Physician-to-Physician eConsultations to Ophthalmologists at an Academic Medical Center

Abstract

Purpose: Physician-to-physician electronic consultation (eConsults) are used across specialties; however, their effectiveness in ophthalmology has not been reported. This study evaluated the feasibility and diagnostic accuracy of a physician-to-physician ophthalmology eConsult program, assessed the timeliness of in-person evaluation, and characterized the clinical questions of non-ophthalmology providers.

Methods: Retrospective review of patients for whom an ophthalmology eConsult was placed to Massachusetts Eye and Ear from February 2019–August 2021. The eConsults were reviewed to identify submission-to-response time, primary diagnoses by eConsultant, and referral outcomes. The eConsults were categorized based on clinical question, urgency, and ophthalmic condition addressed. Demographic data on patients and referring providers were collected.

Results: One hundred ophthalmology eConsults were placed, and 100% were responded to by an ophthalmologist. An average of 1.6 ± 1.9 days elapsed from the time of eConsult to completion. Of the 30 patients who presented for in-person evaluation at an ophthalmology clinic, diagnostic concordance between eConsultant and in-person ophthalmologist was observed in 93% of cases (n = 28 of 30). An average of 28.9 ± 27.4 days from eConsult response to in-person follow-up was observed. The most common clinical inquiries were about appropriate triage/referral (24.4%), management (22%), and diagnosis (19.7%). All eConsults were non-urgent. The most common ophthalmic condition addressed was chalazia/hordeola (14%). Only 5% of patients presented to an emergency department for the same ophthalmic concern addressed by eConsult.

Conclusions: Ophthalmology eConsults provide timely access to nonurgent ocular diagnosis, triage, and management and in our study seem to provide high rates of diagnostic accuracy for nonurgent ophthalmic conditions.

Translational Relevance: Using ophthalmic eConsults can facilitate timely access to specialty care and reduce patient and provider burden.

Introduction

Electronic consultations (eConsults) are asynchronous, provider-to-provider exchanges that occur within shared electronic medical record (EMR) systems or secure online platforms.¹ In eConsults, referring providers submit a clinical question via an EMR, and consulting providers review that question, along with available information in the patient's medical record, and document clinical impressions and recommendations in a clinical note; there is no direct consultant/patient interaction. This asynchronous method of telecommunication confers a range of advantages to patients, providers, and healthcare systems, including increased timely access to specialist consultation, decreased resource waste, improved care coordination, greater patient and provider satisfaction, and cost savings.^2–6 

Through eConsults, providers can determine whether patients require in-person evaluation, which can decrease the number of unnecessary referrals. When patient referral is deemed beneficial, eConsults can streamline the process by identifying the appropriate subspecialty for referral.^1,2 Furthermore, eConsults can determine necessary pre-visit diagnostic workup or imaging, thus optimizing clinical evaluation during the visit and reducing further delays in patient care.^2,7 In addition to care coordination, eConsults can ensure that subspecialty evaluation is achieved. The eConsults can bypass many of the hurdles associated with completing specialist referrals, increase physician follow-up and awareness of referral completion, and remove the burden of seeking specialty care from patients.⁸ By improving the efficiency and effectiveness of interprofessional clinical exchange, eConsults can also improve access to and quality of care.^1,6 However, eConsults also have drawbacks. Namely, there are risks of diagnostic errors and concerns for patient safety given the lack of an in-person examination.⁹ 

Ophthalmology, like many other medical specialties, has increased use of telemedical modalities including eConsults during the COVID-19 pandemic as a means of continuing access to specialty care while limiting in-person exposures.^10,11 Many subspecialties of medicine have reported on the benefits of eConsults.^1,12 Despite the many known advantages of eConsults, the role of eConsults in ophthalmology has yet to be explored. In this study, we describe the use of ophthalmology eConsults in an academic medical center. We report specifically on the diagnostic accuracy, outcomes, and response timeliness of eConsults, as well as the different ophthalmic conditions inquired about through eConsults. In doing so, we characterize the types of clinical questions asked and identify the learning needs of nonophthalmic providers. Finally, we assess the utility of eConsults to safely and comprehensively manage nonurgent ophthalmic conditions remotely. Unlike other literature published on various modalities of ophthalmic telehealth, this study is the first to describe the feasibility and potential impact of eConsults in ophthalmology. 

Material and Methods

Study Design

We conducted a retrospective cohort study of all eConsults submitted to the Ophthalmology Department of Massachusetts Eye and Ear (MEE) from February 11, 2019 through August 18, 2021. 

