Purchase this article with an account.
Christian X Miller, Alexander G Miller, Mark Obri, Joan H Hornik, Douglas Y Rowland, David G Miller; Correlation of Medicare Patient Volume and Online Ratings for Retinal Physicians. Invest. Ophthalmol. Vis. Sci. 2019;60(9):5501.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To evaluate the correlation of retinal physician Medicare patient exams to online ratings in Ohio.
The American Academy of Ophthalmology (AAO; https://secure.aao.org/aao/find-ophthalmologist) and American Society of Retinal Specialist (ASRS; https://www.asrs.org/find-a-specialist) databases were used to generate a list of retinal specialists in Ohio. Subspecialties retina–medical only and retina/vitreous–medical and surgery were used to refine the search. Total Medicare patient exams for each physician in the year 2016 were gathered from the Center of Medicare and Medicaid Services (CMS) database: https://data.cms.gov/utilization-and-payment. Using the sum of the Common Terminology Codes (CPT) for eye exam codes (New patient 92002, 92004, Established patient 92012, 92014) and evaluation and management codes (New patient 99202-99205, Established patient 99211-99215), total CMS patient exams were tabulated. The number of ratings and average value of ratings were recorded from 4 different online physician rating services: google.com, vitals.com, healthgrades.com, and yelp.com. A cumulative summary was calculated for each physician. The statistical analysis used the Spearman rho rank correlation coefficient due to the presence of extreme values with p-values based on testing the null hypothesis of zero correlation.
A total of 90 retinal specialists had their data examined; 88 yielded data usable for analysis. For the cumulative number of ratings (average 30.15, median 16) vs. total CMS exams per physician (average 1,436.91, median 1,454.50), there was a close to significant correlation (0.190; p=0.08). For the average rating vs. total CMS exams per physician, there were significant Spearman rank correlations of about 0.3 found for both Google and Yelp although the latter encompassed only private practice. The cumulative correlations were slightly negative for this. Other correlations were nonsignificant and included some negative correlations for both average and number of ratings.
The cumulative number of CMS patient exams had a marginally significant correlation with the number of online ratings. The cumulative average rating for retinal physicians in Ohio seemed to be unrelated to the number of patients examined, although for each Google and Yelp ratings there were significant correlations, suggesting that different rating systems may have different characteristics.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.
This PDF is available to Subscribers Only