedside for patients who are imminently dying, and a pilot study of MCP in a palliative care setting rendered preliminary positive results.³⁷ This intervention would depend on the patient’s ability to engage in a conversation about meaning. Were there particular meaningful experiences in their life that they want to talk about and share with the clinician? Sharing these meaningful experiences may also allow them to share a meaningful moment with their family and to establish their legacy, which can be very comforting to all as death approaches.

TABLE 3.

Overview of Meaning-Centered Psychotherapy

Session number	Key elements
One: concepts and sources of meaning	Definitions of meaning and purpose, both in general and in relation to a cancer diagnosis. Introduction to the sources of meaning. Become familiar with the patient’s cancer story.
Two: cancer and meaning	Learn how to sustain and consolidate the sense of meaning while battling cancer. Discuss the patient’s identity before and after the cancer diagnosis.
Three: historical sources of meaning	Introduction of the concept of “life as a living legacy”. Discussion of family genealogy, upbringing, traditions, culture, life lessons, and accomplishments. Explore patient’s current meaningful roles and their expectations about what their contribution will be to future generations.
Four: attitudinal sources of meaning	Inquire about prior attitudes toward suffering or life difficulties. Finding meaning through one’s attitude toward suffering. Explore the concept of a meaningful death.
Five: creative sources of meaning	Association between creativity, courage, and responsibility. Introduce the concept of creativity as a mean for transcendence. Creativity in one’s life through work, hobbies, and activism.
Six: experiential sources of meaning	Explore the role of love, humor, beauty, art, and nature as valuable coping tools.
Seven: transitions	Summary of the sources of meaning. Explore patient’s final thoughts about the lessons learned during these sessions and their hopes for the future.

CONCLUSIONS

The terminally ill patient can suffer from a variety of psychiatric conditions such as depression or delirium, but when suicide ideation is present, it is not always due to a major psychiatric illness. When evaluating these patients, it is important to differentiate true suicidal intention from a desire to hasten death, which can be secondary to demoralization, loss of autonomy, frustration, or poor QOL. Standard treatment options such as long-term pharmacotherapy or inpatient psychiatric admission are not feasible in these cases given the time constraints. Instead, management of these cases should focus on symptom relief. Implementation of specific psychotherapeutic techniques like MCP or DT can also be helpful for existential purposes. The possibility of hospice care needs to be considered, and C-L psychiatrists should work closely with palliative care and families to ensure a safe disposition that aligns with the patient’s wishes.

It is important to mention that there is a gap in providing high-quality psychiatric care to individuals at the EOL. Many of the structures of assessment and care either do not apply to or are not well suited to individuals with short prognoses. For instance, there are no operationalized suicide-assessment tools adapted to the needs of individuals with serious medical illnesses. Furthermore, the ethical balance between autonomy, beneficence, and nonmaleficence that clinicians seek to achieve in the management of psychiatric morbidity (e.g., involuntary hospitalization) may be poorly tailored to individuals who are dying. Moreover, an intervention such as an involuntary admission to a psychiatric unit would not yield any tangible benefit to a terminally ill person and, in fact, could cause more harm, as it would prevent the patient from spending their remaining time with their loved ones. This case and others like it highlight the need for C-L psychiatrists to build a more robust base of clinical and research experience in psychiatry at the EOL.