Customize your JAMA Network experience by selecting one or more topics from the list below.
Nikpoor N, Butt AL, Hromas AR, Stone DU. Effect of the Physician Quality Reporting Initiative on Ophthalmologists' Documentation of Practice Patterns. Arch Ophthalmol. 2012;130(10):1351–1352. doi:10.1001/archophthalmol.2012.1459
Author Affiliations: College of Medicine (Ms Nikpoor and Dr Hromas), College of Public Health (Dr Butt), and Department of Ophthalmology, College of Medicine (Dr Stone), University of Oklahoma, and Dean A. McGee Eye Institute (Dr Stone), Oklahoma City.
Several interventions have been proposed as a means of making health care both more affordable and more evenly distributed among users.1-3 Although data exist on the impact of the Physician Quality Reporting Initiative (PQRI) on tracking health care quality with regard to general parameters4 (routine glycated hemoglobin levels, etc) and regarding ophthalmologists' adherence to practice patterns such as optic nerve examinations in patients with glaucoma,5 studies of the effect of pay for performance on ophthalmology are lacking. The purpose of this study was to determine whether implementation of the PQRI was associated with a change in documentation of ophthalmologists' practice patterns.
We performed a retrospective review of patient records generated by diagnosis code. The setting of the study was an academic ophthalmology group (approximately 26 health care providers during the periods being studied). Patients with diagnoses that would qualify for the PQRI were randomly selected from the physicians who qualified for the PQRI bonus during the first full year of implementation (2007), and a group of patients was also selected from nonqualifying physicians. A comparison was made of documentation before and after implementation of the PQRI as well as documentation for patients of PQRI-qualifying and nonqualifying physicians.
The categories were as defined by the PQRI. For the glaucoma group of diagnosis codes, documentation of an optic nerve examination was recorded as yes. Similarly, for diabetes there were 2 data points recorded: communication with the primary care provider and documentation of clinically significant macular edema. For macular degeneration, discussion of Age-Related Eye Disease Study6 vitamins and dilated macular examination were the end points. All statistics were calculated at α = .05 using SAS version 9.2 statistical software (SAS Institute, Inc).
This study was performed in accordance with the tenets of the Declaration of Helsinki, with institutional review board approval of retrospective data collection from the University of Oklahoma Health Science Center Office of Human Research Participant Protection.
Among patients in the PQRI-qualifying group, 1613 unique patients were identified. A total of 140 patients' records were reviewed and included for statistical analysis. The overall univariate (χ2) analysis results after pooling all the diagnosis codes demonstrated a small but statistically significant decline in the documentation of practice patterns in the pre-PQRI period compared with the post-PQRI period. Among the patients, 80.7% had documentation that met PQRI criteria in the year preceding implementation, compared with 75.7% in the first year of PQRI implementation (P < .001).
Among the patients in the nonqualifying physician group, 2774 unique patients were identified. A total of 267 patients' records from the nonqualifying group were reviewed and analyzed. The overall univariate (χ2) analysis results after pooling all the diagnosis codes demonstrated a small but statistically significant decline in the documentation of practice patterns in the pre-PQRI period vs the post-PQRI period. The compliance rates were 80.9% in the pre-PQRI year and 77.9% in the first year of PQRI implementation (P < .001).
Finally, the documentation of practice patterns before and during PQRI implementation was compared between the qualifying and nonqualifying groups using univariate as well as multivariate analysis. Univariate analysis revealed no statistically significant difference (P = .54). For multivariate analysis, we used a logistic regression model. Comparing the qualifying and nonqualifying groups, there was no statistically significant difference in the documentation of practice patterns before (P = .76) or during (P = .56) PQRI implementation.
Our results indicate that implementation of the PQRI did not result in an improvement in the documentation of practice patterns. The absolute numbers reported may not apply to other practice environments and may be biased by the inadequacies of a retrospective study using paper records. However, these findings support the hypothesis that the PQRI did not positively affect documentation and, by extrapolation, may not have influenced clinical care in a meaningful way. Given the tremendous amount of resources being devoted to the reformation of health care delivery in the United States, we propose that policy interventions should undergo the same rigorous outcomes testing and evidence-based implementation expected of any other aspect of health care delivery.7
Correspondence: Dr Stone, Dean A. McGee Eye Institute, 608 Stanton L. Young Blvd, Oklahoma City, OK 73106 (email@example.com).
Financial Disclosure: None reported.
Funding/Support: This study was supported in part by an unrestricted grant from Research to Prevent Blindness to the Department of Ophthalmology, University of Oklahoma and the Dean A. McGee Eye Institute.