7.4 Validated assessment tools for psychological, spiritual, and family issues

Palliative care stands out among other medical disciplines for its foundational commitment to the inclusion of psychological, spiritual, and family issues, premised on founding pioneer Dame Cicely Saunders’ assertion that ‘total pain’ involves suffering in the physical, psychological, spiritual, and social domains (Saunders, 1978).

Saunders further contended that suffering in one domain ‘migrates’ to others. For example, spiritual pain could be somatized or expressed by a person as physical pain. Indeed, research is beginning to identify some of the neurobiological mechanisms for physical expression of pain (García-Campayo et al., 2009; Borkum, 2010). Psychological distress has a reciprocal relationship with social distress, each produced by, and creating, distress in the other. Physical suffering may trigger psychological, spiritual, and social suffering. With the task of the palliative care team being to address suffering in all domains, occasionally resolution of one source will relieve suffering in other domains. Commonly, suffering adopts a life of its own in each domain and needs to be assessed and managed. For instance, an individual in distress who cannot eat due to oesophageal cancer may benefit greatly from palliative surgery. However, the impact of the experience on the individual’s psychological and spiritual states, and their relationships with others, may take longer to resolve and entail anticipatory anxieties or concerns that the situation may recur and worsen in the future. If the individual undergoes palliative surgery successfully but these other domains are unaddressed, he/she remains incompletely managed.

Palliative care providers, therefore, should assess all domains competently and understand how suffering ‘travels’ from one realm to others. The provider must not only assess and address this multi-domain suffering, but measure how well the interventions applied have alleviated it. This applies to the patient and their family, who are considered a unit of care in palliative care, with the provider considering, first and foremost, the patient and then the family.

The patient may experience manifold challenges and losses, starting with the loss of their expectation of health. As each challenge and loss occurs, the patient must absorb the new situation, adapt to it, and reach a new equilibrium (Knight and Emanuel, 2007b). Confronted by serious and rapidly evolving illness, these losses can be overwhelming. Constant reassessment by the provider is essential, as is reconsideration of how to support comprehension, adaptation, and resolution in each of the physical, psychological, spiritual and social domains.

If this were not sufficiently complex, interactions between the patient and their circle of loved ones and community compound it. Understanding how the patient and family impact on one another’s experience is a considerable but necessary challenge. Examples are myriad: a patient with depression due to pancreatic cancer transmits the burden of this depression to those who love them; a family unready to accept the terminal nature of their loved one’s illness may complicate the patient’s acceptance of their own condition and hinder the social tasks of dying that both sides need.

With these interacting dynamics, the provider is greatly assisted by practical assessment approaches and measures of relevant outcomes.

Improving the quality of palliative care is now recognized as a high social priority. Advances in the field’s development, however, are impeded by difficulties achieving reliable assessments and management of the ‘total pain’ domains that matter most in the discipline. These domains are often: (1) subjective, (2) of an impractically large scope, (3) reliant on interviews undertaken at a time when people may be too burdened or ill to answer multiple questions, and (4) difficult to study longitudinally due to the often contracted survival time of this population. Methods to enable measurement of patients’ palliative care needs must be adapted to address these challenges. The availability of psychometrically sound and clinically relevant screening, diagnostic, and outcome evaluation tools that optimize patient evaluation, prognostication, and treatment selection, as well as patient satisfaction and quality of life, are essential to high-quality palliative care.

The following sections describe commonly used assessment and outcome measurement tools in a routine clinical setting—some of which are used, often erroneously, interchangeably. These can be employed in making general patient evaluations across multiple domains, as well as specifically addressing psychological, spiritual, and familial issues. We will discuss their development and, where it exists, validation.

Comparing patient-reported symptoms following open-ended questioning versus a structured survey among 265 advanced disease patients in the United States, Homsi et al. (2006) reported that the median number of symptoms found using the latter was tenfold higher than those volunteered, with most non-volunteered symptoms being moderate or severe and distressing. Effective patient management is therefore based on active, systematic health assessments by providers. These evaluations may improve the accuracy of diagnostic and prognostic determinations and the development of individual treatment plans.

Given its holistic nature, palliative care assessment should include, but extend beyond, traditional medical assessments (Arseven et al., 2005). As suggested by the US-based National Consensus Project for Quality Palliative Care (2013), care plans should involve a comprehensive, timely assessment that is interdisciplinary, coordinated, and include: patient and family interviews; medical record review; discussion with other providers; physical examination and assessment; and relevant laboratory and/or diagnostic tests or procedures. Consultative evaluations should include the patient’s current medical status, adequacy of diagnosis and treatment consistent with review of past history, diagnosis, and treatment, and past treatment responses.

