The study by Ebell and colleagues makes an important contribution to our understanding of the outcomes of hospital cardiopulmonary resuscitation (CPR). Although most prior work has focused on predictors of survival, Ebell et al focused on predictors of a good outcome, which they define as survival to hospital discharge with minimal neurologic disability. In a large group of patients who underwent CPR, they developed and validated the Good Outcome Following Attempted Resuscitation (GO-FAR) score, a prognostic index that has solid accuracy. The GO-FAR score identifies a large proportion of patients (28.3%) who have a low or very low likelihood (<2%) of achieving the good outcome defined by the authors. The variables in this prognostic index are widely available, enhancing the ability to apply this model to most hospitalized patients.
The GO-FAR score can be helpful in clinical decision making. As with any prognostic index, however, clinicians should use it in conjunction with, and not as a replacement for, their clinical judgment. Prognostic indexes provide useful estimates of the probability of various outcomes but often do not identify subtle patient characteristics, such as functional, cognitive, and psychological features, that may render the patient's chance of a good outcome either better or worse than that predicted by the index. Clinicians can and should use their clinical judgment to adjust the risk estimates from a prognostic index upward or downward. In addition, the GO-FAR score developed by Ebell et al needs to be used in conjunction with patient preferences. Although it identifies patients likely to survive with little neurologic disability, many patients would accept CPR if it allowed survival with more moderate or even severe forms of disability.
Cardiopulmonary resuscitation is just one of many preference-sensitive interventions we perform in the hospital. Too often, discussions of patient preferences and goals in the hospital begin and end with “code status.” However, the best clinicians work to understand the outcomes that matter most to their patients and make sure treatments are consistent with patients' overall goals. This requires getting to know which outcomes matter most to our patients. Such an understanding often is very influential in defining the benefits and risks of the interventions that we offer patients in the hospital.
Covinsky KE. Using the GO-FAR Score. JAMA Intern Med. 2013;173(20):1878. doi:10.1001/jamainternmed.2013.10783