Approach to symptom assessment in palliative care

INTRODUCTION — 

Palliative care is an interdisciplinary medical specialty that improves the quality of life of patients and their families facing the problems associated with life-threatening illness, through the prevention and relief of suffering by means of early identification, assessment, and treatment of physical, psychosocial, and spiritual symptoms [1]. Although advanced palliative care skills require subspecialty training to learn and apply, the core elements of palliative care, such as aligning treatment with a patient's goals and basic symptom management, are aspects of the care delivered by all clinicians [2]. Palliative care is not limited to end-of-life care and can be provided at the same time as curative or life-prolonging treatments (figure 1). (See "Benefits, services, and models of subspecialty palliative care".)

Patients in the advanced stage of a serious and/or life-threatening illness typically experience multiple symptoms; the most common include pain, depression, anxiety, confusion, fatigue, breathlessness, insomnia, nausea, constipation, diarrhea, and anorexia (table 1) [3-7]. These symptoms may result from many factors. For example, among patients with advanced cancer, symptoms may result from the disease (eg, bone pain from metastases, dyspnea from pleural effusions), treatment (eg, esophagitis with thoracic radiation therapy), other concurrent conditions (eg, osteoarthritis), or medications (eg, dry mouth, constipation).

This topic review provides a suggested approach to assessing the most common physical and psychological symptoms in patients receiving palliative care for an advanced serious and/or life-threatening illness. The principles underlying the comprehensive palliative care assessment (including domains other than physical and psychologic symptoms) and an overview of treatment for symptoms commonly encountered in a palliative care population are provided elsewhere. (See "Overview of comprehensive patient assessment in palliative care" and "Overview of managing common non-pain symptoms in palliative care".)

GENERAL PRINCIPLES — 

Patient descriptions of physical symptoms and their severity are the primary data for symptom assessment in palliative care. Exploring patients' reports of symptoms requires thoroughness, persistence, and patience, a fundamental aspect of patient-centered care.

Clinicians must assess mental status and stamina before eliciting information about the nature and severity of symptoms. Patients with severe fatigue may only be able to answer a few questions briefly, and more than one visit may be necessary. Patients who are confused or delirious may be unable to cooperate fully, making symptom identification and assessment difficult; failure to recognize delirium may also lead to inaccurate assessments and inappropriate therapies [8]. (See "Diagnosis of delirium and confusional states".)

Patients with advanced illnesses typically have multiple symptoms, and systematic assessment is preferable to relying on spontaneously described symptoms. For instance, a study of patients referred to a palliative medicine program found that there were 10 symptoms for every symptom that a patient volunteered [9]. Health care providers tend to underestimate symptom severity [10,11], leading to missed opportunities for diagnosis and symptom relief. A careful review of symptoms is required, focusing on common problems in palliative care. Comprehensive symptom assessment can significantly improve symptoms and quality of life [12,13].

Each identified symptom should be investigated regarding its Onset, Palliating and Provoking factors, Quality, Response to previous treatments, Related factors/symptoms, Severity, and Temporality (the OPQRST historical structure for evaluating symptoms). Equally but less emphasized is understanding the personal meaning of a symptom and how it affects their psychosocial wellbeing (the MOPQRST mnemonic) (table 2). Sample questions to explore the meaning of specific symptoms include:

●"Of the symptoms that have been bothering you, which symptom bothers you the most [14]?" This question will identify a primary symptom in one-third of patients who would otherwise be missed.

●"How do the symptoms affect you? How much do they interfere with your life (eg, sleep, daily activities, sense of wellbeing)?"

●"What ideas do you have about the meaning of these symptoms?"

●"Do these symptoms make you worry about your health/illness? What are your concerns?"

●"How are these symptoms affecting your family and friends?"

Discussing symptoms with the patient's caregivers may also be informative and provide opportunities to support and educate family and friends [15].

Following the initial assessment, a periodic symptom reassessment is valuable for identifying or modifying treatment goals, monitoring the response to specific symptomatic interventions, and communicating between clinical team members and caregivers.

Time may be the greatest barrier to performing a symptom assessment. In practice, clinicians may have a short mental list of five to six symptoms they routinely use. Experience and research have shown the list should be longer, and efforts to make the process more efficient and rigorous have led to the field of symptom science [16,17].

Many aspects discussed below help the clinician consider and follow up on a patient's answers during symptom assessment.

Dimensions of symptom assessment — Dimensions are different aspects of symptoms, such as severity, frequency, level of interference with activities, and level of distress associated with the symptom.

Symptom severity is a useful guide to treatment (eg, pain) and may also provide some prognostic information (eg, dyspnea). (See 'Performance status, symptoms, and prognosis' below.)

The use of rating instruments completed by patients represents a major change in symptom assessment over the last few decades. In evaluating symptoms, it is helpful to obtain the patient's rating of severity within a given time frame, such as the previous 24 hours. For instance, the approach to rating pain relies exclusively on how the patient reports pain using a verbal or numerical scale. Clinicians can ask the patient to make a mark on a 10 cm straight line (Visual Analog Scale [VAS]) (figure 2) or rate severity on a scale of 0 to 10, where 10 is the worst. These approaches are based on psychophysical studies in the field of subjective sensory physiology, demonstrating that the perception of a stimulus is closely related to the strength of the stimulus [18]. For many symptoms, however, there is no clear evidence of the optimal cut-off values to define symptoms’ severity [19]. (See 'Assessment and rating instruments for symptoms' below.)

Some patients cannot rate their symptom severity on a numerical scale [20]. An alternative approach is to ask patients to rate the associated distress of a symptom rather than its severity, providing categorical responses (eg, bothered not at all, a little bit, somewhat, quite a bit, very much). Patients may rate some symptoms, such as fatigue, as mild in severity but severe in distress. The amount and type of interference with daily activities caused by a symptom can also illustrate symptom severity and provide another means of determining therapeutic success.

To assess the effect of multiple symptoms, the term symptom burden can be used as a summary term [21], whereas “symptom clusters” refers to groupings of symptoms that often occur together . Symptom clusters refers to groupings of symptoms [22,23].

Symptom change — The patient's assessment of symptom relief informs the therapeutic plan. Defining the minimum clinically important differences in symptom instruments for pain, fatigue, and dyspnea has been an important advance in interpreting symptom ratings. It can be considered to be a 2-point change on an 11-point scale (eg, 0 to 10) (figure 2), or any change in categories of a Likert scale, such as the faces pain scale (form 1) [24-28].

Challenges in symptom assessment — Many situations raise particular challenges for symptom assessment in patients receiving palliative care:

●The patient's assessment of symptoms may differ from that of the clinician. In general, clinicians (and caregivers) have greater discordance with patients in assessing the severity of subjective symptoms (eg, pain, anxiety) than for objective states (eg, vomiting). Clinicians tend to rate the severity of symptoms lower than patients and family members and tolerate levels of symptom-related distress and suffering that patients find unacceptable [29-33]. By contrast, family members tend to overrate symptoms [34-36]. However, where pain is concerned, older family members tend to overrate while younger family members tend to underrate pain [34].

●Patients may downplay or not report symptoms (particularly pain) because of fears that worsening symptoms reflect disease progression and that the oncologist will stop treatment. Underreporting may also occur if a patient feels that a symptom such as pain is an inevitable consequence of the disease, its treatment, or dying. In addition, if patients do not receive attention to oft-voiced complaints, they may stop communicating about distress related to that specific complaint.

●Symptoms can change over time.

●Many symptoms are multifactorial in origin. A comprehensive assessment process will help define all the possible factors that may be causing a specific symptom. (See 'Approach to specific symptoms' below.)

●Patients and/or their caregivers may have suffered harm in the healthcare system and lost trust in the system and/or clinicians. This experience may make them particularly reluctant to discuss psychological symptoms. In these cases, the assessment interview with the patient and caregiver may itself have therapeutic value in enhancing the clinician’s relationship with the patient and caregivers, eliciting care preferences, and developing trust. (See "The initial interview in palliative care consultation".)

●Cultural factors may also determine which symptoms are acceptable for discussion between patients and clinicians and how patients perceive their symptoms. For example, women from certain cultures may not be comfortable sharing gynecologic complaints with a male clinician [37,38].

However, at least one study suggests that for most symptoms, culture may have a limited impact on how patient-reported symptoms are ultimately interpreted. In this study, over 1400 patients from five countries (United States, China, Japan, Russia, and Korea) were evaluated for the severity and interference of symptoms related to their cancer and/or treatment using psychometrically validated language versions of the MD Anderson Symptom Inventory [39]. The authors found only limited variations in symptom reporting due to cultural or linguistic variations. The exception to this was with the reporting of nausea and vomiting.

●Language can also pose barriers. In Europe, the assessment of fatigue may be challenging because the word itself is only found in English and French [40].

