- Email: [email protected]
The American Association for Pediatric Ophthalmology and Strabismus workforce distribution project
Major Articles The American Association for Pediatric Ophthalmology and Strabismus workforce distribution project Robert Estes, MD,a Daniel Estes, MS,b Constance West, MD,c Jitka Zobal-Ratner, MD,d Patrick Droster, MD,e and John Simon, MDd PURPOSE METHODS
RESULTS
CONCLUSIONS
To describe data sources and functional utility of the American Association for Pediatric Ophthalmology and Strabismus (AAPOS) workforce database and associated map files. Population data from the 2000 U.S. Census and current listings from the AAPOS and American Academy of Ophthalmology (AAO) databases were organized to demonstrate and analyze practitioner-to-population relationships for metropolitan statistical areas nationwide. An interactive map was developed to provide an intuitive graphical display of the data. A total of 749 active AAPOS members were distributed in 154 of 280 defined metropolitan statistical areas. Within these areas, a 0- to 20-year age subgroup varied from 17.8% to 42.6%, with an average of 30.4%. The AAPOS member-to-million-person ratio varied from 1.3 to 27, with higher numbers generally representing regions with population bases inadequately defined by Census Bureau statistical area definitions. Ratios for a majority of larger, better-defined areas ranged from 3 to 4 AAPOS members per million persons. Sizable areas with no AAPOS member presence were identified and tabulated. AAO members with a specified pediatric practice focus who were not AAPOS members were identified in 103 areas, possibly influencing patient choices and practitioner referrals for these regions. The AAPOS workforce database and related interactive map display practitioner and population data that may assist physicians and planners in targeting practice development and identifying potentially underserved areas. ( J AAPOS 2007;11:325-329)
P
hysician distribution is an issue of concern to practitioners, health care managers, third-party payers, and politicians. Many Americans report having unmet medical needs,1 but the extent to which this involves services provided by members of the American Association for Pediatric Ophthalmology and Strabismus (AAPOS) is not known. Proximate AAPOS member availability would
a Author affiliations: aDepartment of Ophthalmology, Vanderbilt University, Nashville, Tennesee; bDepartment of Biomedical Engineering, University of Michigan, Ann Arbor, Michigan; cDepartment of Ophthalmology, Cincinnati Children’s Hospital, University of Cincinnati Medical Center, Cincinnati, Ohio; dDepartment of Ophthalmology, Albany Medical College/Lion’s Eye Institute, Albany, New York; eDepartment of Ophthalmology, DeVoss Children’s Hospital, Grand Rapids, Michigan Presented at the 32nd Annual Meeting of the American Association for Pediatric Ophthalmology and Strabismus, Keystone, Colorado, March 15–19, 2006. Financial disclosure: No conflicts with software manufacturers or government agencies providing data for this study. Submitted April 17, 2006. Revision accepted August 23, 2006. Reprint requests: Robert Estes, MD, Vanderbilt Eye Institute/Tennesssee Lion’s Eye Center, 1211 21st Ave. S, 104 Medical Arts Bldg., Nashville, TN 37212-2700 (email: [email protected]). Copyright © 2007 by the American Association for Pediatric Ophthalmology and Strabismus. 1091-8531/2007/$35.00 ⫹ 0 doi:10.1016/j.jaapos.2006.08.014
Journal of AAPOS
be expected to facilitate the delivery of specialized pediatric ophthalmologic and strabismus services to targeted populations. The AAPOS directory lists members geographically and detailed population data are available from the U.S. Census Bureau, but comparing practitioner-topopulation ratios for different regions can be a tedious, time-consuming process. The AAPOS demographic database has been organized to show specific relationships between AAPOS members and population demographics in both tabular and map-based formats to more readily address questions about pediatric ophthalmology/strabismus practitioner supply relative to geographic population definitions nationwide and to identify areas potentially needing services.
Materials and Methods U.S. practitioner locations were derived from the current AAPOS membership database as of the publication date of the 2005/2006 directory. Additional practitioners were identified by querying the American Academy of Ophthalmology (AAO) membership database for members listing a pediatric practice focus as of May 5, 2005.
