Areas of Kentucky where the estimated travel time to the nearest ophthalmologist office is within 30, 45, 60, or 90 minutes. Gold dots indicate locations with more than 3000 residents. Circles reflect numbers of optometrist-performed procedures at that location from 2008 through 2014.
Areas of Oklahoma where the estimated travel time to the nearest ophthalmologist office is within 30, 45, 60, or 90 minutes. Gold dots indicate locations with more than 3000 residents. Circles reflect numbers of optometrist-performed procedures at that location from 2008 through 2014.
Areas of New Mexico where the estimated travel time to the nearest ophthalmologist office is within 30, 45, 60, or 90 minutes. Gold dots indicate locations with more than 3000 residents. Circles reflect numbers of optometrist-performed procedures at that location from 2008 through 2014.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Stein JD, Kapoor KG, Tootoo JL, et al. Access to Ophthalmologists in States Where Optometrists Have Expanded Scope of Practice. JAMA Ophthalmol. 2018;136(1):39–45. doi:10.1001/jamaophthalmol.2017.5081
What is the estimated travel time to the nearest ophthalmologist office for persons living in states that have expanded the scope of practice of optometrists?
This population-based cohort study of 10 149 897 residents of Oklahoma, Kentucky, and New Mexico found that more than 75% live within 30 minutes’ estimated travel time of the nearest ophthalmologist office, and 94% to 99% of residents live within 60 minutes of the nearest ophthalmologist office. Roughly half of Medicare beneficiaries in these states who received surgery from optometrists between 2008 and 2014 lived within 30 minutes of the nearest ophthalmologist.
These results may help inform policy makers when weighing the pros and cons of expanding optometrist scope of practice.
As the United States considers how to best structure its health care services, specialty care availability is receiving increased focus. This study assesses whether patients lack reasonable access to ophthalmologists in states where optometrists have been granted expanded scope of practice.
To determine the estimated travel time (ETT) to the nearest ophthalmologist office for persons residing in states that have expanded scope of practice for optometrists, and to quantify ETT to the nearest ophthalmologist for Medicare beneficiaries who received surgical care from optometrists in those states between 2008 and 2014.
Design, Setting, and Participants
This study used data from the 2010 US census, a 2016 American Academy of Ophthalmology member database, and a data set of claims data for a random sample of 20% of beneficiaries enrolled in Medicare nationwide from 2008 to 2014 (n=14 063 725). Combining these sources with geographic information systems analysis, the ETT to the nearest ophthalmologist office was calculated for every resident of Kentucky, Oklahoma, and New Mexico. This study also assessed ETT to the nearest ophthalmologist for Medicare beneficiaries in those states who had received surgery from an optometrist from 2008 to 2014. Data analyses were conducted from July 2016 to July 2017.
Main Outcomes and Measures
The proportion of residents of Kentucky, Oklahoma, and New Mexico who live within an ETT of 10, 30, 45, 60, or 90 minutes of the nearest ophthalmologist office.
The study included 4 339 367 Kentucky residents, 3 751 351 Oklahoma residents, and 2 059 179 New Mexico residents. Of these, 5 140 547 (50.6%) were female. Racial/ethnic composition included 7 154 847 people (70.5%) who were white, 640 608 (6.3%) who were black, and 1 418 246 (14.0%) who were Hispanic. The mean (SD) age was 37.8 (22.8) years. More than 75% of residents in the 3 states lived within an ETT of 30 minutes to the nearest ophthalmology office, and 94% to 99% of residents lived within an ETT of 60 minutes to the nearest ophthalmology office. Among Medicare beneficiaries who received surgery by optometrists, 58.3%, 51.1%, and 46.9% in Kentucky, Oklahoma, and New Mexico, respectively, lived within an ETT of 30 minutes from the nearest ophthalmologist office.
Conclusions and Relevance
In the states where optometrists have expanded scope of practice, most residents lived within an ETT of 30 minutes of the nearest ophthalmologist office, as do half of Medicare beneficiaries who received surgical care from optometrists. These results can help inform policy makers when weighing the pros and cons of scope of practice expansion for optometrists.
