Introduction

Sexual harassment of women is pervasive and is defined as unwelcome sexual advances, requests for sexual favors, and other verbal or physical conduct of a sexual nature.1 Studies from the 1990s reported that female medical students,2 residents,3 and practicing clinicians4, 5 experience sexual harassment by patients and more recent studies confirmed this as an ongoing problem.6,7,8,9 Although all these studies describe the scope of the problem, clear guidance on how to address it is minimal.

Current literature on how to address harassment is limited to two survey-based studies, which categorize common responses used by female providers to combat harassment, and a theoretical framework for how providers could respond to patient-driven harassment.10,11,12 More recently, one intervention published from a surgical training program at Massachusetts General Hospital included some aspects on how to deter gender-based harassment from patients.9 However, none of these publications utilized interviews to obtain participants’ experiences and responses in their own words. They highlight the need for ongoing evaluation and understanding of what providers are enduring and how to address sexual harassment from patients.

Our study purpose was to explore the experiences of female providers with sexual harassment by patients with a focus on how practicing providers address these events.

Methods

We conducted semi-structured interviews with female residents, attendings, and nurse practitioners at a tertiary academic medical center. The purpose of these interviews was to understand provider experiences with sexual harassment from patients and the strategies they use to address these incidents. The study was approved by our IRB. Subjects received no incentive for participating.

Participants and Recruitment

The study was conducted in the fall of 2018. We recruited a convenience sample of Department of Medicine female practitioners via monthly emails that outlined the study purpose. Thirty-five staff providers and forty-eight resident providers were emailed during the recruitment period. Email respondents were contacted by the study coordinator, who explained the project in more detail and arranged interviews by one of five trained interviewers (JLM, CS, AF, CK, SN). All participants provided written informed consent prior to participation.

Interviews

We developed an interview guide (supplement) based on a literature review and discussion among an expert panel of providers. The interviews were piloted among a group of female attending physicians for flow, length, and clarity. The interview questions were open-ended with prompts that interviewers could use as follow-up questions. All interviews were audio-recorded and transcribed verbatim. In preparation, the five interviewers discussed the need for an open approach and three interviewers had completed qualitative methods course work during fellowship training. The interviewers participated in taped practice interviews; these audiotapes were reviewed by one of the coders (JLJ) for interview fidelity and openness of the approach.

Analysis

Two authors (KEF, JLJ) coded the de-identified transcripts for themes using thematic content analysis, an iterative approach with the goal of finding patterns across the dataset.13, 14 The two authors reviewed the first four transcripts together to develop an initial coding scheme through an interactive process of discussion and consensus building. Subsequently, they independently coded transcripts, meeting regularly to compare themes. This process continued until no new themes emerged. In consultation and over several sessions with the two coders (KEF, JLJ), a third author (CS) reviewed the themes and developed a framework for understanding the experience of harassment and the strategies used to address it. Coding was done with NVivo (v4, QSR International).

Results

Our sample consisted of twenty female providers: six resident physicians, thirteen attending physicians, and one nurse practitioner. Two major themes were explored: the experiences with sexual harassment from patients and the strategies used to address sexual harassment. Our analysis highlights the key findings for these two themes and delves into the sub-themes for each.

Thematic Analysis

Theme 1: Experiences with Sexual Harassment

All participants described their experience with sexual harassment from patients and many found these experiences common. From participant discussion, we coded four sub-themes: the descriptions of the types of sexual harassment, the context of the event, the impact of the harassment, and the preparation to address harassment (Table 1).

Table 1 Theme 1: Experiences of Sexual Harassment
Full size table

Descriptions of the Types of Sexual Harassment

We coded six categories of harassment: (1) gender-based behavior, (2) flirting and lewd comments, (3) staring, (4) exposure, (5) physical contact, and (6) solicitation and stalking. Gender-based behavior ranged from being identified as a nurse to comments on physical appearance to defiance in following instructions during a clinical encounter. One participant found that: “If I’m with my male medical students, older men will look toward the male medical student to ask all their questions.” Most of these behaviors were identified as harassment, but some (such as being misidentified as a nurse) were not endorsed as sexual harassment by all providers. Providers found these behaviors bothersome and disruptive.