●Patients may have difficulty communicating. Examples include patients who may be intubated and receiving mechanical ventilation in the intensive care unit, patients who are withdrawn or who have cognitive impairment, and those who are too ill to voice concerns. As patients near death, symptom assessment becomes more difficult because most become noncommunicative, and other symptoms, such as restlessness and confusion, may interfere with effective communication. The study of behavioral changes as clues to the presence of pain (eg, grimacing) is an area of ongoing research, forming the basis for observer-based rating instruments such as the Pain Assessment in Advanced Dementia [PAINAD] tool (table 3). A similar tool (the Respiratory Distress Observation Scale) has been developed for assessing dyspnea (table 4) [41,42]. (See 'Patients unable to self-report' below.)

Clinicians may ask involved family members and caregivers to assess and rate symptoms (proxy ratings). However, disparities may exist between patient and caregiver symptom assessments or ratings [34,43,44]. For patients with impaired communication, the palliative care consultant may have to balance multiple assessments by family, caregivers, and clinicians. (See 'Patients unable to self-report' below.)

●Clinicians may not be aware of the high prevalence of symptoms in these patients, not routinely assess symptoms in their conversations with patients, and lack knowledge about symptom identification and management. The average number of symptoms ranges from 11 to 18 in various illnesses [45-47].

●Patients may be very demoralized, tired of being sick, and feeling helpless and hopeless [48], and they simply stop reporting symptoms. (See 'Demoralization' below.)

ASSESSMENT AND RATING INSTRUMENTS FOR SYMPTOMS — 

Formal multiple symptom assessment tools, especially those that are symptom inventories, like the revised Edmonton Symptom Assessment Scale (table 5), provide a good overview of symptoms in individual patients [49]. In particular, multiple symptom assessment tools are highly effective in measuring unreported symptoms, when combined with further patient interviewing to delineate the details of positive responses. When a specific tool has been chosen, it should be used consistently to ensure reliability in the clinical setting.

Many of these symptom assessment tools have been tested for reliability and validity. However, like any other measurement of health outcomes, there are potential errors in assessment [50]. The possibilities for assessment errors include:

●Motor, hearing, and visual impairments may impede the usefulness of symptom assessment tools in older adults [51].

●Cognitive dysfunction may lead to misleading responses and inaccurate assessment, even when paired with behavioral indicators [8]. The cognitive dysfunction may be quite subtle. (See 'Patients unable to self-report' below.)

●Shorter tools may miss important symptoms, and longer tools may be too tiring for patients to complete. (See 'Challenges in symptom assessment' above.)

●As the patient's condition changes, symptom ratings may change as they accommodate to the new condition, even if the symptom has not changed. This is referred to as the phenomenon of response shift. Ratings may also be influenced by other symptoms, especially anxiety and other emotional reactions, which may change over time. (See 'Symptom change' above.)

●Interpreting the "anchors" (0 and 10 on the numerical 0 to 10 scale) may vary between patients, depending on their experience. Caregivers and clinicians may also interpret the anchors differently. It may be necessary to repeat the definitions of the anchors when asking patients to give a rating.

●For numerical scales, the "cut point" is a number that identifies how to categorize severity. Studies have identified cut points of 4 out of 10 (mild versus moderate) and 7 out of 10 (moderate versus severe) for pain [52]; work is ongoing to determine optimal cut points for other symptoms [19,53].

Despite these limitations, many validated multiple symptom assessment tools are in wide usage in palliative and end-of-life care settings, including:

●The revised Edmonton Symptom Assessment Scale (table 5) [49]

●Memorial Symptom Assessment Scale-Short (MSAS) Form (figure 3) and Condensed MSAS [54-56]

●MD Anderson Brief Symptom Inventory [57,58]

●Rotterdam Symptom Checklist [59]

●Symptom Distress Scale [60]

●Patient-Reported Outcomes Measurement Information System (PROMIS), as developed by the United States National Institutes of Health (see "Evaluation of health-related quality of life (HRQL) in patients with a serious life-threatening illness", section on 'Combined instruments for patient-reported outcomes')

●The Your Symptoms Matter symptoms assessment tool, developed and hosted by Cancer Care Ontario

●The National Comprehensive Cancer Network (NCCN) Distress Thermometer [61]

Instruments are available that focus on one symptom, such as dyspnea or pain. One example is the Brief Pain Inventory (BPI), a patient self-rating scale that assesses pain intensity and interference in various areas of function [62,63]; a short-form version (BPI-SF) is also available online. (See 'History' below.)

However, detailed instruments such as these are not available for all symptoms, and patients may be unable to answer multiple questions from many instruments. Many instruments are now available in languages other than English, which may help assess symptoms in countries where English is not the primary language. The non-profit organization, Mapi Research Trust, created a database of clinical outcome assessments. The website may provide greater access to assessment tools, including access to tools available in multiple languages.

PERFORMANCE STATUS, SYMPTOMS, AND PROGNOSIS — 

Performance status is a critical aspect of patient assessment in palliative care. As a general rule, performance status, as assessed by the Karnofsky [64] or Eastern Cooperative Oncology Group (ECOG) [65] performance status scales (table 6 and table 7) is a key indicator of prognosis in individuals with advanced terminal disease [66] and cancer, respectively. The Palliative Performance Scale (PPS) is gaining wide acceptance as a tool for assessing functional levels and estimating prognosis in patients with advanced cancer [67,68]. The PPS [68,69] may be a better indicator of functional status than other types of performance status scales, including Karnofsky and ECOG, in this population. Comparisons suggest there is considerable overlap [70]. (See "Survival estimates in advanced terminal cancer", section on 'Prognosis in patients receiving palliative care only' and "Survival estimates in advanced terminal cancer", section on 'Performance status'.)

The relationship between symptoms, performance status, and prognosis is less well recognized. For example, symptom number and intensity correlate significantly with performance status and prognosis [71,72]. As a terminal disease progresses, there is often a continual decline in function that is accompanied by an increasing number and severity of symptoms (especially fatigue and cachexia) [73]. (See "Assessment and management of anorexia and cachexia in palliative care".)

Other specific clinical signs and symptoms may represent independent prognostic factors in patients with advanced serious or life-threatening disease. As an example, in a systematic review of survival prediction in terminal cancer, the strongest evidence after performance status for an independent association with death was for dyspnea, dysphagia, weight loss, xerostomia, anorexia, and cognitive impairment (delirium) [74]. The following table shows the range of median survivals in patients with various signs and symptoms in univariate analyses from this and other studies (table 8). These findings suggest that for patients with advanced life-threatening diseases, such as those referred to palliative care programs, the presence of these symptoms may help clinicians estimate patient survival.

Dyspnea is a particularly important indicator of limited survival across many diseases, but especially in patients with cancer who are terminally ill [4,9,54-56,75,76]. In one study, dyspnea was associated with a median survival of less than 30 days [76]. The impact of dyspnea and other symptoms on prognostic estimates in advanced cancer and assessment of dyspnea in patients receiving palliative care are discussed in more detail elsewhere. (See "Survival estimates in advanced terminal cancer", section on 'Clinical signs and symptoms' and "Assessment and management of dyspnea in palliative care", section on 'Symptom assessment'.)

Among patients with dyspnea, other findings may help to predict the likelihood of imminent death. As an example, in a retrospective sample of 122 patients with cancer who presented to an emergency department with acute dyspnea, a respiratory rate greater than 28 breaths per minute, heart rate greater than 110 beats per minute, uncontrolled disease, and the presence of metastases were all predictive of death within two weeks; the relative risk for imminent death among patients with a respiratory rate above 28, compared with the overall group, was 12.7 (95% CI 3.1-52.8) [77].

Noisy breathing, also known as the "death rattle," refers to gurgling breath sounds and is usually a preterminal finding, with death likely in hours to days [78]. (See "Overview of managing common non-pain symptoms in palliative care", section on 'Last hours of life'.)

APPROACH TO SPECIFIC SYMPTOMS — 

The following is a brief discussion about assessing highly prevalent symptoms in patients receiving palliative care. More detailed discussions of some of these symptoms are available in related topic reviews.

Pain — The purpose of pain assessment is to obtain a rating of severity and diagnose the type of pain and its likely cause [79]. Identifying a specific pain syndrome in patients with cancer may inform best clinical practices. (See "Overview of cancer pain syndromes".)

The first step is to ask the patient about the presence of pain and its severity and to believe the answers. As noted above, a great divergence exists between how patients and clinicians evaluate pain severity. In a study of 103 patients with cancer, a comparison of Visual Analogue Scale (VAS) pain ratings by clinicians with those of patients showed lower ratings of severe pain by clinicians more than 70 percent of the time [29].

Yet, pain does not occur only in patients with cancer. It is also experienced by a large percentage of patients with end-stage heart failure [80,81], chronic obstructive pulmonary disease (COPD), and cirrhosis [82,83], and it is particularly associated with arthritis [84]. In fact, the prevalence and severity of pain are remarkably consistent across a variety of serious illnesses [85].