In the past decade, there have been attempts to convince health policy makers to expand the scope of practice of optometrists.1 Members of the optometric community have sought privileges to perform laser surgeries, minor eyelid procedures, and intraocular injections.2,3 A key argument often cited is that many patients reside in communities with few ophthalmologists available to perform these interventions, and that access to care will be improved by permitting optometrists to perform these procedures.2
There are more practicing optometrists in the United States than ophthalmologists, so one would expect that patients have easier access to an optometrist than to an ophthalmologist.1 In fact, Lee et al4 demonstrated that, nationally, the median estimated travel time (ETT) to the nearest optometrist office is slightly less than the median ETT to the nearest ophthalmologist office, and most of the US population lives within an ETT of 14 minutes to the nearest optometrist office and 26 minutes to the nearest ophthalmologist office.4 To our knowledge, however, it is unknown how accessible ophthalmologists are in the states where optometrists have expanded surgical scope of practice, and whether the optometrists who perform surgical procedures are practicing in communities with limited access to ophthalmologists in those states.
To fill in these gaps, this study analyzes the ETT to the nearest ophthalmologist office for residents of Kentucky,2,3 where optometrists have had expanded scope of practice since 2011; Oklahoma,5 where optometrist scope of practice expansion occurred in 1998; and New Mexico,5 where expansion has been in place since 2006. In addition, the study explores the ETT to the nearest ophthalmologist for Medicare beneficiaries residing in these states who underwent surgical procedures by optometrists.
This study used data from the 2010 US census,6 a 2016 American Academy of Ophthalmology (AAO) member database,7 and a data set of claims data8 for a random 20% sample of beneficiaries enrolled in Medicare from 2008 to 2014. These data sets were combined with geographic information systems analysis to determine ETTs.
This study was approved by the University of Michigan Institutional Review Board. The study was exempt from requirements for informed consent procedures because it used deidentified data exclusively.
To estimate the populations of residents living within specific drive times to the nearest ophthalmologist, we accessed 2010 US census block-level data6 for Kentucky, Oklahoma, and New Mexico. census blocks are polygonal features that serve as the base unit for all nested US census area units and are the smallest unit of measure used to report complete data about basic demographic characteristics (such as total population by age, sex, and race/ethnicity) per geographic location.9,10 This source contains information on the number of persons living in a given census block and the demographic profiles of those residents.
Analyses of ETT were based on centroids, the geographic centers of census blocks. Zip code tabulation areas (ZCTAs), which aggregate multiple census blocks, were also used. These are simplified areal representations of US Postal Service zip code service areas, constructed such that their boundaries precisely align with those of census blocks. Both centroids and ZCTAs were used to calculate ETTs to ophthalmologist offices.
The American Academy of Ophthalmology (AAO) is the largest professional ophthalmology specialty society in the world. It has more than 32 000 members, including more than 17 000 who practice in the United States.11 An estimated 93% of ophthalmologists in the United States are AAO members (J. Aguirre, CAE, written communication, November 2016). The AAO conducts a biennial member survey that seeks the primary office address and additional practice addresses of all member physicians. In addition, members are contacted quarterly and encouraged to update their member profiles on the AAO website, which likewise allows a member to list a primary office address and addresses for other practice locations. The organization provided a snapshot of this dynamic database in August 2016.
Data on Medicare enrollees were obtained from the US Centers for Medicare and Medicaid Services.8 This national database included a random sample of 20% of Medicare beneficiaries with Parts A, B, and D coverage from January 1, 2008, through December 31, 2014 (n = 14 063 725). Because the database had incomplete data on persons with noncontinuous enrollment and Medicare Advantage plan enrollees, these persons were excluded. From this data source, we identified all beneficiaries residing in Kentucky, Oklahoma, and New Mexico.
