Special Article
Undertreatment of Glaucoma among Black Americans
N Engl J Med 1991; 325:1418-1422November 14, 1991
- Abstract
- Abstract
Background
Cross-sectional studies and those using national data sets estimate that glaucoma-related blindness is between six and eight times more common among black Americans than among whites. Community-based studies have found that glaucoma is four to six times more prevalent among blacks. It is not known why blacks with glaucoma are more likely to become blind than whites with glaucoma.
Methods
To investigate the possibility that under-treatment of glaucoma is an important factor contributing to this higher rate of blindness, we studied the population-based rates of incisional and laser surgery for open-angle glaucoma among blacks and whites in a 5 percent random sample of Medicare claims for 1986 through 1988.
Results
For all U.S. census divisions combined, the rate of surgery for glaucoma among black Medicare beneficiaries was 2.2 times higher than the rate among white beneficiaries (95 percent confidence interval, 2.1 to 2.3). We calculated an expected rate of treatment among blacks on the basis of the rate of treatment among whites and the assumption that glaucoma is four times more prevalent among blacks — a conservative estimate. The observed rate of glaucoma surgery among blacks was 45 percent lower than the expected rate we calculated, which may in part account for the excess rate of blindness among blacks. The magnitude of this difference in treatment rates varied from 29 percent in the Middle Atlantic states to 50 percent in the South Atlantic states.
Conclusions
Older black Americans are not receiving potentially sight-saving care for open-angle glaucoma at the same rate as older white Americans. (N Engl J Med 1991;325:1418–22.)
Media in This Article
Figure 1
Difference between Expected and Observed Rates of Incisional and Laser Surgery for Glaucoma among Black Medicare Beneficiaries 65 or Older.Table 1
Codes Used to Identify Patients with Glaucoma.Article Activity
- Article
GLAUCOMA is the second leading cause of irreversible blindness in the United States and the most common cause among black Americans.1 2 3 It has long been noted that black patients are overrepresented among those attending glaucoma clinics.2 Data from the 16-state Model Reporting Area indicate that blindness due to glaucoma is eight times more common among blacks,1 and the Baltimore Eye Survey found that the prevalence of blindness due to glaucoma was six times higher among blacks.3 The age-adjusted prevalence of open-angle glaucoma among black residents of Baltimore who are of Medicare age is 7.84 percent, as compared with 1.96 percent among their white neighbors (relative prevalence, 4.0; 95 percent confidence interval, 2.87 to 5.62).4 Mason and coworkers have demonstrated a 12.5 percent prevalence of glaucoma among adult black residents of St. Lucia, West Indies.5 Pilot data from the Barbados Eye Study indicate a 13 percent rate of prevalence among residents over the age of 54.6
To investigate the possibility that the higher rate of blindness due to glaucoma among black Americans might be due in part to undertreatment, we studied the rates of incisional and laser surgery for glaucoma among black and white Medicare beneficiaries. Specifically, we attempted to determine whether the increased prevalence of glaucoma among black Medicare beneficiaries is associated with a corresponding increase in population-based rates of treatment for the disease.
Methods
Study Sample
A 5 percent random sample of Medicare beneficiaries 65 years old or older was selected on the basis of the last two digits of their social security numbers.7 All those with hospital inpatient, hospital outpatient, ambulatory surgical center, or physicians' records indicating that laser, incisional, or cyclodestructive surgery for open-angle glaucoma had been performed during 1986, 1987, or 1988 were potentially eligible for inclusion in the study. Participants in Medicare health maintenance organizations (HMOs) were excluded because complete treatment data were not available.
