**3. Setups for Telehealth Programs**

As discussed previously, telehealth can serve three purposes: (1) screen for glaucoma, (2) evaluate the severity of glaucoma to determine the treatment plan, and (3) to monitor disease progression. How each of these purposes can be achieved should depend on the equipment/facilities available, the patient population (the prevalence of certain types of glaucoma can vary), and the socioeconomic and/or geographic barriers to face-to-face ophthalmologic care.

A number of telehealth screening programs have been implemented and can serve as templates tailored to the needs of the community. For example, the Philadelphia Telemedicine Glaucoma Detection and Follow-up Study [89] executed a program of which people at risk for glaucoma could be screened at a primary care practice or a Federally Qualified Health Center. At the first visit, the participant's medical and family history were recorded. An ophthalmic technician used a nonmydriatic, portable fundus camera to take two fundus photographs (macula and optic nerve) and one anterior segment photograph per eye. The technician measured IOP using the iCare tonometer. If the IOP was ≥30 mm Hg, the participant was referred to an ophthalmologist immediately. All information obtained was sent to glaucoma and retina specialists for reading. If a participant had an IOP of 22–29 mm Hg, an abnormal finding (such as a suspicious optic nerve appearance), or an unreadable image, he/she were contacted to schedule an eye examination at the same primary care practice or health center within 6 months. At visit two, the subjects underwent a slit lamp examination by a glaucoma specialist, along with SAP. The equipment was brought in by a community outreach van. Based on the assessment at this visit, followup testing, appointment, or treatment was recommended. This screening program was conducted over 5 years. Of the 902 people screened, 37% had an abnormal image, 17.2% had an unreadable image, and 6.9% had ocular hypertension; therefore, 59.4% were asked to attend visit 2. Of the people asked to attend visit 2, 64.7% showed up. Of those who showed up, 10.9% had glaucoma, 7.2% had ocular hypertension, and 45.8% were glaucoma suspects. Taken together, 24.6% of the original 902 people screened had glaucoma, ocular hypertension, or were glaucoma suspects in this urban, multiethnic population.

Similar to the Philadelphia screening program, the Manhattan Vision Screening and Follow-up Study in Vulnerable Populations (NYC-SIGHT) [90] is a community-based screening program to be conducted for 5 years, specifically for residents in New York City Housing Authority developments. Because of the COVID-19 pandemic, participants will have medical history obtained and a visual function questionnaire asked over the phone and will be screened for COVID-19 symptoms before being allowed at the community screening site. At the screening, visual acuity check, IOP measurement, and nonmydriatic fundus photography are performed. Participants who fail the vision screening will be scheduled to see an optometrist on-site for a refraction and a nondilated eye examination by a portable slit lamp and a direct ophthalmoscope. Participants with a high IOP, an abnormal fundus image, or a concerning examination finding will be referred to an ophthalmologist in the clinic.

The Michigan Screening and Intervention for Glaucoma and Eye Health Through Telemedicine (MI-SIGHT) [91] and the Alabama Screening and Intervention for Glaucoma and Eye Health Through Telemedicine (AL-SIGHT) [92] programs are designed to be more comprehensive at the first visit. Similarly to the Philadelphia program, federally qualified health centers are used, and people with specific risk factors for glaucoma are eligible. In the Alabama program, an ophthalmic technician checks the visual acuity, performs auto-refraction, measures IOP with iCare rebound tonometer, and takes images using a combined OCT and fundus camera machine (Maestro2, Topcon Medical Systems, Oakland, NJ, USA) and a smartphone with an adapter (D-Eye retinal camera), and performs visual field testing using the Humphrey Visual Field Analyzer and Melbourne Rapid Fields application on a tablet. In the Michigan program, an ophthalmic technician checks the visual acuity and performs refraction, assesses the iridocorneal angle with penlight, assess ocular motility, measures IOP with iCare tonometer, and dilates the participant if

IOP < 30 mm Hg and/or angles are not narrow. Fundus photographs and OCT RNFL images are obtained when the pupils are dilated. In both programs, the data are sent to an ophthalmologist for review in order to determine the appropriate follow-up.

