Abstract TP218: Evaluating Concordance of Simplified Modified Rankin Scale Assessments with Traditional mRS Post IV Thrombolysis and Thrombectomy

Background: Functional outcome measurement using the modified Rankin Scale (mRS) is widely used in post-stroke quality assessments in clinical trials. Abbreviated versions such as the Simplified mRS Questionnaire (smRSq) may improve time efficiency and scalability. This study evaluates agreement between smRSq and traditional mRS. Methods: We retrospectively reviewed follow-up mRS calls placed 75–105 days post-discharge among ischemic stroke patients treated with IV thrombolysis and/or thrombectomy at a comprehensive stroke center (CSC). A total of 147 patients were called: 57 by a remote centralized group of trained regional LVNs (smRSq) and 113 by administered by the local CSC stroke coordinator RNs (traditional mRS). Among these, 54 patients were reached and assessed by both the remote LVN and the local CSC RN. Score agreement was analyzed using frequency distributions, percent agreement, and Cohen’s kappa. Results: Of the 147 patients called, 54 (36.7%) completed both simplified and traditional mRS assessments. This subgroup had a mean age of 69.2 ±11.3 years, 44.4% were female, and 44.4% were Hispanic. Graph 1 A and B depicts the distribution of patients in each category and the score differences between them. SmRSq had a higher mean score than trmRS (1.89± 1.63 vs 1.74± 1.82). Simple agreement was seen in 70.4% (38 of 54) with kappa 0.617 (Z = 8.36, p < .001), indicating substantial inter-method agreement (Table 1). Category-specific kappa ranged from 0.53 to 0.72; highest in mRS 0 (κ = 0.72) and lowest in mRS 1 (κ = 0.53). Agreement could not be calculated for mRS 2 due to low frequency. Conclusion: The simplified mRS demonstrated substantial agreement with traditional mRS assessments. Given the demonstrated agreement between the remote LVN and the local CSC RN, systems may consider using a remote LVN team to capture the 90 day mRS which may help to mitigate costs. These findings support the simplified tool as a valid, scalable method for post-stroke functional outcomes in healthcare systems with limited resources.