Assessment includes documentation of disease status (diagnosis and prognosis); co-morbid medical and psychiatric disorders; physical and psychological symptoms; functional status; and social, cultural, spiritual, and future care planning concerns and preferences. Assessment of children must be conducted with sensitivity to their age and neurocognitive development stage. Patient and family expectations, goals of care and living, understanding of the disease and prognosis, and preferences for the type and site of care, should also be assessed and documented. Regular reassessment and updating are necessary.

Individual clinical assessment approaches have developed organically over time and are often not rigorous, systematic procedures using valid and reliable measurement tools to supplement skilled clinical interviewing. Such instruments exist in abundance, with a systematic literature review by Stiel et al. (2012) identifying 528 different tools (372 targeting patients) for use in palliative care research and clinical practice. Many are not well validated or designed for clinical use and it will likely always remain the case that the clinical interview and some intuitive assessment are essential. Nonetheless, validated instruments should have a role in the clinical setting. The following sections focus on a limited number that have been tested empirically and can facilitate a comprehensive bedside assessment.

The Needs at the End-of-Life Screening Tool (NEST) (Emanuel et al., 2001) is a general assessment tool that uses a holistic approach and 13 questions that are grouped into four thematic dimensions—corresponding, for mnemonic purposes, to each letter of NEST (Needs in the social domain, Existential matters, Symptom management needs, and Therapeutic relationship matters)—to screen patients for needs of concern. The tool was developed using 15 focus groups and six in-depth interviews among patients, family caregivers, and professionals, followed by a survey of a nationally representative sample of 988 patients (650 at 4–6-month follow-up) with any terminal diagnosis of less than 6 months prognosis, except AIDS. The qualitative work identified areas of importance in end-of-life care, which were tested empirically using factor analysis of patients’ survey responses. The tool was assessed further for its feasibility and effectiveness as a screening tool for advanced illness care needs among 451 cancer patients in tertiary care (Scandrett et al., 2010). An adaptation for use in the emergency ward setting is called SPEED (Screen for Palliative and End-of-life needs in the Emergency Department) (Richards et al., 2011).

Measuring palliative care outcomes, and the totality of care provided, is an important mechanism by which high-quality health-care systems are developed and maintained (Pasman et al., 2009). Given that health-care costs are a substantial portion of public and private budgetary expenditures, the medical profession is obliged increasingly to generate quantifiable evidence demonstrating the effectiveness, efficacy, appropriateness, and acceptability of services. In some countries, the commissioning of health-care services is dependent upon evidence from patient-reported outcome measures (PROMs) (Bausewein et al., 2011a), where results of importance from the patient perspective play a critical role in care provision, audit, and research.

Care is focused on achieving positive clinical and satisfaction outcomes for patients and families by using quality services. Correspondingly, four types of health outcome measurement are patient-, carer-, staff-, and service-based (Jocham et al., 2009). Many measures are part of service assessment rather than direct patient care. However, PROMs can be used routinely in care, patient assessment, and as a continuing measure of progress such as effective pain management. Completed measures can be retained in the patient’s medical record to help the clinician monitor change over time.

Identifying good outcome measures, however, is problematic. The utility of a measure is determined by its psychometric properties and how well it relates to its aim. A measure could be referred to as ‘good’ if it is shown to have validity (face, content, criterion, and construct), reliability (inter-rater, test–retest, and internal consistency), appropriateness and acceptability, responsiveness to change, and interpretable results (Bausewein et al., 2011b). Additionally, in the pressured, time-sensitive environment of the clinical setting, a measure should be easy and relatively quick to administer. These are the benchmarks against which the quality of a measure should be evaluated.

A small number of multidimensional palliative care outcome measures have been developed that meet at least some of the above criteria, including the Palliative care Outcomes Scale (POS).

The POS was developed in 1999, for use with patients with advanced disease, and to improve outcome measurement by evaluating essential palliative care components partly from the patients’ perspectives (Hearn and Higginson, 1999). There are currently two versions of the original POS: one for patients to complete, the other for staff. The tool was developed initially across eight palliative care centres across England and Scotland (including inpatient, outpatient, day, home, and primary care settings) among 148 patients for the patient version, and 337 for the staff version. It has since been adapted and validated in multiple regions, from Europe (e.g. Bausewein et al., 2005), to Latin America (Eisenchlas et al., 2008), to Africa (Harding et al., 2010; Powell et al., 2007), and is being adapted for use among children (Downing et al., 2012).

The POS has ten items assessing physical symptoms, emotional, psychological, and spiritual needs and the provision of information and support, with one open question on main problems. Responses are rated on a 5-point Likert-type scale ranging from 0 (‘No effect’) to 4 (‘Overwhelming’).

