Reliability of the Modified Rankin Scale Assessment Using a Simplified Questionnaire in Japanese

We developed a Japanese version of the simplified modified Rankin Scale questionnaire, of which the agreement between patients or their family members and neurologists was verified in a previous single-center study. In the present study, the reliability of the questionnaire was evaluated in another hospital.

The participants were prospectively-registered patients with acute ischemic stroke admitted to Saiseikai Kumamoto Hospital. The questionnaire was administered to the patients or their family members at discharge. At the same time, the attending physician assessed the Japanese version of a guidance scheme for the modified Rankin Scale. The agreement rate between the scores from both assessments was examined using the κ statistic and weighted κ statistic. We examined the factors related to disagreement between the modified Rankin Scale score by the attending physicians and the questionnaire score by patients/family members.

A total of 146 patients (age 77 ± 12 years; 53% men) were included. Agreement between the raters was 39%; the κ statistic was 0.27, but the weighted κ statistic, taking into account the extent of disagreement, was 0.81. In multivariate analysis, the disagreement between the modified Rankin Scale score and the questionnaire score was negatively associated with a modified Rankin Scale score of 5 at discharge (odds ratio, 0.02 per score point; 95% confidence interval, 0.00-0.34).

The questionnaire showed decent reliability and similarity to the modified Rankin Scale assessed by physicians. The questionnaire could be a reliable indicator for assessing stroke functional outcomes even when implemented by non-medical staff.