To the Editor: The incorporation of telemedicine into routine dermatologic care during COVID-19 has created new opportunities to evaluate and optimize our existing teledermatology platforms. Previous pediatric studies have shown that teledermatology improves access to care1, 2 while offering opportunities to improve show rates and reduce wait times.3 We performed a retrospective review of 1110 video visits (live, interactive, patient-to-provider) and 89 e-consults (store-and-forward, provider-to-provider) during the early COVID-19 pandemic (March 18 to May 1, 2020) to acquire key information on continued applications of teledermatology.
In addition to collecting patient demographics (Table I ) and encounter-specific data (eg, diagnoses and referring provider), we reviewed dermatology provider surveys embedded within each virtual encounter. Video visit surveys asked providers about connectivity issues, video quality, and the use of supplementary photographs. Logistic regression was performed to identify associated factors (Table II ). For e-consults, providers indicated whether the encounter was sufficient to assist in diagnosis or provide advice on treatment. Providers also noted when additional workup and/or triage to an in-person visit were recommended.
Table I.
Patient characteristics for all visit types
| Characteristic | Pre-COVID in-person visits (N = 18,188), n (%) | COVID in-person visits (N = 347), n (%) | COVID e-consults (N = 89), n (%) | COVID video visits (N = 1110), n (%) |
|---|---|---|---|---|
| Age group, y (%) | ||||
| 0-1 | 4053 (22.3) | 128 (36.9) | 28 (31.5) | 303 (27.3) |
| 2-7 | 4000 (22.0) | 74 (21.3) | 17 (19.1) | 197 (17.7) |
| 8-13 | 4628 (25.4) | 67 (19.3) | 21 (23.5) | 219 (19.7) |
| 14-18 | 4911 (27.0) | 72 (20.7) | 22 (24.7) | 354 (31.9) |
| Over 18 | 596 (3.3) | 6 (1.7) | 1 (1.1) | 37 (3.3) |
| Sex | ||||
| Female | 10,149 (55.8) | 191 (55.0) | 36 (40.4) | 604 (54.4) |
| Male | 8039 (44.2) | 156 (45.0) | 53 (59.5) | 506 (45.6) |
| Race | ||||
| Black | 4289 (23.6) | 46 (13.3) | 22 (24.7) | 212 (19.1) |
| Other | 5510 (30.2) | 68 (19.6) | 24 (30.3) | 270 (24.5) |
| White | 9970 (54.8) | 233 (67.1) | 40 (44.9) | 683 (61.5) |
| Ethnicity | ||||
| Hispanic or Latino | 1850 (10.2) | 31 (8.9) | 6 (6.7) | 108 (9.7) |
| Not Hispanic or Latino | 16,338 (89.8) | 316 (91.1) | 83 (92.3) | 990 (89.2) |
| Insurance | ||||
| Private | 9037 (49.7) | 242 (69.7) | 61 (68.5) | 771 (69.6) |
| Public | 8822 (48.5) | 83 (23.9) | 27 (30.3) | 337 (30.4) |
| Self-pay | 329 (1.8) | 22 (6.3) | 1 (0.08) | 0 |
Table II.
