Exploring Bias in Online Physician Reviews: The Impact of Gender, Experience, and Geography in Urogynecology

INTRODUCTION: Physician rating websites are highly prevalent in the United States. Emerging data suggest that patients' implicit biases of providers’ gender may impact how they rate their surgeon. Online reviews are increasingly influential on patients as they choose healthcare providers, yet the extent to which they might be biased by non-clinical physician characteristics—such as gender, geographic region, and years of experience—remains underexplored. OBJECTIVE: The goal of this study is to explore potential biases in online physician reviews. METHODS: We identified all physicians in the NPI Registry with the Health Care Provider Taxonomy code for Urogynecology and Pelvic Surgery, 207VF0040X. We then cross-referenced the CMS database to extract self-identified gender, year of medical school graduation, and zip code. Providers were grouped into regions, as defined by the U.S. Department of Labor, based on the location of their current practice. The preliminary analyses reported here focused on physicians in the northeast United States. The name of each physician was queried on the largest commercial review website, healthgrades.com, to identify the number of reviews and number of 1-star, 2-star, 3-star, 4-star, and 5-star reviews they received. Analysis was performed using SPSS, version 27. RESULTS: From the NPI registry, we initially identified 294 providers located in the Northeast. Of these providers, 256 were listed on healthgrades.com, and 239 providers had patient ratings available. Almost 61% (n=145) of providers listed with patient ratings were female. The average medical school graduation year occurred in the 1990s, with the range of 1969–2016. On average, female providers were more recent graduates compared to male providers (2001 vs 1995, p= .001). While not statistically significant, male surgeons in the Northeast had higher overall average ratings compared to their female colleagues (4.24 vs 4.12, p=0.32). Controlling for graduation year, male surgeons scored on average 0.21 points higher on a 1–5 scale, but the difference was not statistically significant (p=0.09). However, years since medical school graduation were significantly correlated to average review score when controlling for gender (p=0.02). For each additional year since medical school, a physician’s average review score decreased by 0.015 points. Gender was not significantly correlated with total number of reviews nor average review score, even when controlling for years since medical school graduation. CONCLUSIONS: This study did not show statistically significant differences in reviews of male and female providers, but this may be due to the geographical limitation of the study. To understand if these findings differ across regions of the United States, additional analysis is currently underway. Further investigation to understand the reasons for differential scoring across specific patient and physician characteristics is warranted.