Prevalence and Risk Factors for Refractive Errors in an Adult Inner City Population

Transcription

1 revalence and Risk Factors for Refractive Errors in an Adult Inner City opulation Joanne Katz,* James M. Tielsch* and Alfred Sommev\ urpose. To estimate the prevalence of refractive errors among adult black and white Americans and to identify risk factors associated with these refractive errors. Methods. Refractive error was measured in a population-based sample of black and white adults age 40 or older residing in east Baltimore from 1985 through Aphakic eyes were excluded from analysis. Results. The prevalence of myopia varied from 10.5% among black men 80 years and older to 42.1% among white women 40 to 49 years of age. Hyperopia ranged from 11.8% among black men 40 to 49 years to 68.1% among white men 80 years of age and older. Astigmatism ranged from 15.8% to 45.2%, and anisometropia ranged from 2.8% to 8.1%, depending on age, race, and gender. Black persons had less myopia, hyperopia, astigmatism, and anisometropia than did white persons. Myopia (< 0.5 diopter [D] spherical equivalent) declined with age, whereas hyperopia (> +0.5 D), astigmatism (>0.5 D of cylinder), and anisometropia (> 1.0 D between eyes) increased with age. Myopia increased with increasing years of education, although this association was stronger for white persons than for black persons and among younger subjects. Hyperopia declined with increasing years of education, and this association was stronger among younger than older subjects. Education was not associated with astigmatism or anisometropia. Conclusions. Black persons had lower rates of refractive error than did white persons, except for hyperopia prevalence, which was comparable in black and white women. Refractive errors are common among adult inner city Americans, but rates vary substantially by age, race, gender, and education levels. Invest Ophthalmol Vis Sci. 1997;38: Although most refractive errors can be corrected through the use of spectacles or contact lenses, the economic burden of correcting these refractive errors is enormous. It has been estimated that correction of refractive errors through the prescription of spectacles and contact lenses is 2 billion dollars per year. 1 Spectacle use can result in discomfort and less than optimal visual outcomes, especially in those with severe refractive errors, whereas contact lens use increases the risk of ulcerative keratitis. 2 ' 3 Options for refractive surgery, not only to correct myopia, but also astigmatism and hyperopia, are increasing rapidly. However, there are From the *Department of International Health, the Johns Hopkins School of Hygiene and ublic Health, Baltimore, Maryland, and the f Dean's Office, the Johns Hopkins School of Hygiene and ublic Health, Baltimore, Maryland. Supported by grants EY-05091, and RR from the National Institutes of Health, Bethesda, Maryland. Submitted for publication July 31, 1996; revised October 7, 1996; accepted October 10, roprietary interest category: N. Reprint requests: Joanne Katz, Johns Hopkins School of Hygiene and ublic Health, 615 N. Wolfe Street, Room 5515, Baltimore, MD few good estimates of the prevalence and severity of refractive errors on which to base treatment demands. Knowing the severity of refractive errors is important, in that better excimer laser results have been obtained among those with mild rather than moderate-to-severe refractive errors, 4 and high myopia has been associated with myopic macular degeneration, cataract, glaucoma, peripheral retinal changes, retinal detachments, and blindness. 5 Although refractive errors are thought to affect a large portion of the U.S. population, there are few population-based studies that document the prevalence of different types of refractive errors in older subjects and describe demographic and other risk factors for these conditions. Some results are available from data collected during the 1970s, 6 " 9 whereas the more recent Beaver Dam study population was almost exclusively white. 10 We report here the distribution and types of refractive errors in a multiracial inner city population of subjects 40 years of age and older. 334 Investigative Ophthalmology & Visual Science, February 1997, Vol. 38, No. 2 Copyright Association for Research in Vision and Ophthalmology

