Teledermatology to Facilitate Patient Care Transitions From Inpatient to Outpatient Dermatology: Mixed Methods Evaluation

Introduction

In 2020, the COVID-19 pandemic drove telemedicine to the forefront of health care [1,2]; dermatology care was no exception. Prepandemic teledermatology had gained popularity in some specific use cases [3,4]; however, the pandemic gave rise to new policies that overcame previous restrictions to ensure continued access to care, facilitating a rapid pivot to telemedicine for outpatients, including patients transitioning from inpatient care [5-7].

The highly visual nature of dermatology is well suited for this cost-effective and efficient care modality [4,8-10], which is well received by clinicians and patients [10-14]. Although convenience and improved access to care are the primary benefits, especially for rural and underserved populations, teledermatology also boasts time and cost savings, greater flexibility for dermatologists and patients, fewer no-shows, and better continuity of care [4,5,10,15-25]. However, teledermatology has some shortcomings that affect care delivery, including suboptimal image quality, patient privacy, diagnostic accuracy, network connectivity, patient technological literacy, and access to digital devices [8,9,11,15]. In addition, the inaccessibility of in-clinic tools and treatments (eg, dermoscopy, biopsy, and cryotherapy) makes managing certain conditions challenging [9]. These limitations may disproportionally affect patients with low socioeconomic status, Medicare beneficiaries, older adults, and non–English-preferring patients [9,11], who may also be at risk for delayed care transitions.

Nevertheless, teledermatology may be particularly beneficial for patients transitioning from inpatient dermatology consultation services to outpatient dermatology care. Currently, high-risk patients who are hospitalized often experience numerous comorbidities and may experience difficulties accessing in-clinic follow-up care. As a result, they risk receiving fragmented care and being lost to follow-up, which could have serious health consequences [26-28]. Teledermatology may improve follow-up access for these patients by increasing the capacity of dermatology clinics and improving the efficiency of scheduling and care provision. As video visits become a fixture in health care expected by patients and clinicians, it is essential to understand whether teledermatology supports timely care transitions. We retrospectively evaluated teledermatology use and its impact on the clinic’s capacity, scheduling efficiency, and timeliness of follow-up care for patients transitioning from inpatient to outpatient dermatology care and explored dermatology clinicians’ and schedulers’ perceptions of teledermatology for this patient population.

Methods

Setting

Stanford University’s Department of Dermatology encompasses 13 outpatient clinics with 16 subspecialties and provides inpatient consultative services in a quaternary hospital; that is, consultation requests placed by the patient’s admitting team, such as general medicine or oncology. Consultations are delivered by 5 dermatologists and 2 dermatology residents on monthly rotations. The team consults >1500 inpatients per year, many of whom have complex, high-risk skin conditions in immunocompromised states and have multiple clinical teams involved in their care. Approximately 40% of these patients require postdischarge outpatient follow-up.

Intervention: Teledermatology

The department rapidly implemented teledermatology across all ambulatory clinics in response to the statewide COVID-19 stay-at-home orders in March 2020 [10]. Clinicians were provided with video visit–enabled hardware to enable the remote provision of teledermatology. All clinicians and staff completed the web-based training developed for the institution’s rollout. Initially, clinicians and staff were encouraged to convert all nonurgent or emergent in-clinic visits into video visits. Once in-clinic capacity began to expand in spring 2020, department-developed clinical criteria guided appropriate video visit use for all patients except (1) patients with high skin cancer risk requiring full skin examination, including melanoma; (2) patients requiring specialized examinations (scalp and genitals); and (3) patients requiring procedural interventions. As of July 2022, teledermatology had remained a fixture and was offered to patients transitioning from inpatient to outpatient care.

Mixed Methods

Overview

Outcomes derived from quantitative scheduling data and qualitative interviews are defined in Table 1. Data were consolidated throughout the analysis and interpreted in parallel to understand converging and diverging issues regarding teledermatology use and its impact on the clinic's capacity, clinical appropriateness, sustainability, and the remaining barriers for patients transitioning from inpatient to outpatient care.