The eConsult Program

Massachusetts General Brigham (MGB) is a tertiary academic health care system with multiple inpatient and ambulatory health centers throughout the state of Massachusetts. The MGB eConsult program was established in 2013 and expanded to include ophthalmology eConsults in 2019.¹³ This program enables providers across all specialties within the MGB healthcare system to submit a clinical question to an ophthalmologist through a shared EMR system, and the ophthalmologist can respond via the EMR without evaluating the patient in person. The eConsult is requested through an EMR order and consists of three fields for referring providers to complete: (1) the reason for eConsult, (2) specific patient care questions, and (3) any additional comments, pertinent patient history, and attachments. Clinical images taken by the referring provider can be uploaded into the EMR. In this study, the term “referring provider” refers to a non-ophthalmology healthcare provider who submits an eConsult to an ophthalmologist on behalf of a patient. The terms “consulting provider” and “eConsultant” are used interchangeably and refer to the ophthalmologist receiving and reviewing the eConsult. The term “referral request” refers to the act of placing an electronic order in the EMR to refer a patient for an in-person evaluation by an ophthalmologist. All ophthalmology eConsults were assigned to and addressed by a comprehensive ophthalmologist, who additionally reviewed the patient's medical history, ophthalmic history, recent progress notes, associated images, and diagnostic tests. The general ophthalmologist asked for clinical advice from appropriate ophthalmology subspecialists to answer eConsult questions as needed. Clinical recommendations were returned to the referring provider through an “eConsult note” within the EMR. The referring provider then determined next steps, including relaying the eConsultant's recommendations to the patient, scheduling appropriate follow-up, and determining which aspects of the eConsultant recommendations act upon. A workflow diagram of the eConsult process is included in Figure 1. There was no direct ophthalmologist-to-patient exchange, and all eConsults included standardized language indicating that ongoing management of the patient's clinical problem is the responsibility of the referring provider and other members of the patient's care team, that eConsults are conducted in the absence of an examination or direct conversation with the patient, and that recommendations made are solely based on the information provided by the referring provider in the eConsult and information available in the EMR. The eConsult program is funded internally by MGB and does not involve charges to patient insurance; rather, MGB pays eConsultants a standard fixed fee for their consulting service. 

Figure 1.

 

Workflow of eConsults. Workflow of the electronic consultation program. Patient presents to a non-ophthalmologist healthcare provider. An eConsult order to ophthalmology is placed in the EMR, which is reviewed by an ophthalmologist. The ophthalmologist reviews and responds to the clinical question and may place a referral for the patient to receive in-person eye care by an ophthalmologist.

Data Extraction and Analysis

All MEE ophthalmology eConsults were retrospectively reviewed through manual review of eConsult encounters in the EMR system, including the reason for eConsult, the date and time an eConsult order was placed and subsequently responded to, the presence of external images of patient eyes, results of diagnostic and screening tests (e.g., fundus photography, newborn vision screens, computed tomography, magnetic resonance imaging, radiography), eConsultant diagnosis, and clinical recommendation (Table 1). Patient demographic data, including patient age at time of eConsult, sex, race, ethnicity, and insurance coverage status, were collected. Demographic data of referring providers (medical specialty, academic degree, and location) were also collected (Table 2). Whether patients had known past ocular history or an established ophthalmologist, defined as any visit with an ophthalmologist in the five years before eConsult, was also recorded. Each eConsult was classified into clinically meaningful diagnostic categories based on the literature.¹⁴ The eConsult questions underwent thematic review to identify the types of clinical questions asked by providers and each question was assigned a category type. Seven question types were identified, including issues of diagnosis, treatment, management, triage/referral, workup, risk assessment and timing of routine screenings, and medication management (Table 2). The eConsults were assessed for whether follow-up with an ophthalmologist was recommended, which party (eConsultant or referring provider) was responsible for placing the referral, whether the referral was placed, and whether a follow-up appointment occurred. Patient charts were reviewed up to six-months following the initial eConsult to assess for subsequent emergency department (ED) encounters associated with the eConsult concern, evidence of eConsult recommendations being relayed to the patient, and patient presentation to an ophthalmologist for in-person evaluation within the MGB system. Clinical details of subsequent in-person ophthalmology visits were assessed, including the diagnoses made during in-person ophthalmology visits which were assessed for concordance with any diagnoses made during eConsults. Institutional level data on ICD-10 codes for patient visits to the MEE ophthalmology-specific ED during the period of this study were obtained through automated EMR reports. This data was used to identify the most common eye diagnoses presenting to the MEE ED, which were used as a comparison for the eConsult eye concerns. Descriptive statistics and chi-square analyses were executed using R version 4.1.0 (R Foundation for Statistical Computing). A P value of α <0.05 was considered to be statistically significant. 

