[Do online ratings reflect structural differences in healthcare? The example of German physician-rating websites]

Previous surveys have shown that patient satisfaction varies with the regional supply of physicians. Online ratings on physician-rating websites represent a relatively new instrument to display patient satisfaction results. The aim of this study was (1) to assess the current state of online ratings for two medical disciplines (dermatologists and ear, nose and throat (ENT) specialists), and (2) to analyze online derived patient satisfaction results according to the physician density in Germany.

We collected online ratings for 420 dermatologists and 450 ear, nose, and throat (ENT) specialists on twelve German physician-rating websites. We analyzed the online ratings according to the physician density (low, medium, high physician density). For this purpose, we collected secondary data from both physician-rating websites and the regional associations of statutory health insurance physicians. Data analysis was performed using Median tests and Chi-square tests.

In total, 10,239 online ratings for dermatologists and 8,168 online ratings for ENT specialists were analyzed. Almost all dermatologists (99.3 %) and ENT specialists (98.9 %) were listed on one of the physician-rating websites. A total of 93.5 % of all listed dermatologists and 96.9 % of ENT-specialists were rated on at least one of the physician-rating websites. Significant differences were found in the distribution (i.e., percentage of listed or rated physicians) of the ratings according to the regional physician density on only one physician-rating website (p<0.001). Furthermore, online ratings were shown to be better in regions with a higher physician density on two physician-rating website. On jameda.de, for example, dermatologist ratings were better in regions with a higher physician density compared to regions with a lower number of physicians (average rating: 2.16 vs. 2.67; p<0.001).

Online ratings of dermatologists and ENT specialists hardly differ in terms of regional physician density. Physician-rating websites thus do not appear to be appropriate to mirror differences in the health service delivery structure. Our findings thus do not confirm the results from previously published studies.