The database uses International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes12 to identify all ocular and nonocular diagnoses and Current Procedural Terminology codes13 to identify all diagnostic and therapeutic procedures performed on each beneficiary during the period included in the data set. In addition, the database included information on demographic information on each Medicare beneficiary (eg, sex, age, and race/ethnicity) and the type of health care professional who provided care at each clinical encounter. This database has been used previously to study patients with ocular diseases.14-16
All data analyses were conducted from July 2016 to July 2017. The AAO database was used to identify the primary and all secondary office addresses for every AAO member licensed to practice in Kentucky, Oklahoma, and New Mexico. Since some patients travel across state lines to neighboring states to receive eye care, the office addresses of all AAO members practicing in states adjacent to the states of interest were identified (for Kentucky: Indiana, Ohio, Illinois, West Virginia, Tennessee, and Virginia; for Oklahoma: Texas, Arkansas, Missouri, Kansas, New Mexico, and Colorado; for New Mexico: Texas, Oklahoma, Colorado, Arizona, and Utah; the nation of Mexico, although adjacent to New Mexico, was not included). All office addresses were geocoded to the street level using the ArcMap feature of ArcGIS version 10.3x (Environmental Systems Research Institute). The ETT from each 2010 US census block centroid within Kentucky, Oklahoma, and New Mexico to the nearest ophthalmologist office was calculated via ArcGIS Network Analyst Extension. The Streetmap Premium 2014 feature of ArcGIS was used as a reference database for geocoding and network calculation. Calculated ETTs were network-based, which means the study assumed travel by car on roads and adherence to specified speed limits. If the nearest ophthalmologist office was in a neighboring state, the model assumed that persons who resided in the states of interest would cross state lines.
The key outcome was the proportion of total persons residing in each state who lived within ETTs of 10, 30, 45, 60, and 90 minutes by car from the census block centroid to the nearest ophthalmologist office. In addition, we determined ETTs of 10, 30, 45, 60, and 90 minutes, stratified by age, race/ethnicity, and urban vs rural residential status. Designation of urbanicity was determined through the US Department of Agriculture’s Rural-Urban Commuting Area Codes.17 Vector maps created in ArcMap depict service areas in each state where the ETT to the nearest ophthalmologist office is within 10, 30, 45, 60, and 90 minutes.
Next, the model used the Medicare claims database to identify the office locations of all optometrists in the 3 states who had performed any of these procedures at least 5 times in the period between January 1, 2008, and December 31, 2014: laser trabeculoplasty, laser capsulotomy, laser peripheral iridotomy, intraocular injection, and eyelid lesion removal. The model determined what proportion of optometrist offices where these procedures were performed at that frequency were located in census blocks with population-weighted centroids within ETTs of 10, 30, 45, 60, and 90 minutes to the nearest ophthalmologist office. Finally, the number of Medicare beneficiaries in these 3 states who received 1 or more of the above-mentioned surgeries by an optometrist from 2008 through 2014 was determined, as was the proportion of these patients who were living in census blocks with a population-weighted centroid within ETTs of 10, 30, 45, 60, and 90 minute to the nearest ophthalmologist office.
Based on the 2010 US census, 4 339 367 persons lived in Kentucky in 2010.6 In the AAO database, the number of office addresses of ophthalmologists practicing in Kentucky was 217, with another 887 offices of ophthalmologists within 90 minutes’ drive time of the Kentucky border in surrounding states. The population in Kentucky within ETTs of 10, 30, 60, and 90 minutes to the nearest ophthalmologist office numbered 1 834 742 (42.3%), 3 385 205 (78.0%), 4 284 905 (98.7%), and 4 337 550 people (99.96%), respectively (Figure 1). Table 1 shows the ETT from each 2010 US census block centroid to the nearest ophthalmologist office for residents of all states of interest, stratified by age and race/ethnicity. Among the 3 823 864 residents living in urban communities, 1 828 346 (47.8%) lived within an ETT of 10 minutes to the nearest ophthalmologist office; for the 515 503 residents who lived in rural parts of the state (12% of the state population), 482 536 (93.6%) lived within an ETT of 60 minutes to the nearest ophthalmologist office (Table 2).