Quantitative: Inclusion Criteria, Data Collection, and Analysis

Patients who received a dermatology consultation in the inpatient or emergency department settings were discharged in 1 of the 3 study periods and potentially needed follow-up with outpatient dermatology. The three study periods were (1) June 1 to September 30, 2019 (baseline [before teledermatology]); (2) June 1 to September 30, 2020 (early teledermatology); and (3) February 1 to May 31, 2021 (sustained teledermatology). Follow-up scheduling and care were recorded for 90 days after discharge; visits scheduled >90 days after discharge were likely unrelated to the patient’s hospitalization. Eligible patients and relevant events were retrospectively identified using the electronic health records and scheduling data. Inpatient dermatology consults were identified using Current Procedural Terminology codes (Multimedia Appendix 1).

Teledermatology use and its impact on the clinic’s capacity, scheduling efficiency, and timeliness of follow-up care were compared across periods and visit modalities (video and in-clinic) using the outcomes described in Table 1. Descriptive statistics were calculated to describe patient characteristics and assess differences across the 3 study periods and by visit modality. Statistical significance was assessed using the Kruskal-Wallis test for patient age, chi-square test for categorical (ie, proportional) outcomes, and generalized linear models for continuous outcomes (eg, days from discharge). Differences in teledermatology use by patient age and distance between patient residence and outpatient dermatology clinic were determined using chi-square tests. Clinically meaningful (<70 years vs ≥70 years) or median-based (<21 miles vs ≥21 miles) categories were used. P values were adjusted for multiple comparisons with an adaptive, 2-stage linear step-up procedure, and significance was set at P<.05 [29].

Qualitative: Data Collection and Analysis

We designed a semistructured interview guide to capture perceptions of teledermatology for follow-up care of patients transitioning from the inpatient setting. Clinicians and schedulers were eligible if they were involved in transitioning patients from inpatient to outpatient dermatology. All eligible clinicians and schedulers (ie, 5 dermatologists, 5 dermatology residents, and 13 schedulers) were invited via email (plus 2 reminders) to participate in a 30-minute phone interview. Ultimately, 15 interviews (5/5, 100% dermatologists; 5/5, 100% residents; and 6/13, 46% schedulers) were conducted between April and May 2021 by 2 experienced qualitative researchers (EAS-G and AA), ranging from 30 to 60 minutes. The interviews were audio recorded and subsequently transcribed.

Data were analyzed deductively and inductively using Microsoft Excel. Deductive codes were derived from the Proctor implementation outcomes [30]. We used multiphase matrix analysis by leveraging rapid analytic procedures to achieve consensus coding of transcripts and extract early themes [31]. EAS-G and AA independently summarized transcripts after each interview; summaries were reviewed, and consensus discussions were held. Summaries were then consolidated into a matrix to identify and compare themes across interviewees. To ensure anonymity, all identifiable information was removed from transcripts, summaries, and reports.

Ethics Approval

This retrospective quality improvement evaluation received a nonresearch determination by Stanford University’s Institutional Review Board (IRB-60382). Interviewees provided informed verbal consent before initiating the interviews and were assured that all responses would remain confidential. Detailed interview notes were taken if consent for recording was not provided.

Results

Patient Characteristics

Patient characteristics are summarized in Table 2. Briefly, 194 patients, 218 patients, and 256 patients were discharged following an inpatient dermatology consultation during the baseline, early teledermatology, and sustained teledermatology phases, respectively. The median patient age was similar across the 3 periods (61.0, 60.5, and 55.5 years for baseline, early teledermatology, and sustained teledermatology, respectively; P=.11). Approximately half of the patients lived ≥21 miles from the dermatology clinic during each study phase, and most had public insurance.

Clinic Capacity

More patients were scheduled for outpatient dermatology visits within 90 days after discharge during the early (n=125) and sustained (n=125) teledermatology phases than at baseline (n=92), indicating a 36% increase (n=125 from 92 and n=125 from 92 patients for early and sustained teledermatology, respectively) in the scheduling capacity (Figure 1). Similarly, the number of follow-up visits completed within 90 days after discharge was higher in the early teledermatology (n=101) and sustained teledermatology (n=100) than at baseline (n=74), indicating a sustained increase in capacity. The proportion of patients who completed their follow-up did not differ across the 3 evaluation periods, as shown in Figure 1.