325
326
Estes
Volume 11 Number 4 / August 2007
Table 1. Summary totals for the 25 most populous metropolitan statistical area derived from Census 2000 MSA area
Population 2000
AAPOS members
Population per AAPOS member
AAPOS members per million
New York-Northern New Jersey-Long Island, NY-NJ-CT-PA Los Angeles-Riverside-Orange County, CA Chicago-Gary-Kenosha, IL-IN-WI Washington-Baltimore, DC-MD-VA-WV San Francisco-Oakland-San Jose, CA Philadelphia-Wilmington-Atlantic City, PA-NJ-DE-MD Boston-Worcester-Lawrence, MA-NH-ME-CT Detroit-Ann Arbor-Flint, MI Dallas-Fort Worth, TX Houston-Galveston-Brazoria, TX Atlanta, GA Miami-Fort Lauderdale, FL Seattle-Tacoma-Bremerton, WA Phoenix-Mesa, AZ Minneapolis-St. Paul, MN-WI Cleveland-Akron, OH San Diego, CA St. Louis, MO-IL Denver-Boulder-Greeley, CO Tampa-St. Petersburg-Clearwater, FL Pittsburgh, PA Portland-Salem, OR-WA Cincinnati-Hamilton, OH-KY-IN Sacramento-Yolo, CA
21199865 16373645 9157540 7608070 7039362 6188463 5819100 5456428 5221801 4669571 4112198 3876380 3554760 3251876 2968806 2945831 2813833 2603607 2581506 2395997 2358695 2265223 1979202 1796857
85 41 31 38 21 32 24 15 16 15 12 13 10 10 9 13 7 7 7 8 7 11 4 3
249410 399357 295405 200212 335208 193389 242463 363762 326363 311305 342683 298183 355476 325188 329867 226602 401976 371944 368787 299500 336956 205929 494801 598952
4.0 2.5 3.4 5.0 3.0 5.2 4.1 2.7 3.1 3.2 2.9 3.4 2.8 3.1 3.0 4.4 2.5 2.7 2.7 3.3 3.0 4.9 2.0 1.7
Population units were chosen as metropolitan or micropolitan statistical areas, defined by the U.S. Office of Management and Budget (OMB) as core areas containing a substantial population nucleus with adjacent communities having a high degree of economic and social integration. Each metropolitan statistical area (MSA) must have at least one urbanized area of 50,000 or more inhabitants whereas a micropolitan statistical area must have at least one urban cluster of at least 10,000 but less than 50,000 persons.2 Population data were obtained from the U.S. Census Bureau’s Web site and organized to selectively display practitioners and population statistics for 276 MSAs defined by the Census Bureau in 2000.3 Four additional areas were derived for 5 individuals practicing in locations defined as micropolitan regions. Data were organized in a Web-compatible format for viewing from the AAPOS Web site4 as well as a graphical display using a mapping application.5
Results The 2000 census established the U.S. population to be 281,421,906. Of this total, 225,981,679 (80%) resided in defined MSAs. The active 749 AAPOS members listed addresses and presumably practiced in 154 of the 280 MSAs. The AAO database included an additional 227 members who specified pediatric ophthalmology / strabismus as a practice focus but did not belong to AAPOS. These individuals were located in 103 of the 280 MSAs. The population-to-AAPOS member ratio was 375,730 (2.7 AAPOS members/million persons) for the entire United States. The ratio was 303,739 (3.3 AAPOS members/million persons) for the 280 MSAs. Limiting popu-
lation totals to only the 154 MSAs with a defined AAPOS member presence, the ratio was 269,796 (3.7 AAPOS members/million persons). Totals for the most populous 25 MSAs are shown in Table 1 (average 3.3 AAPOS members/million persons). The AAPOS member/million person ratio ranged from 1.3 (Las Vegas, NV) to 27 (Iowa City, IA). This latter number was artificially high because of the Census Bureau MSA specification including only Iowa City’s county, leading to a much lower population base than expected for a major regional referral medical center. Similar results were seen for other referral medical centers located in smaller MSAs (Rochester, MN; Lebanon, NH; Morgantown, WV; Madison, WI) as well as a number of onecounty MSAs with populations less than 200,000. The 55,540,227 individuals not included in defined MSAs were generally located in lower population density regions farther removed from central core areas. One state (North Dakota) had no listed AAPOS or non-AAPOS AAO pediatric practice focus practitioner. Many of the lower-density population counties were geographically situated between MSAs with established AAPOS practitioner presences. Patients needing services in these areas could conceivably be evaluated and treated in different MSAs, depending on the choices of the patient, parents, or referring physicians. Subgroups were created based on age definitions specified by guidelines published for early and periodic screening, diagnosis, and treatment of children under Medicaid (0-20 years),6 practice scope statements from the American Academy of Pediatrics,7,8 and Medicare (65⫹ years). 9 The 0- to 20-year subgroup varied from 17.8% (Punta Gorda,
Journal of AAPOS
Volume 11 Number 4 / August 2007
Estes
327
Table 3. Metropolitan statistical areas with populations greater than 250,000 having no listed AAPOS or AAO member specifying a pediatric practice focus
FIG 1. Scatterplot showing age percentage 0 to 20 years for all metropolitan statistical areas (ordered alphabetically). Table 2. Metropolitan statistical areas with pediatric populations 35% or greater and no currently listed AAPOS member
MSA area McAllen-Edinburg-Mission, TX Modesto, CA Visalia-Tulare-Porterville, CA Brownsville-Harlingen-San Benito, TX Odessa-Midland, TX Yakima, WA Merced, CA Laredo, TX Richland-Kennewick-Pasco, WA Las Cruces, NM Bryan-College Station, TX Jacksonville, NC
MSA Population, Total 0 to % 0-20 2000 20 years yearrs 569463 446997 368021 335227 237132 222581 210554 193117 191822 174682 152415 150355
230532 159593 142703 129693 83136 80907 82950 79997 67716 63109 56638 54249
40.5% 35.7% 38.8% 38.7% 35.1% 36.3% 39.4% 41.4% 35.3% 36.1% 37.2% 36.1%
FL) to 42.6% (Provo-Orem, UT) of the total population, with an average of 30.4%. Ninety percent of MSAs fell in the range of 25% to 35% for this population group (Figure 1), with 19 MSAs greater than 35% and 9 MSAs less than 25%. Of the 19 areas with 0- to 20-year populations greater than 35%, only 7 had AAPOS members listed in the current directory (Table 2 shows the 12 areas with no listed AAPOS member). Table 3 lists MSAs with populations greater than 250,000 having no currently listed AAPOS or non-AAPOS AAO member with a specified pediatric practice focus. The 127 MSAs with no AAPOS members encompassed an aggregate population of 24,501,940 (10.8% of the total MSA population).