Table 1.

 

The eConsult Response Time, Specialist Referrals, and Data Content

 

Table 2.

 

eConsult Types of Clinical Questions

 

Institutional Review Board

This study protocol was reviewed by the MGB Institutional Review Board and determined to be a Quality Improvement study. Therefore approval was not required. 

Data Availability

The data used in this study is available on request. 

Results

A total of 100 pediatric and adult ophthalmic eConsults were ordered and completed between February 11, 2019, and August 18, 2021. Ophthalmic eConsults were most frequently ordered by internal medicine providers (67%), followed by family medicine providers (13%), pediatricians (9%), medicine-pediatrics providers (5%), and six other types of specialists (Fig. 2). Nearly one quarter (24%) of eConsults asked two or more clinical questions. Regarding ophthalmic diagnoses, hordeola and chalazia accounted for 14% (14) of eConsults, with a total of 45 different ophthalmic diagnoses being asked about (Table 3). All eConsults were considered nonurgent. Of the 100 eConsults ordered, 52% included a picture, and 8% included diagnostic imaging from previous visits. The average response time and standard deviation (SD) from when the eConsult order was placed to the time the consultation was completed was 1.6 days (SD ±1.9). The average number of days from eConsult response to in-person follow-up was 28.9 (SD ± 27.4) days. Of the 100 eConsults, in-person evaluation was recommended in 62% (n = 62) of cases; roughly 13% of these were due to medication follow-up (e.g., initiation of ophthalmic steroid drops). For patients requiring referral for in person evaluation by an ophthalmic subspecialty, MEE staff volunteered to submit the referral request for 41.9% (n = 26) of patients to ensure rapid referral placement and patient follow-up; referring providers were responsible for submitting a referral request for the remaining 58.1% (n = 36) of patients. However, the rate of actually submitting those referrals varied where MEE staff submitted the referral request 76.9% (n = 20/26) of the time whereas the referring provider submitted the referral request 66.7% (n = 24/36) of the time. No statistically significant difference was found between referring providers and eConsultants in the likelihood of placing the referral request (P = 0.3799). Of the 62 patients recommended for in-person evaluation, 48.4% (n = 30) presented to an ophthalmologist for evaluation. Agreement in diagnostic concordance between the eConsultant clinical diagnosis and in-person diagnosis occurred in 93.3% (n = 28) of cases. Two cases of non-concordant diagnoses were documented (6.9%). The first was a case of light sensitivity and blurred vision secondary to central retinal vein occlusion initially attributed to post-concussion syndrome; the second was a case of orbital fat prolapse initially thought to be a hordeola. Referring providers proceeded to follow eConsult recommendations in 91% (n = 91) of cases. Evidence that the referring provider relayed the ophthalmologist plan to the patient was documented in 75% (n = 75) of cases based on documentation anywhere in the EMR (e.g., visit note, telephone encounter). Only 5% (n = 5) of patients presented to any MGB ED for issues related to the eConsult after the eConsult was placed. Seventy-nine percent of patients presented to any MGB clinic within the six months after the eConsult request, which indicated they were not lost to follow-up in the system. Before the time of eConsult placement, the vast majority of patients (73%, n = 73) had never seen an ophthalmologist or had any documented visit from an ophthalmologist accessible in the EMR and had no known ocular history (62%, n = 62) (Fig. 3). Patient demographics including age, race/ethnicity, and sex are presented in Table 4. 

Figure 2.

 

eConsult referring profession and referring specialty. Proportion of providers placing eConsult orders to ophthalmology. The majority of patients are physicians, and the majority are internal medicine specialists.

Table 3.

 

Diagnosis Groups and Frequency of eConsult (n = 45)

 

Figure 3.

 

Patient ocular history and insurance status. Ocular history and insurance status of patients for which an eConsultation was placed. Nearly one quarter of patients have an established relationship with an eye doctor. The majority do not have any ocular history of eye conditions. Only 1% are uninsured.

Table 4.

