Surrogates' agreement with patients' resuscitation preferences: effect of age, relationship, and SUPPORT intervention. Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatment

Abstract

Purpose: The purpose of this study was to evaluate an intervention to improve patient-surrogate agreement on end-of-life resuscitation preferences.

Materials and methods: Seven hundred seventeen patients with a 50% 6-month survival rate and their surrogate decision-makers were recruited for a randomized clinical trial from five teaching hospitals participating in the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT). Intervention patients (n=386) were assigned specially trained nurses who spent extra time with patients and families explaining prognoses and treatments. Control patients (n=331) received usual care. Patient preferences and surrogate's perception of those preferences at pre- and post-intervention interviews were compared.

Results: Agreement between patients and surrogates was 75.0% at the day 3 interview and 79.6% at the month 2 interview, increasing 4.6% (95% CI: 0.1%, 9.1%). Improvements in agreement from day 3 to month 2 were seen equally in both study groups. A multivariable analysis verified that the intervention did not have an effect on agreement and indicated a decrease in agreement among older patients and among surrogates not in the immediate family.

Conclusions: The SUPPORT intervention was not successful in increasing agreement between patients and surrogates. Because of the complex issues involved in end-of-life decision-making, a more aggressive intervention may be needed. Other findings suggest that improvements in communication are particularly needed when patients are older and when the surrogate is not a patient's immediate relative.