Identification of Patients with Glaucoma
Because Medicare does not cover outpatient prescription medications, we limited our study to incisional, laser, and cyclodestructive surgery for the treatment of open-angle glaucoma. Episodes of surgery for glaucoma were identified primarily on the basis of International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) codes8 found in hospital inpatient files and the Health Care Financing Administration Common Procedure Coding System (HCPCS) codes9 found in ambulatory surgical center and physicians' records (Table 1Table 1
Codes Used to Identify Patients with Glaucoma.). The HCPCS contains all existing Current Procedural Terminology (CPT) codes,10 along with additional codes that have been added for detail or to resolve ambiguity in physicians' billing. We used both ICD-9-CM and HCPCS codes to identify cases of glaucoma treatment in hospital outpatient files, since the coding of outpatient visits changed from ICD-9-CM to HCPCS in July 1987.Beneficiaries were considered to be patients with glaucoma if an episode of glaucoma treatment was identified on the basis of paid claims. Claims denied by the carrier were excluded, as we could not assume that the record contained valid information. A demographic file drawn from Medicare enrollment records for the same 5 percent sample, with information on age, sex, race, state and county of residence, and enrollment status (Part A, Part B, and HMO participation), was used to determine the personal characteristics of the patients with glaucoma and to provide denominators for race and region. Because their numbers were small among the patients and in the denominators, beneficiaries classified as "other nonwhite" rather than "white" or "black" (2.1 percent of the patients and 3.1 percent of the denominators), as well as those whose race was unspecified (1.6 percent of the patients and 1.7 percent of the denominators), were excluded from further analysis.
Statistical Analysis
Age-, sex-, and race-specific rates of glaucoma treatment per 1000 person-years of enrollment were calculated for the entire United States. Race-specific rates were calculated for each U.S. census division and were directly adjusted for age and sex to the 1986–1988 non-HMO Medicare population of beneficiaries 65 years of age or older. We calculated relative risks and 95 percent confidence intervals according to the method of Greenland and Robins,11 using the EPI-INFO software program from the Centers for Disease Control.
We used observed rates of treatment among white and black beneficiaries according to census division and for the entire United States to calculate the difference in rates of treatment. The expected rate of treatment among black beneficiaries was assumed to be the observed rate among white beneficiaries multiplied by the four times (95 percent confidence interval, 2.87 to 5.62) greater prevalence of glaucoma found among blacks in Baltimore. The difference in rates of treatment was the expected rate of treatment among black beneficiaries minus the observed rate of treatment. Since the standard errors of observed rates among black and white beneficiaries were 2x×l0–4 and 4×l0–5, respectively, they introduced negligible error into the calculation of the difference in the rates of treatment. The 95 percent confidence interval for the relative prevalence of glaucoma among blacks and whites was therefore used to estimate the confidence interval for the difference in rates of treatment.
Results
A total of 9481 white beneficiaries and 1821 black beneficiaries were identified as patients with glaucoma during the three-year period from 1986 through 1988. Unadjusted rates of surgery for glaucoma according to age and race are shown in Table 2Table 2
Rates of Surgical Treatment for Open-Angle Glaucoma, According to Age, among White and Black Medicare Beneficiaries, 1986–1988.*. Among the three age groups presented, the rate of surgery was highest among persons 75 to 84 years old (P<0.001). For all age groups, the rate among blacks was higher than among whites, although the relative rate decreased with age. Overall, the rate among blacks was 2.2 times that among whites (P<0.001). The rate among women was higher than that among men (3.7 vs. 2.7 per 1000 person-years of enrollment, P<0.001; data not shown).The age- and sex-adjusted rate of glaucoma treatment was significantly higher among blacks than among whites in all but the Mountain census division (Table 3Table 3
Rates of Incisional and Laser Treatment for Open-Angle Glaucoma, According to Census Division, among White and Black Medicare Beneficiaries, 1986–1988.*), where the sample size was inadequate. For the other eight divisions, the rate of glaucoma surgery was between 2.0 and 2.9 times higher among black beneficiaries than among white beneficiaries. A similar relation between rates of treatment was also observed when laser and incisional surgery were considered separately (data not shown). Black patients were slightly more likely than white patients (45.8 percent vs. 40.3 percent, P<0.001) to have more than one episode of care during the period of observation.We calculated the difference between the observed rate of treatment among black Medicare beneficiaries and the expected rate if black beneficiaries received care for glaucoma at the same rate as white beneficiaries, as adjusted for the fourfold higher prevalence of disease found in blacks in the Baltimore Eye Survey (Fig. 1Figure 1
Difference between Expected and Observed Rates of Incisional and Laser Surgery for Glaucoma among Black Medicare Beneficiaries 65 or Older.). The fourfold difference was chosen because it was the lowest and most conservative estimate of the difference in prevalence that was based on a major cross-sectional study of glaucoma in a racially heterogeneous population. The Baltimore Eye Survey used the strictest possible definition of glaucoma in order to yield the most conservative estimate of prevalence, and comparisons between that survey and other national data sets lend credibility to its results.The analysis suggested a difference of 4.85 per 1000 person-years between the expected rate of treatment among black beneficiaries (10.72 per 1000) and the observed rate of treatment (5.87 per 1000), which corresponds to a rate 45 percent lower than the expected rate, given the prevalence of the underlying disease (Fig. 1). On the basis of the 95 percent confidence interval for the relative prevalence of glaucoma (2.87 to 5.62), the difference between the expected and observed rates of treatment among blacks could range from 24 to 61 percent. The difference in rates of treatment varied from a low of 29 percent in the Middle Atlantic states to a high of 50 percent in the South Atlantic states.