The following sections will briefly summarize the components of the pain evaluation. A more detailed discussion is available elsewhere. (See "Assessment of cancer pain".)

History — In eliciting a pain history, initial questions should include site, nature and severity, interference, and relief. In addition, patients may have had (untreated) chronic pain, such as that related to arthritis, that predates the advanced illness [84].

●Site – To avoid misunderstanding anatomical terms, have the patient point to the site of pain.

●Type of pain – Among patients with cancer, the specific type of pain may help to elucidate the specific pathophysiology. As examples (see "Assessment of cancer pain", section on 'Inferred pathophysiology and treatment implications'):

•Nociceptive pain is sustained predominantly by ongoing tissue injury. Somatic nociceptive pain involves injury to somatic structures like skin, bone, joints, or muscles. It is usually well localized and often described by patients as "aching," "stabbing," or "throbbing" in quality. Visceral nociceptive pain results from distension, injury to, or inflammation of visceral organs. It is usually poorly localized and is characterized as "gnawing" or "crampy" when arising from the obstruction of a hollow viscus (eg, the bowel lumen), or as "aching" or "stabbing" when arising from other somatically innervated visceral structures (eg, organ capsules, parietal pleura). Visceral pain can be referred to somatic structures. Further evaluation often yields an abnormal finding that can be identified on physical examination (such as localized tenderness to palpation) or radiographic scans to explain the symptoms.

•Neuropathic pain is sustained by abnormal somatosensory processing in the peripheral or central nervous system or direct nerve damage. Examples include postherpetic neuralgia, chemotherapy-induced peripheral neuropathy, and phantom pain. Patients may describe unpredictable shooting, burning, numbness, or pruritus. The sensory examination is often helpful if abnormalities, such as lack of sensation, increased sensitivity (hyperesthesia), or pain with nonpainful stimuli (allodynia) can be demonstrated.

Depending on the location and etiology, neuropathic pain may be present with nociceptive pain and may not have its signature features of sensory disturbance or shooting pain but may be gnawing or constant.

●Severity and quality of life impact – Pain severity guides medication choice and dosage. (See "Cancer pain management: General principles and risk management for patients receiving opioids" and "Cancer pain management with opioids: Optimizing analgesia".)

There are several simple ways to assess pain severity, including a faces pain scale (form 1), a VAS (figure 2), or a categorical Likert scale (none, a little, somewhat, quite a bit, very much).

Formal instruments have also been developed to rate pain severity in multiple dimensions. These include the Brief Pain Inventory (BPI; a short-form version of which [BPI-SF] is available online) [62,63], the McGill Pain Questionnaire [86] (a shortened version is available that has been validated in a variety of populations [87,88]), and the Memorial Pain Assessment Card [89]. Both the McGill Pain Questionnaire and the BPI have been translated into several languages. Although many of these tools were developed for patients with cancer, they are equally applicable to patients with other serious and/or life-threatening conditions. (See "Assessment of cancer pain", section on 'Intensity'.)

Another way to evaluate severity is by assessing the impact of pain on physical and psychological function and quality of life. Caregivers may be able to provide important information in this regard. Pain can interfere with quality of life by interfering with daily activities, walking, sleep, and mood. Severe pain may engender a sense of hopelessness and anxiety. Anxiety, in turn, may increase reactions to pain, while fear can decrease pain reactivity [90].

●Aggravating and alleviating factors – This includes response to prior treatment(s).

●Associated symptoms – This includes paresthesias, weakness, nausea.

●Temporal patterns – Temporal assessment of the pain may include information about onset, duration, and fluctuation. Determine both the extent and duration of relief provided by interventions.

Most patients with chronic cancer pain experience periodic flares, often referred to as "breakthrough pain." Breakthrough pain can be caused by inadequate dosing, caused by an unprovoked flare of the underlying pain syndrome, or provoked by a procedure or activity (incident pain). Especially for inpatients, procedural pain is an experience that is often overlooked. Breakthrough pain may result from such routine procedures as dressing changes for pressure ulcers or wounds, positioning for change of bed sheets, or movement in bed for a radiographic study [91]. (See "Assessment of cancer pain", section on 'Pain characteristics' and "Cancer pain management with opioids: Optimizing analgesia", section on 'Management of breakthrough pain'.)

The term "pseudoaddiction" describes a situation in which a patient's legitimate chronic pain condition is undertreated with pain medication, leading the patient to act in a way that resembles addictive behavior (requesting extra medications and demanding attention); such patients are often labeled as demonstrating "drug-seeking behaviors." The situation arises when a clinician is reluctant to provide enough of a controlled drug to provide adequate symptom relief for a patient who requires long-term treatment [92-94]. This response leads to a patient who is either chronically undertreated or intermittently put into withdrawal. The patient responds with a series of "conditioned behaviors" focused on the acquisition of the controlled drug that can appear like aberrant addictive-type behaviors. However, in pseudoaddiction, the behaviors disappear when an adequate amount of the drug is obtained. A careful pain history that emphasizes perceived pain relief may help clarify the issue.

However, it should be noted that the whole concept of pseudoaddiction has been questioned; empirical evidence supporting pseudoaddiction as a diagnosis distinct from opioid addiction has not emerged [95]. The use of the term is accepted for justifying opioid therapy for patients with nonterminal pain with behaviors consistent with addiction but lack a diagnosis of addiction. Psuedoaddiction is a controversial issue, particularly in the era of widespread opioid addiction and overdose. (See "Opioid use disorder: Epidemiology, clinical features, health consequences, screening, and assessment".)

●Patient and caregiver perspective – Patients and their caregivers should be asked about the meaning of the pain, their understanding of its cause, worries about relief, personal and cultural attitudes, concerns about opioid tolerance and addiction, diversion, and expense. Inquiring about personal experiences with pain earlier in the course of the disease or in friends and family members can dispel hidden anxieties and misunderstandings.

Special circumstances

Patients with a history of substance use disorder — Opioids are widely used for treating pain in patients with cancer because of their safety, multiple routes of administration, ease of titration, reliability, and effectiveness for all types of pain (ie, somatic, visceral, neuropathic, mixed). However, opioids also have the potential for misuse. The public health consequences of opioid misuse drive the imperative that all clinicians assume responsibility for risk mitigation and management when these drugs are prescribed for legitimate medical purposes. These issues are discussed elsewhere. (See "Cancer pain management: General principles and risk management for patients receiving opioids", section on 'Risk assessment and management for patients receiving opioids'.)

Assessing and managing pain is difficult in patients who have a history of substance use disorders. Patients with a history of substance use disorder who report pain can cause considerable apprehension for their clinicians because of fears about the potential for deception, addiction, diversion, and overdose [96].

Patients with a remote history of substance use with a stable support system are less likely to misuse opioids [97]. In our own experience, these patients appreciate a careful and objective pain evaluation. They are generally responsible and appropriate in their use of opioids as a component of their medical treatment. Further inquiry into the circumstances of past substance use is helpful. Sometimes they and their caregivers are even more concerned about addiction to pain medications than clinicians based on their own experiences.

Patients who are actively using "street drugs" at the time of initial evaluation should have a careful and objective pain evaluation. Consider the possibility that they are using illicit drugs to obtain pain relief. Even if they have an addiction, their pain is real and assessment must focus on treating their pain and addiction. The initial patient encounter is most important, as failure to establish a relationship will make it challenging to obtain an accurate pain history and render time-limited trials of medications difficult to interpret. These and other issues that arise when considering the need for acute pain management in patients chronically using opioids are addressed in detail separately. (See "Management of acute pain in the patient chronically using opioids for non-cancer pain".)

The risk management principles in these patients are outlined in the table (table 9). Psychologic evaluation may be helpful in evaluating underlying psychiatric or personality disorders. Assessment and recognition of substance use disorders and risk assessment in patients receiving opioids for management of cancer pain are all discussed in more detail elsewhere. (See "Substance use disorders: Clinical assessment" and "Cancer pain management: General principles and risk management for patients receiving opioids", section on 'Risk assessment and management for patients receiving opioids'.)

Patients unable to self-report — Assessment of pain in patients who are unable to self-report is a challenge in symptom assessment. The spectrum of patients who have impaired communication includes patients with permanent deficits, such as dementia, and those with more temporary cognitive impairments such as delirium, sedation, or trauma. Patients with the impaired ability to self-report pain present across the continuum of care, from intensive care units to clinics, home, and long-term care settings. Individuals who are unable to communicate their pain are at greater risk for underrecognition and undertreatment of pain.

Patients with mild dementia are often able to give a description and rating of their pain. Patients with advanced dementia may express pain by distress behaviors (eg, restlessness, grimacing, moaning, guarding, and changes in daily activity) but these behaviors are not specific for pain. When such behaviors occur, a physical examination should be performed with attention to the association of the pain behavior with movement, pressure, toileting, hunger, fear, loneliness, or visual impairment.