Overt flirting, lewd comments, staring, exposure of genitals, and unwanted physical contact were also reported and universally identified as harassment. One provider highlighted: “occasionally patients want a hug and they’re a little grabby with the hug, they invade your personal space inappropriately with sexual overtones.” More egregious violations of the patient-provider relationship, such as overt solicitation for sex and stalking, were also experienced by some participants. One described a patient who made comments to female staff such as “why don’t you come into my room and see me, I could really use your company, it would be nice if you sat on my lap.”

Context of the Event

Participants felt that determining if a comment or behavior was sexual harassment and the severity of such was based on several contextual factors: patient factors, intent, setting, culture, and persistence. Patient factors, such as age and medical and psychiatric conditions, were thought to be important: “Another incident… was just an older, demented patient who just made [comments]: ‘You’re as cute as bow,’... he was demented, it wasn’t intentional.” Interviewees also discussed the intent of the patient as a key component to understanding the interaction, though intent was often difficult to delineate: “Some of my patients, I feel like they don’t mean it in a sexual harassment kind of way. I think they generally are trying to be nice.” The setting (e.g., clinic or hospital ward and the presence of others) and the culture and expectations surrounding acceptance of these behaviors also altered participants’ interpretation of the event. One participant highlighted the importance of the setting: “Especially the ones in the hospital, when we’re doing inpatient wards, that’s a little bit different because I don’t know them as well. So some of those comments do make me feel uncomfortable, especially with residents and medical students around.” Lastly, they noted that persistence of a comment or behavior could change the interpretation from benign to harassment: “The first time, it’s easy maybe to make that mistake. I think what is pretty frustrating is continually being identified as a patient’s nurse.”

Impact

Providers perceived that sexual harassment by patients impacted patient care, providers, and learners, affecting the way they delivered care and how they felt during and after an encounter: “I am looking at the clock to know when I need to cut this person off and send them on their way. Which, actually, makes me feel a little bit guilty, because I feel like I’m not treating all of my patients the same way. But I feel a lack of respect.” In addition to describing guilt, they identified feeling fatigued, degraded, discouraged, and unsafe: “I’m a doctor, I have worked really hard to get to where I am, and you’re basically just taking that away from me.” Another provider remarked: “Even though I was not physically attacked, I was emotionally attacked. I felt as if I was raped and violated.” They also reported self-blame: “‘Did I do something differently this visit? No, I’m dressed as plainly as possible.’ You’re going through the things that most women do when something like this happens. What about me did that?”

Some participants reported a desire to protect learners from harassment but identified a struggle between providing protection and education: “There are patients who seem inappropriate and I have not had a medical student go in if that’s an option. I have had a patient transferred out of the resident clinic for that purpose. So, I’ve tried to protect learners … in one sense, you’re protecting them, right? But then, in the other sense, what are they going to do when they’re attendings?”

Preparation to Address Harassment

Many felt unprepared to address sexual harassment: “I was so taken aback by it that I was scrambling to figure out the appropriate professional response.” Some providers noted a lack of role models, although others commented that prior attendings or peers had helped them develop tactics to address this problem. Several providers reported a lack of training on how to deal with sexual harassment from patients: “I can’t say that I went through any kind of training. I think it’s more [observing] behaviors and listening to advice from other people.” A few mentioned online trainings for how to address harassment in the workplace but noted that these trainings were not effective or applicable to patient situations.

Theme 2: Strategies to Address Sexual Harassment

All participants discussed strategies they used to address sexual harassment from patients. We coded seven sub-themes detailing the strategies used by our participants: indirect strategies, confrontation, modifying the encounter, modifying self, alerting others, debrief, and report (Table 2).