Quality improvement (QI) is a formal, systematic approach to performance analysis that seeks to measure the status quo and identify and implement ways to initiate improvements. Clinical audit can be the first step in the QI cycle; a method of reviewing existing clinical practice against agreed standards of care to identify areas for improvement, and re-evaluating those care practices after a designated period of time, to assess their outcome compared with baseline data.

Gould et al. (2007) reported a QI collaborative that was conducted to develop capacity among health-care providers in New York City to apply QI methodology—delivering 4-year-long implementation cycles with a total of 82 teams—to palliative care services. They found significant improvements in most team projects and substantial gains made in familiarity with continuous QI techniques and in building palliative care programmes and networks.

Measurements that have been developed for use in outcome assessments and QI exercises, and which focus on the particular health-care domains of the psychological, spiritual and familial, are discussed in the following sections.

Advanced medical illness and its treatment are associated with high rates of psychological distress in patients and families. The most common psychological problems occurring near the end of life include depression and anxiety (Mitchell et al., 2011). Rates vary widely across studies and are influenced by multiple factors, including the population assessed, the type and severity of disease, and the treatment setting (Mitchell et al., 2011). Depression ranges from 5.1% to 30.1% in populations with advanced illness, such as cancer, AIDS, chronic obstructive pulmonary disease, and congestive heart failure, and up to 15.4% for anxiety (Mitchell et al., 2011).

This section reviews validated screening tools for the assessment of general psychological distress, depression, and anxiety in populations with advanced illness, discussing multiple considerations for assessment and specific tools that may assist providers in timely and appropriate evaluation. For a review of tools that assess other common psychiatric problems—such as adjustment disorders (Akechi et al., 2006) and delirium (Breitbart et al., 1997)—and family caregiver problems—including bereavement (Zisook et al., 1982) and complicated grief (BrintzenhofeSzoc et al., 1999)—readers may refer to Chapters 17.3, 17.4, 17.6, 17.7 and other published reviews (Stroebe et al., 2007; Thekkumpurath et al., 2008; Grover and Kate, 2012). For diagnosis and management, provide detailed overviews.

Studies show that untreated psychological distress undermines medical decision-making and treatment adherence, which can result in disability and high health-care usage (Colleoni et al., 2000; Prieto et al., 2002). Psychological distress may augment pain perception (Pincus et al., 2002), undermine symptom control (Spiegel and Giese-Davis, 2003), and impair quality of life (Skarstein et al., 2000; Stark et al., 2002). Patients commonly report a perceived loss of meaning or purpose in life, a decrease in engagement in valued activities, and a reduction in pleasure from the social environment (Parker et al., 2003; Hopko et al., 2008). This can impede the ability to cope with the emotional stress of preparing for death and separating from loved ones, and worsen distress in family members (Pessin et al., 2002). Despite effective interventions for psychological distress (Fawzy et al., 1990; Sheard and Maguire, 1999; Donker et al., 2009), it is under-diagnosed (Steinberg et al., 2009; Holland et al., 2010) and screening is rare in practice (Institute of Medicine Committee on Treatment of Posttraumatic Stress Disorder, 2008).

In the past, one barrier has been the lack of validated tools for screening (Vodermaier et al., 2009). The US National Comprehensive Cancer Network (NCCN) has developed guidelines for regular screening and management (NCCN, 2013). The NCCN defines psychological distress as ‘a multifactorial, unpleasant, emotional experience of a psychological (cognitive, behavioural, emotional), social, and/or spiritual nature that may interfere with the ability to cope effectively with cancer, its physical symptoms and its treatment’ (NCCN, 2013). The detection of psychological distress often relies on physician assessment and patient-initiated reports; however, they are frequently discordant (Söllner et al., 2001). Consequently, the following tools can assist providers in more accurate identification.

When selecting an appropriate tool, important considerations include the psychometric properties, particularly sensitivity and specificity, treatment setting, and disease severity. Each tool has its own advantages and limitations. Ideally, validated cut-offs provide high sensitivity (i.e. identify all patients in need of support) and specificity (minimize false positives) (Vodermaier et al., 2009). For example, shorter measures are useful clinically (in terms of cost and time), but have fewer domains and, sometimes, limited specificity. In general, tools are classified as ultrashort (one to four items), short (five to 20 items), and long (21–50 items) (Vodermaier et al., 2009).

One ultrashort screening tool recommended by the NCCN is the Distress Thermometer (NCCN, 2013). A single item measures overall psychological distress in the past week on a 0–10 numeric rating scale, with an accompanying Problem List assessing five domains. It has high sensitivity and specificity with the Hospital Anxiety and Depression Scale (HADS) and the Center for Epidemiologic Studies of Depression Scale (CES-D) (Hurria et al., 2009) and good reliability (Hoffman et al., 2004). While a cut-off score of 4 or higher is meaningful (Hurria et al., 2009), scores of 7 or above may provide optimal sensitivity and specificity (Hegel et al., 2008).

