Abstract WP85: Evaluating the Strengths and Limitations to Patient-Reported Modified Rankin Scale Studies: A Systematic Review

Abstract

Background: Studies evaluating patient-reported modified Rankin scales (mRS) in patients with stroke in different cultural and language settings have not been aggregated to determine validity of these adapted assessments. Objectives: To systematically review all patient-reported mRS studies performed outside of the United States (US). Methods: We searched PubMed and Web of Science for non-US based studies that evaluated patient-reported mRS against in-person clinician mRS among patients with stroke. Studies were analyzed for country of investigation, mRS self-assessment modality, care setting (inpatient vs outpatient), study limitations, and concordance between patient-reported and in-person mRS scores. Results: We identified 13 patient-reported mRS evaluation studies across all continents except Africa. The most common self-reported mRS assessment was the telephone mRS questionnaire. Average time between in-person and self-reported mRS assessments was ~14 days. The weighted kappa agreement between patient-reported and in-person mRS scores ranged from moderate to very good (K w range: 0.56 to 0.90), though unweighted kappa scores were significantly lower (K range: 0.27 to 0.68). Discrepancies between patient-reported and clinician-evaluated mRS scores tended to result from patients at the scale extremes. Patients with good mRS outcomes (mRS < 2) tended to rate themselves better than their clinicians while those with poor outcomes (mRS > 3) tended to rate themselves worse than their clinicians did. Major limitations across studies included sampling bias in the in-patient setting, time delays between in-person and self-reported mRS assessments, and differences in the structure of the clinician-led vs patient-reported mRS assessments. Conclusion: While patient-reported mRS validation studies to date have generally displayed good agreement with clinician-reported outcomes, significant scoring discrepancies exist for patients with very poor and very good outcomes. Future studies should consider limiting time between mRS assessments and understanding reasons for differences in patient and clinician assessments. Further, data on validity of self-report mRS tools are lacking from much of the world, especially Africa.