Long-term survival, quality of life, and quality-adjusted life-years among critically ill elderly patients

Objectives: To assess mortality, quality of life (QOL), and quality-adjusted life-years (QALYS) for critically ill elderly patents. Design: Cross-sectional survey. Setting. A ten-bed medical-surgical intensive care unit (ICU) in a tertiary care university hospital. Patients. The study group included 882 elderly patients ( 65 yrs of age) and 1,827 controls (<65 yrs of age) treated during the period of 1995 to 2000. Intervention: None. Measurements and Main Results., Mortality was assessed during the ICU and hospital stays, and 12, 24, and 36 months after ICU discharge. The cumulative 3-yr mortality rate among the elderly (57%) was higher (p <.05) than that among the controls (40%). The majority (66%) of the elderly nonsurvivors died within 1 month after intensive care discharge. All elderly patients with day-1 Sequential Organ Failure (SOFA) scores > 15 died during the ICU stay. QOL was assessed with EQ-5D and RAND-36 measures from 10 months to 7 yrs after discharge. The majority (88%) of the elderly survivors assessed their present health state as good or satisfactory; 66% found it to be similar or better than 12 months earlier, and 48% similar or better than their preadmission state. QOL measures by RAND-36 revealed that aging decreased their competencies most in physical functioning, physical role limitations, and vitality, but the elderly had better values in mental health than the controls. However, QALYs of the elderly respondents were 21% to 35% lower than the mean QALY minus 2 so units of the age- and gender-adjusted general population. Conclusions. High age alone is not a valid reason to refuse intensive care, but the benefits perceived by intensive care seem to decrease with aging, if reflected as QALYs. However, 97% of the elderly survivors lived at home and 88% of them considered their QOL satisfactory or good after hospital discharge. Therefore, more reliable information on the outcome for the elderly is clearly needed.