eTable 1. Survey Instrument Administered to Emergency Departments
eTable 2. Factors Affecting Perception of the Value of Teleophthalmology for Patient Triage and Full Consultation
Wedekind L, Sainani K, Pershing S. Supply and Perceived Demand for Teleophthalmology in Triage and Consultations in California Emergency Departments. JAMA Ophthalmol. 2016;134(5):537-543. doi:10.1001/jamaophthalmol.2016.0316
Determining the perceived supply and potential demand for teleophthalmology in emergency departments could help mitigate coverage gaps in emergency ophthalmic care.
To evaluate the perceived current need for and availability of ophthalmologist coverage in California emergency departments and the potential effect of telemedicine for ophthalmology triage and consultation.
Design, Setting, and Participants
Surveys were remotely administered to 187 of the 254 emergency departments throughout California via the telephone and Internet from June 30 to September 23, 2014. Emergency department nurse managers and physicians from all emergency departments listed in the California Office of Statewide Health Planning and Development database were individually surveyed to assess facility characteristics and resources as well as the perceived usefulness of teleophthalmology consultation. Data analysis was conducted from June 30, 2014, to March 11, 2015.
Main Outcomes and Measures
Perceived availability of ophthalmology consultation coverage and perceived effect of telemedicine ophthalmology consultation at each facility.
Of the 187 emergency departments surveyed, 18 of 37 rural facilities (48.6%) reported availability of emergency ophthalmology coverage, compared with 112 of 150 nonrural facilities (74.7%). Rural facilities reported a mean (SD) of 23.72 (14.15) miles between the facility and referral location, while nonrural facilities reported a mean of 4.41 (10.23) miles (19.3% difference). On a scale of 1 to 5 (where 1 signifies very low value and 5 signifies very high value), 124 of 187 nurse managers (66.3%) and 80 of 121 physicians (66.1%) rated teleophthalmology as having high or very high value for triage purposes. The most frequently cited potential advantage of emergency teleophthalmology was assistance in patient triage and immediate real-time electronic communication, and the most frequently cited potential disadvantages were unknown cost of contracting and maintenance and concern that eye trauma might make photographs or videos less conclusive.
Conclusions and Relevance
Availability of ophthalmology coverage for emergency eye care is limited, particularly among rural emergency departments in California. Surveyed emergency department nurse managers and physicians indicated moderately high interest and perceived value for a teleophthalmology solution for remote triage and consultation. Overall, the study suggests that teleophthalmology could play a role in mitigating coverage gaps in emergency ophthalmic care and could be further investigated through similar studies in other regions.
Rural emergency departments often have limited resources relative to nonrural emergency departments and patients may face inequities in access to quality emergency care based on their proximity to health care facilities.1 These geographic differences in resources result in coverage gaps, which are more severe for access to specialized care such as ophthalmology.2 Approximately 18 805 board-certified ophthalmologists practice in the United States, with disproportionately more in nonrural vs rural regions.3 Prior studies have demonstrated that the likelihood of adults with diabetes receiving annual dilated eye examinations is reduced in areas with a lower geographic density of eye care professionals.4 Currently, there is a lack of published literature on the unavailability of on-call ophthalmologists for emergency eye care.
Emergency eye care exhibits significant coverage gaps, especially in rural settings with greater distances between emergency care facilities. Patients requiring emergency eye care may be treated by local on-call ophthalmologists or may need to be transferred to another facility with available ophthalmologists. Multiple studies since the latter part of the 20th century have suggested that applying telemedicine to eye care in emergency situations may assist with patient flow in remote emergency departments and help fill coverage gaps in 2 main ways: assistance with triage within and between remote emergency departments, and performance of remote ophthalmic consultations for patients presenting with emergency eye conditions.5- 8
According to the American Telemedicine Association, telemedicine is “the use of medical information exchanged from one site to another via electronic communications to improve a patient’s clinical health status.”9 Teleophthalmology, which involves telemedicine applied to clinical ophthalmology, is already in widespread use for nonemergency eye care. It is well established to screen for diabetic retinopathy10- 13 and retinopathy of prematurity.14- 16 The concept of Internet-based counseling of ophthalmology patients located remotely has existed for more than a decade,17 and its use for urgent consultations has been described internationally, especially in Australia.18- 22 It has also been cited as a teaching tool to educate general practitioners.19,20 In the United States, teleophthalmology has been used by large health care system networks, such as Kaiser Permanente and the US military23 and Veterans Affairs systems. For example, since 2005, Veterans Affairs hospitals have implemented teleretinal imaging, a store-and-forward mode of teleophthalmology that uses digital retina cameras and remote image interpretation18,24; this program has been successful for accurate and sensitive diagnosis and referral.25 Anticipated shortages of health care professionals and new technology may make teleophthalmology increasingly needed and increasingly feasible in broader contexts in the US health care system. Teleophthalmology has been heralded as a potentially radical transformer of care delivery.26,27
Despite the logistical challenges for a feasible teleophthalmology program, previous studies have successfully implemented Internet-based emergency teleophthalmology pilot programs in rural areas, achieving reductions in the need for acute transfers to external tertiary care centers, shortened time to receipt of care, projected increased efficiency, and cost-effective delivery of care.5- 7 When teleophthalmology was applied in an ophthalmic emergency department, images of the anterior segment and posterior pole were found to have high clinical utility (100% agreement between telemedicine and on-site examination diagnoses), while images of the vitreous and peripheral retina had less utility.8 Patient satisfaction was high, with 48 of 49 patients (98%) indicating a preference for teleophthalmology at their next emergency department visit, rather than traditional on-site examination Teleophthalmology evaluation has the potential to expand access to eye care in areas with little to no existing in-person coverage; however, it is not yet the standard of care for addressing emergency eye problems in underserved areas.
