We conducted a qualitative evaluation of inpatient telemedicine use on a COVID-19 designated inpatient unit at a large academic medical center. We analyzed qualitative interviews with stable patients hospitalized for COVID-19 to understand the implementation outcomes related to the practical use and acceptance of the inpatient telemedicine solution from the patients’ perspective.

In March 2020, a large academic health center (Stanford Health Care) designated an inpatient non–intensive care unit to care for admitted patients with confirmed or suspected COVID-19. Each patient received a tablet (iPad; Apple Inc) set up to automatically receive web conference calls (Zoom Video Communications, Inc) from one of two desktops in private workrooms on the unit, all with dual audio and video functionality. Other setting details have been previously described [3] and can be found in Multimedia Appendix 1.

The target population included stable patients who were admitted to the non–intensive care COVID-19 unit. Purposive sampling methods were used to recruit participants representing diversity across age, sex, race and ethnicity, and language until thematic saturation was reached [39].

The semistructured patient interview protocol, analysis, and presentation of findings were informed through deductive themes drawn from prior qualitative work [27] and the Implementation Outcomes Framework (Multimedia Appendix 1) [40]. Interview transcripts were reviewed independently by the lead author (SV), and key themes and inductive coding that emerged were then validated using a consensus approach that included the analytic (DP, MS, JHL, and RG) and interview (SS, EA, and EEW) teams.

The Stanford Institutional Review Board determined that this project did not qualify as human subjects research (protocol 55927). All participants provided verbal consent to proceed and were informed their participation was voluntary and confidential.

A total of 20 interviews were conducted with inpatients undergoing treatment for COVID-19 without intensive care needs. Participants tended to be male (11/20, 55%); aged 50-69 years (15/20, 75%); Latino, White, or Asian, Indian, or Pacific Islander (6/20, 30% each), and spoke primarily English (16/20, 80%) (Multimedia Appendix 1).

Telemedicine use varied on the basis of the clinical context and the type of clinician providing care, though it was broadly accepted by patients given the COVID-19–related isolation precautions (Table 1). Predominant use included nonemergency clinical care and bridging to in-person care—such as when a patient triggered the bedside alert button. In this setting, one nurse was reported to initiate a telemedicine encounter to visually connect with the patient while another donned PPE to evaluate the patient in person.

Typical frequency of virtual encounters ranged from 1 to 3 times a day, though one patient reported being contacted up to 10 times per day via the tablet device and 10 times per day in person. Nurses were reported to both enter the room and call on the tablet more frequently than physicians. However, most patients reported at least one in-person encounter daily from their doctor, with a few reporting multiple in-person encounters when their needs were greater. A few patients reported their physician primarily communicated via the tablet device, and a minority perceived there were some days on which they did not speak with their doctor either in person or via the tablet device.

Patients generally accepted the use of inpatient telemedicine given their circumstances requiring isolation precautions. A desire to prevent infection transmission was recognized by several patients: “It’s helpful that I can talk to multiple doctors without having to expose them to COVID” [Patient 9]. Some felt that telemedicine offered superior convenience, and the visual component was seen as adding value beyond an audio-only telephone encounter. The virtual interactions provided reassurance, which was reported to positively impact patient mood and mental health: “I love to hear from my doctors, it brought such great comfort for them to be updating me” [Patient 6].

Patients broadly reported that the quality of care they received with the integration of telemedicine into their hospitalization was similar to what they might have received without it.

However, some were in favor of an in-person encounter in accordance with the concern that some aspects of their care could be missed, particularly related to the in-person examination and human connection. Some felt that telemedicine changed the way they communicated with their care team: “In person is better…I can have a more in-depth dialogue, ask questions…When they’re physically here, they can be there for me” [Patient 18]. Older patients, in particular, seemed to prefer in-person encounters. Only a handful of patients felt telemedicine did not compromise their connection with their care team.

In general, the technology was reportedly “very simple to use” [Patient 5]. However, certain technical setup considerations impacted patient acceptability, including audio volume, tablet position both within the room and in relation to the patient, and the automatic turn-on feature. Several patients remarked that despite turning the tablet volume up to its maximum capacity, they still had difficulty hearing the clinical team, particularly if someone was speaking further away from the source microphone.

Furthermore, the tablet location and angle relative to the patient sometimes posed challenges. The tablet device was reportedly moved frequently to allow space for nurses doing their clinical work. This lack of a stable location worsened the experience for some patients:

There was not a clear consensus on where the tablet should be placed within the room, as another patient preferred it near the feet to optimize the visual component of the encounter.

Finally, patients were asked to comment on the automatic turn-on feature of the tablet device, which did not allow them to screen calls. Nearly all were comfortable with this feature, likening it to standard hospital care in which a clinician would simply walk into the room, only sometimes following a quick knock. The one patient in his 20s, who expressed some concern over privacy, devised his own solution to turn the camera toward the wall temporarily while he showered.

The clinical team’s use of the technology also impacted patient acceptance, specifically around patient orientation to technology, clinician etiquette regarding encounter introductions and use, and visual connection with each speaker. Patients described minimal or no orientation to the tablet. “They never told me about the iPad; I noticed when they came on-- ‘Hello hello hello’” [Patient 10]. Where an orientation was reported to take place, it was minimal: “They pointed it out to me and said…this is the [tablet] for nurses and doctors to call you” [Patient 11]. While this brief orientation would have been preferred, it did not bother most patients. A preference for an orientation was more prevalent among patients who did not have prior familiarity with using tablet devices or web conferencing technology.

Opportunities were also noted for improved etiquette around introducing each member of the team and positioning the tablet toward the speaker:

This concern was not prevalent among all patients but may have improved acceptability among the few patients who voiced it.

Patients’ experience with telemedicine during an inpatient admission for COVID-19 was largely favorable. While most patients expressed a preference for in-person care where possible, telemedicine was an acceptable alternative given COVID-19 isolation precautions. Both nurses and physicians regularly used the technology to communicate with patients, thus validating past work [27] in which the balance of in-person versus telemedicine workflows was guided by acuity of clinical need. Telemedicine appeared to serve as a supplementary point of contact between in-person encounters, which patients primarily saw as a benefit to their overall experience and mental health. The technology reportedly played a meaningful role in mitigating the fear and anxiety associated with isolation precautions [31-37], though additional evaluation is needed to quantify this impact. Finally, with the exception of a desire by some patients for more hands-on examination and in-person interaction, patients felt their quality of care was no different as a result of inpatient telemedicine.

Direct and implied recommendations to improve inpatient telemedicine include increasing audio volume, allowing for a digital “knock” to serve as a warning prior to automatic turn-on, establishing an unobtrusive stable location for the tablet, adding a call-out button to the nursing team, increasing focus on physical touch, improving clinical team introductions, and standardizing patient orientation to the technology (Table 2).

These data are consistent with the past literature, which suggests that providers may overestimate their communication ability [41]. Given that a physician is required to evaluate each inpatient at least once per day, the frequency of physician encounters less than once per day reported by some patient suggests suboptimal communication and role confusion. A strong introduction includes the name, role, and responsibilities of each member of the care team, information that is often not shared even in nondigital clinical settings [42,43]. Training related to the importance of verbal introductions [44,45] and best practices to foster the patient–provider connection in the virtual setting may help mitigate the challenge of role confusion [46]. Clear badges or face cards [38,43,47,48] can also help, but these tools are not available in the digital setting. Instead, facial recognition technology is being explored to solve other problems in health care [49,50] and could be used to automatically display a virtual name badge.

Some patients were also concerned that aspects of their care may be missed owing to the lack of a hands-on examination. The importance of a hands-on physical examination—both for diagnostic and therapeutic purposes—is evidenced in the literature [51,52], but how to best adapt these learnings to patients are under isolation precautions remains an open question. Nurses were reported to use telemedicine more frequently than physicians, while also maintaining a frequent physical presence within the room, which is perhaps unsurprising given that nurses spend approximately 6-fold more time at the patient bedside than physicians in non–COVID-19 settings [53,54]. These COVID-19 data further validate past qualitative work, which suggests that nurses use telemedicine as a bridge to in-person care and “batch” care activities, such that physical assessment, medication delivery, meal delivery, and sanitation protocols all occur in a single room entry [27,55]. The impact of this shift, alongside the reported reduction in in-person physician assessments, on clinical outcomes is an area for future research.

To this end, inpatient telemedicine risks exacerbating pre-existing clinician reliance on technology over hands-on examination. Substitutes for an in-person examination, such as directing patients toward a self-examination with a digital stethoscope [56], pulse oximeter, or other technologies capable of remote transmission of data, even if appropriate from a diagnostic perspective [57], may be less acceptable to patients in the inpatient setting. Additional evaluation is therefore needed to determine the optimal ratio of in-person to virtual encounters.

Insights from this evaluation are drawn from a small sample of patients within a single institution and therefore cannot be more broadly generalized. These exploratory interviews provide insight into an otherwise difficult-to-reach patient population with currently limited available data, though future work will benefit from expansion to diverse institutions and patient populations. Including patients who required intensive care may be particularly informative; safety and feasibility constraints limited our ability to capture these perspectives in this assessment. In addition, we purposefully sought a diverse set of voices in our sampling protocol, and our final sample overrepresented non-White participants on the basis of local demographics [58]. Increasing sample diversity in terms of languages spoken, given the predominance of English speakers in this sample, may benefit future work. Finally, these qualitative data may also complement and inform ongoing and future quantitative work that explores the impact of inpatient telemedicine on clinician workflows and, by extension, infection control and resource use [59].

Inpatient telemedicine adopted for the purposes of infection control during the COVID-19 pandemic presents a novel use case of the technology, and our understanding of its impact on clinical workflows, patient outcomes, and the patient experience continues to evolve. This qualitative evaluation suggests that while patients still prefer in-person interactions in the hospital setting, inpatient telemedicine is broadly accepted given the need for COVID-19 isolation precautions. Perceived benefits include increased access to the clinical team and reduced anxiety, yet challenges around the technical setup, clinical team introductions, and physical examination remain. Further evaluation is needed to understand the impact on clinical outcomes and the optimal balance between in-person and virtual care in the hospital setting.