Background and Purpose: Although the modified Rankin Score (mRS) is broadly used as a measure of stroke outcome, patients’ valuation of each mRS category has been understudied. Stroke patient/caregiver quality of life (QOL), expressed in 0-1 numerical utility scores, more accurately reflect patient-centered health states. We hypothesized that QOL utility values would not be linearly associated with mRS. Rather, clinically-meaningful transitions with large utility differences would occur among certain mRS scores.
Methods: Utilities were obtained from ischemic stroke patients or proxy family members in the ARTSS-2 randomized trial; clinicaltrials.gov/NCT01464788. A societal perspective for health-state assessments was performed using Time-Trade-Off methodology (EuroQol-5D). Utilities were converted using country-specific tariffs as ARTSS-2 had US & UK sites. Death was assigned a utility=0. Certified raters blinded to utility data assessed 90-day mRS. Mean (SD) utilities were mapped across mRS categories. mRS 5 & 6 were collapsed as 5, bedbound and dependent for all bodily needs, is accepted to be equal or worse than death (mRS=6).
Results: 86 of 90 (96%) patients had utilities at 90-days. Excluding missing and deaths, 70 patients had utilities generated by either the patient (N=48; 70%) or the proxy (N=22; 30%). Of the 22 proxy assessments, 20 (91%) were aphasic, 1 had severe neglect, and 1 remained mechanically ventilated. Utility vales declined in a non-linear fashion with increasing mRS; the largest difference between 3-4 and 4-5+6 (Figure).
Conclusions: Although the mRS is analyzed under a linear assumption that each numerical transition is equal, stroke patient QOL utilities demonstrate important transitions between mRS scores. Future stroke trials should consider replacing mRS with direct patient utilities or convert mRS into utility-weighted scores established by patient-generated values that reflect clinically important health states.