Fig. 12.1
The clinical domains for end-of-life care when caring for people with dementia (Adapted from the EAPC 20014, van der Steen et al. (2014))
12.2 What Is End-of-Life Care?
End-of-life care is a separate and distinctive entity from palliative or terminal care and has its origins in the work of Lorenz et al. (2006, 2008) although use of the term is somewhat ambiguous (O’Connor 2016; Froggatt and Payne 2006; Frogatt et al. 2009). It has been variously used to describe the care given to patients in the last year, months, weeks or days of life, with the UK’s Gold Standards Framework describing it as care for those ‘nearing’ the end of life, although they go on to suggest that it includes care given in the ‘final year’ of life to people with an end-stage illness, though acknowledging that it is not always easy to ascertain when this is the case (http://www.goldstandardsframework.org.uk/ accessed 14th October 2016). The UK’s National Council for Palliative Care (2006) describes it as a general rather than a medical term which may be easily understood by the general public to mean that stage where death can no longer be prevented by treatment and when death is likely to occur sooner rather than later.
Physical indicators in addition to worsening cognitive function include increasing weight loss, frailty, poor appetite, breathlessness or recurring respiratory problems, poor performance status, increasing dependence on others for assistance with activities of daily living, and potentially, increased pain or other terminal symptoms (Porock and Oliver 2007; Porock et al. 2005). The UK’s National Council for Palliative Care (2006) suggests, therefore, that it is ‘the provision of supportive and palliative care in response to the assessed needs of patient and family during the last phase of life’, adding that ‘in most cases, end-of-life care does not begin earlier than 1 year before death’ (p3), though no evidence is provided to support this somewhat arbitrary timescale. Generally speaking, however, this does provide a good indicator as to when end-of-life care should begin for most people, although the cognitive problems associated with dementia mean that end-of-life (or advance care) planning should be started even earlier whilst the individual is still able to make their end-of-life preferences known and be involved in discussions about what these should include to ensure that the individual’s wishes are respected and optimal end-of-life care provided. Hence, knowing when a person is likely to die and recognizing that this is actually happening are crucial in determining the quality of death experienced by a person with dementia as we shall now discuss.
12.3 Prognostication and Timely Recognition of Dying
One of the reasons why care is often suboptimal for those dying of or with dementia is related to the lack of a clear terminal trajectory such as those commonly observed in cancer or other life-limiting conditions including heart failure, chronic obstructive pulmonary disease and renal failure. It is important to note however, that people affected by dementia may also have one or more of these conditions in addition to dementia, so comorbidity is an important ‘confounding variable’ when trying to ascertain whether a resident is approaching the end of their life and how soon this may happen. As a consequence, attempts to predict the remaining lifespan for people with dementia over periods longer than a few weeks or months have met with varying degrees of success since it can be extremely difficult to predict a person’s remaining lifespan with certainty, even though there may be indicators like those listed above, that the individual has entered the final stage of their life. Mitchell et al. (2010), for instance, tried to estimate life expectancy in advanced dementia using the ADEPT score so that nursing home residents could be transferred to hospice to die in a timelier and more appropriate manner; but even when completed at the bedside on a regular basis, the tool only marginally improved nurses’ ability to identify which residents with advanced dementia were likely to die within 6 months and plan their management and care accordingly. This is not surprising for dementia masks many of the signs and symptoms which might otherwise indicate terminal malaise, and residents are usually unable to voice their thoughts or feelings at this time, even though they may have some months left to live.