‎View this figure

Teledermatology Use

Teledermatology use was highest in the early teledermatology period, with 61.4% (62/101) of follow-ups completed via video. This decreased significantly to 47% (47/100) of follow-ups in the sustainability period (P=.04), indicating a more targeted yet still substantial use (Figure 1). Interviewees remarked that patient acceptance of teledermatology and technology capabilities varied during early implementation; however, acceptance increased as it became the standard of care (see exemplary quotes in Textbox 1).

Clinicians and schedulers believed that older patients may prefer clinic visits, whereas patients residing farther from the clinic may favor video visits (Textbox 1). However, video visit use did not differ by patient age during the early or sustained teledermatology phase or by distance from the clinic during the early teledermatology phase (Table 3). However, in the sustained period, significantly more patients residing ≥21 miles away from the clinic had follow-up video visits (26/41, 63% patients) than those living closer (21/59, 36% patients; P=.01), confirming the interviewees’ view that patients living farther away preferred video to in-clinic visits as teledermatology became optional (ie, no longer mandated because of the pandemic).

Clinical Appropriateness

Clinicians expressed interest in continuing to offer video visits to recently discharged patients who have ongoing stable conditions that are well-suited to teledermatologic care or limitations preventing in-clinic care (Textbox 1). Teledermatology was perceived to reduce barriers (eg, lack of time or resources to travel) to attending follow-up in person for high-risk patient populations who may need to attend several follow-up visits and may consequently deprioritize their dermatological issues. A video visit was perceived as better than no follow-up at all for patients who were too debilitated to travel, even if an in-clinic visit was clinically ideal.

Video visits were often considered ideal for quick, simple check-ins for postdischarge patients as they had recently been examined in person, whereas in-clinic visits were considered more appropriate for conditions requiring skin examinations or procedures, laboratory tests, or dressing changes. Nevertheless, a few residents believed that video visits were less suitable for postdischarge patients (Textbox 1). However, most agreed that the choice should ultimately be the patient’s own. According to clinicians, patients appreciate having a choice that meets their priorities, needs, and preferences, and accommodating patients to complete the necessary follow-up is worthwhile.

Teledermatology to Support Timely Care Transitions

To understand the impact of teledermatology on the scheduling efficiency and timeliness of care transitions, differences in days from discharge to initial scheduling, finalized scheduling, and completion of an outpatient dermatology follow-up visit across each study period and visit modality were assessed (Table 1). The proportion of incomplete scheduled visits is also reported by the study period and visit modality.

Scheduling Efficiency

The average days from hospital discharge to initial scheduling of in-clinic visits increased from 4.2 (SD 7.2) days at baseline to 8.5 (SD 10.2) days during early teledermatology (P=.01; Table 4). This returned to 3.0 (SD 4.8) days in the sustained period, similar to baseline (P=.40). In contrast, the average number of days for initial scheduling of video visits was 4.3 (SD 6.8) days and 3.7 (SD 10.1) days for the early and sustainability periods, respectively, similar to the 4.2 (SD 7.2) days at baseline when only in-clinic visits were offered (P=.89 and P=.76, respectively; Table 4). The results were similar for the other efficiency measures; that is, days from discharge to final scheduling (Table 4). These results potentially reflect the decreased availability and increased difficulty in scheduling in-clinic visits in the early implementation period due to pandemic-related restrictions and concerns; however, the increased flexibility and appointment availability of teledermatology enabled comparable timeliness of follow-up scheduling during the pandemic as to before the pandemic, as reported by schedulers (Textbox 1).

Timeliness of Follow-up Visits

The average number of days from hospital discharge to completed follow-up visits did not differ across the periods or by visit modality (Table 4). A higher, although nonsignificant, proportion of patients was seen within 14 days after discharge, which is a local follow-up benchmark, via video than in the clinic (Figure 1). During the early teledermatology period, 53% (33/62) of follow-ups were conducted using teledermatology, whereas 38% (15/39) of follow-ups were conducted in the clinic (P=.15). Similarly, 60% (28/47) of follow-ups were conducted using teledermatology, whereas 49% (26/53) of follow-ups were conducted in the clinic (P=.29) during the sustained teledermatology period (Figure 1). Thus, although teledermatology follow-up visits were scheduled slightly faster (although not significant) than prepandemic in-clinic visits, this did not result in timelier follow-up care.

