Effect of decisions to withhold life support on prolonged survival

Background: The effect on long-term mortality of decisions made to withhold life-supporting therapies (LST) for critically ill patients is unclear. We hypothesized that mortality 60 days after ICU admission is not influenced by a decision to withhold use of LST in the context of otherwise providing all indicated care. Methods: We studied 2,211 consecutive, initial admissions to the adult, medical ICU of a university-affiliated teaching hospital. To achieve balanced groups for comparing outcomes, we created a multivariable regression model for the probability (propensity score [PSI) of having an order initiated in the ICU to withhold LST. Each of the 201 patients with such an order was matched to the patient without such an order having the closest PS; mortality rates were compared between the matched pairs. Cox survival analysis was performed to extend the main analysis. Results: The matched pairs were well balanced with respect to all of the potentially confounding variables. Sixty days after ICU admission, 50.5% of patients who had an order initiated in the ICU to withhold life support had died, compared to 25.8% of those lacking such orders (risk ratio, 2.0; 95% confidence interval, 1.5 to 2.6). Survival analysis indicated that the difference in mortality between the two groups continued to increase for approximately 1 year. Conclusion: Contrary to our hypothesis, decisions made in die ICU to withhold LST were associated with increased mortality rate to at least 60 days after ICU admission.