Conversely, another study by Casarett et al. (2012) of 21074 patients admitted to hospice over a two-and-a-half-year period found that 5562 or 26.4% of those transferred to hospice died within 7 days of admission though this means that nurses and other staff overestimated the individual’s remaining lifespan in three out of every four cases since the prognostic model used in the study, which included the Palliative Performance Scale (PPS) score, proved to be less reliable for people with advanced dementia and those with cancer than conditions such as end-stage pulmonary disease. This clearly demonstrates that end-of-life prognostication for people with dementia is extremely difficult in comparison to other client groups, whether in the long or short term, and another study by Powers and Watson (2008) highlights this difficulty, given that long-term residential care in a nursing home needs to balance the need to help residents live active ‘normal’ lives for as long as possible whilst recognising that at some point in the future, they will need to prepare the individual, though more usually his or her family members, for the imminent likelihood of death. It is a very imprecise science however, and even though dementia is a leading cause of death in American nursing homes, its terminal nature is often forgotten, in spite of the fact that 54.8% of the 323 nursing home residents admitted with advanced dementia in one study died within 18 months of their admission, often within 6 months of that date, though the reasons leading to death may have been multifactorial (Mitchell et al. 2009). This does mean however, that a palliative approach to the management of nursing home residents may be justified from the point of admission to the care home, and there is some evidence that this can actually improve outcomes for such patients no matter how soon they might die once there.
12.4 Optimal Treatment of Symptoms in People with Dementia
In spite of the EAPC’s recommendations, and indications that dementia is at last becoming a policy priority in many countries, the palliative and end-of-life needs of those affected by dementia in all its forms are only now beginning to be fully understood. Studies show that end-of-life care has often been suboptimal, even in those countries with established policies, guidelines or national frameworks for people with the condition. Often, this is because policies or guidelines focus on long-term care needs to the detriment of end-of-life care planning or, as the EAPC suggests, there is too little emphasis on staff education, training or the competencies needed to deliver better end-of-life care and more comprehensive services to those dying with dementia (van der Steen et al. 2014). This is regrettable since everyone affected by dementia is likely to benefit from palliative approaches to care which emphasize comfort, dignity, autonomy, meticulous symptom assessment and timely intervention at every stage of their illness, and not just in their final weeks or months of life. In spite of this, many still fail to receive adequate support from palliative or end-of-life care specialists whilst living in long-term care institutions, and still fewer during the crucial, early stages of the illness when meaningful conversations may still be had about the individual’s anxieties, fears and preferences regarding the end of life before they become unable to express these for themselves more fully (Karikari-Martin et al. 2012).
12.4.1 The Consequences of Poor Prognostication in Dementia Care
The consequences of poor prognostication are clear. Sampson et al. (2011), for instance, demonstrated that people with advanced dementia often receive inadequate end-of-life care, and although small in scope (n = 32), this longitudinal study showed that patients with dementia tended to be frailer than others, with over half (62%) having decubitus ulcers and almost half dying during the 6-month follow-up period, indicating that they should have had an advance care plan in place and been in receipt of intensive end-of-life interventions to promote quality of life in the time which remained to them (O’Connor 2016). Sampson et al.’s study showed that 95% (or 30 of the 32 patients) had unalleviated pain on those occasions when it was assessed by the research team, and only seven had an advance care plan in place at the time of their death, all of which had been written by a carer or close family member rather than the individual themselves. This suggests that the individual’s wishes, needs and preferences with regard to end-of-life care planning had not been considered whilst they were still capable of playing a central role in those decisions or leading the discussion themselves.
Mitchell et al.’s (2009) longitudinal study of 323 nursing home residents identified weight loss and malnutrition as the commonest problems identified in residents with advanced dementia with 85.8% of their sample being unable to maintain an adequate dietary intake to preserve their health or functional status effectively. It is generally understood that parenteral feeding is unlikely to improve on this situation, yet many in the advanced stages of dementia are subjected to futile interventions which have little hope of benefit. Infections too are also common, with 52.6% of residents experiencing a severe febrile event requiring medical intervention during Mitchell et al.’s study, and 41.1% developing pneumonia at some point during this time. After adjusting for age, sex and disease duration, the 6-month mortality rate for residents who developed pneumonia after admission in this study was 46.7%; 44.5% for those developing another form of infection, and 38.6% for those suffering from malnourishment or weight loss, indicating that the commonest symptoms are not necessarily the most life threatening, though all of them offer significant challenges to the care of residents and none should be ignored. Other distressing symptoms include dyspnea (46.0%) and pain severe enough to warrant hospitalization (39.1%), a visit to the emergency room, or other ‘burdensome’ interventions such as tube feeding or invasive medical tests in 40.7% of residents. Interestingly, the likelihood of these interventions occurring was reduced when family members or other proxies, such as nursing home staff or attending physicians, had a good understanding of the individual’s likely prognosis and recognized these problems as indicators that death was imminent, allowing symptomatic management to take place without subjecting the individual to futile measures which were not warranted.