Incomplete Follow-up Visits

Relatively few patients missed their scheduled follow-up visits during the 3 periods (Figure 1). Of the 24 patients who missed their scheduled visit in the early teledermatology period, 12 (50% patients) had a scheduled teledermatology visit and 12 (50%) had scheduled in-clinic visits. A total of 25 patients did not complete their scheduled visit during the sustained teledermatology period, of whom 9 (36% patients) were scheduled for teledermatology and 16 (64% patients) for an in-clinic visit. This aligned with clinician and resident perceptions that teledermatology facilitated the completion of follow-up care but may not improve cancelation rates (Textbox 1). Some perceived that cancelation rates were higher for video visits, and others said they were higher for clinic visits; the small number of missed visits limits our evaluation of these perceptions.

Remaining Barriers to Video Visit Coordination

Interviewees acknowledged the benefits of teledermatology but indicated that care coordination and video visit setup were sometimes challenging (Textbox 1). Access to a smartphone and a means of taking a high-quality photograph were considered essential, especially for at-risk populations, including older adults, who were believed to benefit the most from improved access through reduced travel and risk during the pandemic. Schedulers perceived that these patients frequently needed help in setting up their devices and uploading their photographs before a visit. Despite this assistance, the photographs submitted sometimes lacked sufficient quality. Care coordination via video was further complicated if the patients required laboratory tests. Some interviewees suggested that this coordination should begin during hospitalization at the patient’s bedside to integrate and prioritize the care needs of patients and caregivers in discharge planning and follow-up care scheduling.

Discussion

Principal Findings

Teledermatology was frequently used during the evaluation period; two-thirds of the visits were conducted via teledermatology early in the pandemic, whereas about half of the visits continued to be conducted using teledermatology later in the pandemic, indicating more targeted but nevertheless substantial use. Teledermatology availability increased the clinic’s follow-up scheduling capacity for patients transitioning from inpatient to outpatient dermatology care. Teledermatology also provided a flexible option that increased overall clinic capacity while retaining comparable scheduling efficiency and timeliness of care as before the pandemic, even amid a pandemic and strained health care system. However, the scheduling efficiency and timeliness of care transitions did not improve, suggesting that a wider range of efforts are needed to improve these issues. Interviewees viewed teledermatology as an important care modality for providing accessible care, especially for patients with competing medical priorities and limited ability or availability to travel to the clinic, although important logistical and technological limitations were acknowledged for some patients. Ultimately, interviewees believed that patients should make the final choice between in-clinic or video visits.

Comparison With Prior Work

Teledermatology is an important tool for building clinic capacity, as well as improving scheduling timeliness and completion of care [17,18]. A study in an urban safety net hospital setting found that teledermatology implementation increased the total number of cases evaluated per month by approximately 20% and decreased the time to consultation for new patients from 84.6 days to 6.7 days before the COVID-19 pandemic [17]. Teledermatology has also been shown to increase access to and expedite care for patients in many settings, including referrals from primary to specialty care [19,20], within the Veterans Affairs system [21], in medically underserved populations [22,23], and for those needing inpatient consultations [24]. Our study builds on this literature by demonstrating a sustained increased clinic capacity of 36% after implementing teledermatology in the context of recently discharged patients needing follow-up care; the increase in clinic capacity did not come at the cost of less timely care. In addition, teledermatology services allowed safe access to care during the height of the pandemic when in-clinic care was delayed.

Nevertheless, our study was unable to detect improvements in the timeliness of care, which may be because of the urgency of our patient referrals (desired timeline from discharge to follow-up of only 14 days), as has been reported elsewhere [17-24]. Previous studies that found that video visit implementation improved care timeliness have been conducted in settings where patient referrals were nonurgent [17-24]. Although teledermatology supports increased access to outpatient care for patients of dermatology in general and those needing follow-up care after hospitalization, further research is needed to determine whether telemedicine itself supports more timely scheduling and care provision, particularly for care transitions.

Telemedicine has also been shown to promote visit completion and reduce patient cancelations and no-shows compared with in-clinic visits [16,25]. In the outpatient setting of a large academic health care system, 20% of telemedicine visits were canceled compared with 31% of in-clinic visits [16]. Similarly, a study focusing on dermatological care also showed that a lower percentage of virtual consults, specifically e-consults, were either canceled or not attended (ie, no show) than ambulatory consults (18% vs 39%) [25]. In this study, a few patients missed their scheduled follow-up, of whom 50% (12/24) missed a video visit in the early teledermatology period and 36% (9/25) missed a video visit in the sustained period. This latter result, although a different metric, suggests that recently discharged patients may be less likely to miss scheduled video visits than scheduled in-clinic visits, aligning with previous research [16,25]. However, the small sample size limits interpretability, and additional investigations are needed.

