Mismatch_in_Supply_and_Demand_for_Neuro-Ophthalmic_Care

Original Contribution Section Editors: Clare Fraser, MD Susan Mollan, MD Mismatch in Supply and Demand for Neuro-Ophthalmic Care Adam DeBusk, DO, Prem S. Subramanian, MD, PhD, Molly Scannell Bryan, PhD, Mark L. Moster, MD, Preston C. Calvert, MD, FAAN, Larry P. Frohman, MD Background: Previous research suggests the number of neuro-ophthalmologists in the United States may be below a level that provides sufficient access to neuro-ophthalmic care in much of the United States. However, national estimates of the amount of clinical time spent on neuroophthalmology are lacking. Methods: The North American Neuro-Ophthalmology Society administered a survey on professional time allocation to its active members. Survey response was 95%. The survey characterized the hours each week each respondent allocated to overall work, clinical work, clinical work in ophthalmology/neurology, and clinical work in neuroophthalmology specifically. The survey additionally collected information regarding demographics, current wait times to be seen for new patients, and the difference in clinical time spent in neuro-ophthalmology spent between the current day compared with that shortly after completing clinical training. Linear regression was used to identify potential relationships between the above and average wait time. Results: On average, responding physicians spent 70% of their clinical time on neuro-ophthalmology. In 6 states, there were no reported practicing neuro-ophthalmologists, and in only 8 states was the clinical full-time equivalent to population ratio below the suggested threshold of 1 for every 1.2 million. The median wait time for a new patient was 6 weeks. This wait time was associated with the fraction of clinical time spent in neuro-ophthalmology (0.2 weeks longer wait for a 10 percentage point increase in the fraction of time spent in neuro-ophthalmology; P = 0.02), and suggestively associDepartment of Ophthalmology (AD), Wills Eye Hospital, Sidney Kimmel Medical College, Thomas Jefferson University, Philadelphia, Pennsylvania; Departments of Ophthalmology, Neurology, and Neurosurgery (PSS), Sue Anschutz-Rodgers UCHealth Eye Center, University of Colorado School of Medicine, Aurora, Colorado; Institute for Minority Health Research (MSB), University of Illinois College of Medicine, Chicago, Illinois; Departments of Ophthalmology and Neurology (MLM), Wills Eye Hospital, Sidney Kimmel Medical College, Thomas Jefferson University, Philadelphia, Pennsylvania; Department of Neurology (PCC), Johns Hopkins School of Medicine, Baltimore, Maryland; and Departments of Ophthalmology and Neurosciences (LPF), Rutgers-New Jersey Medical School, Newark, New Jersey. The authors report no conflicts of interest. Address correspondence to Adam DeBusk, DO, Wills Eye Hospital, Suite 930, Philadelphia, PA 19106; E-mail: [e-mail] 62 ated with training (training in ophthalmology was associated with 1.0 week shorter wait time; P = 0.06). Conclusion: The survey suggests that neuro-ophthalmologists are unable to see patients in a timely manner and a decreasing number of clinicians are entering the field. Future interventions should be considered to incentivize neuroophthalmology training in ophthalmology and neurology residents such that the United States population is able to appropriately access neuro-ophthalmic care. Journal of Neuro-Ophthalmology 2022;42:62–67 doi: 10.1097/WNO.0000000000001214 © 2021 by North American Neuro-Ophthalmology Society N euro-ophthalmologists treat patients with complex neurological conditions that affect the visual system. The number of neuro-ophthalmologists has varied over time, and previous estimates have suggested that in the United States, one neuro-ophthalmologist (1 clinical full-time equivalent [CFTE] or FTE) would be required to provide sufficient coverage for 1.2 million individuals (1). However, recent commentaries (2,3) have argued that economic pressures may be creating a scenario where neuro-ophthalmology as a specialty is insufficiently appealing to incentivize medical students and residents to pursue fellowships in the field. Furthermore, many clinicians who are trained in neuro-ophthalmology may also react to the same economic pressures and spend clinical time practicing