Table 2 Theme 2: Strategies to Address Sexual Harassment
Full size table

Indirect Strategies

Indirect strategies were the most common method for addressing gender-based comments and sexual harassment. Almost all providers highlighted these strategies, which included avoiding parts of the patient interview or exam, ignoring the comment, redirecting the conversation, or using humor: “I awkwardly laughed … Because I didn’t really know what else to do.” Most providers found that indirect strategies worked well, and they were able to proceed with patient care. However, several commented that these strategies were not ideal as they did not address the problem and left them feeling distressed: “As I reflect on it I’m disappointed in myself for not responding.” Several interviewees noted that they used indirect strategies because they were surprised by the encounter and had no other strategies in mind.

Confrontation

Many providers discussed instances where they confronted the patient: “I’m way more likely now to say, ‘Hey, that’s inappropriate. Do not do that.’ Or, ‘That’s not okay in my clinic.’” Providers reported struggling between confronting the patient and maintaining good provider-patient relationships. Fear of embarrassing the patient and damaging the relationship was often cited as concerns when confronting patient behavior. Providers discussed two common reactions to the confrontation strategy. Patients who were not aware that their comments were inappropriate usually apologized: “Once I’ve addressed it with them, it wasn’t really an intentional thing to put you down or to make you feel bad. I think just directly communicating with them that that’s not okay is the best way to do it.” Other patients became defensive and told the provider that she was being overly sensitive; however, no provider reported having a patient terminate their relationship after a confrontation event. One participant said: “It has not been a horrible situation. I’ve been ‘fired’ for many reasons but having that conversation has not been one of them.”

Multiple participants commented that they were more apt to confront the patient in situations that included learners. They identified two reasons for this: a desire to set a good example and a need to mitigate the harassment of learners. One provider highlighted: “I didn’t know it was appropriate to say it when I was a student, or an intern, or even a resident. Now I’m just thinking, if I don’t set an example for the team, because I never saw anybody else say anything, then why would they ... it was just going to be the same thing.” Another reported: “Immediately, in front of the learner and the patient, because I was like, ‘Oh no. That cannot happen to my learner.’”

Modify Encounter

Providers identified many ways in which they altered the encounter to deal with sexual harassment. First, they reported being less personable and more distant with these patients: “I usually use last names with patients. I don’t use first names.” Next, providers altered their clinical interview or exam, such as avoiding sexual history questions and avoiding part or all of the physical exam. In situations with a patient known to harass female staff, many participants would leave the door open, enter with another staff member, or have another staff member check in on them during the encounter: “My senior resident and I went to go see him together, just because we had read in the chart that he was already making some inappropriate comments to nursing staff.” Some reported terminating the visit early or transferring the patient’s care to a male provider.

Modify Self

Most participants modified themselves or how they presented themselves to mitigate sexual harassment, including wearing less or no make-up, altering their hair, and avoiding certain clothing, such as skirts, dresses, or heels: “I don’t wear dresses to my men’s clinic anymore ... I always wear pants.” Multiple providers identified using their white coat or a name tag to further identify themselves as a provider and deemphasize their physical appearance: “I probably wore my white coat more often … to kind of place myself in that role, even though wearing a white coat or not doesn’t change the fact that I’m a doctor.” Despite the visual cues of white coats and name tags, patients often struggled to identify them as providers during the visit: “I was always assumed to be a nurse.”

Alert Others

Participants found that documentation in the chart or a verbal warning about inappropriate behavior was beneficial: “It’s helpful when someone has a flag in their chart that this patient shouldn’t be seen by a female provider by herself or he has a history of making sexual comments. Then my guard is up.”

Debrief

Female providers relied on discussing these events with peers for support and assistance with coping: “I like talking to other people so I feel like if someone makes an egregious comment I’ll probably tell one of my coworkers, friends.” Resident providers commented that having preceptors debrief about these incidents was helpful: “We kind of did a debrief and just both sat down and said, ‘You know what? That made me feel really uncomfortable. That was not part of our medical exam’… The debriefing I think was helpful.”