A hierarchical approach has been proposed for this set of patients, whereby multiple sources of information are integrated by the examiner: starting with patient report, reports from observers (proxy reporting of pain), observation of patient behavior, use of behavioral pain assessment tools, and search for potential causes of pain [98]. A number of instruments have been developed to systematically assess pain-related behaviors in nonverbal adults [99-103]. A review of the tools to assess pain in nonverbal adults is available [101]. Sufficient evidence has accumulated to support the use of the Pain Assessment in Advanced Dementia [PAINAD] tool (table 3) [102].

An empiric trial of analgesics may be warranted if, after the initial assessment, it remains uncertain that a distress behavior is indicative of pain in a patient with cognitive dysfunction.

Dyspnea and cough — Dyspnea is a particularly common symptom in patients who have advanced COPD, heart failure, primary lung cancer, or intrathoracic metastatic disease. However, for unclear reasons, it is also a common symptom in patients with no direct lung pathology [104]. As an example, a National Hospice Study found that 24 percent of patients at the end of life experienced dyspnea in the absence of known cardiopulmonary pathology [105]. (See "Assessment and management of dyspnea in palliative care", section on 'Prevalence'.)

As with pain, dyspnea is defined by patient self-report. Attention to the words that patients use in describing their breathing discomfort may provide insight into the underlying clinical condition and physiologic mechanisms producing dyspnea. (See "Assessment and management of dyspnea in palliative care", section on 'Symptom assessment'.)

Objective measures, such as respiratory rate, oxygen saturation, and arterial blood gas determination, may not correlate with nor provide a quantitative measure of the degree of dyspnea [106,107]. Physical examination, pulse oximetry, and chest radiograph do not clarify whether the patient has dyspnea, but they may help the clinician to identify likely causes (table 10). (See "Assessment and management of dyspnea in palliative care", section on 'Testing to determine underlying cause'.)

Many methods for assessing the severity of dyspnea are based upon functional assessment (ie, studies of tasks and effort) and quantify the amount of exertion required to cause dyspnea. Patients experiencing breathlessness inevitably reduce their activity level to accommodate dyspnea, so questions about limitations in performing specific activities can be used to assess the impact of dyspnea. As an example, the Oxygen Cost Diagram (figure 4) [108] asks patients to identify the level of activity they are unable to perform due to dyspnea, giving clinicians valuable information regarding the functional impact of the symptom. (See "Assessment and management of dyspnea in palliative care", section on 'Symptom assessment'.)

Patient-rated scales of dyspnea intensity have been based upon psychophysical descriptions of the relationship between work and perceived exertion. As an example, the modified Borg scale (table 11) measures the severity of dyspnea using a scale from 0 to 10, anchored by descriptive words; it may be used in conjunction with an exertional test [109]. The most common exertional tests involve having the patient walk for 6 or 12 minutes and then rate their dyspnea afterwards. A correlation between the 0 to 10 numerical scale and a four-level categorical verbal descriptor scale for dyspnea has been observed (0, none; 1 to 4, mild; 5 to 8, moderate; 9 to 10, severe) [110].

For patients with COPD, the frequency and severity of dyspnea can be assessed using validated questionnaires such as the Clinical COPD Questionnaire (table 12) or the Chronic Respiratory Questionnaire (CRQ), a shortened available version [111]. Although these approaches are useful for evaluating symptom severity in patients with COPD, neither is appropriate for patients with advanced serious life-threatening illness and severe dyspnea at rest. (See "Chronic obstructive pulmonary disease: Diagnosis and staging" and "Palliative care for adults with nonmalignant chronic lung disease", section on 'Symptom management in chronic lung disease'.)

None of these tools adequately measure the multidimensionality of dyspnea. A Cancer Dyspnea Scale has been developed in Japan for patients with lung cancer and translated into English [112]. Although it appears to reliably measure the sensation and psychologic components of dyspnea, further validation is needed in other palliative care populations.

The prognostic implications of dyspnea, particularly in patients with cancer, are discussed above. (See 'Performance status, symptoms, and prognosis' above.)

Cough is a frequent symptom in palliative care. Asthenia, muscle weakness, and the inability to coordinate an effective swallow near the end of life results in ineffective, persistent coughing in an attempt to clear accumulating airway secretions. Among patients with cancer, cough may be attributed to direct or indirect tumor effects (table 13). Evaluate all patients for other potentially treatable causes of cough. A more in-depth discussion assessing and managing cough and swallowing disorders in patients receiving palliative care is provided separately. (See "Palliative care: Overview of cough, stridor, and hemoptysis in adults" and "Swallowing disorders and aspiration in palliative care: Definition, pathophysiology, etiology, and consequences".)

Patients unable to self-report — A tool, the Respiratory Distress Observation Scale, has been developed for dyspnea assessment in patients who are unable to self-report (table 4) [41].

Fatigue — In the past, the terms "asthenia" and "weakness" were used to describe a subjective sensation of tiredness, while the specific term "fatigue" was used to describe a symptom of tiredness precipitated by effort. However, the terms are often used in the same context, and this broad sense of the term "fatigue" has gained widespread acceptance in the medical literature. (See "Overview of fatigue in palliative care", section on 'Definition'.)

Fatigue is a highly prevalent and poorly understood symptom in patients with advanced serious and/or life-threatening illness. Overall, it is reported by 80 percent of patients with cancer and by up to 99 percent of patients following radiation therapy or chemotherapy [40]. Fatigue is also prevalent in patients receiving palliative care with advanced life-threatening illness other than cancer, including those with HIV, multiple sclerosis, COPD, and heart failure. (See "Cancer-related fatigue: Prevalence, screening, and clinical assessment" and "Palliative care for patients with advanced heart failure: Indications and systems of care" and "Overview of fatigue in palliative care", section on 'Prevalence'.)

The assessment of fatigue in palliative care patients can be complex given its subjective and multidimensional nature. As such, there is a trend towards multidimensional assessment (ie, asking about different aspects of fatigue, such as muscular weakness of tiredness associated with a sleep disturbance) rather than asking a single question, "Are you tired?" (See "Overview of fatigue in palliative care", section on 'Clinical assessment'.)

Instruments have been developed assessing fatigue in patients with cancer, assessing both severity and functional impact (eg, the Brief Fatigue Inventory) (table 14) [113], and this subject is addressed in more detail separately. (See "Cancer-related fatigue: Prevalence, screening, and clinical assessment", section on 'Clinical assessment'.)

A list of assessment tools for fatigue in patients with cancer and other serious life-threatening illnesses is available in the table (table 15). Questions about fatigue are also present in symptom assessment instruments for patients with COPD and heart failure. Though, in such patients, it can be difficult to separate fatigue from dyspnea [114]. A combined dyspnea/fatigue instrument has been developed for patients with heart failure [115]. (See "Palliative care for patients with advanced heart failure: Indications and systems of care".)

Fatigue severity can also be quantified using a 0 to 10 visual analog scale, as provided in the revised Edmonton Symptom Assessment Scale (table 5). Ratings of fatigue severity are clinically meaningful in patients with cancer, focusing on functional interference [116]. (See "Cancer-related fatigue: Prevalence, screening, and clinical assessment", section on 'Clinical assessment' and "Overview of fatigue in palliative care", section on 'Clinical assessment'.)

The cause of fatigue in patients receiving palliative care is typically multifactorial. Distinguish acute fatigue, caused by acute illness such as infection or cardiac decompensation, from more chronic fatigue, which is more typically related to the underlying illness or its treatment (particularly for cancer). Complete a comprehensive history and physical examination to ascertain the various organ systems affected by the underlying disease and the impact of fatigue on activities of daily living and quality of life. Search for potentially reversible or treatable contributory factors to direct the diagnostic workup and review all prescribed and over-the-counter medications, including complementary/alternative therapies. (See "Overview of fatigue in palliative care", section on 'Etiology'.)

Assess factors that are often causally related to fatigue, including anemia, uncontrolled pain, emotional distress, sleep disturbance, weight loss and malnutrition, electrolyte disturbances (table 16), and the presence of other comorbidities (eg, infection, cardiopulmonary, renal, hepatic, neurologic, or endocrine dysfunction). (See "Cancer-related fatigue: Prevalence, screening, and clinical assessment", section on 'Main contributory factors' and "Cancer-related fatigue: Prevalence, screening, and clinical assessment", section on 'Assess for potentially remediable contributing factors'.)

Dry mouth — Xerostomia, the feeling of a dry mouth, is a highly prevalent symptom in palliative care. Risk factors include medications, mouth breathing, advanced age, and, in patients with cancer, a previous history of radiation to the head and neck. Several medical conditions such as Sjögren's syndrome, diabetes mellitus, and anxiety states can contribute. Further, medications and related anticholinergic side effects are a significant cause of xerostomia in palliative care; in one survey, patients enrolled in hospice were on a median number of four drugs that were associated with xerostomia [117]. (See "Clinical manifestations of Sjögren's disease: Exocrine gland disease", section on 'Dry mouth' and "Management of late complications of head and neck cancer and its treatment", section on 'Salivary gland damage and xerostomia' and "Normal aging", section on 'Oropharynx' and "Stopping nutrition and hydration at the end of life", section on 'Fluid deficits at the end of life'.)