In this study, we developed and prospectively administered a structured survey to a comprehensive list of California emergency departments to evaluate current availability of emergency eye care by ophthalmologists and the potential utility of teleophthalmology for patient triage and/or consultations.
Question What is the perceived need for and current availability of ophthalmologist coverage in emergency department settings and what would be the perceived potential effect of teleophthalmology consultation in California emergency departments?
Findings Eighteen of 37 rural facilities (48.6%) reported existing emergency ophthalmology coverage vs 112 of 150 nonrural facilities (74.7%), and 123 of 187 nurse managers (65.8%) and 58 of 121 physicians (47.9%) rated teleophthalmology as having high or very high value for triage.
Meaning Teleophthalmology may play a role in mitigating coverage gaps in emergency ophthalmic care and warrants further study.
We developed a database of all 254 California emergency departments that are included in the outpatient emergency department encounter data published by the California Office of Statewide Health Planning and Development (OSHPD).28 Facilities were identified as rural or nonrural in the OSHPD database based on acute care capacity, local population size, and distance from the most common referral facility. The OSHPD defines a rural hospital as one that is “a general acute care hospital of no more than 76 general acute care beds that is also located in an incorporated place or census tract of 15,000 or less population according to the 1980 census,”29 while nonrural facilities include all that do not meet these criteria. Self-reported outpatient emergency department encounter data, including patient demographic information, patient volume, number of surgical procedures performed, rural hospital status, and disproportionate share hospital status (hospitals serving above a certain proportion of low-income patients and receiving a corresponding adjustment in Medicare payments to offset undercompensated or uncompensated care),30 were available for each facility from submissions through the Medical Information Reporting for California system.31 We excluded 67 emergency departments from the study that did not report data for the January 1 through March 31, 2014, reporting period, and we contacted 187 facilities through this study.
We contacted nurse managers and physicians from each emergency department by telephone and email from June 30 to September 23, 2014, to participate in surveys evaluating the current availability of ophthalmology coverage and potential demand for teleophthalmology at their facilities (eTable 1 in the Supplement). In part 1 of this survey, nurse managers were asked to evaluate the availability of on-call ophthalmologists and retrospectively estimate the volume of referrals of patients requiring emergency eye care to external facilities in 2013 and 2014. In part 2 of the survey, up to 1 nurse manager and 1 physician per facility were asked to rate the potential utility of teleophthalmology for assistance in triage and full emergency consultation. Each respondent reported his or her perception of the value of teleophthalmology for triage and full remote consultation by a teleophthalmologist and selected any potential advantages and disadvantages that contributed to their decisions from a standardized list (or the option “not applicable”). We obtained exemption from the Stanford University Institutional Review Board as no identifiable patient data were used in this study.
Data analysis was conducted from June 30, 2014, to March 11, 2015. We used R, version 3.1.2 (R Foundation for Statistical Computing), to calculate basic descriptive statistics on hospital characteristics, estimates of potential demand for emergency teleophthalmology, and standardized advantages and disadvantages. Means and SDs of each facility characteristic were calculated individually for 3 groups: all facilities, nonrural facilities, and rural facilities. Multivariable linear regression analysis (with a threshold of P < .05 and 95% CI) was used to evaluate the association between reported advantages and disadvantages and subjective ratings of potential usefulness of emergency teleophthalmology. Since rural facilities often face lower per capita availability of ophthalmologists, nonrural facilities and rural facilities were analyzed separately.
