Mark N. Rubin, MD
Chaisinanunkul N, Adeoye O, Lewis RJ, Grotta JC, Broderick J, Jovin TG, et al. Adopting a Patient-Centered Approach to Primary Outcome Analysis of Acute Stroke Trials Using a Utility-Weighted Modified Rankin Scale. Stroke. 2015
The modified Rankin Scale (mRS) is the unequivocal reigning champion of outcome measurement in acute stroke trials, at least in North America but seemingly internationally. It represents a digestible, easy to understand scale that, decades of use later, is the common outcome parlance in the acute stroke treatment literature. The premise of the gradations, however, are based on an evaluation by a trained healthcare professional and the point-by-point changes on the scale are arbitrary; does jumping from a 0 to 1 have the same relative meaning to a patient as jumping from a 3 to a 4?
Many of the titans of acute stroke treatment trials came together in this project to suggest a relative weight to different mRS grades from a patient perspective, and apply these to various acute stroke trials spanning the last twenty years to see how this new scale compares to the more typically employed strategies of dichotomized and/or ordinal shift in mRS. The scale itself was given its patient-centered bona fides by being derived from mapping mRS scales against patient-provided quality of life measurements from the European Quality of Life Scale (EQ-5D) and medical-professional-determined disability weighting from the World Health Organization Global Burden of Disease (WHO-GBD) project. Overall, they found the Utility Weighted mRS (UW-mRS) to perform well, with similar “statistical efficiency” as ordinal analysis of mRS and even better than mRS dichotomization to recognize the benefit (or lack thereof) of the trialed interventions. An added bonus is that statistical analysis was straightforward with t-tests.
The authors present an important point here. Trial outcome reporting has to mirror the shared goals of the patient, clinician and researcher: “how much better will mom do if she undergoes _____?” All discussions are predicated on goals of care, and no one knows the goal better than the patient or surrogate, and it is reasonable to suggest that our studied outcome measures should be weighted with generally accepted views of quality of life and disability.