Discussion The AAPOS workforce database and related map files display area-specific population and practitioner data to facilitate analysis of specific regions. To use the formbased application, a state is chosen from a drop-down list, selecting state-specific MSAs for an adjacent list. An MSA is selected, and Census Bureau defined counties and referenced AAPOS members are listed in separate frames. An
Journal of AAPOS
MSA
Population 2000
% 0-20 years
Fort Wayne, IN Johnson City-Kingsport-Bristol, TN-VA Canton-Massillon, OH Lafayette, LA Beaumont-Port Arthur, TX Rockford, IL Visalia-Tulare-Porterville, CA Appleton-Oshkosh-Neenah, WI Brownsville-Harlingen-San Benito, TX Montgomery, AL Eugene-Springfield, OR Fort Pierce-Port St. Lucie, FL Huntington-Ashland, WV-KY-OH Fayetteville-Springdale-Rogers, AR Columbus, GA-AL Naples, FL
502141 480091 406934 385647 385090 371236 368021 358365 335227 333055 322959 319426 315538 311121 274624 251377
31.9% 25.3% 28.8% 33.4% 30.8% 30.7% 38.8% 30.7% 38.7% 31.0% 28.2% 23.9% 27.1% 31.2% 32.2% 22.8%
Note that some areas are adjacent to MSAs with established AAPOS member practices. Any of these MSAs could have recently trained pediatric ophthalmologists who are not yet listed as AAPOS members, satellites from proximate MSAs, or practitioners who have relocated since the most recent update on which these data are based.
additional frame lists AAO members who have specified a practice focus for pediatric ophthalmology or strabismus. Summary data are displayed below the state and MSA names showing population changes from 1990 to 2000, age breakdowns (0-20 years, 21-64 years, 65⫹ years), the computed population-to-AAPOS member and AAPOS member per million persons ratios for the region.4 Figure 2 shows a grayscale-selected screen from the mapping application. Individual practitioners are identified by color-coded push pins whereas population and AAPOS member/million person ratios are seen by placing a mouse pointer within MSA areas or color-coded circles. A separate file shows individual county populations for comparison within specific regions. The 0- to 20-age subgroup was created based on Centers for Medicare and Medicaid Services (CMS) guidelines for early and periodic screening, diagnosis, and treatment of children Medicaid coverage, which include diagnosis and treatment for defects in vision, including eyeglasses according to a distinct periodicity schedule.6 In addition, the American Academy of Pediatrics has published a policy statement defining the age scope of pediatric practice as 0 to 21 years.7,8 Individual AAPOS members may elect to target their patient age limit to a narrower interval. Census bureau data are divided into 6 separate sections for the 0to 20-year age range. In general, MSAs with greater 0- to 20-year age percentages have higher percentages for smaller age subgroups when different MSAs are compared. A number of qualifications and limitations should be recognized. MSAs generally are better defined for larger regions. The Census Bureau does not consistently define multiple county MSAs for cities smaller than approxi-
328
Estes
Volume 11 Number 4 / August 2007
FIG 2. Grayscale map of AAPOS demographic database showing northern California/San Francisco Bay area. Population totals and pediatric ophthalmologist per million person markers are color-coded with actual numbers activated for view by placing mouse pointer over the specified areas. Individual practitioners are identified by placing mouse pointer over “push-pins,” which are color-coded to differentiate AAPOS members from AAO members who have specified a pediatric practice focus in the AAO database. Mappoint must be installed to view this data representation.
mately 200,000, leading to underestimates for many areas’ potential patient bases. The Bureau changed MSA county inclusions several times while the project evolved, most recently in 2003, when the number of MSAs was increased to more than 360. The newest specification segments the most populous MSAs into smaller areas which may give a better picture of urban population distributions. However, the Bureau has not yet published age-specific breakdowns to accompany the newly specified MSAs. The mapping program uses a transitional, pre-2003 MSA specification delineating 308 areas for which data were adapted for the map display. Because only 280 MSAs are specified in the database, users may encounter numerical differences in population and practitioner ratios when simultaneously viewing certain areas with both programs. Smaller MSAs can be more accurately defined by manually downloading county-specific data from the Census Bureau Web page3 or by using the MSA map in conjunction with a county population map. The mapping program does not allow the county and MSA population figures to be layered on the same map, so users must open both maps and switch between screen views to gather customized population specifications. A number of moderately sized MSAs are adjacent and may better define a population base when combined. Additionally, members may list an address in one MSA while practicing in an adjacent MSA, distorting AAPOS member-to-population statistics for both areas. Satellite locations in different MSAs could not be identified because the directory database allows only one address specification. This problem has been rec-
ognized and plans exist to allow at least two different addresses in future directory listings. If the database is to retain a reasonable degree of accuracy, it will be helpful if members can specify a percentage of time when patients are seen at a satellite location in a different MSA (ie, 20% for 1 day/week). Recently trained fellows are not identified as full AAPOS members for 18 to 24 months, making it difficult to confirm new practitioner activity and location during this transitional period. The AAPOS member/million person ratio will underestimate practitioner availability for regions where these fellows relocate, and this discrepancy can be significant for moderately sized MSAs. Additionally, AAO members who list a pediatric practice focus but are not AAPOS members may affect patient availability and physician referral patterns for certain areas. This designation is chosen voluntarily, so it is not possible to know the actual percentage of these practices devoted to pediatric ophthalmology or strabismus. Individuals exploring practice opportunities should inquire about the nature of physicians’ practices when investigating specific MSAs. Users should recognize that the database will always have an update lag for members who become ill or disabled, move, or retire. Absolute accuracy is an illusion— the data represent a point-in-time snapshot and are best viewed as a baseline approximation that must be examined for real-time validity when a targeted MSA is investigated for practice opportunities. Members should exercise caution when using these data to analyze population needs or practice prospects. Practice viability is influenced by a number of variables: the general