Report

Most providers had not reported prior events of harassment; many felt the harassment did not reach a level which needed to be reported: “I’ve had patients inappropriately stare at me. I would like to hope that I would report a patient who physically touched me in an inappropriate way.” Another described the need to report in terms of patient care: “If I ever felt that the patient-doctor relationship had been compromised to the point that I wasn’t able to do my job safely or mentally, then I think it would be my duty to the patient to report.” Many participants were aware of reporting mechanisms, but some felt unsure how to report sexual harassment from patients specifically.

Discussion

Our qualitative study suggests that sexual harassment of female providers by patients continues to occur and that many feel unprepared to address it. Most harassment was verbal, in the form of patient comments, and the patient’s intent was sometimes difficult to decipher. Moreover, many behaviors had to be judged within the context of patient factors, such as dementia. Providers struggled with a desire to preserve the patient-provider relationship. Consequently, the most common strategy our participants used to deal with harassment was to ignore it and redirect that patient back to the visit’s purpose.

Our results, as well as others recently published,6,7,8, 15 share many similarities to studies published over 25 years ago,5, 11 suggesting that, although more women have entered the field of medicine, little has been done to consistently and systematically address this issue. Providers receive little training about how to address sexual harassment from patients, and workplace sexual harassment training is generally not applicable to these situations.

Sexual harassment is a predictor for provider burnout, depression, and job dissatisfaction.16, 17 Our providers found all types of harassment, including gender-based comments, to be bothersome and disruptive to the patient-provider relationship as well their sense of self. Furthermore, our results and others find that patient care may be disrupted or jeopardized due to these patient-initiated behaviors.10, 18 Hence, there is a need to address this problem for the health and safety of both patients and providers. Unfortunately, most institutions do not have systematic approaches to remedy sexual harassment by patients and provider training is scarce.12, 18

In our study, female providers reported multiple strategies that they have developed or adapted, mostly based on personal experience and role modeling. Our study was not designed to compare these different strategies, but it is important for providers and health systems to understand which strategies are likely to result in the best patient care and provider well-being. An algorithm for how to address sexual harassment from patients was recently published by Viglianti et al..12, 19 This algorithm recommends first assessing for provider safety, then, if the provider feels safe, he/she should confront the patient to stop the behavior. If the behavior is stopped, then patient care can continue; however, if the behavior persists or the provider feels unsafe, the provider should stop patient care, report, and consider transfering care. The results of our study support this algorithm in that direct confrontation was usually effective, rectified the problem, and deterred further issues.

Our results also found that participants valued debriefing, either formally with a supervisor or informally with peers. Debriefing with trainees seems to be of particular importance as many of our participants commented that they would have appreciated it during their own training. Although defriefing is not yet proven to reduce physician burnout, small studies are promising and debriefing is generally used to improve providers’ ability to handle stressful patient care situatios.20,21,22

Additionally, it is important that institutions bring attention to these common occurrences and provide training on how to effectively handle these situations. Some participants were shocked and stunned into silence by patients’ comments and behaviors. Therefore, an open dialogue and formal training may improve providers’ readiness to respond appropriately. Instruction should include bystander training for those who witness harassment.

Our study has limitations. First, selection bias is possible since providers who have encountered harassment may have been more likely to participate. Alternatively, female providers experiencing long-term adverse effects due to harassment may have avoided participation. Additionally, we are a single Midwestern site; our study reflects the experiences at a large tertiary urban hospital and the experiences in other settings could differ. Next, we did not ask our participants the gender of patients from whom they received harassment, but, to the knowledge of our coders, all the patients were referred to as “him.” Lastly, we did not collect demographic information such as age, years of experience, sexual orientation, or gender identity which could have an impact on how harassment events were experienced.

In summary, our participants reported that sexual harassment of female providers by patients is common and disruptive to the patient-provider relationship. It is also a threat to the health and well-being of both provider and patient. Training providers on strategies is an important next step to addressing this problem.