Another ultrashort tool, the four-item Patient Health Questionnaire-4 (PHQ-4) (Kroenke et al., 2009) screens for anxiety and depression. Anxiety items include: ‘Feeling nervous, anxious, or on edge’ and ‘Not being able to stop or control worrying’, two diagnostic criteria for Generalized Anxiety Disorder in the Diagnostic and Statistical Manual of Mental Disorders, fourth edition, text revision (DSM-IV-TR). Depressive items include: ‘Feeling down, depressed, or hopeless’ and ‘Little interest or pleasure in doing things’. Responses are rated on a 4-point Likert-type scale (‘Not at all’ to ‘Nearly every day’), with total subscale scores ranging from 0 to 6; cut-offs of 3 or above are clinically significant (Mitchell et al., 2009; Kroenke et al., 2010), with 0–2 (normal), 3–5 (mild), 6–8 (moderate), and 9–12 (severe). Like other tools, it does not diagnose depression or anxiety. However, its predictive ability and reliability (Kroenke et al., 2009) and brevity are advantages.

The short Brief Profile of Mood States Total Mood Disturbance Score (Brief POMS TMDS) includes 11 items evaluating general distress in cancer patients (breast, pancreatic, gastric cancer, and myeloma) (Cella et al., 1987). It reduced the 58-item TMDS with one factor of general distress. Five mood factors include: tension-anxiety, depression-dejection, anger-hostility, fatigue, and confusion-bewilderment. Items are rated on a 5-point Likert-type scale (‘Not at all’ to ‘Extremely’), with total scores ranging from 0 to 44. It is internally consistent and correlated with the long version. Criterion validity discriminates between pancreatic patients, who typically exhibit higher distress scores, and gastric cancer patients. Other strengths include brevity, sensitivity, and true assessment of psychological distress without the ‘vigour’ mood factor (i.e., measures physical well-being) (McNair et al., 1971; Cella et al. 1987).

Two widely-used, short screening tools for depression are the HADS (Zigmond and Snaith, 1983) and the CES-D (Radloff, 1977). Both are validated in palliative care populations (Le Fevre et al., 1999; Härter et al., 2006; Vodermaier et al., 2009).

The HADS includes two 14-item subscales measuring depressive and anxiety symptoms (HADS-D) (HADS-A). Items are rated on a 4-point Likert-type scale (‘Not at all’ to ‘Very often’), with subscale scores ranging from 0 to 21 (Zigmond and Snaith, 1983). It is clinically relevant, responsive to treatment (Smarr et al., 2011), validated in multiple languages, including Spanish, Chinese, French, and Italian (Smarr et al., 2011), and reliable (Bjelland et al., 2002). However, the cut-offs vary, ranging from 8 to 22 (total score) and 5 to 11 (subscale scores) (Vodermaier et al., 2009). A cut-off of 8 or more provides optimal sensitivity and specificity (Bjelland et al., 2002), with probable major depression rated as: 8–10 (mild), 11–15 (moderate), and 16 or higher (severe) (Snaith and Zigmond, 1994).

The CES-D (Radloff, 1977) includes 20 items measuring depression and identifying individuals at risk of depression. It assesses mood and level of functioning in the past week, and rates symptom frequency across four domains: positive and negative affect, somatic and interpersonal problems. Items are rated on a 4-point Likert-type scale (‘Rarely’ to ‘Most or all of the time’), with total scores ranging from 0 to 60. Internal consistency is good (Radloff, 1977). Cut-off scores greater than 16 suggest probable clinical depression (Katz et al., 2004), yielding a sensitivity of 100% and specificity of 85% in cancer patients (Katz et al., 2004). The increased false-positive risk is balanced by brevity, validation in multiple languages, and free use. Shorter five- and ten-item versions (Shrout and Yager, 1989; Irwin et al., 1999) have improved upon response wording.

While tools such as the HADS screen for depression and anxiety, several provide detailed anxiety assessments. The long State-Trait Anxiety Inventory (Spielberger et al., 1983) includes two 40-item subscales assessing State Anxiety (S-Anxiety) and Trait Anxiety (T-Anxiety). S-Anxiety measures current apprehension and tension on a 4-point Likert-type scale (‘Not at all’ to ‘Very much’). T-Anxiety measures anxiety (i.e. discomfort or worry) during typical daily situations (responses range from ‘Almost never’ to ‘Almost always’; Spielberger et al., 1994). Total subscale scores range from 20 to 80. Cut-off scores of 30 and below (no–low anxiety) or higher than 30 (moderate–high anxiety) (Glozman, 2004) and higher than 54 in geriatric populations have been suggested (Kvaal et al., 2005). Advantages include good reliability (Spielberger et al., 1983) and translation into multiple languages, but use requires purchase (Mind Garden, 2013).

