Abstract
Purpose:
To compare the results of alternative methods of measuring provider performance in the care of patients with open-angle glaucoma (OAG). While chart abstraction is the current gold standard, the expense and need for training limit its utility. We investigated the ability of case vignettes to provide similar information about performance of key examination steps by optometrists enrolled in an ongoing clinical trial.
Methods:
Community-based optometrists who agreed to participate in a trial assessing the effectiveness of electronic methods to improve the quality of glaucoma care completed a baseline survey that included case vignettes. Self-report of examination steps captured in an open-ended response format in the survey was compared to the results of chart abstractions for patients cared for by these providers prior to the initiation of any interventions. We report a comparison of the results for gonioscopy, a critical step in the care of patients with OAG and one that is known to be underperformed by all eye care providers..
Results:
Depending on the criteria used to determine if gonioscopy would be performed, 83 to 93% of respondents reported that they would typically include gonioscopy in their new patient evaluation of open-angle glaucoma. Chart abstractions from the patients of participating providers indicated that gonioscopy could be found on average in less than 2% of charts (SD of 4%) among those who did not indicate they performed routine gonioscopy, compared to 11% (SD of 14%) among those that indicated they would do routinely perform gonioscopy.
Conclusions:
Open-ended clinical vignette responses have been found to be reliable relative to chart abstractions in general medicine. For the performance of gonioscopy by optometrists, self-reported performance is associated with a greater likelihood of actual documentation in patient charts but not with the absolute level of performance. However, while potentially usable as a relative indicator, self-reported responses on vignettes may not be helpful for quantitative assessment. Additional analyses are under way for additional elements of the history and exam elements for OAG patients.
Keywords: 460 clinical (human) or epidemiologic studies: health care delivery/economics/manpower •
468 clinical research methodology