Dehydration from reduced oral intake may be associated with a dry mouth. A dry mouth can also alter taste and make it difficult for patients to eat and swallow. Patients with xerostomia may also complain of mouth pain and difficulty with speaking [118].

On examination, dry lips, dry mucosa, dental caries, a coated tongue, and thick saliva help confirm the impression of xerostomia. Associated relevant findings include the presence of thrush, periodontal disease, and herpetic infection. Patients with erythematous candidiasis may have patchy erythema of the dorsal tongue and other mucosal surfaces and angular cheilitis.

While there are objective tests to confirm the presence of xerostomia (eg, measures of saliva production, such as the unstimulated and stimulated [after eating] whole salivary flow rates), these tests are used primarily for research.

Dysphagia — Dysphagia, which is sometimes associated with odynophagia and/or aspiration, occurs regularly in patients with advanced illness, especially in the setting of neurologic disorders (eg, amyotrophic lateral sclerosis [ALS] and other motor neuron diseases, brain metastases, and leptomeningeal carcinomatosis) and cancers of the head and neck, lung, and esophagus. (See "Swallowing disorders and aspiration in palliative care: Definition, pathophysiology, etiology, and consequences" and "Clinical manifestations, diagnosis, and staging of esophageal cancer", section on 'Clinical manifestations' and "Clinical features of amyotrophic lateral sclerosis and other forms of motor neuron disease", section on 'Bulbar dysfunction' and "Speech and swallowing rehabilitation of the patient with head and neck cancer", section on 'Aspiration'.)

In dysphagia, patients will report the sensation of a bolus getting "stuck" after swallowing often pointing to a specific site. Aspiration should be suspected when patients start coughing after drinking liquids with accompanying hoarseness.

Dysphagia may be classified as oropharyngeal or esophageal, and patients may have difficulty with one or both phases of the swallowing mechanism (see "Oropharyngeal dysphagia: Etiology and pathogenesis", section on 'Physiology of swallowing'):

●Oropharyngeal dysphagia, also called transfer dysphagia, arises from disorders that affect the function of the oropharynx, larynx, and upper esophageal sphincter. Neurogenic and myogenic disorders as well as oropharyngeal tumors are the most common underlying mechanisms for oropharyngeal dysphagia.

●Esophageal dysphagia arises within the body of the esophagus, the lower esophageal sphincter, or cardia, most commonly due to mechanical causes or a motility disturbance.

The most frequent causes of dysphagia in palliative care populations are (see "Swallowing disorders and aspiration in palliative care: Definition, pathophysiology, etiology, and consequences", section on 'Etiology of swallowing disorders in palliative care populations'):

●Neoplasms (especially head-and-neck and esophageal cancers) and posttreatment complications (especially xerostomia)

●Neurodegenerative disorders (Parkinson disease, ALS)

●Other neurologic disorders (eg, stroke, myopathy, multiple sclerosis)

●Chronic illness, general deconditioning, and frailty

●Medical conditions such as autoimmune/inflammatory diseases (eg, polymyositis, systemic sclerosis) or candida esophagitis

●Medications

A separate but related problem, pain upon swallowing (odynophagia), can result from inflammation, infection, neoplasia, or the effects of cancer treatment. (See "Deep neck space infections in adults" and "Pill esophagitis" and "Esophageal candidiasis in adults" and "Herpes simplex virus infection of the esophagus" and "Evaluation of the patient with HIV, odynophagia, and dysphagia", section on 'Etiology' and "Oral toxicity associated with systemic anticancer therapy" and "Management and prevention of complications during initial treatment of head and neck cancer", section on 'Mucositis' and "Clinical manifestations, diagnosis, and staging of esophageal cancer", section on 'Clinical manifestations'.)

Approaches to the differential diagnosis of dysphagia and suggested workup for patients with esophageal dysphagia and oropharyngeal dysphagia are presented in the algorithms (algorithm 1 and algorithm 2 and algorithm 3). (See "Approach to the evaluation of dysphagia in adults", section on 'Symptom-based differential diagnosis' and "Oropharyngeal dysphagia: Clinical features, diagnosis, and management", section on 'Determining the etiology' and "Approach to the evaluation of dysphagia in adults", section on 'Approach to diagnostic testing'.)

Aspiration — A disordered swallowing mechanism predisposes to aspiration. Other predisposing factors include reduced consciousness, which compromises the cough reflex; mechanical disruption of the upper airway or gastrointestinal tract; and patients in a recumbent position receiving large volume tube feedings and feeding gastrostomy. Ask all patients with these conditions about cough, and observe after swallowing. A history of coughing while eating or drinking likely indicates aspiration; however, aspiration may also be clinically silent in 20-30 percent of patients depending on the patient population and underlying conditions [119]. (See "Aspiration pneumonia in adults" and "Swallowing disorders and aspiration in palliative care: Definition, pathophysiology, etiology, and consequences".)

When aspiration is suspected, a bedside assessment includes observing the patient after swallowing a sample of thin and thick liquids and solids. In addition to cough, other signs occur at different intervals: multiple attempts to swallow a bolus at initiation; head or neck repositioning while swallowing; drooling, wet or hoarse vocal quality after swallowing. When evaluated with formal swallowing studies, some of these patients will have laryngeal penetration. In such cases, seek a balance between quality of life (ie, continue eating despite the high risk for aspiration) and safety (mechanical soft or pureed diets, nasogastric tube, gastrostomy tube, or nothing by mouth). Interestingly, many patients who aspirate when swallowing can still swallow small pills without difficulty. (See "Swallowing disorders and aspiration in palliative care: Definition, pathophysiology, etiology, and consequences" and "Swallowing disorders and aspiration in palliative care: Assessment and strategies for management".)

Lack of appetite — Anorexia is frequently seen in patients with many advanced illnesses (eg, heart failure, cancer, COPD, frail older adults) and is often a foremost concern of patients and their caregivers. Simple and practical measures exist. For example, the Functional Assessment of Anorexia/Cachexia Therapy [120,121], including its 5-item version, has been assessed and validated in patients with advanced non-small lung cancer. Further, the 5-item FAACT was used to evaluate the efficacy of ponsegromab in treating patients with cachexia and pancreatic, colorectal, and lung cancer [122]. A cut-off value of ≤37 for the FAACT-A/CS has been suggested [120].

Oftentimes, other symptoms impact appetite; therefore, assessments that capture patient-reported systems may be helpful. The Patient-Generated Subjective Global Assessment is a comprehensive tool designed to assess nutrition status [123] and includes an evaluation of nutritional impact symptoms.

Another instrument, the Subjective Global Assessment of Nutrition instrument, has been correlated with nutritional status and caloric intake [124].

Assessing food intake and nutritional status may further capture anorexia’s impact. A simple vertical VAS measurement has been shown to correlate with nutritional intake in patients with cancer in the last few weeks of life [125]. Simply asking about what the patient eats and drinks in a typical day can estimate caloric intake. Asking if the patient is bothered by lack of appetite will inform its treatment. (See "Assessment and management of anorexia and cachexia in palliative care", section on 'Clinical assessment'.)

Lack of appetite can be a consequence of chronic fatigue or associated with low-grade nausea; xerostomia, taste changes, pain, diarrhea, constipation, and depression may also contribute. An important and treatable etiology is early satiety due to disorders of gastric motility. (See "Gastroparesis: Etiology, clinical manifestations, and diagnosis" and "Diabetic autonomic neuropathy of the gastrointestinal tract".)

Other common considerations include medications (stimulants, antidepressants, antibiotics, antihistamines, digoxin, ranolazine), depression, and food aversions. (See "Assessment and management of anorexia and cachexia in palliative care", section on 'Anorexia'.)

Decreased appetite is a component of the "failure-to-thrive" syndrome, which is described by the National Institute of Aging as "syndrome of weight loss, decreased appetite and poor nutrition, and inactivity, often accompanied by dehydration, depressive symptoms, impaired immune function, and low cholesterol" [126]. In geriatrics, failure to thrive describes a point further along a functional continuum closer to full dependence and death, with "frailty as a mid-point between independence and pre-death" [127]. (See "Failure to thrive in older adults: Evaluation".)

Weight loss — Among patients with advanced illness, weight loss may occur even in the setting of a good appetite and adequate caloric intake, as typified by cachexia associated with advanced heart and lung disease, cancer, HIV, and other chronic infections. (See "Assessment and management of anorexia and cachexia in palliative care", section on 'Definitions'.)