in other subspecialties, such that wait times to be seen for a neuro-ophthalmologic condition may extend beyond clinically desirable intervals. Demand for neuro-ophthalmic care is also increasing because of the “corporatization” of medical practice, with many ophthalmologists and neurologists facing pressure to increase their throughput of patients. Therefore, they refer to neuro-ophthalmologists’ patients they may have worked up and treated themselves in the past, because of the time such an encounter requires. Correspondingly, wait times to see a neuro-ophthalmologist may have increased. A recent single-center study reported wait times of 13 weeks for routine referrals (4). DeBusk et al: J Neuro-Ophthalmol 2022; 42: 62-67 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution However, concrete data are sparse regarding the density, distribution, and practice patterns of neuroophthalmologists throughout the United States. This manuscript describes the results of a survey, which was designed to describe the training and experience of neuroophthalmologists, and also portray the clinical capacity and practice patterns of these clinicians. METHODS Survey Methodology In April 2019, a survey was distributed, via email, to all clinically active members of the North American NeuroOphthalmology Society (NANOS) who currently practice in the United States. The survey queried state of practice, specialties of clinical training and fellowships, time since the completion of training, and time allocation (measured in the number of half-days worked per week) of (1) current overall work, (2) current work spent in clinical time of all clinical subspecialties, (3) current time spent in any ophthalmology/neurology subspecialty, and (4) the time spent in neuro-ophthalmology and neuro-ophthalmic surgery. The survey additionally collected information regarding current wait times to be seen for new patients. Members were prompted to complete the survey by periodic email reminders, and by August of 2019, 386/406 United States-based NANOS member clinicians (95%) had completed the survey. Statistical Approach The data were analyzed to characterize the respondents as above, and further analysis was performed to estimate the reported CFTE in neuro-ophthalmology per population. The non–neuro-ophthalmologic clinical activities that were reported by the survey respondents also were characterized, along with time spent, focusing on time spent in clinical practice, and in neuro-ophthalmology. The reported difference in the fraction of clinical time currently spent in neuroophthalmology compared with that 2–3 years postfellowship was also calculated. Linear regression was used to identify potential relationships between a physician’s reported waiting time for a new patient to obtain an appointment and: the total or fractional time spent in neuro-ophthalmology, the presence of fellowships in non–neuro-ophthalmology subspecialties, or the amount of time since fellowship. RESULTS Summary Of the survey respondents, 236 (61.3%) trained as ophthalmologists, 116 (30.1%) trained as neurologists, and 33 (8.6%) trained as both. The amount of time since the completion of their initial fellowship is graphed in Figure 1, DeBusk et al: J Neuro-Ophthalmol 2022; 42: 62-67 FIG. 1. Self-reported number of years since the end of the for all survey respondents. First 2 categories are .2 years and 2–5 years, and all other categories are in five-year increments. with almost 35% of respondents reporting 10 or fewer years since the end of their initial fellowship training. Although the modal respondent (n = 236; 61.3%) did not pursue fellowships in a specialty other than neuro-ophthalmology (Table 1), orbit or oculoplastic (12.5% of respondents), and pediatric ophthalmology (9.1% of respondents) were common additional fellowships. Time Allocation The modal respondent reported working 10 half-days (i.e.,: 1 FTE) in a week, with a mean 7.3 of those half days devoted to clinical work (Fig. 2). Table 2 details how the survey respondents distributed their clinical time. On average, respondents spent 70% of their clinical time on neuro-ophthalmology, corresponding to 56% of their total work of the week. Other specialties that commanded more than 5% of their clinical time included general neurology (0.4 