Related Experiment Videos
Rehabits: a common language of functional assessment
W P Fisher1, R F Harvey, P Taylor
1Center for Rehabilitation Outcome Analysis, Marianjoy Rehabilitation Hospital and Clinics, Wheaton, IL.
Archives of Physical Medicine and Rehabilitation
|February 1, 1995
Summary
Rasch measurement models can unify functional assessment tools like the Functional Independence Measure (FIM) and Patient Evaluation and Conference System (PECS). This study demonstrated their cocalibration into a single "rehabilitation measuring unit" (rehabit) for accurate conversion and reliable use.
Area of Science:
- Rehabilitation Medicine
- Psychometrics
- Health Services Research
Background:
- Functional assessment instruments, such as the Functional Independence Measure (FIM) and Patient Evaluation and Conference System (PECS), are crucial for evaluating patient progress in rehabilitation.
- Linking different assessment tools into a unified measurement system can enhance data comparability and utility.
- Probabilistic measurement models, like those developed by Rasch, offer a framework for such integration.
Purpose of the Study:
- To cocalibrate the FIM and PECS instruments using Rasch measurement models.
- To establish a common unit of measurement, the 'rehabit' (rehabilitation measuring unit), for both instruments.
- To assess the reliability and validity of the cocalibrated measures for functional assessment.
Main Methods:
- Fifty-four rehabilitation patients (diagnoses: brain injury, neuromuscular, musculoskeletal, spinal cord, stroke) were assessed at admission and discharge using both FIM and PECS.
- Thirteen FIM items and 22 PECS motor skills items were jointly scaled into a 35-item instrument.
- Rasch probabilistic measurement models were applied to cocalibrate the items and derive common scale values.
Main Results:
- The cocalibrated 35-item instrument provided scale values in a common unit (rehabit).
- Separate FIM and PECS measures correlated highly (.94 and .91, p < .0001) with the cocalibrated measures.
- FIM and PECS measures also showed strong correlation with each other (0.91, p < .0001).
Conclusions:
- Cocalibration of FIM and PECS using Rasch models successfully created a unified measurement system.
- The common unit, 'rehabit', allows for easy and accurate conversion between FIM and PECS ratings.
- The stability of item difficulty estimates across diverse populations and settings supports the use of these cocalibrated values as a reliable 'functionometric ruler'.