Discussion
The most important finding of our study was that blacks, who are at least four times more likely to have glaucoma than whites, are only twice as likely to receive surgical treatment for it. The data suggest a difference in the rate of surgical treatment for glaucoma according to race, with lower-thanexpected rates of treatment among black beneficiaries in all areas of the country.
Unexplained racial variation in the treatment of other medical conditions has been noted.12 13 14 Recently, the Health Care Financing Administration published unadjusted rates of hospitalization during 1986 for 14 surgical procedures among elderly non-HMO beneficiaries of Medicare, according to race.15 In all instances, the rate of hospitalization was lower for blacks than for whites.
In considering the difference in rates of treatment, several hypotheses other than undertreatment of blacks must be considered. There may be an appropriate rate of treatment among blacks and considerable overtreatment among whites. In the Baltimore Eye Survey, 50 percent of the white respondents with diagnosed glaucoma were unaware of their condition. Glaucoma is widely thought to be underdiagnosed and undertreated in all races. The rates of treatment among blacks may also be lower than expected because disease progression is slower and more benign or because blacks more often receive surgical treatment before entering the Medicare program. The Baltimore Eye Survey and Model Reporting Area data, however, suggest otherwise, since the reported rates of blindness due to glaucoma among blacks were, respectively, six and eight times higher than the rates among whites.1 , 3
There may be greater uncertainty among providers about appropriate treatment for glaucoma in blacks. Although reports suggest that incisional surgery may be more prone to late failure in black patients with glaucoma,16 , 17 laser surgery is reported to be equally effective in black and white patients, at least for the first two years.18 , 19 If uncertainty among providers is a factor, one would expect a larger difference to be associated with incisional surgery than with laser surgery. In fact, black beneficiaries underwent 2.3 times more incisional surgery, as compared with 2.1 times more laser surgery.
Economic and social factors may contribute to the observed racial differences in treatment relative to the prevalence of disease. Additional research is needed to study the possible effects of factors such as differences in public awareness of glaucoma and its treatment, differences in the willingness to seek or trust in the care of physicians, differences in the proportion with a preference for an optometrist rather than an ophthalmologist as a primary provider of eye care, differences in the ability to pay for the ancillary costs of care (travel, prescription medications, and insurance copayments), and differences in the accessibility of eye care providers.
Not only can patients follow different patterns in seeking care and following recommended treatment plans, but physicians may also alter their practice patterns overtly or subconsciously on the basis of the patient's racial and economic characteristics.20 Wenneker and coworkers have demonstrated differences in patterns of care for ischemic heart disease, based on the source of payment for medical care.21 Although Medicare beneficiaries receive equal reimbursement for equivalent care, the ability to meet copayments and the availability of supplemental insurance may have an effect on physicians' behavior.
The chief strength of this analysis is its ability to derive and compare rates of treatment for glaucoma from a representative sample of elderly black and white Americans. Although any summary data set has inherent inaccuracies, Hsia and coworkers have demonstrated an error rate of 11 percent in the coding of the principal diagnosis in the Part A data base.22 The coding of procedures, especially major ophthalmologic procedures performed by subspecialists, is reported to be 96 percent accurate.23 Even an 11 percent error would not alter our conclusions.
There are three primary limitations to our study. First, our estimate of the magnitude of the difference in rates of treatment is based on the Baltimore Eye Survey finding that glaucoma was four times (95 percent confidence interval, 2.87 to 5.62) more prevalent among blacks than among whites of Medicare age. The Baltimore Eye Survey used a rigorous definition of glaucoma and found a lower prevalence of the disease among blacks than did the Barbados6 and St. Lucia5 studies. The prevalence of glaucoma among white participants in the Baltimore Eye Survey was comparable to that found in the Framingham Eye Study,24 the National Health and Nutrition Examination Survey,25 and the Model Reporting Area Study26 (which contain insufficient numbers of black participants for a reliable comparison). If the relative prevalence of glaucoma among blacks is higher than that predicted by the Baltimore Eye Survey, the difference in rates of treatment would be greater than we have reported.