Journal of AAPOS
Volume 11 Number 4 / August 2007
economic health of a region, including Medicaid and managed care reimbursement, office expenses, and nonmedical provider penetration. It is unrealistic to expect one number to define this globally. In general, a probabilistic approach would suggest that a larger population with a higher percentage of children and fewer subspecialist pediatric practitioners should give better odds for a professionally rewarding and financially stable practice. The RAND report used estimates of treatment need and market demand to generate practitioner supply estimates for ophthalmology based on optometry-first, ophthalmologyfirst, and mixed primary care models prevalent in the managed care environment of the mid-1990s.10,11 Pediatric ophthalmology was projected to have an adequate supply of practitioners for the 2000 to 2010 time frame; however, this number was based on a year 2000 estimate of 558 practitioners, which underestimated the actual number of AAPOS members by more than 100 individuals. The report contained no analysis of geographically based population data. The RAND report modeling has been questioned with suggestions that workforce estimates acknowledge a need for identification and treatment of unrecognized eye disease in vulnerable populations as well as observing differences in patient problem complexity and comprehensiveness of care.12 A separate analysis of oculoplastic surgeons analyzed expected practitioner retirement and concluded that an appropriate number of fellows were being trained to maintain a stable subspecialty practitioner supply.13 These authors speculated that future treatment needs and demands could result in an increased need for fellowshiptrained oculoplastic surgeons in contrast to an oversupply estimated by the RAND report. Although population trends were acknowledged, no specific practitioner-topopulation analysis was referenced. The most recent report of the Council on Graduate Medical Education14 projected an impending shortage of physicians in general and specialty categories based on analysis of the current training levels, increased size of the aging population, anticipated practitioner retirements, and changing work patterns for many providers. Although weighted for expected service demands of older individuals, the report acknowledged potential practitioner shortages for all age categories. This project does not address future demand or need for pediatric ophthalmology services, but it does demonstrate significant variability in proximate pediatric ophthalmologist availability for many areas of the United States. General trends can be inferred by examining statistics for larger MSAs that have supported pediatric ophthalmology practices for a number of years. Table 1 summa-
Journal of AAPOS
Estes
329
rizes data for the 25 most populous MSAs. The average AAPOS member/million person ratio for these areas is 3.3, with a range of 1.7 to 5.4. Most areas cluster in a range of 2.7 to 4.0 AAPOS members/million persons, which could suggest that a range of 3 to 4 AAPOS members/ million persons might be an appropriate target, but only when viewed in context of the financial variables mentioned previously. The report of the Council on Graduate Medical Education points out that there is no single, generally accepted standard for the number of physicians needed by a community in any category.14 The AAPOS workforce database and related mapping files display reasonably accurate workforce and population data that can serve as a baseline to assist physicians and planners in targeting practice development and identifying potentially underserved areas. Understanding pediatric ophthalmology and strabismus practitioner workforce and population data should empower members to more effectively counsel planners and politicians when confronted with access-based healthcare questions. References 1. Sanmartin C, Ng E, Blackwell D, Gentleman J, Martinez M, Simile C. Joint Canada/United States Survey of Health 2002-03. Statistics Canada, 2004. Catalogue 82M0022-XIE. 2. http://www.census.gov/population/www/estimates/aboutmetro.html. Accessed September 11, 2006. 3. http://factfinder.census.gov/servlet/DTGeoSearchByListServlet?ds_ name⫽DEC_2000_SF1_U&_lang⫽en&_ts⫽168713956890. Accessed September 11, 2006. 4. http://www.aapos.org/associations/5371/files/apposdb3.html. Accessed September 11, 2006. 5. Mappoint 2004. Redmond, WA: Microsoft Corporation. 6. http://www.cms.hhs.gov/MedicaidEarlyPeriodicScrn/. Accessed September 11, 2006. 7. Litt IE. Age Limits of Pediatrics, American Academy of Pediatrics, Council on Child Health. Pediatrics 1998;102(Suppl 1)249-50. 8. Policy Statement: AAP Publications Retired and Reaffirmed. Pediatrics 2006;117:1846-7. 9. http://www.cms.hhs.gov/MedicareGenInfo/. Accessed September 11, 2006. 10. Lee P, Jackson C, Relles D: Estimating eye care provider supply and workforce requirements. Santa Monica (CA): RAND Corporation; 1995. Publication MR-516-AAO. 11. Lee P, Relles D, Jackson C. Subspecialty distributions of ophthalmologists in the workforce. Arch Ophthalmol 1998;116:917-20. 12. Bass E. Public health need vs. market demand for ophthalmologists: What are our priorities? Arch Ophthalmol 1998;116:930. 13. Penne R, Lemke B. How many ophthalmic plastic surgeons should be trained? Ophthal Plast Reconstr Surg 2001;17:231-3. 14. Council on Graduate Medical Education. Physician Workforce Policy Guidelines for the US: 2000-2020. Draft Report, August 2004. Rockville, MD: Health and Human Resources Administration, US Department of Health and Human Services; 2004.
