Related Experiment Videos
Pointwise linear regression for evaluation of visual field outcomes and comparison with the advanced glaucoma
Kouros Nouri-Mahdavi1, Joseph Caprioli, Anne L Coleman
1Glaucoma Division, Jules Stein Eye Institute, University of California Los Angeles, 90095-7004, USA.
Archives of Ophthalmology (Chicago, Ill. : 1960)
|February 16, 2005
Summary
Pointwise linear regression (PLR) shows similar glaucoma progression detection to Advanced Glaucoma Intervention Study (AGIS) criteria. PLR may be superior by identifying fewer visual field improvements, aiding in longitudinal visual field evaluation.
Area of Science:
- Ophthalmology
- Medical Statistics
Background:
- Longitudinal evaluation of visual fields is crucial for glaucoma management.
- The Advanced Glaucoma Intervention Study (AGIS) criteria are established for assessing visual field changes.
- Newer statistical methods may offer improved accuracy and efficiency.
Purpose of the Study:
- To investigate the utility of pointwise linear regression (PLR) for longitudinal visual field evaluation.
- To compare the diagnostic performance of PLR with existing AGIS criteria.
Main Methods:
- Analysis of 509 eyes from the AGIS database with at least 3 years of follow-up and 7+ visual field tests.
- Definition of visual field change based on threshold sensitivity (1 dB/y or higher, P<0.01).
- Investigation of various PLR criteria for identifying visual field series progression or improvement.
Main Results:
- Pairwise agreement between AGIS and PLR criteria ranged from 52% to 64% (kappa 0.22-0.30).
- PLR identified 30% of eyes as progressing, compared to 27% by AGIS.
- PLR identified only 2% of eyes as improving, versus 14% by AGIS.
Conclusions:
- Pointwise linear regression (PLR) detects glaucoma progression comparably to AGIS criteria.
- PLR may offer advantages over AGIS by identifying fewer instances of visual field improvement.
- PLR presents a potentially more sensitive and reliable method for longitudinal visual field analysis in glaucoma.