The Utilization and Impact of Live Interactive and Store-and-Forward Teledermatology in a Veterans Affairs Medical Center During the COVID-19 Pandemic

Use of VVC, a platform not previously used by our facility for dermatologic care, scaled up rapidly to meet the gaps in care left by FTF and SFT encounter closures. Wait times for VVC visits and SFT encounters, defined as the number of business days between the day the consult was placed and the day of the appointment or consult completion, were significantly different. The median wait times for VVC, SFT, and FTF encounters were significantly different at 5 days, 3 days, and 16 days, respectively (p < 0.01).

The primary diagnoses addressed in each SFT and VVC encounter were categorized into the categories of benign, infectious, inflammatory, malignant, neoplasm of uncertain behavior, and premalignant (Table 2). The distribution of primary diagnoses addressed was significantly different between VVC and SFT. A majority of VVC encounters involved a primary diagnosis of an inflammatory lesion or condition (66%), while the most frequently addressed primary diagnoses in SFT were benign lesions (35%) and neoplasms of uncertain behavior (25%) (Table 2).

Teledermatology Encounter Outcomes

Rates of recommendations for medications or lesion biopsies were used to assess the impact of SFT and VVC encounters on disease management. Significantly more VVC visits (78%) resulted in a medication recommendation by the teledermatologist compared with a lower rate seen in SFT visits (37%) (Table 3). In contrast, significantly more SFT encounters resulted in biopsy recommendations (23% vs. 2% of VVC visits). Sixty-one percent and 57% of biopsies recommended through SFT and VVC during the study period were performed, respectively.

Diagnostic accuracy of SFT and VVC visits was also analyzed by examining the concordance between teledermatologists’ differential diagnoses and histopathological diagnoses. Of the 71 biopsies that were performed as recommended by SFT telereaders, 76% of clinical teledermatology diagnoses were concordant with histopathological diagnoses. Of the four biopsies performed as recommended by dermatologists over VVC, 50% were concordant with histopathological diagnoses.

Forty-five percent (32 of 71) and 25% (1 of 4) of lesions biopsied per recommendation by SFT and VVC, respectively, were malignancies (Table 4). The single malignancy biopsied per recommendation by VVC was a nonmelanoma skin cancer (NMSC). Among the 32 malignancies biopsied per recommendation by SFT, 27 were NMSCs and 5 were melanomas.

Discussion

The VHA has already been a leader in the field of telehealth, and the COVID-19 pandemic necessitated further utilization of its teledermatology programs. We found that both SFT and VVC teledermatology helped meet patient care needs throughout the year. The median wait times for VVC and SFT (5 days and 3 days, respectively) were shorter than, and in impressive contrast to, the median wait time for FTF encounters during the same time period (16 days). These encounters had high impact on patient care and management; 60% of SFT and 80% of VVC encounters resulted in either medication or biopsy recommendations. Furthermore, 45% and 25% of biopsies recommended through SFT and VVC were malignancies requiring further intervention.

Although VVC was not previously implemented by our facility’s dermatology department, the platform rapidly ramped up when FTF and SFT services were paused. These SFT, VVC, and FTF utilization trends mirror the effects of governmental and SFVAHCS policies that affected in-person appointments. FTF and SFT use increased throughout June through November, mirroring policies during those months in which patients were able to access in-person care.

This then declined again after November, mirroring COVID-19 case surges during these months, limiting and possibly discouraging patients from seeking in-person care. However, VVC appointments reached a steady state of utilization and did not fluctuate along with these trends, raising questions about the factors limiting the reach of this platform.

The differences in demographic characteristics of patients served by these different forms of telehealth provide insight into the specific reach of each program. VVC users were significantly younger than both SFT and FTF patients and had a significantly lower representation of patients from rural areas. Almost half of the veteran population is over 65 years old and older veterans are more likely to reside in rural areas.¹⁰ Usage of the internet and My HealtheVet, the VA’s online personal health record portal, among VHA enrollees is also most prevalent among younger patients.¹¹

Studies have indicated that for patients with complex medical dermatology conditions, outcomes improve with close clinical follow-up and that teledermatology providers prefer a synchronous teledermatology modality for these cases.^8,12 Rural veteran satisfaction with teledermatology care has also been shown to be equivalent to FTF care.¹³ Given the relatively longer wait times for FTF dermatologic care and the barriers that elderly veterans residing in rural areas face, it is imperative that efforts are made to improve their access to VVC to reduce travel costs and deliver efficient care.

Further study of access to a home internet connection and proper videoconferencing equipment of this patient population is warranted, as other studies have shown that veterans in rural areas are less likely to meet the technology requirements of VVC.^10,14 This finding further supports the importance of continuation of VHA initiatives to provide high-need veterans with technological equipment, such as tablets with broadband access, to support access to quality care.¹⁴ In the context of comorbidities being more common in the elderly and also associated with worse COVID-19 health outcomes, expanding access to care that does not require on-site visits and potential exposure to the virus is especially pertinent.