Based on the 2010 US census, 3 751 351 persons lived in Oklahoma.6 The number of office addresses of ophthalmologists practicing in Oklahoma was 177, with another 523 ophthalmologist offices were within 90 minutes’ drive time of the Oklahoma border in surrounding states. In Oklahoma, 1 907 332 (50.8%) of residents lived within 10 minutes (ETT) to the nearest ophthalmologist office; the corresponding numbers were 2 983 285 (79.5%), 3 590 860 (95.7%), and 3 687 215 (98.3%) for residents within ETTs of 30, 60, and 90 minutes, respectively (Figure 2). Among the 3 431 538 residents living in urban areas of Oklahoma, 1 902 789 (55.5%) lived within an ETT of 10 minutes of the nearest ophthalmologist office; for the 319 813 residents living in rural parts of the state (8.5% of the state population), 264 560 (82.7%) lived within an ETT of 60 minutes to the nearest ophthalmologist office (Table 2).
Based on the 2010 US census, New Mexico had 2 059 179 residents in 2010.6 The total number of AAO-documented office addresses of ophthalmologists practicing in the state was 107; another 122 ophthalmologist offices in the surrounding states were within 90 minutes of the border. The New Mexicans living within ETTs of 10, 30, 60, and 90 minutes by car to the nearest ophthalmologist office were 1 118 620 (54.3%), 1 696 341 (82.4%), 1 946 258 (94.5%), and 2 024 793 (98.3%), respectively (Figure 3). Among the 1 928 316 New Mexico residents living in urban areas, 1 117 641 (58.0%) lived within 10 minutes’ ETT to the nearest ophthalmologist office; among the 130 863 rural residents of the state (6.4% of the population), 87 010 (66.5%) lived within an ETT of 60 minutes to an ophthalmologist office (Table 2).
This study found that 34 optometrists had performed 5 or more surgeries each in the defined time period in Kentucky. Of their 368 procedures, 59 (16.0%) were performed at offices in ZCTAs with population-weighted centroids within 10 minutes’ ETT to the nearest ophthalmologist office; 215 (58.4%) were within an ETT of 30 minutes of an ophthalmologist office, and 366 (99.5%) were within an ETT of 60 minutes. In Oklahoma, 64 optometrists performed at least 5 such surgeries. Of the 887 total surgeries performed, 148 (16.7%) were done at offices in ZCTAs with population-weighed centroids within an ETT of 10 minutes to an ophthalmologist office; 444 (50.1%) were within 30 minutes, and 779 (87.8%) were within 60 minutes. The 5 optometrists who performed at least 5 surgeries each in New Mexico performed 23 procedures in total; none occurred at offices in ZCTAs with population-weighted centroids within an ETT of 10 minutes of an ophthalmologist office; 5 (21.7%) and 15 (65.2%) of the procedures were performed within an ETT of 30 and 60 minutes, respectively, of an ophthalmologist office.
Among the 237 patients in Kentucky who underwent 1 or more surgeries performed by optometrists, 41 (17.3%) lived in ZCTAs with population-weighted centroids within an ETT of 10 minutes of the nearest ophthalmologist office, 138 (58.3%) were within 30 minutes’ ETT, and 235 (99.2%) were within 60 minutes’ ETT. Among the 519 patients in Oklahoma who underwent surgery by optometrists, 86 (16.6%) lived in ZCTAs with population-weighted centroids within 10 minutes of an ophthalmologist office, 265 (51.1%) were within 30 minutes, and 451 (86.9%) were within 60 minutes. Among the 32 persons in New Mexico who had optometrist-performed ocular surgeries, 1 (3%) lived in ZCTAs with population-weighted centroids within an ETT of 10 minutes to an ophthalmologist; 15 (46.9%) were within 30 minutes, and 23 (71.9%) were within an ETT of 60 minutes.
Using data from the US Census Bureau, an AAO member database, and a Medicare database, we assessed ETT to the nearest ophthalmologist office for persons residing in 3 states where optometrists have a scope of practice expanded to include surgery. In all 3 states, more than 40% of the population live within an ETT of 10 minutes to the nearest ophthalmologist office; roughly three-quarters live within an ETT of 30 minutes, and more than 90% live within an ETT of 45 minutes.
