Related Experiment Video
Updated: Jun 4, 2026
11:35
The Multiple Sclerosis Performance Test (MSPT): An iPad-Based Disability Assessment Tool
Published on: June 30, 2014
Managing missing scores on the Roland Morris Disability Questionnaire
Peter Kent1, Henrik Hein Lauridsen
1Research Department, Clinical Locomotion Science Network, Spine Centre of Southern Denmark, Hospital Lillebaelt, Middelfart, Denmark. peter.kent@slb.regionsyddanmark
Spine
|February 10, 2011
Summary
The Roland Morris Disability Questionnaire (RMDQ) benefits from proportional recalculation for missing data, offering a more accurate score than simply ignoring unanswered questions, especially with multiple missing responses.
Area of Science:
- Medical assessment and patient-reported outcomes.
- Spinal disorders and disability measurement.
Background:
- The Roland Morris Disability Questionnaire (RMDQ) and Oswestry Disability Index (Oswestry) are commonly used to assess disability.
- Ignoring unanswered questions is a frequent method for RMDQ score calculation.
- Oswestry scores are proportionally recalculated to accommodate missing data.
Purpose of the Study:
- To determine the prevalence of unanswered questions on the RMDQ (23-item version) and Oswestry questionnaires.
- To compare the accuracy of proportional recalculation versus ignoring missing data for RMDQ scores.
Main Methods:
- Prevalence of unanswered RMDQ and Oswestry questions was assessed in research and routine care settings.
- The accuracy of RMDQ proportional recalculation was evaluated using 311 complete questionnaire sets.
- Error was calculated by systematically omitting questions and comparing scores, with Wilcoxon Tests used for analysis.
Main Results:
- Prevalence of unanswered questions: 29.5% (RMDQ, routine care), 20.3% (RMDQ, research), and 13.9% (Oswestry).
- Proportional recalculation yielded more accurate RMDQ sum scores than ignoring missing data when two or more questions were unanswered.
Conclusions:
- Converting RMDQ sum scores to a 0-100 scale using proportional recalculation minimizes error when missing data are present.
- The recommended method involves scaling based on the number of answered questions for accurate RMDQ scoring.