Weight loss greater than 5 percent in the preceding six months (or more than 2 percent with a body mass index of 20 kg/m²) is generally considered significant [128]. Determining whether patients feel they "look like themselves" may lead to discussing their weight loss. Unlike typical weight loss, cachexia is a hypercatabolic syndrome that is characterized by an accelerated loss of skeletal muscle in the context of a chronic inflammatory response, and distinguished from loss of muscle. In advanced illness, weight loss is not remedied by nutritional supplementation alone. Important physical findings that support the diagnosis of cachexia are wasting of the temporal, thenar, and dorsal interosseous muscles. (See "Assessment and management of anorexia and cachexia in palliative care", section on 'Physical examination'.)

While cachexia in advanced illness typically reflects a hypermetabolic disorder of multiple causes in which treatment is largely ineffective, diagnose and treat any potentially reversible causes. Examples include inadequate nutritional intake, malabsorption (ie, pancreatic exocrine insufficiency, short gut syndrome), and poorly controlled symptoms, as described above. In some cases, cancer treatment (eg, androgen deprivation therapy for advanced prostate cancer, tyrosine kinase inhibitors such as sorafenib) may contribute to muscle loss. (See "Assessment and management of anorexia and cachexia in palliative care", section on 'Definitions' and "Pathogenesis, clinical features, and assessment of cancer cachexia", section on 'Contribution of cancer treatment' and "Non-cardiovascular toxicities of molecularly targeted antiangiogenic agents", section on 'Muscle wasting/sarcopenia'.)

Nausea with or without vomiting — Nausea and vomiting are common near the end of life and can cause substantial physical and psychological distress for patients and their caregivers [129]. Although vomiting is more easily quantified, nausea has been routinely measured using a visual analog scale in clinical trials of antiemetics in patients with chemotherapy-induced nausea and vomiting (CINV). Other validated measures used to measure the impact of CINV on the patient’s experience include the Functional Living Index-Emesis (FLIE), and the Multinational Association of Supportive Care in Cancer Antiemesis Tool (MAT) [130,131]. (See "Approach to the adult with nausea and vomiting" and "Prevention of chemotherapy-induced nausea and vomiting in adults".)

A thorough history and physical examination are essential first steps in managing patients to define CINV severity and identify clues to the underlying etiology. Among patients receiving chemotherapy, nausea and vomiting should not be automatically assumed to represent breakthrough CINV, particularly if symptoms develop at a time point inconsistent with chemotherapy administration. (See "Prevention of chemotherapy-induced nausea and vomiting in adults".)

From a pathophysiologic standpoint, nausea and vomiting can be vestibular, vagally mediated through the emesis center in the medulla, systemically mediated through the chemoreceptor trigger zone (CTZ), or centrally mediated (figure 5). The differential diagnosis of nausea and vomiting in adults is broad, including medications and toxic etiologies (eg, chemotherapy and opioids), disorders of the gut and peritoneum (eg, gastroparesis), central nervous system disorders (eg, brain metastases, cranial irradiation), vestibular disorders, and endocrinologic/metabolic causes.

The most common causes of nausea and vomiting in patients receiving palliative care are outlined in the table (table 17). Patients who are nauseated may have more than one contributory factor. This topic is discussed in detail elsewhere. (See "Assessment and management of nausea and vomiting in palliative care", section on 'Patient assessment'.)

Constipation — Constipation is a common symptom that contributes to significant suffering in patients with serious illness, particularly at the end of life. A major contributing factor is opioid-induced constipation, but other factors contribute [132]. (See "Prevention and management of side effects in patients receiving opioids for chronic pain", section on 'Constipation'.)

Risk factors for constipation in patients who have a serious or life-threatening illness include advanced disease, older age, decreased physical activity, reduced fluid intake, low-fiber diet, depression, cognitive impairment, hemorrhoids, and polypharmacy. In addition to opioids, medications which can cause and/or exacerbate constipation include calcium channel blockers, diuretics, anticholinergics, iron, serotonin antagonists, and chemotherapy (eg, vinca alkaloids, thalidomide, vandetanib). There are also neural (eg, epidural spinal cord compression) and metabolic causes (eg, hypercalcemia and hypothyroidism) of constipation.

Determining the presence of constipation can be difficult as the term has varied meanings for different people. For some, it may mean that stools are too hard or too small or that defecation is too difficult or infrequent. The first three complaints are difficult to quantify in clinical practice; the last can be measured and compared with the general population. (See "Etiology and evaluation of chronic constipation in adults".)

Definitions of constipation used in different publications have been variable [133]. Formal criteria are available to define chronic functional constipation (the Rome IV criteria) [134], but they are not useful in palliative care as they require that symptom onset should occur at least six months before diagnosis, and symptoms should be present during the last three months. The Bristol Stool Form Scale can be used to record stool consistency [135] .

However, the updated ROME IV criteria do provide practical diagnostic criteria for opioid-induced constipation, a common problem in palliative care [134]. In addition, a tool for quantifying opioid-induced constipation, the Bowel Function Index (BFI) (table 18), has been developed and validated [136-138]. Using this tool is recommended in consensus-based guidelines from the American Academy of Pain Medicine, which have been endorsed by the American Gastroenterological Association [139]. One of the benefits of using this tool is that specific scores can be utilized to define a threshold for considering prescription therapies in patients with previous or current use of first-line interventions [137,139]. (See "Prevention and management of side effects in patients receiving opioids for chronic pain", section on 'Constipation'.)

Outside of the setting of opioid use, others consider that four domains are pivotal to making a diagnosis of constipation in palliative care [133]:

●Any lifelong history of constipation (using the Rome criteria)

●Evaluation of physical dysfunction that may impede normal defecation

●Subjective sensations (eg, feelings of incomplete defecation or bloating or fullness)

●Objective changes (eg, frequency or consistency of stools)

Ask patients with constipation when they last had a bowel movement and about any associated symptoms, including rectal urgency, rectal pain, fullness or pressure, abdominal distention, and small stool size. In the palliative care setting, patients with constipation may also present with abdominal pain, anorexia, early satiety, nausea, vomiting, diarrhea, or oozing of liquid stool with stool incontinence. Consistency of stool (soft or hard) is a helpful detail.

A comprehensive physical examination should be performed that includes a rectal exam to palpate for hard stool, assess for masses, anal fissures, hemorrhoids, sphincter tone, push effort during attempted defecation, prostatic hypertrophy in males, and posterior vaginal masses in females. Rectal examination may demonstrate impaction, or the rectal vault may be empty. (See "Etiology and evaluation of chronic constipation in adults", section on 'History and physical examination'.)

If the rectal vault is empty and a diagnosis of malignant bowel obstruction is suspected, radiographic imaging is indicated with an abdominal radiograph or computed tomography (CT) scan. (See "Etiologies, clinical manifestations, and diagnosis of mechanical small bowel obstruction in adults".)

There is limited evidence and no consensus as to how the severity of constipation is best addressed in palliative care. The degree of stool retention may be difficult to estimate from history and physical examination. A clinical constipation assessment scale has been developed and validated (table 19) [140,141], but this is semi-quantitative and does not include objective information about the overall stool load.

Are abdominal radiographs needed? — Palliative care clinical guidelines suggest that plain abdominal radiographs may be used as part of the overall assessment of constipation, including severity [142,143]. However, assessing constipation with plain radiographs is poorly validated with available data [144,145], and their utility is debated.

Some have proposed scoring systems for assessing the severity of constipation in patients with advanced cancer based on the amount and distribution of stool in different quadrants of the large bowel [146,147]. Each abdominal quadrant on plain films is assessed for stool content and ranked from 0 to 3, resulting in a severity score from 0 to 12. In these studies, trained palliative care physicians or gastroenterologists performed the radiographic assessment.

Conversely, others have questioned the assessment of constipation severity based on plain radiography in palliative care. One study of 30 patients with constipation receiving palliative care showed little interobserver agreement regarding the degree of fecal loading based on plain abdominal radiographs among four independent clinicians [148]. In addition, there was a poor correlation between clinician-assigned fecal loading scores and objectively measured colon transit times or patient-reported symptoms. Others have noted a similar lack of correlation between radiographic shadowing on plain abdominal imaging and objectively measured colon transit times and with patient self-reports [149,150].

The lack of data to support routinely using plain radiographs to assess the degree of constipation is reinforced by the American College of Gastroenterology Task Force on Chronic Constipation, which did not include plain radiographs in their recommendations for the comprehensive management of constipation. The American College of Colon and Rectal Surgeons recommends against using plain abdominal radiographs to assess constipation [151,152].