This study also examined access to ophthalmologists for persons of different races/ethnicities, ages, and residential status (urban vs rural) in these 3 states. When analyses were stratified to assess access to ophthalmologist offices by race/ethnicity, we found that, in all 3 states, more than 75% of non-Hispanic white, non-Hispanic black, and Hispanic people live within an ETT of 30 minutes of the nearest ophthalmologist office, and greater than 94% of persons in these racial/ethnic groups live within an ETT of 60 minutes of an ophthalmologist office. Likewise, the proportions of people younger than 65 years and those 65 years or older in each of the 3 states residing within an ETT of 30 minutes of an ophthalmologist ranged from 75.8% to 82.7%, and the proportions within an ETT of 60 minutes ranged from 93% to 99%. The proportion of residents of rural communities who live within an ETT of 60 minutes of the nearest ophthalmologist office is 94% for Kentucky, 83% for Oklahoma, and 67% for New Mexico. For optometrists who performed 5 or more surgeries, 0.5%, 12.2%, and 34.8% for Kentucky, Oklahoma, and New Mexico, respectively, occurred in locations where the ETT to an ophthalmologist exceeded 1 hour. Similarly, 0.8%, 13.1%, and 28.1% of persons receiving surgical services from an optometrist in Kentucky, Oklahoma, and New Mexico, respectively, lived in ZCTAs with population-weighted centroids exceeding an ETT of 60 minutes to the nearest ophthalmologist office.
In considering the expansion of optometrist scope of practice to include surgery, policy makers must weigh multiple factors. First, ophthalmologists have considerably more training, experience, and expertise in performing laser procedures, injections, and eyelid procedures compared with optometrists, which might translate into higher quality of care.18,19 Prior research by our group identified that Medicare patients with glaucoma residing in Oklahoma who underwent laser trabeculoplasty by an optometrist had a 189% increased likelihood of requiring additional laser in the same eye compared with those who were treated with lasers by an ophthalmologist (35.9% vs 15.1%; hazard ratio, 2.89; 95% CI, 2.00-4.17; P<.001.).16 While claims data alone cannot clarify why so many more patients who were treated by optometrists required additional laser procedures, 1 possibility is that the surgical care quality differs between optometrists and ophthalmologists.16
Second, what constitutes good access depends on drive time, the sorts of insurance that each practice accepts, the number of days per week when surgeries are performed, and the surgical procedure wait time. These quality-of-care and access factors must be assessed and subsequently balanced in deciding whether to expand optometrists’ surgical scope of practice.
There are some similarities, but also a few key differences, between the present analysis and a related study by Lee et al.4 Both studies used data from the US Census Bureau and Medicare claims data and geographic informations systems analysis to estimate network-based ETT from population centroids to ophthalmologist and optometrist offices. While Lee et al4 identified eye care clinician office locations based on Medicare claims record data, we identified ophthalmologist office locations using AAO member data. Although many AAO members practicing in the states of interest accept Medicare and thus would be captured in both data sets, there might be some AAO members who do not routinely care for Medicare enrollees; there are also other ophthalmologists who care for Medicare beneficiaries but are not AAO members. The Medicare data that Lee et al4 used also captured only the primary office location of each clinician who cared for Medicare enrollees, whereas the AAO member database used in this study can capture multiple locations where an ophthalmologist practices. Unlike the analysis by Lee et al, the calculations of the present study allowed for the possibility that patients can cross state lines to access the nearest ophthalmologist office. (The importance of this consideration to ETTs is exemplified by northern Kentucky, where residents can travel for care to Cincinnati, Ohio, which has many practicing ophthalmologists). Although the studies have some differences in their methods, both clearly demonstrate that most patients residing in the 3 states of interest live close to the office of at least 1 ophthalmologist.
The present study highlights how tapping into big data can be very useful to help answer questions on population health and guide decision making by health policy makers. Moreover, by identifying communities where a sizable number of persons reside (gold dots in Figures 1-3), yet access to ophthalmologists is limited (dark blue and white areas in Figures 1-3), researchers can assist local health care professionals in these communities to implement teleophthalmology20 to screen patients for sight-threatening ocular diseases and identify those who would most benefit from traveling 60 to 90 minutes to an ophthalmologist office for care. Another option policy makers might explore to improve ophthalmologic care access would be to offer financial incentives to ophthalmologists to open offices in underserved locations, in a manner similar to the bonus payments that the US Centers for Medicare and Medicaid Services provides to primary care practitioners who choose to practice in geographic areas where a shortage in primary care health access exists.21
This study has limitations. First, the AAO database lacks information on the types of health insurance each ophthalmologist accepts; thus, our analyses could not account for this. Ophthalmologists listed in the AAO member database might not see patients with certain health insurance types or those who lack health insurance. If some patients possess a form of health insurance that the nearest ophthalmologist office does not accept, they might need to drive further to receive care.
