Teleophthalmology in the Era of COVID-19: Characteristics of Early Adopters at a Large Academic Institution

We analyzed data from 97 ophthalmology faculty to investigate characteristics of telemedicine users. Fifty-six providers (58%) were classified as telemedicine users and 41 providers (42%) were telemedicine nonusers (Table 1). The median number of televisits performed by users was 7 (interquartile range [IQR] = 15.5). By Fisher analysis, telemedicine usage differed significantly by subspecialty (p < 0.0001), years in practice (p = 0.04), and academic rank (p = 0.004) (Table 1). Due to the small sample size and/or all providers being telemedicine users, which would have led to the improbability of concluding a meaningful statistical association, providers in the American Indian or Alaskan Native, small satellite location only, uveitis, oculoplastics, and low vision categories were not included in our binomial logistic regression models (Table 2).

Our univariate model found that female providers were more likely to use telemedicine compared with males [odds ratio, OR, 2.42 (95% confidence interval, CI, 1.03–5.67)]. Compared with Doctor of Optometry (OD) specialists, providers with an MD, MBBCh, or MBBS [OR 0.30 (95% CI 0.10–0.94)] or with a PhD, MPH, MHS, or MS [OR 0.28 (95% CI 0.08–0.99)] were less likely to use telemedicine. Compared with comprehensive ophthalmology specialists, cornea [OR 0.13 (95% CI 0.04–0.47)], glaucoma [OR 0.18 (95% CI 0.04–0.93)], and retina [OR 0.04 (95% CI 0.01–0.17)] specialists were less likely to use telemedicine according to univariate analysis. Clinical assistants, clinical associates, and instructors [OR 12.5 (95% CI 2.63–59.47)] and associate professors [OR 4.38 (95% CI 1.42–13.52)] were more likely to use telemedicine relative to full professors.

The median years in practice for all providers was 18 years (IQR = 24). Compared with providers who have been in practice for ≤11 years, providers who practiced for ≥36 years [OR 0.20 (95% CI 0.06–0.66)] were less likely to use telemedicine according to univariate analysis. After adjusting in a multivariate model, providers who practiced for 12–18 years [adjusted OR 0.22 (95% CI 0.05–0.91)] and ≥36 years [adjusted OR 0.13 (95% CI 0.03–0.68)] were significantly less likely to use telemedicine than those in practice for ≤11 years. Also, after adjusting in a multivariate model, specialists in cornea [adjusted OR 0.10 (95% CI 0.02–0.57)] and retina [adjusted OR 0.01 (95% CI 0.002–0.12)] remained less likely to use telemedicine.

Discussion

Ophthalmology had a low conversion of in-person visits to telemedicine visits compared with other surgical specialties in a large academic health care center with centralized telemedicine support available to all providers. Our univariate model showed that providers who were female or of more junior academic rank were more likely to use telemedicine, and providers who were in practice longer, were cornea, glaucoma, or retina specialists, had a MD, MBBCh, or MBBS, or had a second degree were less likely to use telemedicine. When adjusted for multivariate analysis, providers who were cornea or retina specialists were again less likely to use telemedicine, as well as providers who were in practice longer.

JHM ophthalmology’s decline in in-person visits and use of telemedicine was consistent with trends noticed nationwide during the first 3 months of the pandemic. According to Mehrotra et al. (2020), outpatient visits across all specialties declined by ∼60% from February 1 to mid-March and remained low throughout April.⁴ Telemedicine did not ameliorate this decline as much as health care providers had hoped, as remote encounters accounted for only 30% of ongoing visits. Of the specialties investigated by Mehrotra et al. (2020), ophthalmology experienced the greatest decline in total visits from March 1 to April 5: a 79% decrease. From March 1 to May 10, ophthalmology had an overall decline of 39%, still having the largest decline of all subspecialties but one (pediatrics).⁴ Yet, this 40% increase in visits from April 5 to May 10 was due to resumption of in-person visits rather than telehealth encounters, showing that teleophthalmology did not emerge as a replacement for in-person patient care.⁴

In our study, some provider characteristics emerged as predicters of who will try and who might need more support or technology to adopt telemedicine as the pandemic persists. Our evidence suggests that providers with more years in practice and providers of more senior academic rank (i.e., professors), for example, may need more encouragement to incorporate telemedicine into their practice. A study at the Kellogg Eye Center at the University of Michigan found that the majority of their ophthalmic clinicians were at least somewhat confident about using telemedicine during the pandemic.⁵ Clinicians with high confidence were more likely to use telemedicine and express interest to continue using telemedicine in the future.⁵ Thus, cultivating clinician confidence and interest is imperative to further adoption of teleophthalmology. At JHM, a substantial proportion of the ophthalmology faculty (58%) completed at least one telemedicine encounter, reflecting a willingness to provide telehealth services.