Aggregate reported characteristics of surveyed emergency departments (from OSHPD data) are listed in Table 1. Of the 254 emergency departments, 43 (16.9%) were classified as rural and 211 (83.1%) as nonrural.
Data collected from nurse managers (Table 2) showed that almost all facilities encountered patients requiring emergency eye care, and approximately half had to transfer such patients to another facility for evaluation and management, with higher transfer rates among rural facilities (nonrural, 68 of 150 [45.3%] in 2013 and 2014; rural, 28 of 37 [75.8%] in 2013 and 32 [86.5%] in 2014). Most facilities had at least 1 ophthalmologist on staff, but fewer ophthalmologists available after hospital operating hours. Rural facilities in particular faced challenges in arranging timely evaluation by an ophthalmologist: 18 (48.6%) lacked readily available access to an ophthalmologist.
Of the 187 emergency department nurse managers contacted, all completed parts 1 and 2 of the survey. On a scale of 1 (very low value) to 5 (very high value), nurse managers rated the potential value of emergency teleophthalmology as a mean of 3.78 for assistance in triage (3.63 for nonrural facilities and 4.44 for rural facilities) and 3.51 for obtaining full consultations remotely (3.35 for nonrural facilities and 4.26 for rural facilities). Although rural facilities had lower patient volume and emergency ocular surgical procedures compared with nonrural facilities, in aggregate, the proportion of patients undergoing ocular surgery was slightly higher among rural facilities.
By contrast, of the 187 emergency department physicians contacted, 166 completed parts 1 and 2 of the survey (88.8% completion rate). On a scale of 1 to 5, physicians rated the potential value of emergency teleophthalmology as a mean of 3.37 for assistance in triage (3.26 for nonrural facilities and 3.96 for rural facilities) and 3.19 for obtaining full remote consultations (3.09 for nonrural facilities and 3.67 for rural facilities). On a scale of 1 to 5, a total of 124 nurse managers (66.3%) and 80 of 121 physicians (66.1%) rated teleophthalmology as having high or very high value (4 or 5) for triage purposes (Table 3).
The most commonly cited perceived advantages of teleophthalmology (Table 4) were assistance in patient triage and immediate real-time electronic communication; the most commonly cited perceived disadvantages were unknown cost of contracting and maintenance and concern that eye trauma might make photographs or videos less conclusive. Physicians and nurse managers provided similar responses for perceived disadvantages, but nurse managers were more likely to identify patient triage as a potential advantage of teleophthalmology (172 nurse managers [92.0%] vs 105 physicians [63.3%]). Physicians were more likely than nurse managers to perceive teleophthalmology as a way to attract patients to their emergency department (70 [42.2%] vs 34 [18.2%]). Beyond limited availability, both physicians and nurse managers indicated that their reasons for not already using teleophthalmology included cost concerns and the possibility that it would be less efficient than existing coverage by on-call ophthalmologists.
In multivariable linear regression analysis, identified advantages were significantly predictive of perceived high value and disadvantages were predictive of perceived low value for teleophthalmology (eTable 2 in the Supplement). Increased triage efficiency and providing previously unavailable ophthalmology coverage were most predictive of high perceived value for teleophthalmology among nurse managers (P < .001). Negative prior experiences with other telemedicine programs was most associated with perception of low value for teleophthalmology among nurse managers (P < .01). Among physicians, new or increased ophthalmology coverage was most associated with perceived high value for teleophthalmology (P < .001), and sufficient existing coverage or concerns regarding quality of care with telemedicine were most predictive of perceived low benefit (P < .001). Regression values specifically for rural facilities were not statistically significant.
Our study found there is evidence of a coverage gap for emergency department ophthalmology care, particularly among rural facilities in California. For daytime coverage, less than one-third as many ophthalmologists were available in rural facilities as in nonrural facilities, and on evenings and weekends (after hours) less than half as many ophthalmologists were available. Emergency departments in rural facilities were also much more likely to transfer patients requiring emergency eye care to another facility and tended to be much farther from their most common referral sites (more than 5 times farther than similar transfers from nonrural facilities, taking approximately 30 minutes longer for patients to be seen at the referral site). In addition to higher transfer rates and wait times from rural emergency departments, only 48.6% of rural emergency departments reportedly had ophthalmologists immediately available to examine patients following emergency consultations, compared with 74.7% of nonrural emergency departments. These results imply a need for additional local ophthalmology coverage, possibly with a role for teleophthalmology.