Second, the AAO database captures the primary office address of each ophthalmologist. Ophthalmologists can provide additional office addresses, but this is optional. For this reason, there might be additional office locations of ophthalmologists that were not considered, which would result in an overestimation of distance to the nearest ophthalmologist office.
Third, when calculating ETTs to the nearest ophthalmologist office, the present study assumed that patients traveled by car and not by public transportation or other means. Excluding these alternate means of transportation could result in travel time estimates that are higher or lower than reported here.
Fourth, the data sources used for this study can identify only the subset of clinicians who performed the procedures of interest. Optometrists without a record of performing any of these procedures might have opted not to participate in expanded scope of practice activities, lacked patients seeking such procedures, or lacked the necessary equipment to serve these patients. Likewise, some ophthalmologists practicing in these states may be subspecialists who do not routinely perform certain procedures, lack the equipment to perform these procedures, or do not care for Medicare enrollees. Although this study could not account for these factors, we recognize they can affect access to care.
Finally, this study uses the most recent census data available, as well as the most recent AAO member database and Medicare database; however, the data from these sources were not from the same years. However, since the population of residents and eye care providers in these 3 states did not change substantially from 2008 to 2016, it is doubtful that these differences in data sets substantially affected the findings of this study.
This study provides information about access to ophthalmologists in states with expanded scope of practice for optometrists. More than 40% of the residents in all 3 states live within an ETT of 10 minutes to an ophthalmologist office; even among the small segment of the population in these states who reside in rural communities, most live within an ETT of 1 hour to an ophthalmologist office. In addition, most patients who underwent surgeries by optometrists did not live in rural communities. These sorts of analyses provide researchers and health policy makers with a stronger basis for evaluating whether legislation expanding the scope of practice of optometrists in these states is achieving its stated purpose, which was to expand access. We encourage all states to perform similar analyses to help assess access to ophthalmologists, as these analyses can help health policy makers with making more informed decisions about whether the potential benefits of expanding scope of practice outweigh the potential tradeoffs in quality of care.
Corresponding Author: Joshua D. Stein, MD, MS, Department of Ophthalmology and Visual Sciences, University of Michigan Medical School, W. K. Kellogg Eye Center, 1000 Wall St, Ann Arbor, MI 48105 (firstname.lastname@example.org).
Accepted for Publication: September 28, 2017.
Published Online: November 22, 2017. doi:10.1001/jamaophthalmol.2017.5081
Author Contributions: Dr Stein had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs Stein and Kapoor contributed equally to the drafting of this manuscript.
Study concept and design: Stein, Kapoor, Li, Wagner, Miranda.
Acquisition, analysis, or interpretation of data: Stein, Tootoo, Li, Andrews, Miranda.
Drafting of the manuscript: Stein, Kapoor, Tootoo, Li, Miranda.
Critical revision of the manuscript for important intellectual content: Kapoor, Tootoo, Li, Wagner, Andrews, Miranda.
Statistical analysis: Tootoo, Li, Andrews, Miranda.
Obtained funding: Stein.
Administrative, technical, or material support: Stein, Wagner.
Study supervision: Stein, Kapoor, Tootoo, Wagner, Miranda.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Drs. Stein and Miranda and Messrs. Li and Tootoo have received financial support from the American Academy of Ophthalmology. No other disclosures were reported. The authors have no proprietary interest in any material discussed in the manuscript.
Funding/Support: Drs Stein and Miranda and Messrs Li and Tootoo report support from The American Academy of Ophthalmology during the course of this study; this study was also supported by an unrestricted grant from Research to Prevent Blindness to the University of Michigan.
Role of the Funder/Sponsor: The funders 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 article for publication.
Create a personal account or sign in to: