Purpose: Patient satisfaction survey data and physician reviews are available on the Internet. However, there is no published assessment of the information provided by these sites, relative usage of them, or principles of best practices.
Method: We performed Internet searches in February 2011 for websites that presented patient satisfaction survey data or physician reivews. We expected that high quality sites would report the total number of reviews, confirm that data were obtained from people who were patients of the physician, and have documented methodology. From each site, we systematically abstracted data on 4 characteristics: 1) general characteristics of the site (e.g. whether the site was a healthcare specific site), 2) types of providers reviewed, 3) characteristics of the population providing the reviews, and 4) characteristics of the survey/reviews data. We assessed the US Alexa Traffic Rank for each site and compared the characteristics of the five most frequently visited sites to the other sites using t-tests with Bonferroni correction.
Result: We identified 20 sites that presented patient satisfaction survey or review data for physicians. The US Alexa Traffic Rank ranged from 3 (Yahoo Local) to 398,324 (Find a Doc). Sixteen sites reported patient satisfaction survey data, 16 sites presented physician reviews, and 13 presented both. The five most frequently visited sites were less likely to report the number of total reviews (1/5 vs 4/15 p=0.38), confirm that a reviewer was an actual patient of the physician (0/5 vs 2/14 p=0.87), or have an accompanying description of their methodology (0/5 vs 2/15 p=0.73). The five most frequently visited sites were more likely to be healthcare specific (1/5 vs 4/15 p=0.94), review allied health providers in addition to physicians (4/5 vs 11/15 p=0.19), use a 5 star/point rating system (5/5 vs 8/15 p=0.67), allow the physician to respond to the review (3/5 vs 4/9 p-0.98), have an overall rating score (5/5 vs 11/14 p=0.12), and thave an associated disclaimer (5/5 vs 14/15 p=0.001). The five most frequently visited sites were less likely to allow prose reviews (4/5 vs 12/13 p=0.02).
Conclusion: Traffic to patient satisfaction survey and rating websites is high. Providers need to remain aware of their online reputations. It is concerning that the most highly visited sites do not meet even relatively low quality thresholds.
See more of: The 33rd Annual Meeting of the Society for Medical Decision Making