Second, the race variable in the Medicare files is drawn from self-reported race at the time of application for a social security number. The U.S. population has considerable racial admixture, and beneficiaries may have classified themselves as black or white on the basis of cultural, familial, and other factors. This admixture probably explains some of the difference between the prevalence of glaucoma in Baltimore, which is typical of many American cities, and that in St. Lucia and Barbados, where there has been minimal racial admixture.
Third, we cannot identify beneficiaries who were treated for glaucoma at a Veterans Affairs (VA) hospital or a Medicare HMO. It is currently estimated that 3 to 4 percent of beneficiaries undergo surgery in VA hospitals.27 During our study years (1986 to 1988), 5.6 percent of all person-years of enrollment were in Medicare HMOs. If black Americans seek care in VA or HMO facilities more frequently than white Americans, we have overestimated the difference in rates of treatment relative to the prevalence of disease.
In addition to the racial variation in rates, there is considerable variation according to region (Fig. 1). Because of the regional variation in rates of treatment among whites, there is a resulting variation according to region in the expected rate of treatment among blacks. This is not meant to suggest that the prevalence of glaucoma among blacks or whites necessarily varies from region to region. Regional variation in rates of surgical procedures has been reported by a number of investigators, but a thorough analysis of regional variation in rates of treatment is beyond the scope of this report.28 , 29
The early detection and treatment of glaucoma are thought to prevent or postpone blindness. Avoidable blindness imposes not only devastating personal burdens on the individual patient, family, and community, but also financial burdens on society. We and others have demonstrated the cost effectiveness of some approaches to the prevention of vision loss.30 31 32 33
Our analysis suggests that across the United States there may be substantial undertreatment of glaucoma among black Medicare beneficiaries. If the lower-than-expected rate of glaucoma treatment actually represents an inadequate provision of sight-saving care to black Medicare beneficiaries, it is of major concern for public health. Our findings strongly underscore the need for additional research into the underlying causes of racial variation in the delivery of care for glaucoma and the resulting outcomes of that care. The root causes of this problem are probably complex and must be carefully delineated if an appropriate public health intervention is to be implemented.
Supported in part by grants from the National Eye Institute, the Division of Biological Resources (S10 RR0 6758), National Institutes of Health, and Research to Prevent Blindness, Inc. (R01 EY08805 and R21 EY07744).
The statements contained in this paper are solely those of the authors and do not necessarily reflect the views or policies of the Health Care Financing Administration.
Source Information
From the Worthen Center for Eye Care Research, the Center for Sight, and the Program on Technology Assessment, Department of Community and Family Medicine, Georgetown University, Washington, D.C. (J.C.J.); and the Office of Research (A.M.M., G.A.N., J.D.B., J.L.W.) and the Office of Program Assessment, Health Standards Quality Bureau (H.K.), both in the Health Care Financing Administration, Baltimore. Address reprint requests to Dr. Javitt at the Center for Sight, 3800 Reservoir Rd., NW, Washington, DC 20007.
- References
References
1
Hiller
Web of Science | Medline2
Martin
Web of Science | Medline3
Tielsch
Web of Science | Medline4
Tielsch
CrossRef | Web of Science | Medline5
Mason
Web of Science | Medline6
Leske
CrossRef | Web of Science | Medline7
Lave
Medline8
International classification of diseases, 9th revision, clinical modification: ICD-9-CM. Ann Arbor, Mich.: Commission on Professional and Hospital Activities, 1986.
9
Health Care Financing Administration. HCFA common procedure classification system. Available from Bureau of Data Management and Strategy, HCFA, 6325 Security Blvd., Baltimore, MD 21207.
10
American Medical Association. Physicians' current procedural terminology. 4th ed. (CPT-4). Chicago: American Medical Association, 1977.
11
Greenland
CrossRef | Web of Science | Medline12
Ford
CrossRef | Web of Science | Medline13
Strogatz
CrossRef | Web of Science | Medline14
Mayer
CrossRef | Web of Science | Medline15
Health Care Financing Administration. Health care financing special report: hospital data by geographic area for aged Medicare beneficiaries: selected procedures, 1986. Vol. 2. Baltimore: Health Care Financing Administration, 1990. (HCFA publication no. 03301.)
