Mobile Phone-Based Questionnaire for Assessing 3 Months Modified Rankin Score After Acute Stroke: A Pilot Study

Circ Cardiovasc Qual Outcomes. 2015 Oct;8(6 Suppl 3):S125-30. doi: 10.1161/CIRCOUTCOMES.115.002055.


Background: In many countries, a majority of stroke patients are not assessed for long-term functional outcome owing to limited resources and time. We investigated whether automatic assessment of the modified Rankin Scale (mRS) based on a mobile phone questionnaire may serve as an alternative to mRS assessments at clinical visits after stroke.

Methods and results: We enrolled 62 acute stroke patients admitted to our stroke unit during March to May 2014. Forty-eight patients completed the study. During the stay, patients and/or caregivers were equipped with a mobile phone application in their personal mobile phones. The mobile phone application contained a set of 20 questions, based on the Rankin Focused Assessment, which we previously tested in a pilot study. Three months after inclusion, the mobile phone application automatically prompted the study participants to answer the mRS questionnaire in the mobile phones. Each question or a group of questions in the questionnaire corresponded to a certain mRS score. Using a predefined protocol, the highest mRS score question where the study participant had answered yes was deemed the final mobile mRS score. A few days later, a study personnel performed a clinical visit mRS assessment. The 2 assessments were compared using quadratic weighing κ-statistics. Mean age was 67 years (38% females), and median baseline National Institutes of Health Stroke Scale (NIHSS) score was 5 (interquartile range 2-10.5, range 0-23). Median and mean clinical visit mRS at 3 months was 2 and 2.3, respectively. We found a 62.5% agreement between clinical visit and mobile mRS assessment, weighted kappa 0.89 (95% confidence interval 0.82-0.96), and unweighted kappa 0.53 (95% confidence interval 0.36-0.70). Dichotomized mRS outcome separating functionally independent (mRS score 0-2) from dependent (mRS score 3-5) showed 83% agreement and unweighted kappa of 0.66 (95% confidence interval 0.45-0.87).

Conclusions: Mobile phone-based automatic assessments of mRS performed well in comparison with clinical visit mRS and could be used as an alternative in stroke follow-up.

Keywords: cell phones; outcomes research; prediction; questionnaires; stroke.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Acute Disease
  • Aged
  • Cell Phone*
  • Female
  • Humans
  • Male
  • Mobile Applications*
  • Pilot Projects
  • Reproducibility of Results
  • Research Design
  • Self-Assessment
  • Stroke / diagnosis*
  • Stroke / epidemiology
  • Surveys and Questionnaires
  • Sweden
  • Time Factors