Outcomes of a virtual craniofacial clinic for assessing plagiocephaly during the COVID-19 pandemic.

Objective: The aim of this study was to carry out a quantitative analysis of a virtual craniofacial clinic during the COVID-19 pandemic.
Methods: The charts of 90 patients from a single institution were reviewed. Of these patients, 45 visited the virtual clinic during the COVID-19 pandemic. The other 45 patients visited the clinic in the 3 months prior to COVID-19. Demographics including the mean age at the visit, chief complaint, visit diagnosis, appointment duration, helmet usage, accuracy of the diagnosis, need for a CT scan, and the need for a follow-up appointment were assessed. Diagnostic accuracy, the frequency of follow-up appointments, and patient satisfaction (via survey), as well as additional associated factors, were analyzed to determine the efficacy and satisfaction associated with the virtual clinic approach.
Results: The mean patient age at time of the visit was 5.6 and 7.3 months (p = 0.244), and the mean time from referral to appointment was 19.2 and 19 days (p = 0.934), in the in-person and virtual cohorts, respectively. There was no significant difference in the variety of chief complaints between the in-person and virtual visits, with 97.8% and 93.3% of patients' parents reporting abnormal head shape, respectively, and the remainder reporting more infrequent complaints (p = 0.435). The visit diagnosis was plagiocephaly in 93.3% of the in-person cohort and 80.0% of the virtual cohort (p = 0.118). The final diagnosis exhibited a similar pattern, with 95.6% of the in-person cohort and 88.9% of the virtual cohort observed as positional plagiocephaly; the remaining diagnoses were more infrequent (p = 0.434). The most common alternative diagnosis in the virtual visit cohort was a metopic ridge (8.4%). In the in-person visit cohort, the most common alternative diagnosis was equally a benign enlargement of the subarachnoid space in infancy, scalp mass, and skull lesion (2.2% each). None of the patients in either cohort were diagnosed with synostosis. Eighty percent of the in-person visits were 15 to 30 minutes in duration, with the remaining 20% being 31 minutes or longer; virtual visits were all 30 minutes or less, with 95.6% being 15 to 30 minutes (p = 0.002). Helmets were prescribed for 2 patients in the in-person cohort and no patients in the virtual cohort (p = 0.494). Alterations in diagnosis were made in 2.2% of in-person visits and 6.7% of virtual visits (p = 0.616). Follow-up was required in 15.6% of the in-person visits and 31.1% of the virtual visits (p = 0.134). CT was only utilized twice, once in the in-person visit cohort and once in the virtual visit cohort.
Conclusions: Virtual clinic encounters resulted in comparable diagnostic accuracy. The trend toward frequent follow-up assessments and changes in the final diagnosis in the virtual clinic cohort has indicated a level of diagnostic uncertainty via the virtual interface, which required in-person assessment for confirmation. This finding did not contribute toward diagnostic inaccuracy with respect to missed synostosis. The study results have indicated that telemedicine can be an effective modality in assessing craniofacial pathology.