Time to shut down the acute care conveyor belt?

A rapid response system may be an appropriate model for meeting the urgent need for more suitable care for patients at the end of life

Hospitals can be dangerous places where people can unexpectedly die. Hospitals can also be dangerous places because people are not allowed to die. When they eventually die, it can be a prolonged and demeaning experience.1,2

The population of the world is increasing. People are living longer. An increasing number of aged people are spending their last few days, weeks or months in acute hospitals,3 many of whom will die in intensive care units (ICUs).4 Almost a third of Americans will spend time in an ICU during the last month of their life.4 However, most people want to die at home, not in an acute care hospital.

Discussions around the end of life (EOL) are ubiquitous and the term can be interpreted in many different ways. For the purposes of this article, we limit the term EOL to older people with significant comorbidities who, based on existing evidence, have less than 1 year to live. If EOL care in acute hospitals is one of the largest contributors to health care costs and if our society does not want it, how did it happen and how can we manage EOL care more appropriately and in line with what people want?

The acute care conveyor belt

The current situation is akin to older people with illness being placed on a conveyor belt, beginning in the community and eventually taking them to an acute hospital and then possibly to an ICU. Dying and EOL care have slowly and almost imperceptibly become medicalised. As we have developed impressive ways of keeping people alive, it has become very difficult to exercise the choice to die naturally and not be surrounded by machines and well-meaning people.

When people suffer sudden injury or illness in the community, an ambulance is often called, whether the illness is potentially reversible or simply a minor deterioration in someone with only days to live. The conveyor belt operates largely because of uncertainty of the patient’s prognosis, the lack of practical and readily available alternatives, and the failure of people to have stated their wishes regarding EOL care. Even if people’s wishes are expressed clearly, uncertainty can be a barrier to those wishes being carried out. Patients with severe chronic diseases marked by acute exacerbations become used to the idea that the medical system can always rescue them from life-threatening deteriorations. Of course, rescue is sometimes possible, until it is not. Well-meaning clinicians may convince patients that there is a large, potentially reversible component to their deterioration and that they deserve every chance to leave hospital alive. As a result, the rate of emergency admissions to hospitals and ICUs is dramatically increasing.5-7 Many of these patients are in the last year of life.6 Once patients are in the emergency department, it is easier to admit them to hospital rather than to embark on the time-consuming task of discussing options with them and their caregivers and finding a more appropriate place of care.6

After admission to the hospital, patients are confronted with further pressures to keep them on the conveyor belt. Hospital medicine has become increasingly specialised and fragmented.8 Hospital specialists are experts in their own particular part of the patient’s clinical status. However, the population of patients in acute hospitals has changed and elderly patients rarely have single-organ problems.9 They are now older, with multiple comorbidities. The ability to diagnose dying patients or to identify patients at the EOL, or understand the bigger picture of the patient’s circumstances, is often overlooked by single-organ specialists.10 All in all, the management of the dying process is not done well in acute hospitals.1,2

The final step in this process is to an ICU, where drugs and machines can, in theory, prolong life, no matter what the ultimate prognosis is.

The conveyor belt is also maintained and oiled by those who make the decision about whether the value of further medical intervention is futile. The diagnosis of dying, as defined by the well worn tenet “medicine at this time has nothing more to offer”, has moved from being made by general practitioners in community settings to hospital-based specialists. Deaths in ICUs are rarely sudden and unexpected.11 They are usually orchestrated as a result of withdrawing and withholding treatment after it has been decided, together with the patient and his or her carers, that further medical intervention has nothing else to give. Often, the underlying “disease” or combination of medicalised problems is simply the end result of normal and expected ageing.

