Abstract
BACKGROUND: While patients in the United States generally have final say in their code status, discussion with their physician plays an important role in decision-making. However, physicians do not discuss code status with every patient, and do not consistently mention patients' prognosis following cardiopulmonary resuscitation (CPR). Understanding how physicians perceive patients' CPR decisions is prerequisite to improving code status discussions. METHODS: We report a planned secondary analysis from a prospective randomized controlled trial of 102 English-speaking adults aged ≥65 evaluating whether "Allow Natural Death" was preferred to "Do Not Resuscitate" as the "no code" option in code status discussions. We measured physician agreement/disagreement with patient code status decisions and the correlation with objective outcome measures. Two clinically validated instruments-measuring likelihood of surviving resuscitation (Good Outcomes Following Attempted Resuscitations (GO-FAR)) and morbidity level/1- and 10-year mortality (Charlson Comorbidity Index (CCI))-were calculated for each participant. RESULTS: Physicians agreed with patients' code status decisions 88.3% of the time. Physician agreement with code status was not correlated with GO-FAR or CCI scores. GO-FAR and CCI scores do not always align, indicating that illness severity and CPR outcome are not directly linked. CONCLUSIONS: This study highlights that while physicians tend to agree with patient's code status, their decisions do not align with data from clinically validated predictors of coding success or illness severity/mortality prediction. Further research is required as to how physicians perceive whether attempting CPR is appropriate or not.