In our view, the use of plain abdominal radiographs should be limited to the following clinical scenarios:

●Concern for bowel obstruction or ischemia

●A patient with liquid stool in whom severe constipation is thought to be responsible (ie, overflow diarrhea) [153]

●A patient with a lifelong history of constipation in whom neurogenic causes of reduced bowel motility with a concern of megacolon or megarectum

●Occasionally to monitor the response to laxatives in patients with fecal retention

Peripheral edema — Peripheral edema, palpable swelling that is produced by expanding interstitial fluid volume, is a common complication in patients with advanced organ dysfunction. When massive and generalized, the excess fluid accumulation is called “anasarca.” Anasarca or edema limited to the lower extremities can interfere with ambulation and self-care and may cause pain from stretching of the skin. It may also directly affect the patient's sense of body image. (See "Clinical manifestations and evaluation of edema in adults".)

Assessment of edema starts with the history that should include questions about area(s) of involvement, associated symptoms (eg, pain), use of medications associated with edema (eg, nonsteroidal antiinflammatory drugs [NSAIDs] and gabapentin/pregabalin) or used to treat edema (eg, diuretics), progression of symptoms, and history of medical conditions associated with lymphedema (eg, thrombosis, surgery, radiation, travel, or infection). (See "Pathophysiology and etiology of edema in adults", section on 'Drug-induced edema'.)

The physical examination should establish the edema pattern (pulmonary, peripheral, ascites, localized) and assess the central venous pressure. The results can indicate the likely cause of edema (table 20). (See "Clinical features, diagnosis, and staging of peripheral lymphedema", section on 'Diagnosis' and "Pathophysiology and etiology of edema in adults", section on 'Etiology'.)

For patients with peripheral lymphedema, clinical girth measurements or estimates of lymphedema volume are necessary to establish a baseline and track changes during treatment. Methods to obtain clinical measurements are described in detail elsewhere. (See "Clinical features, diagnosis, and staging of peripheral lymphedema".)

Various clinical conditions are associated with developing edema, including heart failure, cirrhosis, hypothyroidism, hypoalbuminemia, and nephrotic syndrome; associated local conditions include venous and lymphatic disease and/or malignant ascites (table 21). Among patients who are terminally ill with a serious and/or life-threatening illness, the most common causes of chronic lower-extremity edema are thrombosis, venous obstruction, hypoalbuminemia, decreased mobility, fluid retention, and medications (eg, corticosteroids). (See "Pathophysiology and etiology of edema in adults".)

Lymphedema, a specific type of peripheral edema, may represent a complication of a cancer or its treatment. Lymphedema occurs when the lymphatic load exceeds the transport capacity of the lymphatic system, causing filtered protein-rich fluid to accumulate in the interstitium [154]. Lymphedema is generally due to disrupting lymphatic flow, which may be secondary to injury, surgery, radiation, infection, or malignancy. Compared to generalized edematous states, the capillary filtration rate is normal in patients with lymphedema. (See "Pathophysiology, classification, and causes of lymphedema", section on 'Pathophysiology'.)

Chronic lymphedema affects physical and psychological wellbeing as a result of the interruption of normal daily activities. Patients with cancer and lymphedema are more likely to experience greater disability, poorer quality of life, and greater psychological distress as compared to patients without lymphedema [155]. It may reduce mobility and function, causing patients to experience mental suffering and discomfort [156,157]. (See "Clinical features, diagnosis, and staging of peripheral lymphedema".)

Insomnia — Insomnia is a condition of impaired sleep, with difficulties in initiating or maintaining sleep and/or experiencing sleep as nonrestorative and unrefreshing despite having the appropriate opportunity for sleep to occur. According to the third edition of the International Classification of Sleep Disorders Text Revision (ICSD-3-TR), insomnia disorder is confirmed when all four of the following criteria are met: [158] (see "Evaluation and diagnosis of insomnia in adults", section on 'Diagnostic criteria'):

●The patient reports difficulty initiating sleep, difficulty maintaining asleep, or final waking earlier than desired. In individuals with dementia, the sleep disturbance may manifest as resistance to going to bed at the appropriate time or difficulty in sleeping without caregiver assistance.

●Sleep difficulties occur despite adequate opportunity and circumstances for sleep.

●The patient describes daytime impairment that is attributable to the sleep difficulties. This may include fatigue or malaise; attention, concentration, or memory impairment; social dysfunction, family dysfunction, vocational dysfunction, or poor school performance; mood disturbance or irritability; daytime sleepiness; motivation, energy, or initiative reduction; errors or accidents at work or while driving; and concerns or worries about sleep.

●The sleep-wake difficulty is not solely explained by a current sleep disorder, medical disorder, mental disorder, or medication/substance use.

Insomnia is a very common medical complaint in patients receiving palliative care, with an estimated prevalence of 60 percent or more [159-161]. Insomnia is associated with multiple consequences that negatively impact quality of life and the ability to perform normal functions. In addition, insomnia is a strong predictor of the development of psychiatric disorders. (See "Risk factors, comorbidities, and consequences of insomnia in adults", section on 'Adverse outcomes'.)

In the context of palliative care, insomnia often coexists with other symptoms (eg, pain, depression, and anxiety) that often exacerbates the other, contributing to a decrease in quality of life [162]. In addition to physical symptoms, environmental factors, daytime naps, emotional distress, and medication side effects may contribute and should be addressed. This subject is discussed in detail separately. (See "Insomnia in palliative care".)

Psychological and cognitive symptoms — The key psychological symptoms requiring assessment in palliative care are depression (as distinct from adjustment reaction and grief), anxiety, and delirium. Clinicians should screen for the multiple factors that increase the prevalence and/or severity of psychological distress (rapidly advancing disease, medications [eg, interferon or glucocorticoids], metabolic abnormalities, uncontrolled pain, impaired cognition, financial distress) and address them, if possible. Psychological distress frequently lessens with adequate pain and symptom relief [163].

Sadness, grief, depression, and demoralization — There is a high frequency of mood disorders in patients with a serious and/or life-threatening illness. As an example, in one meta-analysis, there was a 25 percent combined prevalence of all types of depression and a 38 percent prevalence of mood disorders among patients receiving palliative care [164]. The spectrum of mood disorders and conditions seen in palliative care includes sadness, normal grief reaction (including anticipatory grief), pathologic grief, adjustment disorder with depressed features, minor as well as major depression, and demoralization.

Grief versus depression — Grief, a natural and expected reaction to having a life-limiting illness and loss, can and should be distinguished from major depression [165].

A summary of some of the contrasting characteristics of normal grief versus depression in patients with terminal illness is provided in a table (table 22), and this subject is discussed in detail elsewhere. (See "Assessment and management of depression in palliative care", section on 'Assessment and diagnosis' and "Bereavement and grief in adults: Clinical features", section on 'Major depression'.)

Unfortunately, depression is frequently unrecognized and undiagnosed in palliative care [30,166]. The prevalence, burden, and challenges in recognizing and treating depression in palliative care patients are addressed in detail elsewhere. (See "Assessment and management of depression in palliative care", section on 'Prevalence of depression in palliative care' and "Assessment and management of depression in palliative care", section on 'Burden of depression'.)

Patients with advanced illness may display many of the signs of depression (sleep disorder, poor concentration, fatigue, lack of appetite) without being depressed since a medical condition or its treatment can cause such symptoms. Patients facing the end of life may also demonstrate significant guilt and preoccupation with death without being depressed. Nevertheless, the presence of these signs in the absence of other causes should lead to the consideration of depression. Indicators of poor emotional wellbeing, such as a sense of pervasive hopelessness, helplessness, worthlessness, or guilt, may be more useful in diagnosing and assessing suicide risk. Unrelieved symptoms, particularly pain, may contribute to a sense of hopelessness and lead to depression. Loss of independence can contribute to guilt and worthlessness. Among patients with severe pain or other symptoms, a diagnosis of depression should be reconsidered until symptoms are controlled.

The impact of mood disorders in palliative care is underscored by one study, which found that mortality rates in patients with cancer are up to 25 percent higher in those who experience depressive symptoms [167]. In addition, individuals who suffer from depression are at increased risk of suicide. Passive suicidal ideation ("I'd be better off dead") should be distinguished from active suicidal ideation ("I'd be better off dead, and this is how I'm going to do it"). Do not consider an expressed desire for a hastened death as a request for euthanasia but as a marker for intense distress requiring further exploration [168]. (See "Assessment and management of depression in palliative care", section on 'Burden of depression' and "Medical aid in dying: Ethical and legal issues".)

Screening tools — A variety of screening tools for symptoms of depression are available:

●Abbreviated screening instruments (table 23) appear to be as effective as longer instruments to screen for depression. As an example, screening using a simple one-question assessment ("Are you depressed?" or "Have you been depressed most of the time for the past two weeks?") has a negative predictive value (NPV) of 94 percent [169,170] for identifying patients in whom a diagnosis of depression can be excluded based on a comprehensive psychiatric interview. Adding a second question that addresses anhedonia ("Have you experienced loss of interest in things or activities that you would normally enjoy?") increases the NPV to 98 percent [170].

●Another simple screening strategy for mood disorder and psychosocial distress utilizes the Distress and Impact Thermometer [171] or other ultra-short methods [172].