There was a significant association between gender and teledermatology care platform use; significantly more females utilized VVC, with 16% of patients identifying as female. Our VVC patients had a median age of 63, so a contributing factor may be that the middle-age group (45–64 years old) represents the largest group of women veterans.¹⁵ Nevertheless, current projections by the VHA indicate that women compose 10% of the veteran population and that this representation continues to grow rapidly each year.^15,16

Women have historically underutilized VHA services for a variety of reasons beyond barriers involving technology requirements.^17,18 Unique challenges faced by women veterans include negative experiences within VHA grounds, with one-fourth of a sample of women veterans reporting experiencing harassment in VHA health care grounds.¹⁸ Previous studies have also indicated that women veterans view videoconferencing as an acceptable platform for medical care and that they may prefer seeking dermatologic care from dermatologists instead of their primary care providers.¹⁹ Thus, further investigation on how the VVC platform could possibly be better at meeting the needs of women veterans is warranted.

Our study also showed that the SFT platform was more commonly utilized for consultation on neoplasms, while the VVC platform was more commonly utilized for inflammatory lesions, which is consistent with a decision tree algorithm that was utilized by the SFVAHCS dermatology service for patient triage and diversion away from FTF appointments during times of shutdown. This is also consistent with studies that have shown similar diagnostic category distribution.^8,12

Given the higher resolution of digital images submitted for SFT compared with the more variable resolution during VVC, we have found the SFT platform to be more appropriate for detailed review of neoplasm morphologies. In contrast, VVC has been more effective for inflammatory rashes given the ability of teledermatologists to gather more details of the patient’s history through interactive interview and provide direct counseling on management. Our clinical outcome measures of rates of medication and biopsy recommendations are as expected and consistent with the diagnostic categories associated with each teledermatology platform.

Rates of biopsy recommendations were higher in the SFT group given that a majority of encounters were associated with requests for neoplasm evaluation. Rates of medication recommendations were higher in the VVC group given that a majority of encounters were associated with inflammatory conditions requiring medication interventions. Furthermore, VVC establishes a physician–patient relationship, allowing the teledermatologist to directly prescribe recommended medications. In contrast, our dermatologists serve as strict teleconsultants when using the SFT platform by providing recommendations to the referring providers, who then enter orders themselves and provide necessary patient counseling.

There are inherently more workflow-related challenges for the SFT system with the intermediary referring provider. Although the rate of prescription of teledermatologist-recommended treatments by our referring providers is approximately consistent with rates published by other groups, we are actively working to continuously improve this rate.²⁰ For example, the SFVAHCS has piloted a system encouraging more active management by consultants while easing workflow-related challenges for referring providers by having teledermatologists prepare recommended orders within the electronic health record system, which are then reviewed and signed by the referring provider.

Other quality improvement interventions being piloted within the VHA system include teledermatologists directly prescribing low-risk medications.²¹ Further work and investigation of the impact of these interventions are warranted and could improve our teledermatology program’s impact on clinical outcomes.

Our data on biopsy recommendations are relatively limited, but still provide useful insight. We show our rate of follow-through on biopsy recommendations (61% and 57% of biopsies recommended through SFT and VVC were performed, respectively) to be lower than other published work performed within the VHA (79%).²² However, it is important to note that this prior study examined the time period of 2009 to 2011.

Our lower rates of teledermatology recommendation adherence were likely affected by unprecedented system-level and individual-level challenges that contributed to patients ultimately being unable to come in for their recommended biopsies, including limited FTF visits or fear of risking exposure to the SARS-CoV-2 virus in health care settings during the pandemic. The rates of diagnostic accuracy seen within our small sample size (76% and 50% for SFT and VVC, respectively) are slightly lower than diagnostic accuracy rates reported in other studies, which have ranged from 51% to 92% agreement.²³

These are also lower than rates observed in prior studies conducted on SFT and live interactive teledermatology programs within the VHA system.^4,24,25 This may be due to a lack of implemented standardization for clinical image capture or the technological requirements within our own health system, limiting the assessments performed by our teledermatologists; however, further investigation is needed to elucidate contributing factors.

Limitations of our study include its focus on a single institution serving a limited region and lack of comparison with a control group from previous years.

Conclusions

Our study indicates that teledermatology programs contributed to meeting veteran care needs during the pandemic, made an impact on clinical outcomes, and had significantly shorter wait times relative to FTF care. The demographic characteristics of patients who utilized these care modalities shed light on factors to consider in efforts to improve access as teledermatology programs continue to expand, especially during this period of accelerated implementation and growth due to the COVID-19 pandemic’s effects on health care delivery.

Telehealth is a continuously evolving model of care delivery. Given its theoretical ability to increase specialty care access and the recent expansions of telemedicine reimbursement, it is clear that teledermatology use will continue to grow. Understanding the utilization and impact of these programs will continue to be crucial to ensuring the quality and cost-effectiveness of teledermatology used to meet veteran care needs.

Disclosure Statement

No competing financial interests exist. The views expressed in this work are those of the authors and do not reflect the position or policy of Veterans Affairs or the U.S. Government.

Funding Information

This work was supported in part by the U.S. Department of Veterans Affairs, Office of Rural Health (to D.H.O.) and in part by Award #1I01 HX002961 from the U.S. Department of Veterans Affairs, Health Services Research and Development Service (to D.H.O.).