Using Computerized Adaptive Testing to Measure Disability in Low Back Pain Patients
Author Information
Author(s): Elhan Atilla Halil, Öztuna Derya, Kutlay Şehim, Küçükdeveci Ayşe A, Tennant Alan
Primary Institution: Ankara University
Hypothesis
Can computerized adaptive testing (CAT) effectively measure disability in patients with low back pain?
Conclusion
The study demonstrated that CAT can effectively calibrate items onto a single metric for measuring disability in low back pain patients.
Supporting Evidence
- Factor analysis identified two dimensions of disability: body functions and activity-participation.
- Reliability exceeded 0.90 for both dimensions.
- Disability levels from CAT were highly correlated with those from original questionnaires.
Takeaway
This study shows that a computer program can help doctors understand how much back pain affects a person's daily life without asking too many questions.
Methodology
The study involved 399 patients who completed various questionnaires, followed by exploratory factor analysis and Rasch analysis to develop a CAT.
Potential Biases
Potential bias due to the educational and income gradient affecting patient responses.
Limitations
The study's sample size may be lower than average for CAT standards, and the dimensionality of the item banks showed some fragility.
Participant Demographics
Mean age of 52.2 years, 16% men, with a mean complaint time of 8.24 years.
Statistical Information
P-Value
0.0136
Confidence Interval
4.2%–9.4%
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website