Teledermatology is well suited and highly accepted in dermatology, even for high-risk, recently discharged patients [10-14]. In this study, clinicians and scheduling staff recognized that teledermatology is convenient for patients experiencing difficulties related to their current health, with competing medical and care needs, or limited time and resources to access in-person care. In fact, even as in-clinic visits became more available, patients who lived farther from the clinic were significantly more likely to use teledermatology care than those living closer to the clinic, aligning with previous research reporting on the convenience of telemedicine for rural and underserved populations [10-14]. The widely reported flexibility of teledermatology [4,8-10] was recognized to support care transitions and continuity by interviewed clinicians and schedulers. Ultimately, telemedicine may be particularly well suited to this highly visual specialty, as reported here and elsewhere [4,8-10], perhaps even more so for follow-up care of patients recently physically examined and for whom the clinician expects but wants to confirm an improvement in their condition.

Although video visits are a well-accepted and widely used technology, some patients, clinicians, and schedulers still prefer in-person visits due to their limitations [32-34]. As reported here and previously, teledermatology continues to have shortcomings that affect care delivery, including incomplete previsit preparation, poor quality images, limited patient technological literacy, inability to access certain in-person tools and procedures, and patients’ lack of capabilities with digital devices [8,9,11,15]. Patient privacy concerns, diagnostic accuracy, and network connectivity are also well-recognized limitations [8,9,11,15]. The reported inequities in access to telemedicine care, known as the digital divide [9,11], compound the limitations of teledermatology. Although patients with complex medical issues may particularly benefit from the convenience and flexibility of teledermatology care, they may also lack access to a smartphone and the technological capacity to, for example, take and submit high-quality photographs. Thus, efforts to improve care transitions need to not only be attuned to patients who may benefit from teledermatology but also be able to assess whether such patients have access to the needed skills and technology; if not, alternatives or appropriate support for skills and technology must be provided to facilitate equitable access to care for all. Bedside communication for more patient-centered care [35] or employing dedicated care coordinator teams [28,36,37] to ensure that patients’ care needs are met could not only better support timely care transitions but also ensure high levels of patient and caregiver satisfaction, improved patient outcomes, and lower readmissions.

Limitations

This retrospective study has 3 main limitations. First, it was conducted in a single health care setting. Second, Current Procedural Terminology codes (Multimedia Appendix 1) associated with inpatient dermatology consultations were used as a surrogate to identify patients potentially needing follow-up care; ideally, such data would be based directly on clinical recommendations, including the follow-up timeline; however, such data were not systematically available. Third, 2 of the 3 study periods were during the COVID-19 pandemic. We considered many factors that could have influenced our findings, including seasonality, COVID-19 pandemic surges and restrictions, and the presence of other quality improvement initiatives, to identify comparable periods. However, this retrospective evaluation was not able to account for all potential confounding factors, including regular policy changes and vaccine availability.

Conclusions and Future Directions

Telemedicine has moved to the forefront of health care delivery and is anticipated to continue to expand. As telemedicine becomes an established care modality, additional evaluation of its quality, acceptability, and appropriateness for specific use cases and patient populations is needed to ensure the provision and sustainability of appropriate, high-quality care without continuing to widen the care access divide. Teledermatology was viewed as an important tool for maintaining accessible, flexible, and convenient care for patients transitioning from inpatient to outpatient dermatology care. Despite its shortcomings, including photograph quality and varying patient technological capabilities, teledermatology is predicted to be a standard option for patients. However, teledermatology alone does not completely solve care transition delays; it must be coupled with other efforts to improve communication between patients and care teams, patient access to and comfort with video technology, and workflows that support timely and equitable access to follow-up care. Care transitions are a vulnerable time for patients who may easily slip through the cracks and remain a challenge in health care systems [26,28,38-41]. Continued evaluation of alternate approaches to care delivery during care transitions, including telemedicine, as well as reporting of these efforts, is needed to understand their impact on this risky time in the patient care continuum.