The Beck Anxiety Inventory (Beck et al., 1988) includes 21 items assessing cognitive and somatic symptoms. Items are rated on a 4-point Likert-type scale (‘Not at all’ to ‘Severely’), with total scores ranging from 0 to 63. A cut-off score greater than 10 provides high sensitivity and specificity (Hopko et al., 2008), with 0–9 (normal), mild to moderate (10–18), moderate to severe (19–29), and severe (30–63). While brief, easily administered, and validated in oncology populations (Hopko et al., 2008), specific drawbacks include the multitude of somatic symptoms, which can overlap with medical conditions (Julian et al., 2011). Although internally consistent (Fydrich et al. 1992), further validation in palliative populations is suggested (Vodermaier et al., 2009).

The extent of the family carer’s burden and impact are becoming better understood and characterized. There are many potential stressors associated with caring, including managing psychological and physical needs, instrumental assistance with a range of practical and medical tasks, such as adherence to medical regimens and decision-making and providing seemingly constant vigilance. Qualitative data reveal common themes in subjective caregiver impact (Dellon et al., 2010; Braithwaite et al., 2011) and several questionnaires report both positive and negative consequences of caregiving (Glajchen et al., 2005). Although carers are often inadequately trained, under-supported, misunderstood, and distressed, there are also positive aspects of providing care to a family member, including satisfaction in caring for a loved one (Glajchen et al., 2012).

When confronted by suffering, distress, and mortality across the disease trajectory, patients may articulate spiritual concerns. Despite professional sensitivity regarding the investigation of patients’ spiritual distress and the feasibility of addressing issues identified in the typically brief clinical encounter, the spiritual concerns of patients with a life-limiting illness can impact upon treatment choices (Balboni et al., 2013; Peteet and Balboni, 2013). However, research in the spiritual domain remains one of the least investigated in palliative and hospice care provision, in contrast to the physical-functional aspects.

Whilst instruments to assess religious/spiritual (R/S) issues within the discipline exist—with more being developed due to a growing appreciation of their need (for a more comprehensive list of existing R/S tools, see a review by Büssing (2012))—much of the work has three significant shortcomings: lack of a generally accepted conceptual definition of spirituality (Puchalski et al., 2009), resulting in confusion (la Cour and Götke, 2012); a paucity of tools validated in cross-cultural palliative care populations (Selman et al., 2011); and the dearth, or lack of reporting, of methodologically rigorous supportive validation studies per se, especially for clinical tools (Galek et al., 2011).

Three categories of formal assessment tool in the spiritual domain are screening, history taking, and assessment (Puchalski, 2010). These categories are aligned with normal medical practice and are the best current models for interdisciplinary spiritual care (Puchalski et al., 2009). For a fuller description of this model, see Chapter 4.5.

Spiritual screening questions are designed for insertion into routine assessment or screening processes to identify patients with significant spiritual needs. The widely used Distress Thermometer (NCCN, 2012)—a patient-rated tool whose specificity and sensitivity have been demonstrated among bone marrow transplant patients (Ransom et al., 2006) and among a heterogeneous mix of Swedish oncology patients (Thalén-Lindström et al., 2013)—includes ‘spiritual/religious concerns’ on its problem list. However, no research exists determining what this category means to patients, and therefore what it is measuring.

Some clinicians use the one-question screen ‘Are you at peace?’, with a 5-point Likert-type scale, that was evaluated by Steinhauser et al. (2006). Although the question was found to provide a brief gateway to assessing spiritual concerns, to date the screen has only been tested with patients with advanced, life-limiting illnesses such as stage IV cancer, congestive heart failure, or chronic obstructive pulmonary disease, despite research indicating such concerns can influence medical decision-making throughout the disease trajectory.

The only multi-question screening instrument supported by a published partial validation study was developed by Fitchett and Risk (2009). The brief screening protocol, subsequently named the Rush Religious Struggle Screening Protocol and used in a QI pilot project among oncology patients (Blanchard et al., 2012) and blood and marrow transplant patients (King et al., 2013), focuses on the identification of patients who may be experiencing R/S struggle—a well-researched concept that has been shown to be related to several significant health outcomes, including mortality (Pargament et al., 2004)—and patents who would like a visit from a chaplain. The triage method consists of three questions that can be administered by anyone on the health-care team without significant training, while the protocol for making referrals to a chaplain is simple and transparent. In a pilot study among non-chaplain health colleagues in an 18-bed, Chicago-based acute medical rehabilitation unit, the protocol identified 7% (n = 12) of patients possibly experiencing R/S struggle, with all but one of these confirmed in need following a chaplain assessment (Fitchett and Risk, 2009).

