LAUSR

We used the Standards for Quality Improvement Reporting Excellence (SQUIRE) reporting guidelines for quality improvement studies.3 This study was exempt from review by the University of California, San Francisco, institutional review board. We identified the top 50 US hospitals using 2020 Newsweek rankings, which represent a geographically diverse group of academic and community hospitals.4 We conducted an online search of publicly available documents and determined that the internal, employee-centered policies we sought were not readily available. Between September and November 2020, we emailed the chief medical officers (or equivalent), requesting their policies on sexual harassment by patients and/or their families toward physicians. Institutions were contacted up to 3 times. If institutions were unable to share documents, we sent follow-up emails requesting policy details. Two investigators (G.E.R. and S.R.A.) used deductive coding of all documents and emails. In cases of disagreement, the study team met to reach consensus. Any policy reference to “staff” or a similar descriptor (eg, “provider,” “faculty,” “employee”) was assumed to apply to physicians. We categorized policies as sexual harassment, inappropriate behavior, patient rights/responsibilities, employee workplace safety, or other patient-facing material. We extracted whether the policy explicitly mentioned any of the following behaviors: sexual harassment, physical assault/violence, verbal harassment/intimidation, or discrimination. We used the standard definition of discrimination as differential negative treatment based on race, gender, sexual orientation, ability, or other protected category. We coded whether policies addressed the harassment of health care staff by patients or family specifically. We noted any description of institutional response. For single documents applicable to multiple hospitals in one system, the document was counted once for each institution.