2 Refractive Errors 335 METHODS The Baltimore Eye Survey was a population-based survey of ocular disorders among noninstitutionalized subjects 40 years of age and older living in east Baltimore. This survey was conducted from January 1985 through November The methods for selection of the population have been described previously. 1 '~ ld Briefly, a stratified cluster sample of 16 clusters was selected for participation in the study. All noninstitutionalized residents 40 years of age and older in selected clusters were eligible to participate. Clusters were stratified to ensure an approximately equal sample of black and white adults. A short interview (response rate, 95.8%) was conducted at the household of each eligible participant. This included demographic information such as age, race, and gender, as well as years of education completed. Those identified as eligible (residents 40 years of age or older) were invited to a neighborhood screening site, where an ophthalmic screening examination was conducted that included refraction, visual acuity, other ocular measures, and a personal interview, including a history of prior ocular surgery. The response to screening was 79.2%. At the screening site, all subjects underwent subjective refraction with the AO Reichert SR-TV rogrammed Subjective Refractor (Reichert Instruments, Charlotte, NC). Refractions were refined as needed for possible overcorrection to eliminate the effect of instrument accommodation. Those who could not respond adequately to automated refraction were refracted using retinoscopy or with manual techniques. Subjects did not undergo cycloplegia. Subjects who had a history of cataract surgery or who were found to be aphakic or pseudophakic on examination were excluded from the analysis of refractive errors. Because the correlation of refractive error between fellow eyes was very high (0.86), the refractive error of the right eye only was used to construct the distribution of refractive errors in the population. The right eye only was used to estimate the prevalence of myopia, hyperopia, and astigmatism. Myopia was defined as < 0.5 diopter (D) spherical equivalent. Hyperopia was defined as >+0.5 D spherical equivalent. Astigmatism was defined as >0.5 D of cylinder. Anisometropia was denned as a difference of >1.0 D of spherical equivalence between right and left eyes. Logistic regression was used to identify risk factors associated with myopia, hyperopia, astigmatism, and anisometropia. All models included age, race, gender, and education; interactions were included only if they were substantial in size and statistically significant at the 5% level. The tenets of the Declaration of Helsinki were observed in this study. Ethical approval was obtained from the Joint Committee on Clinical Investigations Spherical Equivalent (diopters) FIGURE l. Unadjusted distribution of refractive error (spherical equivalence in right eyes) among black and white adults in the Baltimore Eye Survey, 1985 through of the Johns Hopkins University School of Medicine. Written informed consent was obtained from all participants. RESULTS A total of 5308 black and white adults were examined at the neighborhood screening site. There were 272 participants who had undergone cataract surgery and were excluded from this analysis. Among the remaining 5036 subjects who were phakic, data on refractive errors in the right eye were unavailable for 177 subject. An additional eight subjects had refractive error measurements in the right eye but not in the left eye. Hence, estimates of the prevalence of anisometropia were based on 5028 subjects. The distribution of refractive error for white persons and black persons is shown in Figure 1. Although the mean refractive error was similar, black persons had less negative and positive deviations from emmetropia than did white persons. The distribution of spherical equivalent by age and race is given in Table 1. The rates of moderate ( 3.0 to 6.0 D) and high myopia (< 6.0 D) and of hyperopia of >2.0 D were greater for white persons than for black persons at all ages. The mean refractive error became more positive with increasing age for both black and white adults. Cylindrical error was expressed using the median and the interquartile range rather than the means and standard deviations because of the skewed nature of the cylinder distributions. The mean difference in spherical equivalence between eyes was small regardless of age or race, with age- and race-specific means ranging in absolute value from 0.12 to 0.01 D. The prevalence of myopia (< 0.5 D) declined with age among men and women, black and white (Table 2). The prevalence of hyperopia (>+0.5 D)

