Related Experiment Videos
Interobserver agreement on visual field progression in glaucoma: a comparison of methods
A C Viswanathan1, D P Crabb, A I McNaught
1Institute of Ophthalmology, London, UK.
The British Journal of Ophthalmology
|May 29, 2003
Summary
Clinician agreement on visual field progression is poor with standard Humphrey printouts. Using the PROGRESSOR software significantly improves agreement for assessing visual field deterioration.
Area of Science:
- Ophthalmology
- Clinical Optometry
Background:
- Assessing progressive visual field deterioration is crucial for managing glaucoma and other optic neuropathies.
- Current methods, like standard Humphrey printouts, may lead to subjective interpretations and poor inter-clinician agreement.
Purpose of the Study:
- To compare the level of agreement between clinicians when assessing visual field series using standard Humphrey printouts versus the PROGRESSOR software.
- To determine if a specific analysis method enhances diagnostic consistency.
Main Methods:
- Five expert clinicians evaluated 27 visual field series (16 fields each, excluding the first three).
- Assessments were performed using both standard Humphrey printouts and the PROGRESSOR (pointwise linear regression) software.
- Inter-clinician agreement was measured using the weighted kappa statistic.
Main Results:
- Clinician agreement was significantly higher when using PROGRESSOR (median kappa = 0.59) compared to standard Humphrey printouts (median kappa = 0.32).
- This difference in agreement was statistically significant (p = 0.006).
Conclusions:
- Standard Humphrey printouts result in poor agreement among expert clinicians for visual field progression assessment, even with ideal data.
- The PROGRESSOR software demonstrates superior performance, leading to closer agreement on visual field progression status under ideal conditions.