half-days/week or 5.9% of clinical time), orbit and oculoplastic (0.5 half-days/ week; 6.3% of clinical time), and cataract/comprehensive ophthalmology (0.7 half-days/week; 9.2% of clinical time). Although survey responses varied substantially (Fig. 3), on average, the respondents do not currently report spending less time on neuro-ophthalmology than 2 or 3 years after their practice began. Geographic Distribution of Survey Respondents The survey respondents reported practicing neuroophthalmology in all but 6 states: New Mexico, Maine, Montana, Delaware, South Dakota, and Wyoming (Table 3). Across the country, there are 1.7 million residents per reported CFTE in neuro-ophthalmology. Only 8 states (Vermont, Maryland, Massachusetts, New York, New Hampshire, Minnesota, Michigan, and Connecticut) and Washington, D.C. report a CFTE to state population ratio below 1:1.2 million persons. 63 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 1. The reported fellowship-level secondary training of survey respondents Other Fellowship None Orbit and/or oculoplastic Pediatric ophthalmology Other neurology fellowship Other ophthalmology fellowship Retina Glaucoma Neuro-immunology Multiple sclerosis Ocular pathology Musculoskeletal diseases Uveitis Cornea Count (%) 236 48 35 22 12 11 9 7 6 4 2 2 1 (61.3%) (12.5%) (9.1%) (5.7%) (3.1%) (2.9%) (2.3%) (1.8%) (1.6%) (1.0%) (0.5%) (0.5%) (0.3%) The respondent’s parent specialty also did not significantly affect the wait times. Training in ophthalmology was associated with 1.0 week shorter wait time (P = 0.06); training in neurology was associated with a 0.53week longer wait time (P = 0.35); whereas training in both was associated with a 1.6-week longer wait time (P = 0.08). Respondents who reported fellowships in specialties besides neuro-ophthalmology did not appreciably differ in their wait times from respondents who reported only neuroophthalmology fellowships (b = 0.6 weeks shorter wait for those with other fellowships; P = 0.23). Wait times also did not seem to be affected by the amount of time since the respondent finished their initial fellowship (F-statistic P-value: 0.10). Wait Times CONCLUSION The median wait time for a new patient to obtain an appointment was reported to be 6 weeks (Fig. 4), although 89 respondents (23%) reported wait times of 12 weeks or longer. The wait time for a new patient to be seen was not strongly associated with the total weekly time the respondent reported working in clinical neuro-ophthalmology, described by total half-days (b = 0.15 weeks longer for each additional half-day worked in neuro-ophthalmology; P = 0.14). Nonetheless, the fraction of all clinical time spent in neuro-ophthalmology was positively associated with longer wait times (b = 0.2 weeks longer wait for a 10% point increase in the fraction of time spent in neuroophthalmology; P = 0.02). A key finding of the survey is that the median wait time to see a neuro-ophthalmologist is 6 weeks, and for almost a quarter of the USA, the wait time is 12 weeks or longer. Neuro-ophthalmologists see patients with acute disease that may be vision-threatening or even life-threatening. These conditions include giant cell arteritis, optic neuritis, CN III palsy from aneurysm, and papilledema. The inability of a neuro-ophthalmologist to see these patients quickly impairs the quality of their care. They are often seen in an emergency room or by health care personnel with limited neuro-ophthalmologic expertise. This lack of expertise may result in inappropriate diagnosis (5), testing, and management. In fact, it may lead to undirected, unnecessary, and FIG. 2. Professional time use reported by survey respondents. The total number of half-days worked (top), and the total number of half-days in which the work was clinical (bottom). 