16
Welsh
CrossRef | Web of Science | Medline17
Miller
Medline18
Krapin
Web of Science | Medline19
Schwartz
Web of Science | Medline20
Eisenberg
Web of Science | Medline21
Wenneker
CrossRef | Web of Science | Medline22
Hsia
Full Text | Web of Science | Medline23
Fisher 24
Podgor
Web of Science | Medline25
Roberts J, Ludford J. Monocular visual acuity of persons 4–74 years, United States, 1971–1972. Vital and health statistics. Series 11. No. 201. Washington, D.C.: Government Printing Office, 1977. (DHEW publication no. (HRA) 77–1646.)
26
Kahn HA, Moorhead HB. Statistics on blindness in the model reporting area 1969–1970. Washington, D.C.: National Eye Institute, 1973. (DHEW publication no. (NIH) 73–427.)
27
Roos
CrossRef | Web of Science | Medline28
Wennberg
CrossRef | Web of Science | Medline29
Chassin
Full Text | Web of Science | Medline30
Javitt
Web of Science | Medline31
Javitt
Web of Science | Medline32
Javitt
Web of Science | Medline33
Dasbach
CrossRef | Web of Science | Medline
- Citing Articles (30)
Citing Articles
1
Rohit Varma, Lie-Ju Hwang, John W. Grunden, Gerald W. Bean, Marla B. Sultan. (2009) Assessing the Efficacy of Latanoprost vs Timolol Using an Alternate Efficacy Parameter: The Intervisit Intraocular Pressure Range. American Journal of Ophthalmology 148:2, 221-226
CrossRef2
Robert L Stamper, Marc F Lieberman, Michael V Drake. 2009. Genetics of glaucoma. , 330-338.
CrossRef3
David S. Friedman, Steven R. Hahn, Laurie Gelb, Jason Tan, Sonali N. Shah, Elizabeth E. Kim, Thom J. Zimmerman, Harry A. Quigley. (2008) Doctor–Patient Communication, Health-Related Beliefs, and Adherence in Glaucoma. Ophthalmology 115:8, 1320-1327.e3
CrossRef4
Rohit Varma, Lie-Ju Hwang, John W. Grunden, Gerald W. Bean. (2008) Inter-visit Intraocular Pressure Range: An Alternative Parameter for Assessing Intraocular Pressure Control in Clinical Trials. American Journal of Ophthalmology 145:2, 336-342
CrossRef5
Anne L. Coleman, Fei Yu. (2008) Eye-Related Medicare Costs for Patients with Age-Related Macular Degeneration from 1995 to 1999. Ophthalmology 115:1, 18-25.e1
CrossRef6
Anne L. Coleman, Fei Yu, Stacy J. Evans. (2006) Use of Gonioscopy in Medicare Beneficiaries Before Glaucoma Surgery. Journal of Glaucoma 15:6, 486-493
CrossRef7
Robert J. Noecker, Melissa L. Earl, Thomas K. Mundorf, Steven M. Silverstein, Michael P. Phillips. (2006) Comparing bimatoprost and travoprost in black Americans. Current Medical Research and Opinion 22:11, 2175-2180
CrossRef8
Mark R. Cullen, Sally Vegso, Linda Cantley, Deron Galusha, Peter Rabinowitz, Oyebode Taiwo, Martha Fiellin, David Wennberg, Joanne Iennaco, Martin D. Slade, Kanta Sircar. (2006) Use of Medical Insurance Claims Data for Occupational Health Research. Journal of Occupational and Environmental Medicine 48:10, 1054-1061
CrossRef9
David S. Friedman, Beth Nordstrom, Essy Mozaffari, Harry A. Quigley. (2005) Variations in Treatment among Adult-Onset Open-Angle Glaucoma Patients. Ophthalmology 112:9, 1494-1499
CrossRef10
Kevin C. Greenidge. (2005) Race, politics, ethnicity, and science. American Journal of Ophthalmology 139:4, 704-706
CrossRef11
Anne L. Coleman, Fei Yu, Susannah Rowe. (2005) Visual field testing in glaucoma Medicare beneficiaries before surgery. Ophthalmology 112:3, 401-406
CrossRef12
JOSÉ J. ESCARCE, KANIKA KAPUR. (2003) Racial and Ethnic Differences in Public and Private Medical Care Expenditures among Aged Medicare Beneficiaries. Milbank Quarterly 81:2, 249-275