 

Patient Demographics

 

Discussion

In this study we found that eConsults in ophthalmology provide for high diagnostic accuracy, are useful across a range of clinical diagnoses and concerns, and result in timely responses. Additionally, this study thematically describes eConsult questions and topic areas. Although eConsults have not previously been reported within the field of ophthalmology, this study supports the efficacy and feasibility of eConsults to provide asynchronous specialty advice to non-ophthalmic providers.^1,12 

Clinical Outcomes

The results of this study demonstrate the feasibility of eConsults to accurately diagnose nonurgent ophthalmic conditions. Of the patients recommended for an in-person evaluation with an ophthalmologist and subsequently followed up, concordance between the clinical assessment of the eConsultant and in-person ophthalmologist occurred in 93.1% of cases, with only two cases of missed diagnoses identified (central retinal vein occlusion and orbital fat prolapse as described in the results section). The first of the two cases of missed diagnosis deserves special attention. This case was a patient who presented two months after a motor vehicle accident that resulted in a concussion with light sensitivity and blurred vision. The patient’s symptoms were initially thought to be related to post-concussion syndrome; however, the eConsultant recommended the patient be evaluated to rule out other serious and vision-threatening etiologies. This patient was given an in-person appointment within three weeks of the eConsult response, during which this patient was found to have multiple peripheral retinal hemorrhages and elevated intraocular pressure and was referred for evaluation of a central retinal vein occlusion by the retina service. This scenario highlights the importance of referral for in-person evaluation to fully evaluate concerning clinical presentations and to rule out serious pathology. The growing practice of eConsultations will help build a body of evidence that can be leveraged to provide guidance in the future on such cases. The rate of diagnostic accuracy found in this study is similar to prior reports of high diagnostic accuracy when eConsults were used in other subspecialties such as dermatology.¹⁵ This is encouraging because access to ophthalmic care has become increasingly challenging given the growing shortage of ophthalmologists, persistent geographic barriers to ophthalmic care, and widening socioeconomic inequities in accessing specialty care.^16–19 

Ophthalmology Referrals to Subspecialty Services

The eConsults allow prompt ophthalmology evaluation to be achieved for patients presenting with ocular concerns in non-ophthalmic ambulatory care settings. Rates of completion for referrals to in-person specialty care vary tremendously from 30% to 80% across specialties, making eConsults an attractive alternative.^20,21 Among ophthalmology referrals for in-person evaluation specifically, reported rates of completion are similarly variable, ranging from 5% to 75% within the recommended referral period.^22,23 Patients also report that the cost of visits, insurance status, distance to clinics, lack of transportation, language barriers and limited health literacy, and work schedule conflicts lead to delayed or complete inaccessibility to ophthalmic care.^18,23,24 The eConsults address many of these barriers through the elimination of time, cost, or travel to achieve an initial subspecialty evaluation. Additionally, eConsults create a direct line of communication among all members of the patient’s care team, delivering well-coordinated, high-quality patient-centered care.² In the present study, 100% of submitted eConsults were completed, and 75% of cases had evidence that the ophthalmologist's recommendations were communicated to the patient. This reflects the suitability of eConsults as a mechanism to ensure patient access to ophthalmic care. 

Timing of eConsults and Subsequent Evaluation

Importantly, eConsults have a demonstrated benefit of decreasing excessive wait times to reach specialty care.^1,9,25 In the literature, the estimated median time from when a specialist referral is placed to the completion of an in-person visit is 7.5 to 8.7 weeks.^26,27 In our study, the average response time to complete an eConsult was 1.6 days. This is consistent with the results of other eConsult programs, including dermatology, allergy/immunology, endocrinology, and rheumatology, where eConsult were completed within three days.^13,28–30 This is an especially notable benefit of eConsults because lengthy waiting periods are a common deterrent to seeking follow-up care, and the wait times for specialty appointments are steadily increasing.^26,31 The eConsults also improve referral quality, ensuring that patients requiring in-person evaluation are scheduled with the appropriate subspecialist.^1,2 In our study, 50% of subsequent referrals were made to ophthalmology subspecialties, including retina, pediatrics, neuro-ophthalmology, and oculoplastics. The eConsult program thus bypassed the standard practice of an initial in-person evaluation by a comprehensive ophthalmologist prior to receiving a subspecialist appointment. Of the many benefits that can be inferred from this, key among them is expedition of time to care, decreased unnecessary patient visits, and associated patient cost savings. The ability of eConsults to decrease unnecessary specialist referrals and potential economic advantages to the healthcare system merits further investigation in future studies. 