Importantly, these brief tools are designed for screening patients who may subsequently need a full diagnostic evaluation for depression. The subject of screening for depression in palliative care is discussed in more detail elsewhere. (See "Assessment and management of depression in palliative care", section on 'Simple screening instruments'.)

Demoralization — Demoralization is a common problem among patients with serious illness. One systematic review found a 15 percent prevalence of demoralization in patients with cancer and progressive disease [173] and can be precipitated by existential distress [48]. By contrast, depression is a loss of pleasure and interest in the present moment (generalized anhedonia). (See "Assessment and management of depression in palliative care", section on 'Existential concerns and spirituality'.)

Demoralization is a common problem among patients with serious illness. One systematic review found a 15 percent prevalence of demoralization in patients with cancer and progressive disease [174]. Despite this prevalence, there is a general lack of awareness of demoralization and how and when to intervene [175]. Factors consistently associated with demoralization include poorly controlled physical symptoms, inadequately treated depression and anxiety, reduced social functioning, unemployment, single status [173], and possibly age and sex [176].

The Demoralization Scale has been used to identify affected patients [177]; it has been revised and revalidated (table 24) [178]. An abbreviated tool to assess for demoralization in palliative care patients has also been developed (the Short Demoralization Scale [179]), which measures five dimensions of demoralization (loss of meaning, helplessness, disheartenment, dysphoria, and sense of failure). However, this tool has not yet been independently validated.

Treatment of demoralization in patients receiving palliative care can include restoring some control (eg, better pain management) or providing meaning (eg, cognitive behavioral therapy, life review, dignity therapy, and meaning-centered psychotherapy); more research is needed in this area [48].

Anxiety — Anxiety is common in patients with an advanced serious or life-threatening illness. As an example, the reported prevalence of anxiety among patients with advanced cancer is 18 percent [180]. (See "Overview of anxiety in palliative care".)

Anxiety may also be related to a preexisting anxiety disorder, substance use disorder, delirium, or under-treated symptoms, most commonly pain. A single aggregate question ("Have you felt anxious, nervous, uneasy, tense, or frightened in the recent days?") can be utilized to screen for symptoms of anxiety.

Delirium — Delirium and confusional states are among the most common neuropsychiatric disorders encountered in patients with medical illness, particularly older individuals. According to the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5), the diagnosis of delirium requires each of the following criteria [181]:

●Disturbance in attention and awareness that develops quickly (usually hours to days) and tends to fluctuate each day.

●Disturbance in cognition (eg, memory, orientation, language, visual-spatial ability, and/or perception).

●The disturbances in attention, awareness, and cognition are not better explained by another neurocognitive disorder and do not occur in the context of coma.

●Evidence from the history, physical examination, or laboratory findings indicate that the disturbances are caused by a general medical condition, substance intoxication or withdrawal, and/or medication side effect.

Subtypes of delirium have been delineated based on the patient's psychomotor behavior and level of arousal [181,182]:

●Hypoactive subtype – Psychomotor slowing, lethargy, and decreased level of arousal.

●Hyperactive subtype – Restlessness, agitation, and hypervigilance. Mood lability, failure to cooperate with care, and psychotic features may also be present.

The mnemonic FACT provides a summary of the diagnostic criteria of delirium:

●Fluctuating cognitive deficit(s) with acute onset

●Attention deficits, and either

●Consciousness level disturbance; or

●Thought disorganization

Delirium is the most common neuropsychiatric complication seen in patients with advanced cancer nearing the end of life and can result in distress for patients, family members, and clinicians. (See "Palliative care: The last hours and days of life", section on 'Delirium'.)

It is also a marker for shortened survival in patients who are hospitalized and institutionalized [89,183,184]. (See 'Performance status, symptoms, and prognosis' above.)

Delirium in patients with cancer is frequently missed and is often misdiagnosed as akathisia, anxiety, dementia, depression, or psychosis [185,186].

For many patients approaching the end of life with COPD, end-stage kidney disease, heart failure, and other serious life-threatening diseases, global cerebral dysfunction without a definable reversible cause, manifested as delirium, is the final common complication that precedes death [187,188]. However, for many patients receiving palliative care, many factors may contribute to delirium, many of which are potentially reversible [189]. Some of the most frequent contributors are [190]:

●Brain tumor/metastases/cerebral edema

●Cancer treatment (chemotherapy, radiation therapy) (see "Overview of neurologic complications of conventional non-platinum cancer chemotherapy" and "Patients with cancer: Overview of the clinical features and diagnosis of psychiatric disorders")

●Psychotropic drugs (tricyclic antidepressants, benzodiazepines) and anticholinergics

●Metabolic (increased calcium, decreased sodium, renal failure) (see "Hypercalcemia of malignancy: Mechanisms" and "Clinical manifestations of hypercalcemia" and "Overview of the treatment of hyponatremia in adults" and "Diagnostic evaluation of adults with hyponatremia")

●Disturbed sleep (in the intensive care unit [ICU] setting), and other causes of sleep deprivation

●Opioid-induced toxicity (see "Prevention and management of side effects in patients receiving opioids for chronic pain", section on 'Somnolence and mental clouding')

●Sepsis

●Paraneoplastic neurologic syndromes (see "Overview of paraneoplastic syndromes of the nervous system" and "Autoimmune (including paraneoplastic) encephalitis: Clinical features and diagnosis")

Many of the risk factors for delirium are already present in patients with advanced terminal illnesses, although many patients improve with treatment of specific causes. (See "Overview of managing common non-pain symptoms in palliative care", section on 'Delirium'.)

The fact that delirium may be reversible in up to 50 percent of cases [189] underscores the importance of early recognition and aggressive evaluation and management.

Multiple instruments have been developed for screening for delirium:

●The Confusion Assessment Method (CAM) (table 25) has become a standard screening device clinical studies of delirium, conducted across multiple settings including emergency departments and in long-term care [191]. A version is available (CAM-ICU) validated in patients in the ICU [192,193]. (See "Diagnosis of delirium and confusional states".)

●The Memorial Delirium Assessment Scale (MDAS) (form 2A-B) is a 10-item, four-point (0 to 3) observer-rated scale that was designed to quantify the severity of delirium in medically ill patients with serial observations [194,195]. It includes an assessment of disturbances in awareness, orientation, short-term memory, digit span, attention capacity, organized thinking, perception, delusions, psychomotor activity, and arousal to reflect all the main diagnostic criteria according to the Diagnostic and Statistical Manual for Mental Illness.

The MDAS has been used and validated for screening and diagnosis of delirium in patients with cancer [195] and can distinguish patients with delirium from those with other cognitive or noncognitive psychiatric disorders. In one study, a cutoff score of 7/30 on the MDAS yielded the highest sensitivity (98 percent) and specificity (96 percent) for the diagnosis of delirium [195]. In many institutions, this tool has replaced the Mini-Mental Status Examination (MMSE) for assessing and monitoring delirium.

●The bedside confusion scale (ability to recite the 12 months in reverse order and assessment of consciousness state) is another excellent screening tool and longitudinally assesses delirium in palliative care [196]. Serial-sevens and spelling a word such as "farm" or "world" backward are other simple tests of attention. Another bedside test of attention is outlined in the table (table 26). (See "Diagnosis of delirium and confusional states" and "Patients with cancer: Overview of the clinical features and diagnosis of psychiatric disorders".)

SOCIETY GUIDELINE LINKS — 

Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Palliative care".)

SUMMARY

●Symptom dimensions – Patient descriptions and symptom ratings are the primary data for overall assessment. Using different dimensions (severity, frequency, distress, interference) may help patients and caregivers characterize symptoms. (See 'Dimensions of symptom assessment' above.)

●Common symptoms – Frequently encountered symptoms in patients with an advanced serious and/or life-threatening illness include pain, dyspnea, fatigue, dry mouth, dysphagia and aspiration, loss of appetite and weight, nausea/vomiting, constipation, edema, depression, anxiety, demoralization, and delirium. An approach to assessment of these symptoms is outlined in the sections above. (See 'Approach to specific symptoms' above.)

●Performance status – Performance status (table 6 and table 7) is associated with symptom severity, and is a key indicator of prognosis in individuals with advanced terminal disease. For patients with advanced, serious, life-threatening diseases, such as those referred to palliative care programs, the presence or absence of certain symptoms, particularly dyspnea, may help clinicians provide a more refined estimate of patient survival. (See 'Performance status, symptoms, and prognosis' above.)

●Assessment tools – A number of validated symptom assessment tools may be useful in palliative care, including those that evaluate multiple symptoms, such as the revised Edmonton Symptom Assessment Scale (ESAS) (table 5). Many symptom-specific tools are available, such as the Brief Pain Inventory (BPI). Multisymptom assessment tools yield more symptoms and can be used to inform on the multifaceted nature of complex symptoms and inform multimodal clinical interventions (See 'Assessment and rating instruments for symptoms' above.)