Gender and racial biases in Press Ganey patient satisfaction surveys
Are we collecting the best data, and enough of it, to evaluate physician performance based on patient satisfaction?
Controlling the narrative for the future: Proposed strategies
The rapid, widespread adoption of the Press Ganey survey across specialties, clinical care settings, and geographic areas may have been largely due to the ease and operational benefits for hospitals rather than after rigorous study and validation. For example, repeated use of a specific measurement tool may facilitate comparison across areas within a hospital but also across institutions, which can help assess performance at a national level. Hospitals also may have a financial incentive to work with a single third-party or vendor instead of using multiple options across multiple vendors. However, the impact of adoption of novel measures of performance should be evaluated prior to widespread adoption and utilization.
A similar example of an emergence of a technological advancement that has changed the field of medicine and how we provide care is the EMR. Epic is now the most commonly used medical record system and holds the market share of the industry, covering 78% of patients in the United States.31 While there are certainly many potential benefits of a common EMR, such as ease of information sharing and standardization of formatting, opportunities are identified in real time and require product adjustment. For example, modifications have been made to accurately represent gender outside of the previously used dichotomous options. Diagnoses such as cervical cancer screening can now be used even if the patient gender is listed as male.
Similarly, the Press Ganey and other patient satisfaction questionnaires should be evaluated and modified to address existing societal biases. The World Health Organization estimates that it will take 300 years to fix gender inequality,32 but we have an opportunity now to control the narrative and improve patient feedback.
Future research avenues
Ultimately, there is a need to further explore currently available methods of evaluating clinical encounters to better understand the inherent biases and limitations. We hope this review will encourage other physicians to examine their specialties and hospitals and require similar analyses from vendors of such satisfaction rating products prior to using them. At the very least, health systems should be willing to partner with vendors and physicians on an ongoing basis to better understand the biases involved in these survey results and make modifications as needed. Patients also obtain information from and contribute to self-reported, publicly available websites; therefore, additional exploration into a nationalized standard for assessing patient satisfaction also may serve as an opportunity to standardize the information patients evaluate.33 Further assessment of the potential financial and emotional impact of using the currently available patient-reported surveys on female physicians, Asian physicians, young physicians, and physicians who see young patients is needed. It is time to put pressure on a broken patient satisfaction system and improve on a national level to avoid undue negative consequences on our physicians. Use of patient satisfaction survey data should be limited until we all understand and account for the biases that are evident. ●
- Appeal to the Press Ganey corporation with the results of recent data and other studies to ensure they are aware of the biases that exist in their product
- Appeal to hospital-level administration to refrain from using Press Ganey scores as a tool to dictate reimbursement; instead rely on other more objective measures of performance (such as publications, presentations, research accomplishments, patient and surgical outcomes, promotion, committees, national leadership roles, etc)
- Apply a “corrective factor” or “adjustment factor” to eliminate the baseline discrepancy between scores for men and women
- Consider moving to an alternative survey methodology
- Provide patient education to define “performance” (ie, frame what a patient can expect from a provider such as being on time, courteous, and empathetic; caution against asking patients to assess competence and knowledge)
