Predicting whether the ICU can help older patients: score needed

Because the need for intensive care exceeds its availability in several countries, intensivists must admit those patients most likely to benefit. Intensive care unit admissions of elderly patients will increase substantially in the near future. Decreased self sufficiency and quality of life are common after hospitalization in older patients and they may require discharge to a nursing home, although some patients feel that life in a nursing home would be worse than dying. We have much to learn about matching the use of life-supporting treatments to the health-related values of older patients. A specific outcome-prediction score for older patients would help improve quality of care.

Because the need for intensive care exceeds its availability in several countries, intensivists must admit those patients most likely to benefit. Intensive care unit admissions of elderly patients will increase substantially in the near future. Decreased self sufficiency and quality of life are common after hospitalization in older patients and they may require discharge to a nursing home, although some patients feel that life in a nursing home would be worse than dying. We have much to learn about matching the use of life-supporting treatments to the health-related values of older patients. A specific outcome-prediction score for older patients would help improve quality of care.
The demographic revolution that is sweeping across industrialized countries will dramatically increase the absolute number of older individuals over the next 40 years. Identifying patients most likely to benefit from critical care is essential, both to prevent suffering related to unnecessary treatments and to ensure optimal use of finite resources. Quality of life may be of greater concern for the elderly than for younger individuals.
In this issue of Critical Care, De Rootj et al. [1] review recent data on outcomes of elderly patients admitted to intensive care units (ICUs), discussing the impact of age, diagnosis, pre-admission functional status and patients' preferences regarding life-sustaining treatments. They suggest that developing a specific score for predicting long-term outcomes in elderly ICU patients would help intensivists identify those patients most likely to benefit from ICU admission and would aid decision-making with the patient or relatives. The outcome of elderly patients after ICU admission is a complex issue and the quality of survival may be as important as the quantity of survival. As highlighted by De Rootj et al., available outcome studies are heterogeneous, have included various age groups [2][3][4][5] and have rarely evaluated the influence of pre-hospital disability [4].
In this editorial, we will discuss two important aspects. Firstly, as emphasized by De Rootj et al., studies that have assessed outcomes of elderly patients have included admitted patients instead of patients triaged for ICU admission. Patients with severe impairments and the very oldest may be underrepresented among admitted patients, which would bias the results toward better outcomes. Although age per se is an important determinant of hospital mortality, the most important determinant of ICU mortality is severity of the acute illness, as shown by studies taking this important factor into account (see, for example [2]). Age and severe disability have major effects on long-term outcomes [4]. In studies of triaged patients [6][7][8][9][10], age was an independent factor of ICU refusal with non-surgical disease and self sufficiency. The impact of triage, the lack of scoring systems specifically designed for elderly patients and the poor performance of available scores [11,12] complicate comparisons of outcome studies. Evaluating the lack of chance possibly associated with denial of ICU admission in older patients is difficult, given the limited reliability of severity scores for predicting hospital mortality. In the overall triaged population, patients with intermediate severity of illness who are denied ICU admission have high standardized mortality ratios (ratio of observed mortality over predicted mortality) [6,7]. Triage is extremely challenging in older patients, as there is often a complex combination of acute and chronic diseases associated with physical impairments and psychological distress. In this setting, a specific score would be of great interest to quantify two key aspects: the lack of chance associated with ICU refusal and prediction of outcomes, including functional dependency, after ICU admission.
The second important aspect concerns one of the greatest challenges regarding care of the elderly, which is to assist patients in deciding what treatments they want to receive should they experience a life-threatening condition. Older

Maité Garrouste-Orgeas and Jean Carlet
Medical-Surgical ICU, Saint Joseph Hospital, 75014 Paris, France patients who survive critical illness at the cost of losing their self sufficiency are frequently distressed by the burden they feel that they impose on their family. As reported by De Rootj et al., the patient's wishes are usually unknown at the time of triaging to the ICU, for reasons including acute cognitive impairment, absence of advance directives, absence of a primary-care physician [13] or under-recognition of preexisting cognitive impairment [14]. Inaccurate estimates by intensivists of patients' willingness to receive life-sustaining treatment may adversely affect quality of care [15,16]. Characteristics of the patient, including self sufficiency and quality of life, show no correlation with wishes regarding ICU admission [17]. Even patients with only a small chance of recovery may want to be admitted [17]. In countries where medical paternalism remains an obstacle to patient autonomy, physicians must work harder on encouraging patients with chronic diseases to think about their preferences and to communicate their wishes to their doctors or relatives, in order to improve the match between what they want and what they receive.
Both the general public and healthcare professionals are interested in the epidemiology of outcomes, social burdens related to care and quality of life in older patients. Data on these points are useful for planning post-ICU rehabilitation programs. An improved scoring system would provide new knowledge on the prognosis of older patients with critical illness, thereby improving quality of care.