Even in the ICU, the decision to continue active management may not be appropriate, as many of the survivors of the ICU do not leave hospital alive and many survivors of intensive care die within 12 months of discharge.12 For the cohort that does survive and leave hospital, there is a high incidence of a severe decrease in quality of life and of symptoms similar to those of posttraumatic stress disorder.13,14 Even if patients at the EOL survive the hospital intervention, little may have been done to improve posthospital survival or quality of life. Apart from the failure of the hospital system to recognise when people are at the EOL, there are other drivers pushing them along the conveyor belt. Society is bombarded with daily reports of medical miracles.15 There is little honest discussion about ageing and dying. Physicians seem to be complicit in this. It is sobering to reflect on the finding that most patients with terminal cancer may not be aware that the palliative chemotherapy they receive is unlikely to be curative.16

Rapid response systems and end-of-life care

Rapid response systems (RRSs) were established as a patient safety system to improve patients’ outcomes in acute hospitals.17 Their key features are vital sign and observational abnormalities which identify seriously ill patients and, in turn, trigger an urgent response by an individual or team with the appropriate skills, knowledge and experience to deal with any hospital emergency.17

Initially, they were established to identify seriously ill patients with potentially preventable illnesses. However, an RRS will also identify patients who are predictably and normally dying.18 Up to a third of all RRS calls are for patients who require limitations of treatment as a result of being at the EOL.18 This has important implications. It tells us that acute hospitals do not necessarily recognise patients at the EOL.9 The rapid response team becomes the surrogate “dying” team. The poor prognosis of the patient then has to be brought to the attention of the admitting team. It does not seem acceptable that patients only become aware of their parlous state when they are so close to death. The inappropriate management of patients at the EOL not only largely contributes to the unsustainable cost of health care but it is also not in the interest of patients and their carers to be denied rational and personal choices based on the prognosis.

Another approach to patients at the end of life

Managing patients in acute hospitals who are at the EOL may benefit from an approach similar to that used in RRSs. This would require identification of the patient followed by an appropriate response. There are currently attempts to identify such patients.19 Whatever tool is developed must deal with uncertainty, which is an integral part of medicine. However, a prognostic tool would equip the patient and the caring physicians with information such as the possibility of dying within, say, months or a year. We would not consider withholding poor prognostic information from a 20-year-old patient with a terminal brain tumour. The estimate of exactly how long that patient had to live would be uncertain in terms of weeks, months or even years in some instances. However, the same honest discussions usually do not occur with aged, frail patients who have had multiple admissions to hospital and who may have a similar prognosis to that young patient with the brain tumour.

It is early days in predicting prognosis in frail older people, but there are features (such as weight loss, significant decrease in mobility, gait speed, increasing weakness and exhaustion) which may indicate that a person has a high probability of having less than 12 months to live.19 Once a patient is identified as being in an at-risk group, there would need to be an urgent and appropriate response. As with RRSs, the response to the seriously ill, deteriorating patient would need to involve a clinician with appropriate skills, knowledge and experience, as well as the time to carry out lengthy and complex discussions with the patient and his or her carers. Currently, most admitting teams do not have that level of training.18 Moreover, the home team is often busy caring for the more conventional aspects of medical care. One could argue that we need an extensive rethink of our undergraduate and postgraduate training in order to give every clinician, including attending doctors and nurses, the skills to appropriately manage EOL care. Alternatively, or until we undergo that radical retraining, a response could be in the form of someone specifically trained in EOL care such as a palliative care nurse.

The response would obviously begin with involvement of the patient and their carers. Discussions would then be centred on the patients’ wishes, such as whether they want to be continually admitted to acute hospitals or whether they would prefer more home and community-based care. It would be essential to involve GPs in continuing management plans. These choices and the patient’s condition may, of course, change with time.

Other responses could include the offer of palliative care, perhaps in parallel with continued active treatment. Choices would include how patients would like to spend their last few months. Where they would like to spend them? What sort of support do they have? And what support do they need?

During these discussions, there would also need to be close communication with others in the palliative care team as well as members of the admitting team.

Conclusion

The current medicalisation of dying can be compared with the medicalisation of birthing in the 1950s, when mothers’ legs were put up in stirrups, babies were delivered and removed from their mothers and put in a large room together with other babies, and mothers were only allowed to see their babies for feeding. And, of course, fathers were excluded from the process altogether. Perhaps there is hope that the baby boomers, who have changed the way society operates in many ways, will demand greater choice and involvement in the way that their dying is handled.

We seem to have lost the ability to be honest with society about the limitations of modern medicine and to recognise that people will inevitably age and die badly. Currently, many patients at the EOL are placed on a conveyor belt where futile treatment and less than full disclosure have replaced genuine and appropriate care for patients and their carers.