3 336 Investigative Ophthalmology & Visual Science, February 1997, Vol. 38, No. 2 TABLE l. Distribution of Refractive Errors by Age and Race (%) (N= 613) (N= 666) (N = 568) (N= 287) (N= 66) Black ersons Spherical equivalents <-6.0D <-3.0 D to -6.0 D <-0.5 D to -3.0 D >0.5 D to =s0.5 D >0.5 D to 2.0 D >2.0D Mean refractive error (SD) Median cylinder (IQR) Mean difference (SD) between eyes (1.48) -8.75, (0.50) 0.00, (0.72) -6.63, (1.81) , (0.50) 0.00, (0.91) -9.00, (1.86) , (0.75) 0.00, (0.63) -4.50, (2.31) , (0.75) 0.00, (1.38) -9.63, (2.07) -5.50, (0.75) 0.00, (1.17) -4.75, 3.88 (N = 531) (N = 598) (N= 852) (N = 538) (N= 140) White ersons Spherical equivalents <-6.0 D <-3.0 D to -6.0 D <-0.5 D to -3.0 D s0.5 D to ^0.5 D >0.5 D to 2.0 D >2.0D Mean refractive error (SD) Median cylinder (IQR) Mean difference (SD) between eyes (2.26) , (0.50) 0.00, (0.85) -5.25, (2.06) -9.13, (0.50) 0.00, (0.96) -4.88, (2.50) , (0.75) 0.00, (1.30) , (2.69) , (1.00) 0.00, (1.64) , (2.69) , (1.25) 0.00, (1.92) , 7.00 increased with age among men and women, white persons and black persons. The prevalence of hyperopia was higher for white men than for black men, but the rates were similar by race for women of similar age. The prevalence of astigmatism also increased with age, and rates of astigmatism were higher for white persons than for black persons at all ages. revalence rates of anisometropia (>1.0 D of spherical equivalence between eyes) increased with age for white persons and black persons, men and women. The prevalence was slightly higher for white persons than for black persons. The prevalence of myopia among black persons increased with years of education until the completion of high school but leveled off thereafter. Similarly, the decline in hyperopia was not seen for those with education beyond high school (Fig. 2A). However, small sample sizes in the highest education level precluded accurate estimation of prevalence for black persons. Hence, all education beyond high school has been combined in Figure 2A. Among white persons, the prevalence of myopia increased and the prevalence of hyperopia decreased with higher levels of education (Fig. 2B). Astigmatism declined slightly with increasing years of education among black persons but not among white persons. The prevalence of anisometropia was not related to education levels among black persons or white persons. Results from the multivariate models are presented in Table 3. For black and white adults, the prevalence of myopia increased with years of education, and this relation was stronger among younger than older subjects (odds ratio = 0.995, = for an education-by-age interaction). Furthermore, the association between education and myopia prevalence was stronger for white persons than for black

4 Refractive Errors 337 TABLE 2. revalence and 95% Confidence Intervals for Myopia, Hyperopia, Astigmatism, and Anisometropia by Age, Race, and Gender Female Male Myopia Hyperopia Astigmatism Anisometropia Myopia Hyperopia Astigmatism Anisometwpia Blacks Whites , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , 25.1 CI = confidence interval. persons (odds ratio = 1.05, = 0.04 for an educationby-race interaction). Black men had half the rate of hyperopia than did women (odds ratio = 0.48, = ), but this gender difference was not present for white persons (odds ratio = 1.88, = for an interaction between race and gender). Hyperopia decreased with increasing education, and this relation was stronger among younger white persons and black persons. The association between education and hyperopia was similar for black and white adults as evidenced by a lack of interaction between education and race. Astigmatism increased slightly with age, white persons had more than did black persons, and men had more than did women. There was no association between education and astigmatism. There were no significant interactions in the prediction of astigmatism among age, race, gender, and education. Anisometropia also increased with age by 4% per year and was significantly higher (35%, = 0.01) among white persons than among black persons. However, the rates were similar for men and women and increased slightly (4% per year) but not statistically significantly with years of education. There were no interactions for anisometropia. DISCUSSION These data provide the first population-based comparison of refractive error among black and white Americans in 20 years. Data from the National Health and Nutrition Examination Survey (NHANES) from 1971 to 1972 provided rates of refractive error in a nationally representative sample of the U.S. population. 