64 DeBusk et al: J Neuro-Ophthalmol 2022; 42: 62-67 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 2. Reported time spent in clinical practice for various activities, as a percentage of total work time and as a percentage of total clinical work time Time Specialization Summary Statistic Half days/week spent on neuro-ophthalmology As a percentage of total work time As a percentage of total clinical time Half days/week spent on neuro-ophthalmic surgery As a percentage of total work time As a percentage of total clinical time Half days/week spent on neuro-immunology As a percentage of total work time As a percentage of total clinical time Half days/week spent on multiple sclerosis As a percentage of total work time As a percentage of total clinical time Half days/week spent on musculoskeletal diseases As a percentage of total work time As a percentage of total clinical time Half days/week spent on other neurologic subspecialty As a percentage of total work time As a percentage of total clinical time Half days/week spent on general neurology As a percentage of total work time As a percentage of total clinical time Half days/week spent on glaucoma As a percentage of total work time As a percentage of total clinical time Half days/week spent on orbit and/or oculoplastic As a percentage of total work time As a percentage of total clinical time Half days/week spent on retina As a percentage of total work time As a percentage of total clinical time Half days/week spent on uveitis As a percentage of total work time As a percentage of total clinical time Half days/week spent on pediatric ophthalmology As a percentage of total work time As a percentage of total clinical time Half days/week spent on cornea As a percentage of total work time As a percentage of total clinical time Half days/week spent on ocular pathology As a percentage of total work time As a percentage of total clinical time Half days/week spent on cataract/comprehensive ophthalmology As a percentage of total work time As a percentage of total clinical time sometimes invasive workups that result in excessive health care costs and delayed diagnosis and management. A key part of any strategy to improve access involves enticing more residents to train in neuro-ophthalmology. In our experience ophthalmology residents are disincentivized to pursue neuro-ophthalmology compared with other fields of ophthalmology, because it is perceived to be less financially rewarding than other areas. Many residents choose ophthalmology because it is a surgical specialty DeBusk et al: J Neuro-Ophthalmol 2022; 42: 62-67 Mean (SD): 4.9 (2.6); median: Mean (SD): 56.3 (29.6) Mean (SD): 70.2 (31.7) Mean (SD): 0.3 (0.8); median: Mean (SD): 3.1 (10.5) Mean (SD): 3.6 (11.0) Mean (SD): 0.2 (1.0); median: Mean (SD): 2.9 (12.6) Mean (SD): 3.2 (14.2) Mean (SD): 0.3 (0.8); median: Mean (SD): 3.6 (12.5) Mean (SD): 4.3 (15.0) Mean (SD): 0.1 (0.5); median: Mean (SD): 1.5 (8.7) Mean (SD): 1.7 (10.4) Mean (SD): 0.3 (1.0); median: Mean (SD): 3.7 (13.5) Mean (SD): 4.5 (16.3) Mean (SD): 0.4 (1.2); median: Mean (SD): 5.4 (16.1) Mean (SD): 5.9 (16.3) Mean (SD): 0.2 (0.9); median: Mean (SD): 2.7 (11.0) Mean (SD): 2.9 (11.6) Mean (SD): 0.5 (1.5); median: Mean (SD): 5.7 (16.8) Mean (SD): 6.3 (18.5) Mean (SD): 0.1 (0.8); median: Mean (SD): 1.7 (9.1) Mean (SD): 1.7 (9.1) Mean (SD): 0.0 (0.1); median: Mean (SD): 0.2 (1.5) Mean (SD): 0.3 (2.2) Mean (SD): 0.3 (1.2); median: Mean (SD): 3.9 (13.7) Mean (SD): 4.9 (16.4) Mean (SD): 0.0 (0.3); median: Mean (SD): 0.7 (5.8) Mean (SD): 0.7 (6.1) Mean (SD): 0.0 (0.2); median: Mean (SD): 0.2 (2.2) Mean (SD): 0.2 (2.2) Mean (SD): 0.7 (1.7); median: Mean (SD): 8.4 (19.3) Mean (SD): 9.2 (21.1) 5.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 and may consider neuro-ophthalmology as lacking a surgical footing and thus unappealing. Furthermore, prior surveys by NANOS (not published) identified the perceived difficulty of the specialty as a reason that residents did not choose to train in neuro-ophthalmology. Other reasons reported in a recent survey of 96 US PGY-4 ophthalmology residents graduating in 2018 or 2019 (conducted online74% of respondents were pursuing a fellowship but none in neuro-ophthalmology) include the perception that patients 65 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 3. State-by-state comparison of the number of NO CFTE’s with that state’s population recommended ratio is 1 CFTE for every 1.2 million population FIG. 3. Comparison of the number of half-days spent working in N-O at the time of the survey with the number of half-days the respondent recalls working in N-O 2–3 years after their fellowship. The mean change was not statistically different from zero, and the median change was 0 half days. in a neuro-ophthalmology practice may have complex needs, leading