RESULTS
CONCLUSIONS
To describe data sources and functional utility of the American Association for Pediatric Ophthalmology and Strabismus (AAPOS) workforce database and associated map files. Population data from the 2000 U.S. Census and current listings from the AAPOS and American Academy of Ophthalmology (AAO) databases were organized to demonstrate and analyze practitioner-to-population relationships for metropolitan statistical areas nationwide. An interactive map was developed to provide an intuitive graphical display of the data. A total of 749 active AAPOS members were distributed in 154 of 280 defined metropolitan statistical areas. Within these areas, a 0- to 20-year age subgroup varied from 17.8% to 42.6%, with an average of 30.4%. The AAPOS member-to-million-person ratio varied from 1.3 to 27, with higher numbers generally representing regions with population bases inadequately defined by Census Bureau statistical area definitions. Ratios for a majority of larger, better-defined areas ranged from 3 to 4 AAPOS members per million persons. Sizable areas with no AAPOS member presence were identified and tabulated. AAO members with a specified pediatric practice focus who were not AAPOS members were identified in 103 areas, possibly influencing patient choices and practitioner referrals for these regions. The AAPOS workforce database and related interactive map display practitioner and population data that may assist physicians and planners in targeting practice development and identifying potentially underserved areas. ( J AAPOS 2007;11:325-329)
P
hysician distribution is an issue of concern to practitioners, health care managers, third-party payers, and politicians. Many Americans report having unmet medical needs,1 but the extent to which this involves services provided by members of the American Association for Pediatric Ophthalmology and Strabismus (AAPOS) is not known. Proximate AAPOS member availability would
a Author affiliations: aDepartment of Ophthalmology, Vanderbilt University, Nashville, Tennesee; bDepartment of Biomedical Engineering, University of Michigan, Ann Arbor, Michigan; cDepartment of Ophthalmology, Cincinnati Children’s Hospital, University of Cincinnati Medical Center, Cincinnati, Ohio; dDepartment of Ophthalmology, Albany Medical College/Lion’s Eye Institute, Albany, New York; eDepartment of Ophthalmology, DeVoss Children’s Hospital, Grand Rapids, Michigan Presented at the 32nd Annual Meeting of the American Association for Pediatric Ophthalmology and Strabismus, Keystone, Colorado, March 15–19, 2006. Financial disclosure: No conflicts with software manufacturers or government agencies providing data for this study. Submitted April 17, 2006. Revision accepted August 23, 2006. Reprint requests: Robert Estes, MD, Vanderbilt Eye Institute/Tennesssee Lion’s Eye Center, 1211 21st Ave. S, 104 Medical Arts Bldg., Nashville, TN 37212-2700 (email: [email protected]). Copyright © 2007 by the American Association for Pediatric Ophthalmology and Strabismus. 1091-8531/2007/$35.00 ⫹ 0 doi:10.1016/j.jaapos.2006.08.014
Journal of AAPOS
be expected to facilitate the delivery of specialized pediatric ophthalmologic and strabismus services to targeted populations. The AAPOS directory lists members geographically and detailed population data are available from the U.S. Census Bureau, but comparing practitioner-topopulation ratios for different regions can be a tedious, time-consuming process. The AAPOS demographic database has been organized to show specific relationships between AAPOS members and population demographics in both tabular and map-based formats to more readily address questions about pediatric ophthalmology/strabismus practitioner supply relative to geographic population definitions nationwide and to identify areas potentially needing services.
Materials and Methods U.S. practitioner locations were derived from the current AAPOS membership database as of the publication date of the 2005/2006 directory. Additional practitioners were identified by querying the American Academy of Ophthalmology (AAO) membership database for members listing a pediatric practice focus as of May 5, 2005.
325
326
Estes
Volume 11 Number 4 / August 2007
Table 1. Summary totals for the 25 most populous metropolitan statistical area derived from Census 2000 MSA area
Population 2000
AAPOS members
Population per AAPOS member
AAPOS members per million
New York-Northern New Jersey-Long Island, NY-NJ-CT-PA Los Angeles-Riverside-Orange County, CA Chicago-Gary-Kenosha, IL-IN-WI Washington-Baltimore, DC-MD-VA-WV San Francisco-Oakland-San Jose, CA Philadelphia-Wilmington-Atlantic City, PA-NJ-DE-MD Boston-Worcester-Lawrence, MA-NH-ME-CT Detroit-Ann Arbor-Flint, MI Dallas-Fort Worth, TX Houston-Galveston-Brazoria, TX Atlanta, GA Miami-Fort Lauderdale, FL Seattle-Tacoma-Bremerton, WA Phoenix-Mesa, AZ Minneapolis-St. Paul, MN-WI Cleveland-Akron, OH San Diego, CA St. Louis, MO-IL Denver-Boulder-Greeley, CO Tampa-St. Petersburg-Clearwater, FL Pittsburgh, PA Portland-Salem, OR-WA Cincinnati-Hamilton, OH-KY-IN Sacramento-Yolo, CA
21199865 16373645 9157540 7608070 7039362 6188463 5819100 5456428 5221801 4669571 4112198 3876380 3554760 3251876 2968806 2945831 2813833 2603607 2581506 2395997 2358695 2265223 1979202 1796857
85 41 31 38 21 32 24 15 16 15 12 13 10 10 9 13 7 7 7 8 7 11 4 3
249410 399357 295405 200212 335208 193389 242463 363762 326363 311305 342683 298183 355476 325188 329867 226602 401976 371944 368787 299500 336956 205929 494801 598952
4.0 2.5 3.4 5.0 3.0 5.2 4.1 2.7 3.1 3.2 2.9 3.4 2.8 3.1 3.0 4.4 2.5 2.7 2.7 3.3 3.0 4.9 2.0 1.7
Population units were chosen as metropolitan or micropolitan statistical areas, defined by the U.S. Office of Management and Budget (OMB) as core areas containing a substantial population nucleus with adjacent communities having a high degree of economic and social integration. Each metropolitan statistical area (MSA) must have at least one urbanized area of 50,000 or more inhabitants whereas a micropolitan statistical area must have at least one urban cluster of at least 10,000 but less than 50,000 persons.2 Population data were obtained from the U.S. Census Bureau’s Web site and organized to selectively display practitioners and population statistics for 276 MSAs defined by the Census Bureau in 2000.3 Four additional areas were derived for 5 individuals practicing in locations defined as micropolitan regions. Data were organized in a Web-compatible format for viewing from the AAPOS Web site4 as well as a graphical display using a mapping application.5
Results The 2000 census established the U.S. population to be 281,421,906. Of this total, 225,981,679 (80%) resided in defined MSAs. The active 749 AAPOS members listed addresses and presumably practiced in 154 of the 280 MSAs. The AAO database included an additional 227 members who specified pediatric ophthalmology / strabismus as a practice focus but did not belong to AAPOS. These individuals were located in 103 of the 280 MSAs. The population-to-AAPOS member ratio was 375,730 (2.7 AAPOS members/million persons) for the entire United States. The ratio was 303,739 (3.3 AAPOS members/million persons) for the 280 MSAs. Limiting popu-
lation totals to only the 154 MSAs with a defined AAPOS member presence, the ratio was 269,796 (3.7 AAPOS members/million persons). Totals for the most populous 25 MSAs are shown in Table 1 (average 3.3 AAPOS members/million persons). The AAPOS member/million person ratio ranged from 1.3 (Las Vegas, NV) to 27 (Iowa City, IA). This latter number was artificially high because of the Census Bureau MSA specification including only Iowa City’s county, leading to a much lower population base than expected for a major regional referral medical center. Similar results were seen for other referral medical centers located in smaller MSAs (Rochester, MN; Lebanon, NH; Morgantown, WV; Madison, WI) as well as a number of onecounty MSAs with populations less than 200,000. The 55,540,227 individuals not included in defined MSAs were generally located in lower population density regions farther removed from central core areas. One state (North Dakota) had no listed AAPOS or non-AAPOS AAO pediatric practice focus practitioner. Many of the lower-density population counties were geographically situated between MSAs with established AAPOS practitioner presences. Patients needing services in these areas could conceivably be evaluated and treated in different MSAs, depending on the choices of the patient, parents, or referring physicians. Subgroups were created based on age definitions specified by guidelines published for early and periodic screening, diagnosis, and treatment of children under Medicaid (0-20 years),6 practice scope statements from the American Academy of Pediatrics,7,8 and Medicare (65⫹ years). 9 The 0- to 20-year subgroup varied from 17.8% (Punta Gorda,
Journal of AAPOS
Volume 11 Number 4 / August 2007
Estes
327
Table 3. Metropolitan statistical areas with populations greater than 250,000 having no listed AAPOS or AAO member specifying a pediatric practice focus
FIG 1. Scatterplot showing age percentage 0 to 20 years for all metropolitan statistical areas (ordered alphabetically). Table 2. Metropolitan statistical areas with pediatric populations 35% or greater and no currently listed AAPOS member
MSA area McAllen-Edinburg-Mission, TX Modesto, CA Visalia-Tulare-Porterville, CA Brownsville-Harlingen-San Benito, TX Odessa-Midland, TX Yakima, WA Merced, CA Laredo, TX Richland-Kennewick-Pasco, WA Las Cruces, NM Bryan-College Station, TX Jacksonville, NC
MSA Population, Total 0 to % 0-20 2000 20 years yearrs 569463 446997 368021 335227 237132 222581 210554 193117 191822 174682 152415 150355
230532 159593 142703 129693 83136 80907 82950 79997 67716 63109 56638 54249
40.5% 35.7% 38.8% 38.7% 35.1% 36.3% 39.4% 41.4% 35.3% 36.1% 37.2% 36.1%
FL) to 42.6% (Provo-Orem, UT) of the total population, with an average of 30.4%. Ninety percent of MSAs fell in the range of 25% to 35% for this population group (Figure 1), with 19 MSAs greater than 35% and 9 MSAs less than 25%. Of the 19 areas with 0- to 20-year populations greater than 35%, only 7 had AAPOS members listed in the current directory (Table 2 shows the 12 areas with no listed AAPOS member). Table 3 lists MSAs with populations greater than 250,000 having no currently listed AAPOS or non-AAPOS AAO member with a specified pediatric practice focus. The 127 MSAs with no AAPOS members encompassed an aggregate population of 24,501,940 (10.8% of the total MSA population).
Discussion The AAPOS workforce database and related map files display area-specific population and practitioner data to facilitate analysis of specific regions. To use the formbased application, a state is chosen from a drop-down list, selecting state-specific MSAs for an adjacent list. An MSA is selected, and Census Bureau defined counties and referenced AAPOS members are listed in separate frames. An
Journal of AAPOS
MSA
Population 2000
% 0-20 years
Fort Wayne, IN Johnson City-Kingsport-Bristol, TN-VA Canton-Massillon, OH Lafayette, LA Beaumont-Port Arthur, TX Rockford, IL Visalia-Tulare-Porterville, CA Appleton-Oshkosh-Neenah, WI Brownsville-Harlingen-San Benito, TX Montgomery, AL Eugene-Springfield, OR Fort Pierce-Port St. Lucie, FL Huntington-Ashland, WV-KY-OH Fayetteville-Springdale-Rogers, AR Columbus, GA-AL Naples, FL
502141 480091 406934 385647 385090 371236 368021 358365 335227 333055 322959 319426 315538 311121 274624 251377
31.9% 25.3% 28.8% 33.4% 30.8% 30.7% 38.8% 30.7% 38.7% 31.0% 28.2% 23.9% 27.1% 31.2% 32.2% 22.8%
Note that some areas are adjacent to MSAs with established AAPOS member practices. Any of these MSAs could have recently trained pediatric ophthalmologists who are not yet listed as AAPOS members, satellites from proximate MSAs, or practitioners who have relocated since the most recent update on which these data are based.
additional frame lists AAO members who have specified a practice focus for pediatric ophthalmology or strabismus. Summary data are displayed below the state and MSA names showing population changes from 1990 to 2000, age breakdowns (0-20 years, 21-64 years, 65⫹ years), the computed population-to-AAPOS member and AAPOS member per million persons ratios for the region.4 Figure 2 shows a grayscale-selected screen from the mapping application. Individual practitioners are identified by color-coded push pins whereas population and AAPOS member/million person ratios are seen by placing a mouse pointer within MSA areas or color-coded circles. A separate file shows individual county populations for comparison within specific regions. The 0- to 20-age subgroup was created based on Centers for Medicare and Medicaid Services (CMS) guidelines for early and periodic screening, diagnosis, and treatment of children Medicaid coverage, which include diagnosis and treatment for defects in vision, including eyeglasses according to a distinct periodicity schedule.6 In addition, the American Academy of Pediatrics has published a policy statement defining the age scope of pediatric practice as 0 to 21 years.7,8 Individual AAPOS members may elect to target their patient age limit to a narrower interval. Census bureau data are divided into 6 separate sections for the 0to 20-year age range. In general, MSAs with greater 0- to 20-year age percentages have higher percentages for smaller age subgroups when different MSAs are compared. A number of qualifications and limitations should be recognized. MSAs generally are better defined for larger regions. The Census Bureau does not consistently define multiple county MSAs for cities smaller than approxi-
328
Estes
Volume 11 Number 4 / August 2007
FIG 2. Grayscale map of AAPOS demographic database showing northern California/San Francisco Bay area. Population totals and pediatric ophthalmologist per million person markers are color-coded with actual numbers activated for view by placing mouse pointer over the specified areas. Individual practitioners are identified by placing mouse pointer over “push-pins,” which are color-coded to differentiate AAPOS members from AAO members who have specified a pediatric practice focus in the AAO database. Mappoint must be installed to view this data representation.
mately 200,000, leading to underestimates for many areas’ potential patient bases. The Bureau changed MSA county inclusions several times while the project evolved, most recently in 2003, when the number of MSAs was increased to more than 360. The newest specification segments the most populous MSAs into smaller areas which may give a better picture of urban population distributions. However, the Bureau has not yet published age-specific breakdowns to accompany the newly specified MSAs. The mapping program uses a transitional, pre-2003 MSA specification delineating 308 areas for which data were adapted for the map display. Because only 280 MSAs are specified in the database, users may encounter numerical differences in population and practitioner ratios when simultaneously viewing certain areas with both programs. Smaller MSAs can be more accurately defined by manually downloading county-specific data from the Census Bureau Web page3 or by using the MSA map in conjunction with a county population map. The mapping program does not allow the county and MSA population figures to be layered on the same map, so users must open both maps and switch between screen views to gather customized population specifications. A number of moderately sized MSAs are adjacent and may better define a population base when combined. Additionally, members may list an address in one MSA while practicing in an adjacent MSA, distorting AAPOS member-to-population statistics for both areas. Satellite locations in different MSAs could not be identified because the directory database allows only one address specification. This problem has been rec-
ognized and plans exist to allow at least two different addresses in future directory listings. If the database is to retain a reasonable degree of accuracy, it will be helpful if members can specify a percentage of time when patients are seen at a satellite location in a different MSA (ie, 20% for 1 day/week). Recently trained fellows are not identified as full AAPOS members for 18 to 24 months, making it difficult to confirm new practitioner activity and location during this transitional period. The AAPOS member/million person ratio will underestimate practitioner availability for regions where these fellows relocate, and this discrepancy can be significant for moderately sized MSAs. Additionally, AAO members who list a pediatric practice focus but are not AAPOS members may affect patient availability and physician referral patterns for certain areas. This designation is chosen voluntarily, so it is not possible to know the actual percentage of these practices devoted to pediatric ophthalmology or strabismus. Individuals exploring practice opportunities should inquire about the nature of physicians’ practices when investigating specific MSAs. Users should recognize that the database will always have an update lag for members who become ill or disabled, move, or retire. Absolute accuracy is an illusion— the data represent a point-in-time snapshot and are best viewed as a baseline approximation that must be examined for real-time validity when a targeted MSA is investigated for practice opportunities. Members should exercise caution when using these data to analyze population needs or practice prospects. Practice viability is influenced by a number of variables: the general