6 ' 7 In that survey, myopia was defined as any minus refractive error (spherical equivalent) based on measuring refractive error in those whose visual acuity was worse than 20/40 and lensometry among those whose visual acuity was 20/40 or better. Using this definition, the prevalence of myopia in the Baltimore Eye Survey was 48.1% among white persons 45 to 54 years of age compared with 25.5% among white NHANES participants of the same age. The prevalence of myopia among black persons in this same age group was 42.2% in the Baltimore Survey compared with 17.3% among black NHANES participants of the same age. Although published data were not available to permit adjustment for education, we were able to compare the prevalence of myopia by years of education among all NHANES participants 45 to 54 years of age. Among those with less than 5 years of education, 2.7% of NHANES participants, 21.4% of Baltimore black persons, and 41.7% of Baltimore white persons were myopic. Among those with more than a high school education, 39.5% of NHANES participants, 40.1% of Baltimore black persons, and 52.2% of Baltimore white persons were myopic. The Framingham Eye Survey was conducted between 1973 and 1975 in a predominantly white population. Refractive error was measured in those whose visual acuity was 20/30 or worse. 8 The refractive error

5 338 Investigative Ophthalmology & Visual Science, February 1997, Vol. 38, No. 2 Blacks 60 -, A Whites 70 -, Education 10-,._-«- 0-5 B Education - -Myopia a - Hyperopia Astigmatism - Anisometropia Myopia»- Hyperopia Astigmatism * - Anisometropia FIGURE 2. (A) revalence of myopia (<-0.5 diopter [D]), hyperopia (>+0.5 D), and astigmatism (>0.5 D of cylinder) in the right eye, and anisometropia (>1.0 D of difference between eyes) by level of education among black persons in the Baltimore Eye Survey, 1985 through (B) revalence of myopia (< 0.5 D), hyperopia (>+0.5 D), and astigmatism (>0.5 D of cylinder) in the right eye, and anisometropia (>1.0 D of difference between eyes) by level of education among white persons in the Baltimore Eye Survey, 1985 through of those whose visual acuities were better than 20/30 were obtained by lensometry. Those without glasses were considered emmetropic. Based on this definition, the prevalence of any myopia (any negative diopters' spherical equivalent) in the Framingham population was 17.5%. Using the same definition of myopia, the age- and gender-adjusted prevalence was 28.0% among Baltimore white persons and 27.6% among Baltimore black persons. The estimates of myopia prevalence from both NHANES and Framingham appear to be lower than the Baltimore rates among both white persons and black persons of similar age. This comparison between Framingham and Baltimore was not adjusted for education because refractive error by education level was not available in the Framingham monograph. However, the education level in the Baltimore sample was low, with 77.7% of black persons and 65.4% of white persons not having completed high school. Hence, an education-adjusted comparison of Baltimore and Framingham rates likely would have increased the observed differences in prevalence between these two studies. The more recent Beaver Dam Eye Survey conducted from 1988 through 1990 in a predominantly white population in southern Wisconsin refracted participants with an automated refractor. 