to longer clinical encounters and lower patient throughput. They also expressed concern that neuro-ophthalmic patients have poorer long-term outcomes. Some residents were dissuaded by the fact that many neuro-ophthalmology jobs exist in academia rather than private practice settings (6). Neurology residents often do not get adequate exposure to neuro-ophthalmology during their training because it is not a required area of instruction/clinical rotation, and thus few are familiar with its actual practice and do not enter the field. Increasingly limited exposure to ophthalmology in medical school also may make neurologists uncomfortable doing detailed ophthalmic examinations. For both ophthalmologists and neurologists, exams are complex and time-consuming, and fewer patients can be seen in a day. Thus, trainees rightly discern that specialties with briefer encounters/more volume, and/or more procedures are more financially rewarding and are dissuaded from choosing one of the more cognitive specialties of medicine. In this respect, neuroophthalmology suffers from the same challenges under the current reimbursement models as other nonprocedural specialties of medicine (i.e., Rheumatology, Endocrinology, Infectious Disease), which is leading to a shortage of practicing physicians in all of these crucial fields (7). Reimbursement models that better reflect the efforts of complex examinations, writing consults, phone consultations, and reviewing ancillary testing are critical for the continued recruitment of trainees into the cognitive specialties. One option could be a higher fixed salary rather than a lower base salary with a productivity-based incentive. The disproportionately low reimbursement for time and work required for a neuro-ophthalmologic examination makes a productivity-based incentive challenging for most of the physicians in the field. The relative value unit is a 66 Principal State of Practice State Population State Population Per CFTE Vermont Washington, DC Maryland Massachusetts New York New Hampshire Minnesota Michigan Connecticut Nebraska Oregon North Dakota Illinois Oklahoma Arkansas Washington Pennsylvania Wisconsin Kansas Florida Utah Nevada California North Carolina West Virginia Iowa New Jersey Idaho South Carolina Missouri Alaska Virginia Georgia Ohio Colorado Texas Alabama Arizona Louisiana Tennessee Kentucky Mississippi Indiana Puerto Rico Rhode Island Hawaii New Mexico* Maine* Montana* Delaware* South Dakota Wyoming* 626,299 702,455 6,042,718 6,902,149 19,542,209 1,356,458 5,611,179 9,995,915 3,572,665 1,929,268 4,190,713 760,077 12,741,080 3,943,079 3,013,825 7,535,591 12,807,060 5,813,568 2,911,505 21,299,325 3,161,105 3,034,392 39,557,045 10,383,620 1,805,832 3,156,145 8,908,520 1,754,208 5,084,127 6,126,452 737,438 8,517,685 10,519,475 11,689,442 5,695,564 28,701,845 4,887,871 7,171,646 4,659,978 6,770,010 4,468,402 2,986,530 6,691,878 3,195,153 1,057,315 1,420,491 2,095,428 1,338,404 1,062,305 967,171 882,235 577,737 447,356 540,350 728,038 841,725 876,332 968,898 1,039,107 1,148,955 1,152,472 1,205,792 1,232,562 1,266,795 1,313,513 1,359,682 1,369,920 1,477,566 1,600,882 1,709,872 1,712,650 1,805,027 1,859,473 1,896,495 1,901,781 1,996,850 2,006,480 2,104,096 2,172,809 2,192,760 2,421,012 2,450,580 2,458,126 2,581,116 2,629,868 2,656,691 2,712,173 2,759,792 3,258,580 3,259,839 3,328,555 3,563,163 3,723,668 3,733,162 4,779,912 6,390,306 10,573,150 14,204,910 *New Mexico, Maine, Montana, Delaware, South Dakota, and Wyoming each contained no reported NANOS members. CFTE, clinical full-time equivalent; NANOS, North American Neuro-Ophthalmology Society. DeBusk et al: J Neuro-Ophthalmol 2022; 42: 62-67 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 4. Reported amount of time before a new patient could be seen in the practice. The mean wait was almost 7 weeks (SD: 5.1 weeks), and the median wait was 6 weeks. One respondent reported a wait time of more than 6 months. poor measure of productivity for the neuro-ophthalmologist because it also undervalues complex examinations and nonprocedural work. Another possibility could be additional funding for clinicians specializing in underrepresented fields of medicine (8). Better reimbursement models for neuro-ophthalmology, and other cognitive specialties, could