Prognosis-based Futility Guidelines
Does Anyone Win?
Published in: Journal of the American Geriatrics Society, v. 42, No. 11, Nov 1994, p. 1202-1207
Posted on RAND.org on December 31, 1993
OBJECTIVE: Advocates for health care reform and others claim that significant savings could be achieved if "futile" care were eliminated. Our objective was to provide an initial estimate of the effects of a public policy that would preclude futile life-sustaining treatments, defined as those employed despite < or = 1% chance of surviving for 2 months. DESIGN: Simulation using data from an observational cohort study. SETTING: Five academic medical centers. PATIENTS: Seriously ill hospitalized adults enrolled in the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatment (SUPPORT). METHODS: We examined the impact of prognosis-based futility guidelines on survival and hospital length of stay on a cohort of seriously ill adults. We calculated the number of days of hospitalization that would not be used if, on the third study day, life-sustaining treatment had been stopped or not initiated for subjects with estimated 2-month survival probability of < or = 1%. RESULTS: Of the 4301 patients, 115 (2.7%) had an estimated chance of 2-month survival of < or = 1%. All but one of these 115 subjects died within 6 months. Almost 86% died within 5 days of prognosis. At the time of death, 92 subjects (80.0%) had had no attempt at resuscitation; 35 (30.4%) had had a life-sustaining mechanical ventilator withdrawn. A Do-Not-Resuscitate order was written either before (n = 61) or within 5 days (n = 18) of reaching this prognosis for 68.6% of the patients. These 115 subjects had total hospital charges of $8.8 million. By forgoing or withdrawing life-sustaining treatment in accord with a strict 1% futility guideline, 199 of 1,688 hospital days (10.8%) would be forgone, with estimated savings of $1.2 million in hospital charges. Nearly 75% of the savings in hospital days would have resulted from stopping treatment for 12 patients, six of whom were under 51 years old, and one of whom lived 10 months. CONCLUSIONS: Patients at a high risk of dying can be identified prospectively. Implementation of a strict, prognosis-based futility guideline on the third day of a serious illness would result in modest savings.