Suddenly Becoming a “Virtual Doctor”: Experiences of Psychiatrists Transitioning to Telemedicine During the COVID-19 Pandemic

From March 31 to April 9, 2020, we conducted 20 semistructured interviews with psychiatrists practicing in outpatient settings. To recruit these participants, we worked with a research firm with access to a panel of 730,000 physicians. This panel has been used in previous federally funded research studies and comprises physicians who have joined an online platform to access clinical content (news, condition and drug information, and journal articles), continuing medical education activities, and clinical tools (10, 11).

Psychiatrists on the panel were sent an eight-item screener survey to assess eligibility for participation. We used three inclusion criteria based on this initial survey to identify outpatient psychiatrists who were transitioning to telemedicine for the first time in the context of the COVID-19 pandemic: board-certified psychiatrist currently practicing in one of several select states with significant COVID-19 activity in late March 2020 (New York, California, Washington State, New Jersey, and Louisiana), limited use of telemedicine before the pandemic (<10% of patient encounters), and spent >50% of working hours in an outpatient setting. We excluded psychiatrists who were active duty military and those who were employed in integrated delivery systems with extensive use of telemedicine (i.e., Department of Veterans Affairs and Kaiser Permanente). Those deemed eligible were invited to participate in a 30-minute telephone interview with the study team. Of the 43 psychiatrists who started the screener survey, 20 (47%) were found eligible and consented to be interviewed. We recruited participants until we reached thematic saturation, defined as the point at which new interviews did not uncover new themes or patterns.

Interviews followed a semistructured protocol. Topics included details on practice setting and patient population, experience and perceptions of telemedicine prior to the COVID-19 pandemic, nature of telemedicine use since March 2020 (e.g., modalities, volume, and platforms), barriers encountered in transitioning to telemedicine, perceived impact of telemedicine on the quality of patient interactions, and future plans for telemedicine. Four members of the study team trained in qualitative research conducted the interviews, which were recorded and transcribed. Interviewees were given a $100 gift card for their participation, and they provided verbal informed consent. This study was approved by Harvard University’s institutional review board.

We conducted a rapid qualitative analysis to ensure that study results could be published in time to inform the COVID-19 response. Rapid research is designed to address the need for timely results in rapidly changing situations. Although some experts have noted challenges in regard to maintaining rigor, multiple studies have shown comparable results between rapid and more in-depth analyses (12, 13).

We conducted the analysis in two steps. First, we developed a templated summary of each interview that was organized by codes mapped to key research questions covered in the interview protocol, as well as novel topics that emerged. We populated the summary with data extracted from interview transcripts and included illustrative quotes. We then conducted a supplemental matrix analysis, in which rows contained all participants and columns listed the salient categories we developed from the codes included in the summaries (14). Such matrices have been used in qualitative data analysis to efficiently identify similarities, differences, and trends in responses across groups of informants (15). A matrix provides a visual data display that enables the search for and a detailed analysis of patterns, themes, and other relationships and informs conclusions (16). Here, the matrix allowed us to interpret each participant’s comments in the context of the particular telemedicine modalities and platforms the participant was using. We identified themes through well-established techniques, including repetition (e.g., if a concept was expressed more than three times) and emphasis (e.g., if respondents particularly engaged with or dedicated significant time to a concept).

Just under half of the interview participants had some experience with telemedicine before the COVID-19 pandemic. These psychiatrists generally offered telemedicine in select cases to patients who had moved away, were traveling, or had unique circumstances that prevented in-person visits (e.g., a pregnant patient on bed rest). These telemedicine visits represented only a small fraction of their visit volume before March 2020.

At the time of the interviews, all psychiatrists in our sample had been delivering telemedicine services for 2–4 weeks, including video visits, phone visits, or a combination of the two modalities (Table 2). Also, although psychiatry is considered an essential service and can be provided in person in all the states in our sample, most of the psychiatrists had transitioned to fully virtual practices. Only a quarter of the participants were seeing any patients in person.

Most participants reported conducting video visits with most of their patients, although they generally offered both video and phone visits. Among those offering more than one modality, most allowed their patients to choose the modality. Interview participants noted that when given the option, some patients opted for phone visits, including older patients or patients who were self-conscious about their appearance, had social anxiety disorder, or lacked suitable devices or had only limited broadband connection. Some psychiatrists also reported having to switch to phone visits on encountering technical difficulties during video visits.

Approximately one-third of the interviewed psychiatrists reported using the telephone for most or all visits. Reasons for not offering video visits included confidence that the telephone will work with no technical difficulties, lack of compatible devices or Internet access among underserved patient populations and older adults, familiarity of patients with the telephone, and lack of patient e-mail contact information to send a video link. Most psychiatrists who conducted most of their patient visits by telephone were not actively planning to offer video visits in the coming weeks.

Psychiatrists used several platforms for the video visits, including Zoom, Doxy.me, FaceTime, Skype, Google Meet, WhatsApp, Clocktree, and thera-LINK. Multiple participants mentioned minor technical issues with one or more of these platforms that led them to experiment with new platforms or to offer phone visits. Several appreciated the broadened flexibility to use non–HIPAA-compliant platforms, especially when HIPAA-compliant platforms were overloaded such as during the first weeks of the pandemic.

Psychiatrists identified numerous positive and negative effects of telemedicine on their practice (Table 3). Positive impacts included being able to see the patient’s home environment and greater ease and access to care for some patients. Negative impacts were a reduced ability to observe nonverbal cues providing information for a diagnosis and treatment, less patient privacy, difficulty hearing patients clearly by phone or video, more distractions for patients in their home environment, inability to conduct a physical exam and assess vital signs, difficulty assessing extrapyramidal symptoms from antipsychotics, shorter visits, and challenges in managing time during the visit.

Most participants argued that given the unprecedented circumstances, the transition to telemedicine went more smoothly than they had expected, and they were pleasantly surprised that they could meet patients’ needs via telemedicine. As noted by a psychiatrist practicing at a health system in Washington State, “I’m actually stunned at how amazingly well it’s gone. . . . It has surprised me that I have been able to feel connected with patients on video.” A psychiatrist in private practice in New Jersey said, “I didn’t really like technology at all, but it’s working quite well for me. I really felt after I did [Zoom sessions], that I could possibly have done this [before COVID-19], because clinically it’s working fine.”

Most participants said that their patients were responding positively to the switch and had provided good feedback about telemedicine; however, they also pointed out that the positive response may have been driven by patient fears about ongoing access to care during the emergency, rather than by their general acceptance of telemedicine visits.

A psychiatrist in private practice in California reported that patients’ responses had been “[U]niformly positive. . . .People are so grateful that I am continuing to be available.” A psychiatrist practicing in a New York hospital said, “[Patients] appreciate it because they felt like everything was going to be canceled, they wouldn’t be called . . . their meds would not be refilled. . . . So when we call, they feel so appreciative.” As a psychiatrist from a nonprofit clinic in New York summarized, “Patients have been very happy that they’ve been able to get seen or treated in any manner, shape, or form . . . and not having to go into the doctor’s office.”

Psychiatrists in private practice expressed more concerns about the impact of telemedicine on revenue and on the sustainability of the delivery model, most likely because this group of psychiatrists has a more direct role in managing billing than do those in other practice settings. Several participants in private practice mentioned that the payers they worked with had not been transparent about reimbursement. Lack of clarity on what treatment would be covered, coupled with the fact that these participants had not yet submitted claims for March at the time of the interview, created uncertainty about the impact of telemedicine on practice revenue. As explained by a psychiatrist in private practice in California, “I haven’t heard any specific feedback yet from the insurances that I’m dealing with as to whether or not their reimbursements are going to be any different for video services than they would’ve been for in-office visits. I’ve actually been holding off on doing my claims submissions for March until I get some clarification about that. And I’m going to need to send those claims in momentarily. [I need] to try to investigate what the coding changes need to be and what the likely reimbursement changes are going to be.”

A different psychiatrist in private practice in California stated, “[A payer] just sent an e-mail . . . that was written in legalese. [It] said, ‘We will cover this, but essentially some plans may not cover it.’ What are you supposed to do with that information?”

Multiple participants across practice settings commented that they were currently busy with existing patients and that it was challenging for them to engage new patients via telemedicine. Most participants were accepting new patients and successfully evaluating them via video or phone. However, some were not, and they pointed out that continued reliance on telemedicine could threaten sustainability of their practice in the long term. According to a psychiatrist in private practice in Washington State, “I haven’t done any new patients yet because I can’t quite figure out how to evaluate somebody over the phone.” A psychiatrist in private practice in California said, “I can sustain my practice now, but [growing the practice] will be hard.”

The psychiatrists in our sample expressed a strong preference to return to in-person care after the pandemic. Reasons for this preference included the ritual of going to an office, the fact that the office is a private and safe space, and, for some, the perceived inferior quality of physician-patient interactions via telemedicine. One psychiatrist employed by a university hospital in Washington mentioned that her hospital will likely return to requiring in-person visits to recoup a facility fee. Many of the psychiatrists in private practice, nonetheless, expressed an interest in continuing with some telemedicine, explaining that in the future they may offer it to patients who face logistical challenges of presenting in person. As explained by a psychiatrist in private practice in New York, “I definitely would want to go back to only using it when necessary and still would prefer patients to come to my office. But I do think I’m more comfortable with it. If someone needed to reschedule, I might [say], ‘Well, we can just do a video session,’ because I know it can work.”

Psychiatrists shared lessons learned in rapidly transitioning to telemedicine that may be informative for clinicians who have yet to transition to it. Table 4 presents eight strategies that interview participants credited with improving the quality and conduct of telemedicine visits. One strategy includes identifying patients at risk for having difficulty with video visits (e.g., older adults and adults with cognitive impairments) before the visit and exploring whether someone in their environment could aid them. Another would be to start each visit by asking patients whether they are concerned about privacy, and if yes, help them identify a private place, such as a car, to conduct the visit.