incentivize physicians to dedicate more of their existing clinical time to the field. This could increase CFTE and possibly reduce wait times and increase access to care. It could also help recruit new people into the specialty because there is a shortage of new physicians to replace the ones exiting the field and retiring. Recruitment at schools of osteopathic and allopathic medicine may attract additional candidates. At the current CFTE per neuro-ophthalmologist (average around 0.5), there would need to be approximately 180 additional physicians added to the field to achieve a goal CFTE of 1:1.2 million individuals. Of course, if reimbursement models changed to incentivize practitioners already trained in neuro-ophthalmology to dedicate more clinical time to this specialty, the number of additional needed physicians would be less. Limitations These responses may be prone to selection bias given that those who were eligible for the survey were NANOS members. Practitioners who are invested enough in the discipline that they continue to be active NANOS members may be more motivated to continue to devote substantial time to neuro-ophthalmology. It is possible that practitioners who reduced the percentage of time they spent on neuro-ophthalmology over their careers may also decide to DeBusk et al: J Neuro-Ophthalmol 2022; 42: 62-67 no longer retain membership in NANOS, and therefore this survey may under-count reductions in time spent in neuroophthalmology over a career. In conclusion, the current supply of neuroophthalmologists is inadequate to meet the needs of the US population, as demonstrated by the long wait times for new patient care and poor geographic distribution of specialists in most of the United States. At least 1/3 of respondents reported being 25+ years beyond fellowship training, suggesting that access will worsen if a robust training pipeline is not created immediately. Neuro-ophthalmology is a cognitively challenging and rewarding field, but if physicians perceive they are not recognized appropriately for their efforts, they will choose other lines of work that are both intellectually and financially rewarding. STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: A. DeBusk, P. S. Subramanian, M. Scannell Bryan, M. L. Moster, P. C. Calvert, and L. P. Frohman; b. Acquisition of data: A. DeBusk, P. S. Subramanian, M. Scannell Bryan, M. L. Moster, P. C. Calvert, and L. P. Frohman; c. Analysis and interpretation of data: A. DeBusk, P. S. Subramanian, M. Scannell Bryan, M. L. Moster, P. C. Calvert, and L. P. Frohman. Category 2: a. Drafting the manuscript: A. DeBusk, P. S. Subramanian, M. Scannell Bryan, M. L. Moster, P. C. Calvert, and L. P. Frohman; b. Revising it for intellectual content: A. DeBusk, P. S. Subramanian, M. Scannell Bryan, M. L. Moster, P. C. Calvert, and L. P. Frohman. Category 3: a. Final approval of the completed manuscript: A. DeBusk, P. S. Subramanian, M. Scannell Bryan, M. L. Moster, P. C. Calvert, and L. P. Frohman. REFERENCES 1. Frohman LP. The human resource crisis in neuro-ophthalmology. J Neuroophthalmol. 2008;28:231. 2. Frohman L, Digre K. Elimination of consult codes in neuroophthalmology: another blow to our subspecialty? J Neuroophthalmol. 2010;30:210. 3. Frohman LP. How can we assure that neuro-ophthalmology will survive? Ophthalmology 2005;112:741–743. 4. Stunkel L, Mackay D, Bruce B, Newman N, Biousse V. Referral patterns in neuro-ophthalmology. J Neuroophthalmol. 2020;40:485–493. 5. Stunkel L, Newman NJ, Biousse V. Diagnostic error and neuroophthalmology. Curr Opin Neurol. 2019;32:62–67. 6. Francis C, Patel V. Recruiting ophthalmologists into neuroophthalmology. Presented at the Annual Meeting of the North American Neuro-Ophthalmology Society, March 2019, Las Vegas, NV. 7. Battafarano DF, Ditmyer M, Bolster MB, Fitzgerald JD, Deal C, Bass AR, Molina R, Erickson AR, Hausmann JS, Klein-Gitelman M, Imundo LF, Smith BJ, Jones K, Greene K, Monrad SU. 2015 American Colllege of Rheumatology Workforce Study: supply and demand projections of adult Rheumatology workforce, 2015– 2030. Arthritis Care Res. 2018;70:617–626. 8. Frohman LP. Neuro-ophthalmology: transitioning from old to new models of health care delivery. J Neuroophthalmol. 2017;37:206–209. 67 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.