Using residential aged care data to understand natural deaths

Everybody dies. This truth is unbreakable. Until we learn to thwart death, humans will continue to exit life unexpectedly or predictably.

As clinicians, we try to delay mortality for as long as possible. Our success in this endeavour is evidenced by the number of people who now survive into their 9th and 10th decades. However, every patient eventually reaches a point where extending life is no longer a rational treatment goal. At this point our obligation as clinicians is simple. We should do all we can to support a good death. Recognising when a patient has reached this point is difficult and, for many patients, the shift to a palliative approach comes far too late.¹

The decision to cease active treatment is particularly difficult in people experiencing age-related terminal decline. How long is it reasonable to continue offering interventions that might prolong survival but provide no symptomatic relief? How do we decide which treatments are appropriate or inappropriate when patients can no longer discuss their preferences? And when patients do understand their prognosis, why are we so uncomfortable discussing their imminent death?^2,3

To improve the quality of end-of-life care we offer frail older people, we need better knowledge of the trajectory of natural death, particularly in people with dementia as their primary diagnosis. Knowing with confidence the signs of an irreversible terminal decline in these patients would reassure us when ceasing active treatment that we are practising compassionate acceptance of reality rather than creating a self-fulfilling prophecy.

Aged residential care (ARC) is the optimal setting to study death in older people. ARC facilities in Aotearoa New Zealand and Australia house a rapidly expanding cohort of people who have reached the natural end of their lives. In September 2021, in Aotearoa New Zealand, 35 254 people were occupying an aged residential care bed.⁴ This total comprised approximately 5% of the 75- to 84-year-old population and >25% of the over-84-year-old population.⁵ Australia reports similar rates of ARC occupancy.⁶ The median age of entry to residential care is 86 years, and the median survival time from admission is 18 months.^7,8 International data on place of death indicate that one in three of us will enter ARC shortly before we die.⁹

Research requiring the active participation of people living in ARC can be challenging. People with cognitive impairment severe enough to require residential care often do not have the capacity to provide informed consent to participate in research. However, large volumes of clinical data consented for research use are collected regularly from every person who receives government-funded age-care services in Aotearoa New Zealand.¹⁰ In addition, many large residential aged care providers in Aotearoa New Zealand and Australia maintain a fully digital patient record comprised of regular comprehensive nursing assessments and, in most cases, a complete electronic record of all medicines charted and administered. These records contain detailed, structured, accessible information about functional changes and clinical events experienced by individuals during the last months, weeks, and days of their lives. These data are a rich untapped resource that could help us understand end-of-life trajectories in people experiencing age-related terminal decline.

We must use these data for research designed to improve quality of life and quality of death for people living in ARC. A particularly pressing question is how well we currently recognise impending mortality in people living with dementia. Recent research activity into modelling indicators of reversible decline in people living in ARC is extremely valuable and will reduce preventable deaths;^11,12 however, it is equally important to recognise signs of irreversible decline in this population. By doing so, we could avoid futile treatments that reduce quality of remaining life in people in whom death is imminent and inevitable.¹³

In the absence of a reliable evidence base, we risk misjudging indicators of imminent death such as appetite loss, altered behaviour, and increasingly frequent falls as treatable causes of mortality. Well-meaning interventions such as unpalatable modified diets, harmful psychotropic medicines, or poorly designed exercise programs may not only be futile, but also cause harm, if they reduce the quality of remaining life.¹⁴

We must use healthcare data from people living in ARC to improve their end-of-life care. First, however, we must accept that a good death is a perfectly rational treatment goal for people who have reached the natural end of their lives.

Data sharing is not applicable as no new data were generated or analysed during this study.

Kathleen Potter works as a research facilitator for Ryman Healthcare, a large private provider of residential aged care services in Aotearoa New Zealand and Australia. The authors are co-investigators on the MRFF 2022 Quality, Safety and Effectiveness of Medicine Use and Medicine Intervention by Pharmacists Grant MRFMMIP000022: “Pharmacist Review to Optimise Medicines in Residential Aged Care: PROMPT-RC”.

This research did not receive any specific funding.