Factors associated with provider-reported connectivity issues, video quality, and submission of photographs∗
| Connectivity issues (Y/N) (N = 1042) |
Video quality adequate (Y/N) (N = 1035) |
Photographs submitted (Y/N) (N = 1054) |
|||||||
|---|---|---|---|---|---|---|---|---|---|
| n | OR (95% CI) | P | n | OR (95% CI) | P | n | OR (95% CI) | P | |
| Age group, y | |||||||||
| 0-1 | 198 | Reference | - | 197 | Reference | - | 200 | Reference | - |
| 1-7 | 278 | 1.66 (1-2.75) | .051 | 276 | 0.6 (0.36-0.99) | .045 | 279 | 1.70 (0.91-3.15) | .095 |
| 7-13 | 332 | 1.43 (0.8-2.58) | .229 | 197 | 0.64 (0.32-1.29) | .151 | 201 | 0.74 (0.38-1.40) | .352 |
| 13-18 | 36 | 1.58 (0.87-2.87) | .134 | 329 | 0.66 (0.37-1.17) | .223 | 337 | 0.30 (0.12-0.74) | .009 |
| Over 18 | 198 | 2.52 (1.01-6.29) | .047 | 36 | 1.14 (0.39-3.28) | .812 | 37 | 0.89 (0.46-1.69) | .729 |
| Insurance | |||||||||
| Commercial | 722 | Reference | - | 717 | Reference | - | 732 | Reference | - |
| Public | 320 | 0.95 (0.69-1.3) | .728 | 318 | 0.7 (0.51-0.95) | .022 | 322 | 0.74 (0.53-1.05) | .087 |
| Diagnosis | |||||||||
| Acne | 225 | Reference | - | 223 | Reference | - | 228 | Reference | - |
| Adnexal skin disorder | 24 | 3.16 (1.23-8.09) | .017 | 24 | 1.43 (0.49-4.15) | .51 | 25 | 0.49 (0.17-1.4) | .181 |
| Alopecia | 63 | 1.01 (0.49-2.08) | .984 | 63 | 4.7 (1.71-12.94) | .003 | 63 | 0.35 (0.17-0.71) | .003 |
| Dermatitis | 295 | 1.37 (0.81-2.3) | .237 | 295 | 1.1 (0.65-1.84) | .728 | 297 | 0.52 (0.3-0.89) | .018 |
| Hemangioma | 131 | 1.18 (0.57-2.45) | .647 | 130 | 1.11 (0.54-2.3) | .774 | 132 | 1.4 (0.58-3.34) | .454 |
| Infection or infestation | 63 | 2.07 (1.02-4.22) | .044 | 63 | 0.59 (0.29-1.18) | .137 | 64 | 0.93 (0.37-2.33) | .881 |
| Melanocytic nevus | 62 | 1.71 (0.83-3.54) | .148 | 62 | 0.59 (0.29-1.18) | .133 | 63 | 1.51 (0.57-4.02) | .408 |
| Other | 78 | 0.99 (0.51-1.93) | .971 | 77 | 1.11 (0.58-2.12) | .759 | 78 | 0.6 (0.31-1.15) | .126 |
| Pigmentary disorder | 24 | 0.96 (0.32-2.87) | .945 | 23 | 1.38 (0.46-4.11) | .563 | 26 | 2.61 (0.56-12.14) | .221 |
| Psoriasis | 31 | 1.02 (0.42-2.51) | .963 | 31 | 1.09 (0.45-2.67) | .845 | 31 | 0.66 (0.27-1.64) | .373 |
| Rash | 46 | 0.42 (0.15-1.17) | .097 | 44 | 0.87 (0.39-1.93) | .734 | 47 | 0.73 (0.3-1.77) | .486 |
| Study week | |||||||||
| Week 1 | 32 | Reference | - | 31 | Reference | - | 38 | Reference | - |
| Week 2 | 132 | 0.79 (0.35-1.79) | .575 | 130 | 1.09 (0.46-2.6) | .841 | 137 | 0.65 (0.29-1.43) | .282 |
| Week 3 | 142 | 0.7 (0.31-1.57) | .382 | 140 | 1.26 (0.53-3) | .599 | 143 | 1.86 (0.81-4.25) | .144 |
| Week 4 | 168 | 0.87 (0.39-1.93) | .735 | 167 | 1.45 (0.62-3.4) | .396 | 168 | 2.44 (1.07-5.57) | .033 |
| Week 5 | 223 | 0.34 (0.15-0.75) | .007 | 222 | 1.54 (0.67-3.57) | .312 | 223 | 2.12 (0.97-4.62) | .059 |
| Week 6 | 260 | 0.26 (0.12-0.57) | .001 | 260 | 3.06 (1.31-7.17) | .01 | 260 | 2.14 (0.99-4.64) | .053 |
| Week 7 | 85 | 0.18 (0.07-0.48) | .001 | 85 | 2.63 (1.01-6.88) | .049 | 85 | 2.17 (0.87-5.41) | .095 |
| Visit type | |||||||||
| Follow-up | 695 | Reference | - | 691 | Reference | - | 703 | Reference | - |
| New patient | 347 | 0.89 (0.64-1.24) | .489 | 344 | 0.64 (0.47-0.88) | .006 | 351 | 2.31 (1.54-3.45) | <.001 |
OR, Odds ratio.