Rather than screening the community for glaucoma, a telemedicine program [93] in Northern Alberta served as a glaucoma consult service. Patients seen by an optometrist, ophthalmologist, or family physician were referred to the program if they had risk factors for glaucoma or suspicious-looking optic discs or visual field test. At each office, a tonometer, corneal pachymeter, visual field machine, and a retinal camera were available for use by technicians. A glaucoma specialist at the University of Alberta then reviewed the data remotely and gave recommendations for management and follow-up.

In addition to screening for glaucoma, telemedicine can be used to monitor for the development of glaucoma. The Kaiser Permanente Eye Monitoring Center conducted a 2-year telemedicine program [94] to monitor low-risk glaucoma suspects. Each year, a technician checked the visual acuity, measured the IOP using a handheld applanation tonometer, and took OCT RNFL images at a local ophthalmology clinic. Different from other telemedicine programs, the data were sent to a trained technician first, rather than a glaucoma specialist. If there was a decline in visual acuity of at least two lines, an IOP elevation ≥ 5 mm Hg, or a significant change in the RNFL thickness in the superotemporal or inferotemporal region (defined as ≥10-micron reduction or transition into the abnormal red range), the technician would send the patient data to a glaucoma specialist for review remotely. Of the 225 glaucoma suspects enrolled in the program, five were referred for examination by an ophthalmologist due to concern for progression on OCT. Of those five patients, two were started on glaucoma medications. This program demonstrated that telemedicine is a viable option for monitoring glaucoma suspects and can capture the small number of patients who develop glaucoma and need treatment.

Telehealth programs can be used to monitor patients with an established diagnosis of glaucoma as well. Rutgers New Jersey Medical School conducted a program [95] on patients who had glaucoma or were glaucoma suspects. The patients went through the tele-glaucoma setup in the following order: (1) medical history intake; (2) IOP measurement with puff tonometer; (3) auto-refraction; (4) OCT imaging of the iridocorneal angle (and central corneal thickness measurement), cup/disc ratio, RNFL, and ganglion cell complex; (5) nonmydriatic color photography of the anterior segment and fundus, as well as autofluorescence imaging of the fundus. A glaucoma specialist then reviewed the data remotely and gave recommendations on management and follow-up. To compare the accuracy of the data and assess the program with a clinical examination, the subjects underwent a comprehensive eye examination on the same day by an ophthalmologist. IOP was measured by Goldmann applanation and the slit lamp was used to examine the anterior segment and fundus. OCT and visual field testing were performed under the discretion of the ophthalmologist. When comparing the tele-glaucoma program with the clinical examination, there were strong correlations in IOP measurements and cup/disc ratios. The recommended follow-up time was shorter for the tele-glaucoma program (2.7 months vs. 3.9 months). The clinical examination was better at identifying exotropia, iridotomy, iris neovascularization, and trabeculectomy. The tele-glaucoma program was better at identifying narrow angles, age-related macular degeneration, macular edema, diabetic retinopathy, retinal vein occlusion, choroidal nevus, and splinter hemorrhages.

A similar study [96] in London compared a "virtual clinic" staffed by an ophthalmic nurse versus an examination by an ophthalmologist for patients with open angle glaucoma. In the nurse clinic, a technician checked visual acuity, conducted SAP, took fundus photos, and performed scanning laser ophthalmoscopy with Heidelberg Retina Tomography. The nurse performed Goldmann applanation tonometry and a slit lamp examination of the anterior segment. The patient then was examined by an ophthalmologist and the assessment was recorded. One year later, the same ophthalmologists who took part in the study were asked to review the data from the nurse "virtual clinic" a year ago and classify whether the patient was stable or unstable based on just these data. The study found that 3.4%

of patients were misclassified as "stable" by review of the "virtual clinic" data when in fact they were "unstable" according to the in-person assessment by the ophthalmologist. The authors concluded that 3.4% was an acceptably low misclassification rate and that a "virtual clinic" run by ophthalmic nurses can be a viable option for managing relatively stable glaucoma patients.

