Mapping PROMIS physical function and pain interference to the modified low back pain disability questionnaire.

Area of Science:

• Biomedical research
• Health outcomes assessment
• Musculoskeletal health

Background:

• The Modified Low Back Pain Disability Questionnaire (MDQ) is a standard tool for assessing low back pain (LBP) functional status.
• The Patient-Reported Outcomes Measurement Information System (PROMIS) offers an alternative platform for measuring patient-reported health in LBP.
• Mapping between these instruments is crucial for data integration and comparative analysis in LBP research.

Purpose of the Study:

• To establish reliable mapping equations between the MDQ and PROMIS Physical Function (PROMIS-PF) and PROMIS Pain Interference (PROMIS-PI) scales.
• To compare the accuracy of different statistical methods for score transformation.
• To provide researchers with tools for translating scores between the MDQ and PROMIS scales for LBP patients.

Main Methods:

• Retrospective analysis of 13,585 LBP patients' data from Cleveland Clinic.
• Application of linear regression, linear equating, and equipercentile equating to map scores between MDQ and PROMIS scales.
• Internal validation of mapping equations using root mean squared error (RMSE), mean absolute error (MAE), and correlation coefficients.

Main Results:

• Mapping methods incorporating both PROMIS-PF and PROMIS-PI demonstrated superior accuracy in estimating MDQ total scores.
• Equipercentile equating using MDQ items proved most effective for estimating individual PROMIS-PF and PROMIS-PI T-scores.
• The study successfully generated and validated equations for score translation between the MDQ and PROMIS scales.

Conclusions:

• Validated mapping equations between the MDQ and PROMIS-PF/PROMIS-PI scales were successfully developed.
• These equations facilitate the translation of scores, enabling researchers to integrate data from different measurement tools.
• The findings support the use of these validated maps for enhanced LBP research and data analysis.