Spiritual history questions are recommended for inclusion in the social history section of the routine history. The HOPE tool (sources of Hope, Organized religion, Personal spirituality, Effects on care) (Anandarajah and Hight, 2001), and the SPIRIT tool (Spiritual belief, Personal spirituality, Integration with community, Rituals, Implications for care, Terminal events planning) (Maugens, 1996) are often used.

However, the FICA (Faith, Importance, Community, and address in Care) spiritual history tool (Puchalski, 2006) is the only such instrument partly supported by a validation study (Borneman et al., 2010). Developed in a consensus expert-review process, and then modified based on anecdotal feedback from users, this tool is not intended to be used as a checklist, but rather as a guide to initiate dialogue with patients to explore deeper questions their relationship to spirituality, their spiritual beliefs, and their goals for spiritual health. One of the weaknesses of the limited validation conducted on the tool is the lack of heterogeneity in the R/S preferences of its sample, with over 70% declaring themselves Catholic or Protestant.

Spiritual assessment is, ‘A more extensive [in-depth, on-going] process of active listening to a patient’s story as it unfolds in a relationship with a professional chaplain and summarizing the needs and resources that emerge in that process (see Chapters 4.5 and 17.1). The summary includes a spiritual care plan with expected outcomes which should be communicated to the rest of the treatment team’ (Fitchett and Canada, 2010).

However, although intended, in North America and elsewhere, to guide the work of professional chaplains in detailed discussions with patients about their R/S strengths and issues and how they might affect their care, all existing spiritual assessment tools are unvalidated. Moreover, all require significant training to use and are best reserved for patients whose R/S concerns affect their coping and care planning significantly. Among these unvalidated instruments, the recommended include the following:

Like spiritual assessment tools, most spiritual measurement instruments are unvalidated. Two exceptions that have been used widely among physically ill populations are the Brief RCOPE and the Functional Assessment of Chronic Illness Therapy—Spiritual (FACIT-Sp).

The Brief RCOPE (Pargament et al., 2000) features 14 items across two scales: positive religious coping and negative religious coping, recognizing that religion can support or impede coping. It identifies spiritual adjustment; that is, the extent to which conflict, self-blame, or anger at God is present.

The tool was tested initially on a sample of 540 college students coping with a significant negative life event. They were predominantly white (93%), single (99%), female (69%), and Catholic (45%) or Protestant (41%). Factor analysis of the results produced factors largely consistent with the conceptualization and construction of the subscales. Subsequently confirmatory factor analysis of the RCOPE was undertaken in a second sample composed of 551 elderly hospital patients, and found to be moderately supportive of the initial factor structure. Studies have adjusted the tool for individuals with a Buddhist background (Phillips et al., 2009) and adapted the underlying coping concept for use in cancer patients (Zwingmann et al., 2006).

The tool can be very helpful to the clinician or researcher trying to identify patient use of religion in coping and contains a good deal of religious language. As such, it is more appropriate for populations with a theistic and Western, largely Christian, background (Pargament et al., 2011).

The FACIT-Sp (Peterman et al., 2002) focuses on spiritual well-being, another useful concept for clinicians, and is available in several variations aimed at specific populations, such as pain patients. Developed among cancer patients, and avoiding religious language, two subscales, measuring a sense of meaning and peace and the other assessing the role of faith in illness, were determined from the 12-item tool. A total score for spiritual well-being is produced. Based upon two studies, the tool has demonstrated good internal consistency and reliability, a significant association with quality of life, and convergent validity with five other measures of religion and spirituality in a sample of patients with mixed early-stage and metastatic cancer diagnoses. A subsequent study among 240 long-term (average of 10 years post-diagnosis) white, female survivors of cancer supported a three-factor model (meaning, peace, and faith) (Canada et al., 2008). However, this was later found to have less stability as a model across 8805 cancer survivors from a much more ethnically diverse sample, suggesting that items and their constructs might be understood differently across groups (Murphy et al., 2010). Consequently, because the variations of the scale are generally based on specific populations, users should determine whether the version they use is appropriate for their population.