A review of these glaucoma telehealth programs shows that a variety of setups can be used to screen for glaucoma and monitor for disease progression. At the bare minimum, a technician should record the patient's medical history, visual acuity, IOP, and take a fundus photograph. Technological advances have allowed IOP measurements without the use of topical anesthetic and fundus imaging without pharmacologic pupillary dilation. SAP to detect visual field scotomas and OCT to detect structural nerve fiber layer loss can provide additional valuable data, but these bulky machines are unlikely available outside of the ophthalmology clinic setting. As deep learning artificial intelligence technology matures, fundus imaging may be all that is needed to accurately predict RNFL thickness and visual field loss. Artificial intelligence will play a significant role in reducing the amount of equipment required for glaucoma screening and monitoring through telehealth. A summary of the components of a glaucoma telehealth examination is listed in Table 1. Even at its current state, without reliance on artificial intelligence, telehealth has shown to be cost-effective. An analysis of remote glaucoma screening in rural Alberta, Canada revealed that teleglaucoma costs an average of CAD 867 per patient, which was dramatically less than the average CAD 4420 per patient for in-person screening [97]. In order to control healthcare costs while providing access to care, especially in rural regions, telehealth will become an important tool in the screening and monitoring of chronic diseases such as glaucoma.


**Table 1.** Components of a glaucoma telehealth examination.


**Table 1.** *Cont.*

#### **4. How the Coronavirus Pandemic Shaped Telehealth**

In December 2019, a novel respiratory illness COVID-19 emerged in Wuhan, China. Because of the disease's highly contagious nature, it quickly spread globally, and a pandemic was declared by the World Health Organization on 11 March 2020. Governments worldwide imposed lockdowns to stop the spread of disease, as COVID-19 overwhelmed hospital systems with vast numbers of people requiring ventilators. In the United States, state governments issued stay-at-home orders and social distancing guidelines. People were asked to work from home and to avoid venturing outside except for essential activities. On 18 March 2020, the American Academy of Ophthalmology (AAO) recommended the cessation of elective surgery and routine clinic visits to protect patients from catching COVID-19 and to conserve personal protective equipment (PPE).

Because many ophthalmology practices closed their offices, telehealth through video visits became a necessary way for patients to see their doctors. In the United States, the Centers for Medicare and Medicaid Services (CMS) relaxed the requirements to bill for telehealth visits; thus, allowing practices to be reimbursed for remote patient care. Many practices implemented a telemedicine program for the first time and had to develop protocols to address patient needs virtually. Saleem et al. [98] depicted a workflow diagram as a reference to implement an ophthalmology telemedicine program. Essentially, the front desk staff reaches out to patients who had their appointments canceled and offers them a telephone or video visit for non-urgent problems. If the patient describes an issue that appears to be an emergency, the physician is contacted to determine whether the problem can be addressed remotely or the patient must be examined in person.

A major hurdle in managing glaucoma patients through video visits is that glaucoma, for the most part, is an asymptomatic disease, unless there is a substantial increase in IOP causing eye pain or rapid visual field loss causing noticeable constriction in vision. A video visit does not allow for IOP measurement, visual field testing, or the visualization of the optic nerve. A crude method for the patient to estimate IOP is via finger palpation on the eye through the eyelid. A more accurate way than digital palpation is for the patient to use the iCare HOME rebound tonometer on him/herself. The tonometer is easy to use, comfortable, and requires no topical anesthetic. Because the device is expensive, companies such as MyEYES (myeyes.net) and Enlivened (enlivened.com) offer rentals for a fee. Patients are taught how to use the device and borrow it for one or more weeks. The downside, however, is that the IOP readings are not displayed; the patient must return the device to the office to extract the IOP diurnal curve. An alternative to using the iCare HOME is to wear the Sensimed Triggerfish® Contact Lens, which makes automated corneoscleral dimensional measurements for 24 h. However, the patient is required to have a circular antenna taped around the eye, wear a recording device hanging from the neck, and return to the office the next day to extract the diurnal curve. A new contact lens being developed in South Korea allows for convenient IOP monitoring using a smartphone. Implantable devices such as the Eyemate® and Injectsense can provide IOP monitoring as well. If there is concern for visual field progression, the patient can use the computer-based Peristat Online Perimetry, the tablet-based Melbourne Rapid Fields program, or virtual reality perimetry to generate a visual field report and send it to the physician.

As pandemic lockdown restrictions loosened, ophthalmology practices reopened with the implementation of new protocols for the safety of the patients and staff members. Vinod et al. [99] described methods that practices used to enforce social distancing and enhance safety, such as limiting the number of appointments, rearranging chairs in the waiting rooms, asking patients to remain in their cars outside the clinic until they are called, mandating everyone to wear masks, and installing large breath shields on slit lamps. However, some patients are still uncomfortable with in-person examinations and prefer telehealth until the pandemic ends.