Across all subspecialties, basic routine assessments such as visual acuity, light reflexes, pupillary response, ocular motility, confrontation visual fields, and eye alignment can still be evaluated via televisits.⁶ Saleem et al. (2020) also explained that the subspecialties of oculoplastics, pediatric ophthalmology, and neuro-ophthalmology may be better suited for televisits since the majority of examinations in these fields can be performed externally.⁶ Similarly, we found that for oculoplastics and pediatric ophthalmology, more providers used telemedicine than did not. On the contrary, our study showed that subspecialties such as cornea and glaucoma were less likely to use telemedicine, in line with Saleem et al. 2020’s discussion that these subspecialties are less amenable to virtual practice.⁶ Cornea specialists (unable to closely examine the eye surface virtually) typically experience great challenges, unless the pathology is very obvious externally, such as a viral conjunctivitis.⁶ Glaucoma specialists face barriers to obtaining tonometric measurements remotely, although televisits could be useful to assess glaucoma medication tolerance for medication reconciliation.⁶ Some providers in these subspecialties, however, have been able to overcome many obstacles by instructing and guiding patients on how to self-perform certain portions of eye examinations and supply providers with needed ocular estimates.⁵

Although utilization of telehealth by ophthalmology as a whole appreciably lagged behind all other surgical specialties, teleophthalmology uptake varied based on several characteristics, suggesting that there may be ways to expand the use of teleophthalmology. As the COVID-19 pandemic progresses, leaders and innovators in ophthalmology should determine ways to increase access to teleophthalmology care and heighten patient satisfaction through introduction of improved teleophthalmology platforms and perhaps more use of hybrid encounters. Hybrid visits could encourage teleophthalmology adoption, especially among retina specialists who were the least likely to use telemedicine in our study. A strategy for hybrid visits may be the implementation of widespread home-based posterior segment imaging (can be self-administered by patients and integrated with smart phones for data collection, storing, and transmission).⁷ In addition, clinic-based imaging equipment could be driven to patients’ homes with a van or used by certified ophthalmic photographers at other locations outside of the clinic/hospital.⁷

Moving forward, even in the post-COVID-19 era, ophthalmology may need to better adopt telemedicine to meet evolving patient preferences. A recent survey of 1,000 ophthalmology patients revealed that about 75% of respondents had a virtual visit during the current pandemic and over 75% of these patients reported that they were highly satisfied with their experience. The results from this survey show that patients embrace the opportunity of telemedicine and would most likely endorse teleophthalmology as a more permanent option.⁸ On April 30, 2020, the Centers for Medicare & Medicaid Services temporarily expanded coverage of four teleophthalmology modalities under Medicare Part B: virtual two-way communication, physician/patient phone calls, online digital service e-visits, and evaluation of video and images.⁹ These benefits applied to both existing and new patients, with reimbursement the same as an in-person visit. Due to this change, telehealth became more accessible to patients in need and want of teleophthalmology care. Prior lack of insurance coverage of these services may explain some of the low telehealth utilization in ophthalmology relative to other medical specialties. If this policy is made permanent and payment remains adequate, the means to provide remote eye care for patients will improve.¹⁰

JHM has a dedicated office of telemedicine that provides the technical infrastructure, technical training for staff and providers, as well as ongoing updates on guidelines for billing and regulations regarding telemedicine to all JHM faculty to support use and adoption of telemedicine. This served as a consistent practice environment and allowed us to compare all surgical specialties through the same lens. However, our study has several limitations. First, the low numbers of visits completed by ophthalmology might be patient driven as well as provider driven. For example, since ophthalmology patients tend to be older and many of them may have visual impairments, they may not be able to engage with telehealth as easily. This limitation needs to be addressed by introducing simpler platforms for communication. Another limitation was report from a single institution, so our results may not directly reflect utilization rates at other institutions. Future efforts to compare utilization rates at other institutions may provide insight into telemedicine adoption more broadly.

Conclusion

Utilization of telehealth by ophthalmology significantly lagged compared with all other surgical specialties. However, uptake did vary based on characteristics such as subspecialty, suggesting there may be ways to broaden teleophthalmology. Informed by this study, leaders and innovators in ophthalmology need to determine ways to increase access to teleophthalmology care and heighten patient satisfaction through improved platforms and more use of hybrid encounters.

While our study’s findings define early telemedicine adopters, study over a longer time period may reveal whether all other providers achieve similar levels or if early adoption correlates with persistent usage. As the COVID-19 pandemic continues to unfold, such findings may prove very useful to guide other departments as they attempt to expand use of telemedicine and support their faculty in adoption.

Authors’ Contributions

All authors contributed to the conception of the work or the acquisition and interpretation of the data, drafting the piece or critically revising it, and the final version of the article.

Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.