In the second paper in The Lancet Series on Critical Care, the large variations in end-of-life care in the ICU are discussed, along with the dynamics of the doctor-family interactions that directly affect care. The authors, led by Professor J Randall Curtis, Immediate Past President of The American Thoracic Society, say that doctors must be vigilant about decisions to withdraw life support possibly becoming 'routine', and that 'clinicians must be careful to guard against the subtle institutional pressures to withdraw life support' while at the same time, ensure patients don't receive care that they would not want.
Critical care is an integral part of hospital care, and the intensive care unit (ICU) is the setting where patients are given the most technologically advanced life-sustaining treatments. These treatments are expensive and resource-intensive, but can sustain life despite severe and multiple organ dysfunction. The ICU is, however, also a setting where death is common and end-of-life care is frequently provided. Since the focus in ICUs is on sustaining life, the delivery of high quality end-of-life care can be particularly challenging, and clinicians often find the dual responsibilities of saving lives and delivering end-of-life care difficult.
Possible reasons for variability of care in this setting could be differences in religion, legislation and culture, organisation of care in the ICU, attitudes of physicians toward end-of-life care, severity of illness and case mix, and the physician's predictions of prognosis and future quality of life. Variability also exists within countries and between intensivists within hospitals.
As discussed in paper 1, ageing populations will increase the demands on ICU care. A fifth of all deaths in the USA occur in the ICU, and this proportion does not decrease with increasing age until age 85 years. The authors say: "Society and countries need to develop approaches to address the appropriate delivery of critical care to the increasing population of elderly people, especially those with chronic life-limiting disease."
There is wide variation in levels of interaction within medical teams and between those teams and families dependent on location. For example, in a questionnaire study of 1961 intensivists from 21 countries, for a hypothetical patient without any family, 62% of physicians from northern and central Europe would involve nurses in end of- life discussions compared with only 32% of physicians in southern Europe, 39% in Japan, 38% in Brazil, and 29% in the USA. Poor interdisciplinary collaboration about end-of-life care is associated with increased symptoms of burnout, depression, and post-traumatic stress among clinicians working in the ICU.
In the Ethicus study, done in 37 ICUs in 17 European countries, end-of-life decisions were discussed with the family more commonly in northern (84%) and central (66%) than in southern (47%) Europe. Huge variations have been reported in family involvement—from 100% in India, 98% in Hong Kong, 79% in Lebanon, 72% in Spain, to just 44% in France.
The authors say: "Physicians need to be aware of the variety and complexity of attitudes present in our increasingly multicultural society, and adapt their approach to the situation." They propose a potential approach to decision making in the ICU, in which first the doctor assesses the prognosis and the certainty of the prognosis. The worse the prognosis and the greater the certainty, the more physicians should be willing to take on the burden of decision-making. The preferred role of the family should be assessed next, and then the final approach is based on the results of these first two steps. Ideally, shared decision making with joint responsibility will be the outcome, but this should be adapted to the patient circumstances and family preferences. Communication is vital for all patients in ICU, not just those expected to die. The authors report that, paradoxically, Families of patients who survive are less satisfied with communication from ICU clinicians than are those of patients who die.
The authors also discuss key elements of what would be a successful interaction between the ICU care team and the patient's family, including ensuring a private location for discussions, doctors allowing family members to ask questions, assuring them that the patient will not be abandoned, and focusing on the patient's own treatment preferences and values.
There also differences in some doctors' attitudes to withdrawing life support already initiated, compared with withholding treatment yet to start. One study showed that, in northern Europe, 47% of deaths follow withdrawal or withholding treatment compared to just 18% in southern Europe.
Religion is an important determinant of attitudes toward dying, death, and end-of-life care, and includes the religion of patients, their families, and their clinicians. For example, in the Ethicus study, treatment was withheld more often than it was withdrawn if the physician was Jewish (81%), Greek orthodox (78%), or Muslim (63%), whereas withdrawal occurred more often when physicians were Catholic (53%), Protestant (49%), or had no religious affiliation (47%). Religion is also an important determinant of acceptance of brain death, a state that is widely, but not universally, accepted.
The authors note that increasing numbers of deaths are occurring in ICUs and there is a lack of definitive guidelines to help make vital decisions on withholding or withdrawing life support. They say: "Withdrawal of these treatments is a clinical procedure that deserves the same preparation and expectation of quality as do other procedures. These decisions can become routine for clinicians working in the ICU, and,as such, clinicians must be careful to guard against the subtle institutional pressures to withdraw life-sustaining treatments. Rationale for the decision to withdraw life support should be noted in the medical record."
The authors conclude: "Development of global consensus about end-of-life care, to the extent possible, will require open and continued discussion of these issues in international forums. In all regions, the delivery of ethical and high-quality critical care requires training and emphasis on ethical decision making, communication and collaboration throughout the interdisciplinary team, effective communication with patients and families, and identification and resolution of conflict within the team and with patients and families."
Professor J Randall Curtis, Immediate Past President of The American Thoracic Society and Harborview Medical Center, University of Washington, WA, USA. T) +1 206 744 3356 E) jrc@u.washington.edu
For full Series paper 2, see: http://press.thelancet.com/cc2.pdf
NOTE: THE ABOVE LINK IS FOR JOURNALISTS ONLY; IF YOU WISH TO PROVIDE A LINK TO THE FREE ABSTRACT OF THIS PAPER FOR YOUR READERS, PLEASE USE THE FOLLOWING, WHICH WILL GO LIVE AT THE TIME THE EMBARGO LIFTS:
http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(10)60143-2/abstract
Journal
The Lancet