Tara Lagu, Katherine Metayer, Michael Moran, Leidy Ortiz, Aruna Priya, Sarah L. Goff, Peter K. Lindenauer. Website Characteristics and Physician Reviews on Commercial Physician-Rating Websites. JAMA. 2017;317(7):766–768. doi:10.1001/jama.2016.18553
Patients are increasingly seeking information about physicians online. Nearly 60% report that online reviews are important when choosing a physician.1 Because publicly reported quality data are not reported at the physician level, patients must consult physician-rating websites to find such reviews.2 The purpose of this cross-sectional study was to describe the structure of commercial physician-rating websites and the quantity of physician reviews on these sites.
During September 2016, we searched Google for websites that allowed patients to review physicians in the United States, using search terms such as rate my doctor. We included active sites that were written in English, available to the public, allowed patients to leave reviews, did not require a subscription, and allowed searching by physician name. We excluded websites that were affiliated with an insurance company or health system or were limited to a single specialty. We cross-referenced search results against a published list,3 added any websites that met inclusion criteria, and recorded website characteristics. We then used publicly available lists of registered and active physicians to identify a random sample of 600 physicians from 3 metropolitan areas (Boston, Massachusetts; Portland, Oregon; and Dallas, Texas). We searched each website for reviews and calculated mean and median number of reviews per physician per site using SAS (SAS Institute), version 9.3.
We identified 66 potential sites, of which 28 met inclusion criteria (Table 1). We excluded 5 nonfunctional sites, 8 without reviews, 9 that required subscriptions, 7 that did not allow searching by name, and 9 that were limited to a single specialty or insurance plan. Most sites (n = 26; 93%) included an overall “star” rating (hereafter referred to as quantitative reviews) and collected narrative comments (n = 28; 100%). Two sites that collected narratives did not post them. Few sites allowed the user to search by clinical condition (n = 5; 18%), sex of physician (n = 4; 14%), hospital affiliation (n = 15; 54%), languages spoken (n = 3; 11%), or insurance accepted (n = 9; 32%).
Across 28 websites, there were 8133 quantitative reviews (1784 with narrative comments) for the 600 physicians. Among physicians with at least 1 review on any site, the median number was 7 reviews (interquartile range [IQR], 2-20) per physician across all sites. We report median number of reviews per physician per site in Table 2. One-third of sampled physicians (34%) did not have a review on any site.
Commercial physician-rating websites have significant limitations. Search mechanisms are cumbersome and reviews scarce. The number of physician reviews online appears to be increasing4 (a similar 2009 study revealed only 190 reviews for 300 physicians across 33 sites, with 73% of physicians having no review on any site).2 However, the increase in number of reviews that we observed was not meaningful; most physicians in 2016 still had no more than 1 review on any site. This study is limited by small sample size, inability to limit the denominator to practicing physicians, and inability to assess the effect of reviews on patients or physicians. We did not attempt to analyze quantitative ratings because of the paucity of reviews per physician and variation in scales across sites. Despite these limitations, these results demonstrate that it is difficult for a prospective patient to find (for any given physician on any commercial physician-rating website) a quantity of reviews that would accurately relay the experience of care with that physician.
Given the demand by consumers for information about physicians, other methods for publishing patient feedback are being developed,5 and some health systems are beginning to report quantitative reviews and narratives drawn from patient experience surveys.6 Because of the scarcity of reviews on commercial sites, one of these other methods of publishing patient feedback may emerge as the dominant route by which patients seek reviews about physicians. Methods that use systematic data collection (eg, surveys) may have a greater chance of amassing a sufficient quantity and quality of reviews to allow patients to make inferences about patient experience of care.
Corresponding Author: Tara Lagu, MD, MPH, Center for Quality of Care Research, Baystate Medical Center, 280 Chestnut St, Third Floor, Springfield, MA 01199 (firstname.lastname@example.org).
Author Contributions: Dr Lagu had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Lagu, Metayer, Moran, Goff, Lindenauer.
Acquisition, analysis, or interpretation of data: Lagu, Metayer, Moran, Ortiz, Priya.
Drafting of the manuscript: Lagu, Metayer, Moran, Ortiz.
Critical revision of the manuscript for important intellectual content: Lagu, Metayer, Ortiz, Priya, Goff, Lindenauer.
Statistical analysis: Metayer, Moran, Priya.
Obtained funding: Lagu.
Administrative, technical, or material support: Lagu, Lindenauer.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Lagu reports receiving personal fees from the Institute for Healthcare Improvement and the Journal of Hospital Medicine. Dr Lindenauer reports receiving salary support from the Yale New Haven Hospital and Center for Outcomes Research and Evaluation under contract from the Centers for Medicare & Medicaid Services to develop hospital outcome measures. No other disclosures were reported.
Funding/Support: This work is supported by grants K01HL114745 (Dr Lagu) and 1K24HL132008 (Dr Lindenauer) from the National Heart, Lung, and Blood Institute and grant K23HD080870-01A1 from the National Institute of Child Health and Development (Dr Goff) of the National Institutes of Health.
Role of the Funder/Sponsor: The funders had no role in design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and no role in the decision to submit the manuscript for publication.
Additional Contributions: We thank Caroline Norton; Lindsey Russo, BS; and Jessica Meyers, BS (all from University of Massachusetts), for their help with manuscript preparation and submission. All contributors received compensation.