Many of these measurement tools also have been used as spiritual screening tools, in part because of the dearth of validated tools designed explicitly for screening. However, clinicians are cautioned that even the varyingly validated tools have not been validated for clinical screening purposes. Additionally, since these are paper and pencil instruments, they are limited by the language they use. For instance, several use references to ‘God’ or other higher powers which may be unhelpful to an increasing number of patients, such as in the United States, who perceive themselves as not religious (Pew Research Center’s Forum on Religion and Public Life, 2012). Conversely, other instruments use no explicitly religious language, which may not resonate with patients who only express their R/S issues in such terms. Lastly, the terms ‘spiritual risk’, ‘spiritual struggle’, and ‘spiritual pain’ are not well defined or differentiated in the extant literature, although ‘risk’ is normally understood as a condition which renders the patient susceptible to ‘pain’ or ‘struggle’. Clinicians and investigators should, therefore, be aware of possible ambiguity in the use of these terms.

Support for the family is particularly important in palliative care. The needs of family members, many of whom contribute as carers, can shape the patient’s experience and ultimately become the focus of bereavement care. Assessment of family members must address their role as carers as well as members of a larger family. While many tools exist, identifying appropriate tools and integrating them into routine clinical management or research is not standard, is poorly tested, and therefore problematic for practical implementation.

The types of caring roles family members adopt stem from diverse patient needs. Family carers can find themselves supporting a patient’s physical, psychological, existential, and social needs. An abundant supply of carer burden scales and questionnaires exist that investigate the multidimensional impact on the carer’s daily life and well-being. In a review of instruments related to carers in palliative care, the following categories were identified:

A unitary comprehensive ‘gold standard’ assessment tool has not emerged, and is unlikely to do so. Although the diversity of existing instruments is potentially valuable—providing the clinician and researcher with choices based upon specific characteristics of a given disease site or caregiving burden domain of interest—the plethora of instruments carries the risk of ‘confusing the marketplace’ with a multitude of divergent questions, rating scales, and aggregated scores. Despite the wide availability of such instruments, few providers use them in routine practice.

Moreover, challenges exist to using assessment tools that evaluate the family itself and the specific roles of family members of a palliative care patient. There are varying contextual components that impact on the focus of assessment tools. For example, the relationship of a patient to his/her family and family structure can vary considerably depending on whether the patient is an infant, adolescent, adult, or elderly person. Similarly, assessing the family needs of a household with multiple young children differs from assessing those of a patient and their spouse, or adult child. Categories of existing family assessment tools based on their contextual focus include characteristics of the patient (e.g. spouse vs parent vs child); attention to the family as a unit versus members/caregivers within the family; disease categories, for example, patients with cancer versus those with Alzheimer’s disease; location of patient, such as home versus at the hospital; and timing of illness, such as during chronic phases of illness versus end-of-life versus bereavement.

Few existing instruments that focus on families have been evaluated for their utility in guiding clinical practice. Rather, many have been developed based on a particular theoretical framework and validated in specific clinical settings (Neabel et al., 2000). How and when to use such tools clinically, and the mechanisms for implementing interventions based on their results, remain poorly established. While this area is clearly underdeveloped, the following sections discuss tools for assessing and measuring the family unit. (For a more comprehensive list of existing family assessment tools, see a review by Neabel et al. (2000) and of instruments related to family caregivers of palliative care patients by Hudson et al. (2010).)

The Needs Assessment Tool Progressive Disease (NAT-PD, nd) or the Needs Assessment Tool—Progressive Disease Cancer (NAT-PD-C, nd) is recommended by the Centre for Palliative Care’s practice guidelines (Hudson et al., 2010) to assess carers’ needs. These tools were developed by an expert consensus panel of 66 leaders and key stakeholders in palliative care, based on an extensive literature review. Initially, face and content validity, acceptability, and feasibility were confirmed by health-care professionals who viewed video-recorded, simulated consultations involving actors as patients (Waller et al., 2008). Further testing was undertaken in a clinical setting to determine the inter-rater reliability, validity, and feasibility of the tool (Waller et al., 2010). The impact of using this tool in the clinical setting was assessed for cancer patients and associated with significant reduction in health system, information, patient care, and support needs (Waller et al., 2012). These screening tools include six items to assess the ability of the carer/family to care for the patient across the physical, daily living, psychological, information, financial and legal, and family relationship domains as well as two items to assess the carer’s well-being in relation to their own physical, psychological, and bereavement issues.

The Family Crisis Oriented Personal Evaluation Scales (F-COPES) is a 29-item self-report questionnaire intended to identify problem-solving and behavioural strategies used by families during crises (McCubbin et al., 1987). It measures five subscales related to acquiring social support, reframing, seeking spiritual support, mobilizing family to acquire and accept help, and passive appraisal. The tool can help health-care teams identify which families may need additional assistance with managing crises.

