This quality improvement project received a nonresearch determination by Stanford University’s Institutional Review Board (IRB-55644). It was conducted at Stanford University’s Department of Neurology and Neurological Sciences, which includes 11 ambulatory subspecialties staffed by 60 physicians and 8 advanced practice providers (APPs). As previously described [27], at the beginning of the local COVID-19 stay-at-home directive in March 2020, approximately 50 in-clinic computers were video visit–enabled, and 50 additional computer devices were readied for remote use by providers. Clinic staff converted >90% of scheduled in-person visits to video. In-person visits were reserved for procedures such as autonomic testing, implanted device programming, injections, electromyography, and electroencephalograms, as well as patients determined either during previsit screening or initial video visit to require a full examination.

Out of the 68 clinicians in the department’s 11 ambulatory subspecialties, 66 clinicians conducted regular video visits during the 8-week implementation period and were included in analyses (two clinicians were on leave and thus excluded). Table 2 summarizes the number of surveyed and interviewed clinicians, by subspecialty. In total, 48 (73%) clinicians responded to the survey and 30 (45%) were interviewed. In total, 21 (32%) clinicians participated in both the survey and interviews. Only 9 (14%) did not participate in either modality.

During the transition week, the clinician champion onboarded 30 of the 66 physicians included in this study, and 129 video visits were conducted. Prior to this transition week, clinicians in these ambulatory neurology clinics had not used video visits. During the 8-week implementation period, video visit adoption was high based on both percentage of clinicians using them, and percentage of visits completed via video. Within the first two weeks, 52 (79%) clinicians integrated video into their practice, increasing to 65 (98%) clinicians by week 6. Almost all (92%) visits were conducted via video and adoption of video visits was high for both new and return patient visits (93% and 91% of completed visits, respectively). The total number of completed visits, conservatively estimated, was almost 80% of expected visits based on the 2019 comparison period (Figure 1A). Clinics recuperated more expected return patient (83%) than new patient visit volumes (68%; Figures 1B-C).

Overall acceptability of video visits was high, but clinicians experienced technological and scheduling issues, and identified additional needs. Two themes emerged related to acceptance: technology and workflow efficiency.

Three themes emerged related to appropriateness of video visits: benefits and barriers, physical examination, and continuity of care, each of which are described below.

Most clinicians were positive toward video visits and believed they could incorporate video into their practice long-term, although several insisted that video cannot replace a full in-person examination. Moreover, 40 (83%) surveyed clinicians agreed or strongly agreed that video visits supported their overall well-being. The general view was that for long-term sustainability, patient video visits will need to be selected carefully to optimize care and respect preferences. Patients’ suitability for video visits would need to be determined during scheduling based on several criteria (eg, physical examination needs, patient’s technological capacity and demographics, new versus return). However, previsit screening and triage to determine which patients are best suited for video was considered burdensome, which could compromise their sustainability. Seldom was the video visit itself considered a good tool for triage.

Most survey respondents (n=39; 81%) agreed that video visits should be supplemented with in-person visits. The recommended frequency of supplemental in-person visits varied among the respondents: 6 (15%) recommended quarterly, 11 (28%) biannually, 17 (44%) annually, and 5 (13%) every 2 years. A concern mentioned by a minority of interviewees and 7 (15%) surveyed clinicians was that patients may find video too convenient and opt out of recommended in-person visits. In contrast, 9 (19%) survey respondents with representation from epilepsy, memory, headache, stroke, and neuro-oncology, indicated that an all-video practice would be feasible for their patient population.

Video visits were rolled out rapidly at Stanford University’s Department of Neurology and Neurological Sciences during the COVID-19 pandemic. A high proportion of visits were conducted via video at volumes near prepandemic volumes, indicating robust and rapid adoption by both clinicians and patients. This necessary and widespread adoption allowed for thorough assessment of new opportunities and barriers in video-based care across almost all ambulatory neurology subspecialties.

Clinicians recognized the patient benefits of video visits, including saving patients unnecessary travel and increasing access to vulnerable populations, as is well-documented [1,3-8]. As patients with chronic progressive neurologic conditions lose function, in-person visits become increasingly challenging and care by video is a welcome new supplemental resource. Video visits could improve the experience for these patients and others, but various challenges remain. As clinicians noted, the most vulnerable high-need patients were often the hardest to care for virtually. Vulnerability is not uniform across patients and could be related to diagnosis (eg, patients with sensory or cognitive deficits), support (eg, caregiver status), socioeconomic status, or access (eg, poor access to technology). New creative solutions are being explored to address these challenges at various levels, with the goal of rendering virtual care an effective catchment for all. For instance, a pilot program delivering tablets to patients cared for by the US Department of Veterans Affairs has been successful in addressing concerns related to diagnosis and technology [40]. Further, municipality-level internet and broadband for all could address access issues [41].

Clinicians reported that video visits were beneficial for seeing the patient’s home and meeting caregivers and family members. Observing patients at home is a documented benefit of video visits; it can reduce patient and caregiver anxiety, and allow assessment of fall hazards and habitual behaviors that are not always possible to capture during an in-person visit [6,42]. These benefits were rarely identified within the top 3 benefits of video visits in the survey, which is likely related to the many perceived benefits of video visits as well as the diverse needs of neurology subspecialties’ patient populations. Interviewed clinicians emphasized that the presence of a caregiver is often essential to a successful video visit. For example, a caregiver is essential for collateral history when the patient has cognitive impairment or for safely examining gait for a complaint of parkinsonism, but not for a younger patient presenting with headache. Communicating the importance of including caregivers in certain video visits is thus important for optimal care provision.