12.4.2 Polypharmacy and Inappropriate Interventions at the End of Life
Parsons et al. (2011, 2012) reviewed the medication administration records of 115 care home residents and followed this up with a review of 112 and 105 records on two subsequent occasions. They found that approximately two-thirds of residents (66.9%) were regularly given one or more psychotropic medications with more than 10% of these having a high sedative load score at baseline (12.2%) increasing to 14.3% at time points 2 and 3. Multiple drugs, particularly serotonin reuptake inhibitors, were most often used, but these drugs are not without side effects, including sedation. The study confirmed the findings of others such as Blass et al. (2008) and Tjia et al. (2010) which question the overuse of non-beneficial medications in advanced dementia as death approaches which may be compounded by the lack of monitoring for adverse effects. Thus, whilst medication undoubtedly does have a role to play in advanced dementia, it should always be tailored to the primary goals of care, namely the provision of supportive and palliative care in response to the assessed needs of both the patient and family members during the last phase of life. This encompasses continuous assessment, goal setting, the promotion of comfort and dignity, psychosocial and spiritual care, good symptom management, excellent communication and family involvement in the delivery of coordinated care which is person centred and based on shared decision making (EAPC 2013).
Notwithstanding this, Teno et al.’s (2011a) survey of 486 family members asked to provide information on the circumstances surrounding the death of a loved one from dementia showed that 10.8% died with a feeding tube in place, 17.6% made a decision not to use a feeding tube and 71.6% reported playing no part in any decision about the use of feeding tubes. Of those respondents whose family member died with a feeding tube in place, 13.7% stated that there was no discussion with them about its insertion, and 41.6% of those who did reported that the discussion lasted less than 15 minutes. The risks associated with the use of feeding tubes were not discussed with them in one-third of cases, and 12.6% felt pressured into agreeing to a feeding tube by the attending physician. Residents were physically restrained in order to insert the feeding tube in 25.9% of cases and pharmacologically restrained in 29.2%. It is perhaps not surprising therefore, that family members of individuals who died with a feeding tube in place were less likely to rate the care of their loved one as being ‘excellent’ at the end of life than those treated more palliatively without a feeding tube in place.
Pressure to insert a feeding tube in patients with advanced dementia is by no means unusual, with Sharp and Shega (2009) demonstrating that 56% of the healthcare professionals questioned in their survey (n = 326) had recommended the insertion of a percutaneous endoscopic gastrostomy (PEG) tube to feed patients with advanced dementia and dysphagia in spite of there being little evidence that they improve nutritional status or increase the survival of the individual concerned, and the fact that they did not consider them useful in improving functional status or the individual’s quality of life! Interestingly, whilst almost 40% believed, in spite of their own negative experiences, that tube insertion was the recommended standard of care which should be offered routinely, only 15% believed it should be, and only 11% said that they would want such an intervention themselves if they had advanced dementia. Another study in Italy by Toscani et al. (2013) looking at the incidence and intensity of prevailing signs and symptoms, treatment options and the use of measures such as parenteral hydration and nutrition, cardiopulmonary resuscitation and life-sustaining drugs in the last 48 hours of life showed that 71.6% were inappropriately prescribed antibiotics, i.e. with curative rather than palliative (symptomatic) intent, and 29 patients (20.5%) were receiving nasogastric or percutaneous endoscopic feeding. Two-thirds (66.6%) were receiving parenteral fluids which may worsen pulmonary symptoms at the time of death, and 58.2% were immobilized by bed rails or other immobilizers which in the view of most experts, including many in the EAPC, should not be used at all. Not surprisingly, therefore, almost half the residents in the study were suffering from decubitus ulcers at the time of data collection, and many had still not formally been classified as dying during the final 48 hours of their life, meaning that palliative or end-of-life measures which may have been needed were not being considered, including those of a psychosocial or spiritual nature within hours of death.
Apart from the human cost of such failures, futile interventions on dying residents are unlikely to be cost effective, though relatively few economic analyses of the cost of these interventions have been conducted (Goldfeld et al. 2012). In contrast, when a palliative approach to end-of-life care is taken, a relatively limited pharmacopeia of palliative drugs can elicit better patient outcomes when titrated to individual symptoms and assessment of need which, together with the provision of information and advice about the dying process, can improve care outcomes for those who are dying and, equally importantly, those who remain behind after the death. The range of drugs needed for the effective management of terminal symptoms is even more limited, and can usually be managed quite effectively in the care home with the support of the resident’s GP or community nursing team, especially when given via the preferred routes, namely, oral, transdermal, sublingual, buccal or subcutaneous (via a ‘butterfly cannula’ or syringe driver) as follows (see Table 12.1):
Table 12.1
Drugs commonly used at the end of life in dementia
Opioid drugs for pain | Alfentanil Rapifen® Buprenorphine for non-malignant pain unresponsive to non-opioid analgesics, e.g. BuTrans® Transtec® Diamorphine hydrochloride Fentanyl – as transdermal patches for severe chronic pain, e.g. Fencino® Fentalis® Matrifen® Mezolar® Opiodur® Osmanil® Tilofyl® Victanyl® Fentanyl – as sublingual tablets for breakthrough pain in adults where other short-acting opioids are unsuitable, e.g. Abstral® Recivit® Fentanyl – as buccal tablets/lozenges for breakthrough pain in adults where other short-acting opioids are unsuitable, e.g. Effentora® Actiq® Morphine sulphate MST® MXL® Oramorph® Sevredol® Zomorph® Oxycodone hydrochloride OxyNorm® OxyContin® |
Anti-emetics (to control nausea and vomiting) | Metoclopramide Haloperidol Cyclizine Levomepromazine |
Anxiolytics (to control anxiety) | Diazepam Lorazepam Midazolam |
Anti-muscarinic/antispasmodic drugs (to control terminal secretions or muscle spasms) | Hyoscine hydrobromide Hyoscine butylbromide Glycopyrronium bromide |
12.4.3 The Benefits of Hospice or Specialist Palliative Care Interventions at the End of Life in Dementia
A study by Miller et al. (2012) of 4344 nursing home residents with advanced dementia demonstrated that residents referred for hospice or specialist palliative care support as part of their management plan were prescribed fewer inappropriate medications, invasive intravenous or intramuscular injections, feeding tubes or intravenous fluids, and received better management of agitation and restlessness in the last 90 days of their life in comparison to those not referred to palliative care services. They also had a lower incidence of dyspnea, meaning that respiratory symptoms were better controlled; and they were also less likely to die in an unfamiliar hospital environment as opposed to the care environment they were most familiar with and felt safest in.
A controlled study of 538 family members of people officially certified as dying of dementia by Teno et al. (2011b) found that those being treated with palliative intent by hospice service services (260 or 48.3% of their sample) reported fewer unmet needs or concerns regarding the quality of the care received, and felt that the process of dying had been more peaceful for their loved one, accentuating the importance of recognizing when a resident with dementia is actively dying in order to determine the most appropriate management plan for the final weeks or months of their life. Parsons et al. (2012) found that older people in general, and specifically care home residents, are at greater risk of suboptimal or inappropriate prescribing behaviors by physicians in comparison to other client groups. The researchers found that over two-fifths of older people with dementia residing in six residential care homes in England were prescribed at least one potentially inappropriate medication at each point in their data collection which might normally be discontinued as life draws to a close had a more palliative approach been taken.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