Journal of AAPOS
Volume 11 Number 4 / August 2007
economic health of a region, including Medicaid and managed care reimbursement, office expenses, and nonmedical provider penetration. It is unrealistic to expect one number to define this globally. In general, a probabilistic approach would suggest that a larger population with a higher percentage of children and fewer subspecialist pediatric practitioners should give better odds for a professionally rewarding and financially stable practice. The RAND report used estimates of treatment need and market demand to generate practitioner supply estimates for ophthalmology based on optometry-first, ophthalmologyfirst, and mixed primary care models prevalent in the managed care environment of the mid-1990s.10,11 Pediatric ophthalmology was projected to have an adequate supply of practitioners for the 2000 to 2010 time frame; however, this number was based on a year 2000 estimate of 558 practitioners, which underestimated the actual number of AAPOS members by more than 100 individuals. The report contained no analysis of geographically based population data. The RAND report modeling has been questioned with suggestions that workforce estimates acknowledge a need for identification and treatment of unrecognized eye disease in vulnerable populations as well as observing differences in patient problem complexity and comprehensiveness of care.12 A separate analysis of oculoplastic surgeons analyzed expected practitioner retirement and concluded that an appropriate number of fellows were being trained to maintain a stable subspecialty practitioner supply.13 These authors speculated that future treatment needs and demands could result in an increased need for fellowshiptrained oculoplastic surgeons in contrast to an oversupply estimated by the RAND report. Although population trends were acknowledged, no specific practitioner-topopulation analysis was referenced. The most recent report of the Council on Graduate Medical Education14 projected an impending shortage of physicians in general and specialty categories based on analysis of the current training levels, increased size of the aging population, anticipated practitioner retirements, and changing work patterns for many providers. Although weighted for expected service demands of older individuals, the report acknowledged potential practitioner shortages for all age categories. This project does not address future demand or need for pediatric ophthalmology services, but it does demonstrate significant variability in proximate pediatric ophthalmologist availability for many areas of the United States. General trends can be inferred by examining statistics for larger MSAs that have supported pediatric ophthalmology practices for a number of years. Table 1 summa-
Journal of AAPOS
Estes
329
rizes data for the 25 most populous MSAs. The average AAPOS member/million person ratio for these areas is 3.3, with a range of 1.7 to 5.4. Most areas cluster in a range of 2.7 to 4.0 AAPOS members/million persons, which could suggest that a range of 3 to 4 AAPOS members/ million persons might be an appropriate target, but only when viewed in context of the financial variables mentioned previously. The report of the Council on Graduate Medical Education points out that there is no single, generally accepted standard for the number of physicians needed by a community in any category.14 The AAPOS workforce database and related mapping files display reasonably accurate workforce and population data that can serve as a baseline to assist physicians and planners in targeting practice development and identifying potentially underserved areas. Understanding pediatric ophthalmology and strabismus practitioner workforce and population data should empower members to more effectively counsel planners and politicians when confronted with access-based healthcare questions. References 1. Sanmartin C, Ng E, Blackwell D, Gentleman J, Martinez M, Simile C. Joint Canada/United States Survey of Health 2002-03. Statistics Canada, 2004. Catalogue 82M0022-XIE. 2. http://www.census.gov/population/www/estimates/aboutmetro.html. Accessed September 11, 2006. 3. http://factfinder.census.gov/servlet/DTGeoSearchByListServlet?ds_ name⫽DEC_2000_SF1_U&_lang⫽en&_ts⫽168713956890. Accessed September 11, 2006. 4. http://www.aapos.org/associations/5371/files/apposdb3.html. Accessed September 11, 2006. 5. Mappoint 2004. Redmond, WA: Microsoft Corporation. 6. http://www.cms.hhs.gov/MedicaidEarlyPeriodicScrn/. Accessed September 11, 2006. 7. Litt IE. Age Limits of Pediatrics, American Academy of Pediatrics, Council on Child Health. Pediatrics 1998;102(Suppl 1)249-50. 8. Policy Statement: AAP Publications Retired and Reaffirmed. Pediatrics 2006;117:1846-7. 9. http://www.cms.hhs.gov/MedicareGenInfo/. Accessed September 11, 2006. 10. Lee P, Jackson C, Relles D: Estimating eye care provider supply and workforce requirements. Santa Monica (CA): RAND Corporation; 1995. Publication MR-516-AAO. 11. Lee P, Relles D, Jackson C. Subspecialty distributions of ophthalmologists in the workforce. Arch Ophthalmol 1998;116:917-20. 12. Bass E. Public health need vs. market demand for ophthalmologists: What are our priorities? Arch Ophthalmol 1998;116:930. 13. Penne R, Lemke B. How many ophthalmic plastic surgeons should be trained? Ophthal Plast Reconstr Surg 2001;17:231-3. 14. Council on Graduate Medical Education. Physician Workforce Policy Guidelines for the US: 2000-2020. Draft Report, August 2004. Rockville, MD: Health and Human Resources Administration, US Department of Health and Human Services; 2004.