Previously, the perceived limitations of virtual physical examinations have been a source of hesitant incorporation of telemedicine into neurology practice [10-12]; however, virtual physical examinations are possible, especially to supplement heavily history-based presentations [1,6,43]. A previous investigation demonstrated that virtual exams were adequate even for many patients with multiple sclerosis and neuroimmunology issues [44]. Presence of a caregiver can further facilitate a detailed and complete medical history and examination [28]. In this study, several clinicians reported that video cannot replace a full in-person examination, but others reported that the virtual physical examination was more feasible than expected. Best practices that were followed across clinics included consenting patients and explaining video limitations prior to the visit, and scheduling patients for in-person follow-up as soon as they are able to do so safely whenever an in-person examination was deemed important during the video visit. As clinics adjust to new routines and workflows, measures are also being taken to further optimize the virtual exam experience. For instance, the department has organized teleneurology professional development webinars on optimal virtual physical examination techniques [23,45], and guidance documents on this topic have also been developed internally. Together with time and experience, these efforts may further increase clinician comfort with the virtual physical examination.

Clinicians were keen about the ability to work remotely, reduction in uncompensated work, and flexible scheduling facilitated by video visit use. In the current configuration, video visits are not constrained by clinic spaces nor specific staffing needs, thereby allowing clinicians flexibility to vary timing and length of both their overall clinic session and individual patient appointments. This is a clear benefit, as lack of flexibility in work is a key driver of burnout [46], and even more salient given increased family and childcare duties for many due to the pandemic. Furthermore, reducing uncompensated work has potential well-being benefits. Historically, clinicians with full clinic schedules answered patient messages and phone calls uncompensated outside of normal clinical hours; this is another driver of physician burnout [47]. Video visits have relatively fewer constraints than in-person visits that are often booked out for months, and could enable the ability to address a new patient concern in short order “on the clock” and compensated. Further, most respondents also reported that video visits support their well-being, which is key in a specialty where an estimated 60% of clinicians reported at least one symptom of burnout even before the pandemic [48].

Perceived sustainability of video visits was high; clinicians emphasized a desire for video visits to remain a permanent fixture in their practice. However, the proportion of video visits and in-person visits within their practice was anticipated to change in their postpandemic practice and vary by individual providers and subspecialty. Most clinicians preferred to have a practice that uses both video and in-person visits, but more than half (57%) expressed a preference for less frequent in-person visits (ie, every 1 or 2 years). Perhaps most surprisingly, nearly 20% of clinicians from a variety of neurology subspecialties indicated that an all-virtual practice was possible. This variation in preferences may reflect the wide variation in the nature and examination demands of neurologic subspecialties, with subspecialties known to be especially exam-dependent (eg, movement and neuromuscular disorders) notably absent from those indicating feasibility of an entirely virtual practice.

The capabilities of mobile technology and remote monitoring have facilitated great advances in telemedicine [2,6,8,14]; however, several challenges remain and need to be addressed to enable long-term sustainability of video visits. Audio and video connectivity issues may be partially ameliorated by software updates and through increased experience with telemedicine, particularly for clinicians. On the other hand, some technological issues experienced by patients could potentially be addressed with targeted efforts, including more previsit planning and education with a staff member. To enhance the video visit experience for both patients and clinicians, pertinent features included virtual waiting rooms and multiuser interfaces to promote incorporation of trainees, interpretation services, and multidisciplinary care. Some of these features were added in subsequent updates and parallel workflows were developed to use other HIPAA-compliant platforms for visits including trainees or interpreters. In our experience, this parallel platform and approach were important for resolving some issues of functionality and service; however, the lack of integration and interoperability with standard EMR video functionality could impact sustainability.

Further analysis of the workflow around telemedicine in ambulatory neurology is needed as there is growing recognition that workflow is nontrivial and complex [49]. In this study, initial workflows for video visits did not include MAs taking patient vitals and performing medication reconciliation, key aspects of in-person visits for new patients. Clinic processes should reincorporate MAs in video previsits, perhaps through phone or video visits in the days prior to the clinical video visit. As clinical workflows evolve to sustain video visit use and meet the needs of the most vulnerable patient populations, these video previsits could potentially also provide patient education on optimal lighting, how to position devices, space, and presence of caregivers.

Perceptions of clinicians using video visits were captured in real time during early rapid implementation through interviews and/or a survey across many ambulatory neurology subspecialties. However, given its nascent state, operational scheduling data only captured data of video visits conducted via the EPIC Hyperspace Platform and did not consistently reflect visits conducted on other HIPAA-compliant software, such as Zoom. Therefore, the data presented herein likely underestimate the actual total number of completed visits. Further, when video visits were diverted to phone calls or other software during the 8-week implementation period, the scheduling system incorrectly categorized the encounter as cancelled, no show, or patient left without being seen. This limitation has been recently addressed, presenting an opportunity to investigate the impact of video visits on clinic utilization in future investigations. Finally, views of other essential health care staff, such as patient care coordinators and MAs, who performed essential activities (eg, scheduling and confirming patients’ technology capabilities), as well as residents and patients/caregivers, are not presented here.

Video visit adoption was rapid at Stanford’s ambulatory neurology clinics. Almost all clinicians conducted video visits by the eighth week of implementation and achieved near-normal patient volumes during the COVID-19 pandemic stay-at-home order. Despite the sudden change in workflow, clinicians largely expressed positive views toward video visits; clinicians supported permanent integration of video visits and noted them to be conducive to physician well-being. However, overall clinician satisfaction was impacted by technological issues, limitations with the physical examination, and challenges accessing vulnerable patient populations. Although our mixed methods evaluation confirmed the success of video visits in all subspecialty neurology clinics across many dimensions, innovations must be developed to address their limitations. Additional solutions are also needed for the most vulnerable patient populations. A crucial next step for optimization is to understand patients’ experiences and preferences.