Palliative care provision in the context of older age needs to consider some specific issues including confusion, dementia and/or communication difficulties.  Mild Cognitive Impairment (MCI) is generally defined as significant memory loss without the loss of other cognitive functions.  Cognitive impairment carries an increased risk of mortality. The increase in mortality increases with severity of dementia, even though the mechanism for this is not fully understood.  Dementia is an incurable, life-limiting condition. 
Dementia is the leading cause of cognitive decline and the second leading cause of death and disability in Australia. [3, 4] Dementia isn't a specific disease. Instead, it describes a group of symptoms affecting intellectual and social abilities severe enough to interfere with daily functioning.  As the community ages, dementia is becoming increasingly common. [1,3] The estimated prevalence of dementia in Australia in 2012 was 353,800 persons, and this is predicted to rise to 900,000 by 2050.  It seems that adequate and appropriate end-of-life or palliative care is not as commonly offered to this cohort as it might be to people dying from cancer.  The diagnosis of early dementia provides the opportunity for patients and their carers to undertake advance care planning. 
In total three systematic reviews were retrieved with specific focus to the palliative management of older adults with cognitive decline. [6,8] A systematic review by Hines et al.  addressed the issues around palliative care provision to older adults with advanced dementia. Following a comprehensive search, 107 manuscripts were synthesised under three key themes. This review presents mostly evidence of moderate quality. The systematic reviews by van der Steen et al.  and Parpa et al.  incorporate studies of mixed methodology of low-moderate quality mainly due to the reporting of bias and the quality of included studies.
The effectiveness of a palliative approach
Care for people with advanced dementia should be based on a palliative approach and involve a palliative care service if indicated.  Most of the studies reviewed by Hines et al.  were supportive of a palliative approach for the management of older adults with advanced dementia. The studies primarily measured effectiveness through the absence of aggressive or attempts at curative treatments and the presence of advance care planning.
In a number of studies evaluated by Hines et al.  palliative care was associated with less symptoms of pain, restlessness and agitation and a higher likelihood of the person dying in their location of choice. Palliative care was discussed in the only randomised controlled trial (RCT)  which found that families contacted to screen for palliative care were more likely to accept this care option and were more satisfied with the level of care overall.  Fewer acute admissions to hospital were required by the recipients of palliative care. 
Residential aged care is likely to be the place of death for many residents so it is important that staff are able to provide a palliative approach to care. Education on how to provide palliative care in these settings will be important for the quality of life for the person. [6,9] In this instance, training staff to refocus care needs on comfort care rather than curative care is important. 
Absence of aggressive treatments
In a Dutch prospective/observational study, findings suggested that a palliative care approach in end-of-life was equally effective for older adults with severely dementia and older adults without cognitively impairment. This was measured using the Global Deterioration scale to determine the severity of cognitive impairment and the Palliative Outcome scale to assess the impact of symptoms and the disease process on quality of life and coping.  Patients with severe dementia and those with milder cognitive impairment both received similar interventions despite the fact that this may be contrary to good clinical practice, given the poor prognosis of patients with severe dementia. 
A retrospective cohort study conducted in the USA, discussed in the review by Hines et al. , found that despite over 80% of older adults with advanced dementia having a ‘do not hospitalise’ order, hospital transfers still occurred in the last six months of life for a quarter of these older adults. While admission to hospital appears to initiate some level of advance care planning, it is suggested that these discussions should take place much earlier in the care pathway and should include preference for hospitalisation should it be required. For older adults with dementia and an advance care directive (ACD) in place, the focus of the ACD seems to be more on anticipation of death and symptom relief rather than disease modification and treatment choices.  Several other American studies found low engagement of advance care planning for older adults with dementia, particularly those in residential aged care. For these adults, there was an increase in hospitalisation, a higher likelihood of use of feeding tubes, and a higher reporting of pain and other burdensome symptoms.  In the systematic review by Van der Steen et al.  the initiation of advance care planning for older adults with dementia appears to be dominated by the family’s preference or willingness to engage in discussions or the health professional’s willingness to initiate discussion. The disease trajectory being able to be quantified or personal experience of seeing cognitive decline seems to initiate end-of-life care decisions, but often this can be in the midst of a crisis triggered by a rapid decline in health and is not optimally timed and may therefore start too late for smooth coordination of services.  Overall there is a lack of research investigating when and how to best time to initiate end-of-life planning for an older adult with dementia.
The appropriateness of a palliative approach for dementia
Hines et al.  found no evidence to suggest that palliative care is inappropriate for older adults with dementia. Appropriateness was measured through the general acceptability of treatment for people with dementia as perceived by their carers or relatives. These findings were from 11 studies of moderate to weak methodological quality.
One study in this review concluded that the recognition of the futility of treatment is the first step in the successful delivery of palliative end-of-life care. The authors recommend all aged care residents of advanced age should have an appropriate end-of-life care plan. 
A significant difference approaches in dementia management between Western countries and Asian countries was noted. Certain value systems in Asia, age remains to be honoured and life is to be preserved. Studies conducted in the US and Europe, show that families are more likely to engage in advance care planning and to pursue care through comfort and maintenance of dignity. It was noted that in Asian cultures families which observed long term-suffering of a relative may begin to question their cultural values. 
The types of care appropriate in a palliative approach to dementia
The use of feeding tubes is generally inconsistent with a palliative approach. [5,6] Studies in the review by Hines et al.  suggest that feeding tubes are more likely to be used for cognitively impaired older adults, in the following cases: in urban ‘for profit’ residential aged care facilities, or in the absences of an ACD, or when care is not delivered by a specialised dementia unit.  The majority of the evidence suggests that there is no benefit to the use of feeding tubes; careful hand feeding is preferential and promotes quality of life.  Healthcare providers may also lack evidence-based knowledge about artificial nutrition and hydration affecting informed discussions with families and surrogate decision makers. 
‘Do not hospitalise’ (DNH) orders
Facilities with specialised dementia care centres implement DNH orders more frequently than non-specialised care facilities. While symptoms indicating palliative management of dementia are present from up to 12 months before death, comfort care is often not implemented until the last month of life. The presence of a DNH order is often a predictor for the surrogate decisions to not provide aggressive treatments. 
Several studies evaluated the measure of discomfort for older adults with advanced dementia treated with or without antibiotic therapy in the case of pneumonia. Findings suggest that discomfort was high regardless of whether antibiotic therapy was initiated. In the case of withholding this treatment, life was shortened by less than 24 hours in 46% cases and by one month or more in 24% cases. A palliative approach appears to be more usual practice in Europe than in American and may be influenced by the presence of advance care directives or input from a person’s substitute decision-maker.  Overall intravenous antibiotic therapy is thought to be more distressing for the person than beneficial. [5,6] Antibiotics may be considered as a palliative measure in the terminal stages of dementia, but this needs an individual assessment. 
Symptoms such as depression, agitation, pain, aggression are common in end-stage dementia.  Difficulty with communication can make management of these symptoms more challenging.  Management of symptoms such as pain and agitation were found to be the most benefit to older adults with dementia in the systematic review by Hines et al.  Documenting care preferences and family meetings can help these decisions to be made with respect for the person’s wishes and low burden to the family. [4,5]
In end-stage cancer and other life-limiting diseases, cognitive impairment can be overlooked or mistaken for symptoms of delirium.  Delirium is an acute state of confusion causing a change to cognition and can be difficult to diagnose.  The risk of developing delirium is particularly high in end-stage diseases, and pre-existing dementia can increase risk of onset. Despite being difficult to distinguish, differences between delirium and dementia include fluctuating level of consciousness, rapid appearance of symptoms with incoherent speech. The cause of delirium can be multifactorial and includes opioid use and dehydration, both of which can be reversed with medical attention. Where older adults with dementia experience delirium, it is often hard to discern symptoms of each and delirium is often highly attributed to mortality. 
The burden of cognitive decline can be significant particularly as the disease trajectory is difficult to predict and timely advance care planning does not always occur. Palliative care appears to be appropriate, where the intention is to improve quality of life for the person and his/her family and carers. Aggressive management of symptoms of end-stage dementia does not seem to be warranted and may cause more distress without extending life or offering quality of life. Symptom management should be the focus, and ideally this would have been communicated and discussed with the family before decision making regarding withdrawal of treatments e.g. antibiotics or fluids, is required. Where behaviour or cognition deteriorates quickly, screening and appropriate treatment of delirium where appropriate may reduce mortality. The use of hospice services and a coordinated palliative care team approach to the management of people with dementia is desirable however this does not appear to be occur routinely.
Page updated 07 August 2017