Bold indicates statistical significance.
Primary care providers placed the most e-consults (36.0%), followed by inpatient providers (30.3%), emergency department providers (22.5%), and other subspecialty services (11.2%). The overall mean turnaround time was 5.84 hours (range, 0.07-188.13 hours), though the emergency department typically received responses within 90 minutes. Providers reported assisting in diagnosis and advising on treatment in more than 90% of the e-consults. Further workup and/or triage to an in-person visit were recommended about half the time (48.8%).
Providers reported issues with connectivity (26.5%) and inadequate video quality (25.5%) in about one-fourth of video visits. Most video visit providers (76%) reported using parent-submitted photographs. When photographs were not submitted, providers said they would have helped with the diagnosis most of the time (73.4%).
The prevalence of public insurance across teledermatology encounters was significantly lower than our practice baseline, and during video visits with patients on Medicaid, providers were more likely to report inadequate video quality. We also observed a significant reduction in Black patients receiving care. These findings are important because disparities in access to dermatologic care disproportionally affect minority children and those enrolled in Medicaid.4 A platform like e-consults, which relies on providers and does not require families to have internet access, could address these inequities, although widespread implementation would require reimbursement policies that cover several forms of teledermatology.
We learned that e-consult providers primarily assisted in diagnosis or treatment recommendations. Because the turnaround time was relatively quick (6 hours) and the median wait time to see a pediatric dermatologist in our region is approximately 4 months,3 we see e-consults as an opportunity to expedite care, enhance knowledge among requesting providers, and limit redundant consults.
Providers reported connectivity issues about one-fourth of the time, although we saw a significant reduction in connectivity issues and improvement in the perceived video quality over time (likely reflecting provider and patient acclimation). Given that previous studies have shown both store-and-forward and live interactive teledermatology to be comparable diagnostically,2, 5 a hybrid model may be ideal as store-and-forward modalities eliminate connectivity issues altogether by not requiring a live interactive experience.
Conflicts of interest
None disclosed.
Acknowledgments
We thank Emily Drinkwater, Joy Ukaigwe, and all the Pediatric Dermatology providers in our practice for their assistance in gathering data for this study.
Footnotes
Funding sources: None.
IRB approval status: Not applicable.
Previous presentation: An abstract of this study was presented at the Society of Pediatric Dermatology Annual Meeting, which took place virtually from July 10-12, 2020.
References
- 1.Utidjian L., Abramson E. Pediatric telehealth: opportunities and challenges. Pediatr Clin North Am. 2016;63(2):367–378. doi: 10.1016/j.pcl.2015.11.006. [DOI] [PubMed] [Google Scholar]
- 2.O’Connor D.M., Jew O.S., Perman M.J., Castelo-Soccio L.A., Winston F.K., McMahon P.J. Diagnostic accuracy of pediatric teledermatology using parent-submitted photographs: a randomized clinical trial. JAMA Dermatol. 2017;153(12):1243–1248. doi: 10.1001/jamadermatol.2017.4280. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Stephens M.R., Murthy A.S., McMahon P.J. Wait times, health care touchpoints, and nonattendance in an academic pediatric dermatology clinic. Pediatr Dermatol. 2019;36(6):893–897. doi: 10.1111/pde.13943. [DOI] [PubMed] [Google Scholar]
- 4.Naka F., Makkar H., Lu J. Teledermatology: Kids are not just little people. Clin Dermatol. 2017;35(6):594–600. doi: 10.1016/j.clindermatol.2017.08.009. [DOI] [PubMed] [Google Scholar]
- 5.Heffner V.A., Lyon V.B., Brousseau D.C., Holland K.E., Yen K. Store-and-forward teledermatology versus in-person visits: a comparison in pediatric teledermatology clinic. J Am Acad Dermatol. 2009;60(6):956–961. doi: 10.1016/j.jaad.2008.11.026. [DOI] [PubMed] [Google Scholar]