Nearly half of surveyed emergency departments (regardless of rural status) reported using some form of telemedicine, but none reported current use of teleophthalmology for photograph or video transmission to address coverage gaps in emergency eye care. Most nurse managers and physicians believed that teleophthalmology would be a valuable addition to patient care, and both groups rated teleophthalmology as more valuable for triage than for full patient consultations. Nurse managers tended to rate teleophthalmology with slightly more potential benefit than did physicians, and both groups rated teleophthalmology higher at rural facilities than at nonrural facilities.
Key perceived advantages of teleophthalmology included improved triage and timeliness of evaluation. Nurse managers in particular believed that teleophthalmology could improve their emergency department’s efficiency by assisting with triage (92.0%). Barriers to future adoption included potential costs of installation, maintenance, and contracting payments for teleophthalmologists (cited by 165 nurse managers [88.2%] and 131 physicians [78.9%]) and effect on workflow efficiency (cited by 84 nurse managers [44.9%] and 65 physicians [39.2%]). Six nurse managers (3.2%) but no physicians indicated negative prior experiences with other telemedicine programs (not necessarily ophthalmology related). Nurse managers were also more likely to be concerned regarding quality and ability of teleophthalmology to interpret eye images in the setting of ocular trauma (125 nurse managers [66.8%] vs 95 physicians [57.2%]). Anticipated increased triage efficiency and need for new or increased ophthalmology coverage were most predictive of high perceived value for teleophthalmology.
One limitation of our study is that 67 of the 254 emergency departments indexed by the OSHPD were not included in the study owing to lack of available data at the time of assessment. Other limitations included that only 1 nurse manager and 1 physician were surveyed from each emergency department. Although contacting more individuals at each facility was not feasible given study size, we did attempt to speak to nurses and physicians in administrative or leadership roles and asked that they consider the perspective of the staff at large. If the requested nurse manager or physician was unavailable or did not start the survey, the emergency department was contacted up to 3 subsequent times to administer the survey to any emergency department nurse manager or physician, respectively.
We achieved a 100% survey completion rate for nurse managers, but only approximately 88.8% of contacted physicians completed the survey. Although this response rate limits our results, it is above average for survey response rates in general.32 Emergency departments with responses from both nurse managers and physicians had characteristics similar to aggregate characteristics for all facilities recorded in OSHPD data, suggesting a representative sample. However, rural facilities exhibited lower numbers of statistically significant data points, suggesting that the lower number of rural facilities (compared with nonrural facilities), the many variables evaluated, and/or confounding factors may have affected results.
Teleophthalmology theoretically involves provision of medical eye care to patients who are at a distant location from ophthalmologists via advice to local health care professionals or communication directly to patients.33 It has been successfully used in specific settings, such as the Veterans Affairs health care system, for retinal screening25 and has been found to be cost-effective for care delivery: in an Australian study, teleophthalmology was less expensive than traditional options for a rural clinic, at a threshold of 128 patients.34
Although teleophthalmology is not currently widespread or considered the standard of care (except in specific settings such as screening for diabetic retinopathy), it may have value for acute patient triage and consultation, particularly in rural settings. Recognizing the inevitable challenges in integrating teleophthalmology into the clinical workflow,35 further development and investigation into teleophthalmology technology, delivery, and reimbursement systems is warranted, particularly in the current policy landscape with continued interest in reducing access disparities, caring for a growing patient population, and providing high-value, efficient health care.
Submitted for Publication: September 18, 2016; final revision received January 26, 2016; accepted February 3, 2016.
Corresponding Author: Suzann Pershing, MD, MS, Byers Eye Institute at Stanford University, 2452 Watson Ct, Palo Alto, CA 94303 (firstname.lastname@example.org).
Published Online: March 24, 2016. doi:10.1001/jamaophthalmol.2016.0316.
Author Contributions: Dr Pershing and Ms Wedekind had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Wedekind, Pershing.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Wedekind, Pershing.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Wedekind, Pershing.
Study supervision: Sainani, Pershing.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Pershing reported serving as a consultant/advisory board member for Digisight Technologies. No other disclosures were reported.
Funding/Support: The Human Biology Research Experience Grant Program of the Stanford University Department of Human Biology supported Ms Wedekind’s residence at Stanford University during the data collection phase.
Role of the Funder/Sponsor: The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Previous Presentations: Presented at American Association for the Advancement of Science Annual Meeting; February 14, 2015; San Jose, California; youth + tech + health Innovation Live; April 28, 2015; San Francisco, California; Stanford Women in STEM Symposium; February 21, 2015; Stanford, California; US Agency for International Development Higher Education Solutions Network TechCon; November 10, 2014; Berkeley, California; UCSF Health Disparities Research Symposium; October 17, 2014; San Francisco, California; and Stanford Research and Public Service Symposium; October 23, 2014; Stanford, California.