Advantages of Ophthalmology eConsults

The eConsults also confer potential advantages across other care settings. An average of two million eye-related ED visits occur annually, with nearly half of those visits being for nonurgent conditions that can safely be managed in the outpatient setting.^32,33 EDs across the country are poorly positioned to provide optimal ophthalmic care because of inadequate or nonexistent ophthalmology service coverage and insufficient ophthalmic training for ED providers—limitations that are exacerbated in rural and underserved populations.^34,35 These limitations have also been associated with ED provider discomfort and inaccuracy evaluating ocular concerns.³⁵ At the national level, hordeola are the second-most common cause of nonurgent ocular presentation to an ED.³² At the institutional level, 6% of all ophthalmic related presentations to the MEE ED (approximately 935 visits annually) from May 2019 to June 2021 had a primary diagnosis of hordeola and chalazia. In our study, hordeola and chalazia were the most common ophthalmic conditions inquired about in eConsults, accounting for 14% of the submitted eConsults. All eConsults related to hordeola and chalazia were managed conservatively with symptom improvement or resolution documented in patients’ charts, except in the case of a missed diagnosis described above. None of these eConsults were followed by an ED presentation, which suggests that nonurgent ocular conditions can safely and accurately be diagnosed and managed remotely and are a potential source of avoidable ED visits. Evidence suggests that the cost of managing nonurgent ophthalmic conditions in the ED can be anywhere from two to four times higher than in an ambulatory care setting.^32,33 Ophthalmic eConsults can help patients and healthcare systems avoid unnecessary costs for nonurgent, non-vision-threatening ophthalmic issues.³³ Additionally, reduction in nonurgent ocular ED visits can help decrease overall ED crowding and allow resources to be directed toward patients with urgent ophthalmic and medical conditions.³³ 

The eConsult offers educational advantages to providers and medical educators.^36–38 By communicating a plan of care back to the referring provider, eConsults create an opportunity for immediate feedback and education of the referring provider as it relates to ophthalmic issues.^3,38 This is in contrast to the traditional referral process where providers may not be aware of the specialist recommendation or plan of care for prolonged periods of time. Furthermore, providers will be able to apply information learned in the eConsult to similar cases in the future. Understanding the types of clinical problems and content areas providers most commonly ask questions about is necessary to better guide educational efforts. For example, this study suggests that further education on hordeolum/chalazion identification and management may be beneficial to physicians in ambulatory care settings. 

Limitations

There are a few notable limitations in the use of eConsults in our study. First, the present study included only 100 eConsults over a 30-month period submitted by 74 individual providers, which suggests that many of the nearly 1400 primary care providers in the MGB system were unaware of or did not use the service.³⁹ Understanding attitudes toward ophthalmic eConsult practices and seeking a broader sample size would be advantageous to inform widespread implementation and adoption of such programs. Second, the role of eConsults in decreasing or exacerbating disparities in ophthalmic care has yet to be established.⁴⁰ An understanding of how eConsults can be used to decrease disparities in the provision of healthcare would be useful. Another important limitation is the retrospective nature of this study, because follow-up of patient outcomes was limited by chart review and documentation. Additionally, the percentage of patients who presented to an ophthalmologist after receiving a referral for an in-person evaluation was low, thus limiting our ability to assess the true rates of diagnostic concordance. Furthermore, all eConsults were answered by a single ophthalmologist at a single academic institution. Patients who sought subsequent eye care outside of this institution were unidentifiable in this study, because they would not be identifiable in the institutional EMR. Future research should evaluate the use of eConsults in various practice sites and locations, which would improve the generalizability of the findings. This study did not collect survey data on patient perspectives regarding confidence in, comfort with, or perceived benefits of eConsults. As a result, this study is unable to comment on patient perspectives of the ophthalmology eConsult program and the potential impact it could have on seeking care elsewhere or delaying care if symptoms progress. Future studies should consider a patient survey component. Last, this study focuses solely on eConsults and is therefore unable to compare the advantages and limitations of eConsults to other synchronous and asynchronous forms of telehealth. 

In this study, ophthalmic eConsults were associated with timely response back to the referring provider, and with a high rate of diagnostic accuracy among a subset of patients subsequently seen for an in-person ophthalmology visit. Our results support the use of eConsults as an effective telehealth modality to obtain timely diagnosis, access, and management of nonurgent eye conditions. Our study also demonstrates that eConsults enhance patient quality of care by soliciting specialist input, ensuring that timely ophthalmology evaluation is achieved, coordinating appropriate referral management and triage, and collaborating across interdisciplinary care teams. 

Acknowledgments

The authors thank Frances McDonald and Anne Murphy of the Information Systems department of Massachusetts Eye and Ear for their help in extracting clinical information from the EMR. 

Disclosure: N.A. Sherif, None; A.C. Lorch, None; G.W. Armstrong, Ocular Technologies Inc, McKinsey & Company, Kriya Therapeutics, Xenon Ophthalmics, Chart Biopsy, Dynamed, Optomed, Genentech (C)