CrossRef13
Lyne Racette, M.Roy Wilson, Linda M Zangwill, Robert N Weinreb, Pamela A Sample. (2003) Primary Open-Angle Glaucoma in Blacks: A Review. Survey of Ophthalmology 48:3, 295-313
CrossRef14
Robert J. Noecker, Melissa L. Earl, Tom Mundorf, James Peace, Robert D. Williams. (2003) Bimatoprost 0.03% versus travoprost 0.004% in Black Americans with glaucoma or ocular hypertension. Advances in Therapy 20:2, 121-128
CrossRef15
Harry A Quigley, Chan Kee Park, Patricia A Tracey, Irvin P Pollack. (2002) Community screening for eye disease by laypersons: the Hoffberger program. American Journal of Ophthalmology 133:3, 386-392
CrossRef16
Beth A Virnig, Marshall McBean. (2001) A
DMINISTRATIVE DATA FOR PUBLIC HEALTH SURVEILLANCE AND PLANNING . Annual Review of Public Health 22:1, 213-230
CrossRef17
David A. Ganz, Robert J. Glynn, Helen Mogun, Eric L. Knight, Rhonda L. Bohn, Jerry Avorn. (2000) Adherence to Guidelines for Oral Anticoagulation after Venous Thrombosis and Pulmonary Embolism. Journal of General Internal Medicine 15:11, 776-781
CrossRef18
Debra A. Schaumberg, William G. Christen, Robert J. Glynn, Julie E. Buring. (2000) Demographic Predictors of Eye Care Utilization Among Women. Medical Care 38:6, 638-646
CrossRef19
S. I. White-Means. (2000) Racial Patterns in Disabled Elderly Persons' Use of Medical Services. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences 55:2, S76-S89
CrossRef20
David C. Aron, Howard S. Gordon, David L. DiGiuseppe, Dwain L. Harper, Gary E. Rosenthal. (2000) Variations in Risk-Adjusted Cesarean Delivery Rates According to Race and Health Insurance. Medical Care 38:1, 35-44
CrossRef21
Omofolasade Kosoko, Harry A Quigley, Susan Vitale, Cheryl Enger, Lisa Kerrigan, James M Tielsch. (1998) Risk factors for noncompliance with glaucoma follow-up visits in a residents’ eye clinic. Ophthalmology 105:11, 2105-2111
CrossRef22
Fang Wang, Jonathan C. Javitt, James M. Tielsch. (1997) Racial variations in treatment for glaucoma and cataract among Medicare recipients. Ophthalmic Epidemiology 4:2, 89-100
CrossRef23
Barbara Crawley, Roberta Scherer, Patricia Langenberg, Kay Dickersin, The IONDT Study Group. (1997) Participation in the Ischemic Optic Neuropathy Decompression Trial: sex, race, and age. Ophthalmic Epidemiology 4:3, 157-173
CrossRef24
Cornelia H. M. vanIneveld, Deborah J. Cook, Sheri-Lynn C. Kane, Derek King, . (1996) Discrimination and abuse in internal medicine residency. Journal of General Internal Medicine 11:7, 401-405
CrossRef25
K. Robin Yabroff, Benjamin P. Linas, Kevin Schulman. (1996) Evaluation of quality of life for diverse patient populations. Breast Cancer Research and Treatment 40:1, 87-104
CrossRef26
Jonathan C. Javitt. (1995) Preventing blindness in americans: the need for eyehealth education. Survey of Ophthalmology 40:1, 41-44
CrossRef27
Jonathan C. Javitt, Hugh R. Taylor. (1995) Cataract and latitude. Documenta Ophthalmologica 88:3-4, 307-325
CrossRef28
A. Marshall McBean, Joan L. Warren, J. Daniel Babish. (1994) Measuring the incidence of cancer in elderly americans using medicare claims data. Cancer 73:9, 2417-2425
CrossRef29
Quigley, Harry A.. (1993) Open-Angle Glaucoma. New England Journal of Medicine 328:15, 1097-1106
Full Text30
Seddon, Johanna M., . (1991) The Differential Burden of Blindness in the United States. New England Journal of Medicine 325:20, 1440-1442
Full Text