Like assessment tools, identifying measurement tools for carers and families requires consideration of the specific contextual components important to the families assessed and the area of interest. For example, some tools are developed specifically for use with cancer patients. Tools measuring the management of carer psychological distress differ from those addressing a family’s ability to cope. Moreover, the questions of interest for the health-care teams and organizations will determine which tool to use and how to implement identified tools in the clinical setting. Measuring family needs consecutively over time may be necessary to determine the efficacy of the palliative care team in addressing them. Alternatively, it may be beneficial to measure carer bereavement following the death of their loved one, to consider whether palliative care services are providing long-term outcomes for family members. Needs assessment tools can be useful also for health-care organizations in determining resource allocation. Finally, many measures of families and carers have been developed and studied in non-palliative care settings, meriting a degree of caution if employed in palliative care situations.

Broad categories of relevance for measuring families and carers in palliative care include carer satisfaction with service delivery, carer quality of life, psychological impact, and bereavement. Many tools used to measure psychological domains for patients can also be used for carers.

The Quality of Life in Life-Threatening Illness—Family Carer Version (QOLLTI-F) is a 16-item tool measuring caregiver quality of life and includes the carer’s perception of patients’ conditions. There are seven subscales: environment, patient state, carer’s own state, carer’s outlook, quality of care, relationships, and financial worries. It was developed from qualitative research, asking carers what is important to their own quality of life, rather than focusing on the changes or burdens related to caring, with most domains responsive to change (Cohen et al., 2006). All items, the subscale scores and the total score (the mean of the seven subscale scores) have a scoring range from ‘0’ to ‘10’; to ensure ‘0’ always indicates the worst situation and ‘10’ the best, item numbers 3, 4, 14, 15, and 16 must have their scores transposed.

The Family Appraisal of Caregiving Questionnaire for Palliative Care (FACQ-PC) is a 25-item tool that measures four theoretically derived domains: caregiver strain, positive caregiving appraisals, caregiver distress, and family well-being. Among 160 adult primary carers of a relative with cancer receiving home-based palliative care, it showed acceptable psychometric properties (internal consistency, factorial validity, and convergent and discriminant validity), can be used in clinical assessment, and has potential use in evaluating the effectiveness of palliative care (Cooper et al., 2006). However, one limitation is that the content validation process did not include carers, but rather a small group of professionals (n = 5).

Based upon the FAMCARE scale (Kristjanson, 1993), which was developed using qualitative research to identify indicators of family care satisfaction and their subsequent reduction using Q-sort study to the most salient, the FAMCARE-2 was developed among 497 carers from 29 Australian inpatient- and community-based palliative care services. FAMCARE-2 includes 17 items and is measured using a 5-point Likert-like scale (from 1 = ‘Very satisfied’, to 5 = ‘Very dissatisfied’) and an additional ‘Not relevant to my situation’ option.

The tools available for the assessment and measurement of psychological, spiritual, and family care issues remain largely inadequate, both in terms of their psychometric validation and applicability in divergent diagnostic, ethnic and age groups, service delivery settings, and cultural contexts. As understanding of the nature of needs and care in the setting of serious illness deepens, we should expect improvements both in the tools available and how to integrate them into processes of care and QI.

Additional concepts need to be developed and integrated into mainstream palliative care. For instance, work exists on the nature of the state of peace that some attain near the end of life and that many aspire to (Emanuel et al., 2010). Yet, delineation of this state of peace as a construct that can be assessed, measured, and addressed has not developed to the point that tools are integrated into care delivery. Similarly, further understanding of what makes for resilience and successful adaptation are necessary if more patients are to achieve the best death possible for themselves, their families, and society (Knight and Emanuel, 2007a).

All this will be possible only if the culture of palliative care remains seriously committed to the inclusion of psychological, spiritual, and family care within palliative care. While the specialty may still be better at this than other medical disciplines, it is also the case that advances in the clinical sciences of physical symptoms have outpaced other areas. Approaches to address mental health are improving; information on anxiety, depression, delirium, and other psychological suffering encountered in palliative care is increasing, but still there is far to go. Similarly, counselling and psychotherapy for patients and families facing serious illness is a developing area, as are the economics at the household level of palliative care. However, to date there are limited data and clinical guidance available for either. Most striking of all is the sparse empirical knowledge of spiritual care in palliative care. Readers are advised to keep abreast of advances in all of these fields and participate in the culture of palliative care that insists on their importance. If patients are assessed for needs in these areas, and if care outcomes in these areas are measured, increased understanding and improvement will be possible.

Dr Dhingra wishes to thank Jack Chen, Huiyan Ye, and Anna Gordon for their assistance with literature reviews and references.

Complete references for this chapter are available online at <http://www.oxfordmedicine.com>.