10 They excluded subjects who were aphakic and anyone whose visual acuity in at least one eye was better than 20/40. Myopia was defined as a refractive error of > 0.5 D. Hyperopia was defined as refractive error greater than +0.5 D. Using these definitions and adjusting for age, gender, and educational level, we found the prevalence of myopia was 19.4% among Baltimore black persons and 28.1% among Baltimore white persons, compared with 26.2% among Beaver Dam participants. In Baltimore, the age-, gender-, and educationadjusted prevalence of hyperopia was 41.0% among black persons and 43.9% among white persons, compared with 49.0% among Beaver Dam participants. The Baltimore rates of myopia and hyperopia among white persons are comparable to those of Beaver Dam, but the prevalence of myopia among black persons was lower, even after adjustment for age, gender, and education. Based on the comparison with NHANES, Framingham, and Beaver Dam, it appears likely that myopia has increased from the early 1970s to the late 1980s among middle-aged and older Americans. The prevalence of myopia declined with age, whereas the prevalence of hyperopia, astigmatism, and anisometropia increased with age. Similar patterns were observed in NHANES, Framingham, and Beaver Dam, 6 " 810 as well as in studies that were not population based. 16 " 18 It is unclear whether the risk of these conditions changes as individuals grow older or whether these reflect a cohort effect. In Beaver Dam, a small gender difference was seen in the rates of myopia, but not hyperopia. There was no gender difference in the prevalence of myopia or hyperopia among white persons in Baltimore, but the prevalence of hyperopia was higher among black women than among black men after adjustment for age and education. The response to the screening examination was 79.2%. This rate was slightly higher among black persons than among white persons and lower among older than younger subjects. 11 Differences might be observed if the patterns of prevalence of refractive errors in respondents were different from those of nonrespondents. Based on the demographic data available on nonrespondents in this study, it is unclear whether older nonrespondents were more or less

6 Refractive Errors 339 TABLE 3. Results of Logistic Regression With Myopia, Hyperopia, Astigmatism, and Anisometropia as Outcomes, and Age, Race, Gender, and Year of Education as Covariates Myopia Hyperopia Astigmatism Anisometropia Covariate Race Gender Education (years) Education by age Education by race Race by gender (0.99, 1.03) 0.81 (0.50, 1.30) 1.06 (0.92, 1.22) 1.36 (1.20, 1.54) (0.993, 0.997) 1.05 (1.00, 1.10) (0.99, 1.03) (0.86, 1.18) 0.48 (0.40, 0.59) 0.67 (0.60, 0.76) (1.004, 1.008) 1.88 (1.45, 2.44) (, 1.45 (1.27, 1.20 (1.05, (0.99, 1.03) 1.66) 1.36) 1.03) (1.03, 1.35 (1.07, 1.19 (0.90, 1.04 (0.97, 1.05) 1.72) 1.40) 1.11) CI = confidence interval. likely to have refractive errors than respondents of a similar age. Hence, it is difficult to speculate whether the nonresponse might affect the prevalence and risk factor results described in this article. Education has been associated strongly with the risk of myopia in many studies. 6 ' 7 ' 1019 " 27 The association between years of education and myopia was stronger among younger than older subjects and was stronger among white persons than black persons. It is unclear why this might be the case. However, the level of education declined with age among black and white adults, and black persons had less education than did white persons at all ages. Hence, the smaller variation in education among older black subjects made it more difficult to detect an association between education and myopia in this group. The prevalence of hyperopia declined with increasing years of education among black and white adults. Hence, the distribution of refractive errors was narrower and weighted more heavily in the center among those with less education than those with more education. There are no data from population-based studies of older persons that have examined the relation between hyperopia and education. However, hyperopia was more common among 20- to 30-year-old Australian aborigines than among those of European descent who had a higher level of education. 28 Education was not associated with astigmatism, although myopia and astigmatism each have been associated with education in previous studies of younger populations. 28 ' 29 Refractive errors such as myopia, hyperopia, astigmatism, and anisometropia were common among older Americans in east Baltimore, but varied substantially with age, gender, race, and education levels. Near vision was not measured in this study, but difficulty with near vision tasks and reading, in particular, was reported to be the most common vision-specific problem among black and white adults in this population. 15 The prevalence of reading difficulties was highest among those in their 40s and 80s but lower among those between 50 and 80 years of age. This likely reflects the impact of the onset of presbyopia in the younger population and ocular disease in the older subjects. The data provided in this study give more recent estimates of refractive errors among black and white adults from a population-based survey that would be helpful in planning service delivery for older Americans. Key Words anisometropia, astigmatism, hyperopia, myopia, refractive error References 1. National Advisor)' Eye Council. Vision research. A national plan: Bethesda, MD: United States Department of Health and Human Services, National Institutes of Health. 2. Schein OD, Glynn RJ, oggio EC, Seddon JM, Kenyon KR, the Microbial Keratitis Study Group. The relative risk of ulcerative keratitis among users of daily-wear and extended-wear soft contact lenses: a case-control study. NEnglJMed. 1989;321: Schein OD, Buehler O, Stamler JF, Verdier DD, Katz J. The impact of overnight wear on the risk of contact lens-associated ulcerative keratitis. Arch Ophthalmol. 1994;112: Taylor HR, McCarty CA, Aldred GF, for the Melbourne Excimer Laser Group. redictability of excimer laser treatment of myopia. Arch Ophthalmol. 1996; 114:

7 340 Investigative Ophthalmology & Visual Science, February 1997, Vol. 38, No Cur tin B. The Myopias. Basic Science and Clinical Management. hiladelphia: Harper and Row, ExpEyeRes. 1974; 18: Brown N. The change in the lens curvature with age. 6. Sperduto RD, Seigel D, Roberts J, Rowland M. revalence of myopia in the United States. Arch Ophthalmol. emmetropizing mechanism for the adult eye? Am J 17. Grosvenor T. Reduction in axial length with age: an 1983; 101: Optomhysiol Opt. 1987; 64: Refractive Status and Motility Defects of ersons Hemenger R, Garner LF, Ooi CS. Change with age Years, United States, U.S. Department of of the refractive index gradient of the human ocular Health, Education, and Welfare ublication No. lens. Invest Ophthalmol Vis Sci. 1995;36: (HS) Hyattsville, MD: National Center for 19. arssinen O. The wearing of spectacles in different Health Statistics, social and educational groups in a sample of the population of central Finland. Scand J Soc Med. 1987; 8. Leibowitz HM, Krueger DE, Maunder LR, et al. The Framingham Eye Study Monograph: An ophthalmological and Epidemiological study of cataract, glau- 20. Rosner M, Belkin M. Intelligence, education and myo- 15: coma, diabetic retinopathy, macular degeneration, pia in males. Arch Ophthalmol. 1987; 105: and visual acuity in a general population of Teasdale TW, Goldschmidt E. Myopia and its relationship to education, intelligence and height. Ada Oph- adults. Surv Ophthalmol. 1980;24(suppl): Hyams SW, akotilo E, Shkurko G. revalence of refractive errors in adults over 40: A survey of 8,102 eyes. 22. Lin LLK, Chen CJ, Hung T, Ko LS. Nation-wide surthalmol. 1988;185(S): Brf Ophthalmol. 1977;61: vey of myopia among schoolchildren in Taiwan, Wang Q, Klein BEK, Klein R, Moss SE. Refractive status in die Beaver Dam Eye Study. Invest Ophthalmol Vis Ada Ophthalmol. 1988(S): Bear JC, Richler A, Burke G. Near work and familial Sci. 1994; 35: resemblances in ocular refraction: a population based study in Newfoundland. Clin Genetics. 1981; 19: Tielsch JM, Sommer A, Witt K, Katz J, Royall RM, Baltimore Eye Survey Research Group. Blindness and visual Wong L, Coggon D, Cruddas M, Hwang CH. Education, reading, and familial tendency as risk factors for impairment in an American urban population. The Baltimore Eye Survey. Arch Ophthalmol. 1990; 108: myopia in Hong Kong fishermen. /Epidemiol Community Health. 1993;47: Tielsch JM, Sommer A, Katz J, et al. Socioeconomic status and visual impairment among urban Americans. 25. Zylbermann R, Landau D, Berson D. The influence Arch Ophthalmol. 1991; 109: of study habits on myopia in Jewish teenagers. Jediatr 13. Sommer A, Tielsch JM, KatzJ, Quigley HA, etal. Racial Ophthalmol Strabismus. 1993; 30: differences in the cause specific prevalence of blindness in East Baltimore. NEnglfMed. 1991;325:1412- tion. A population study in Newfoundland. Ada Oph- 26. Richler A, Bear JC. Refraction, nearwork and educa thalmol. 1980; 58: Varma R, Tielsch JM, Hilton SC, et al. Race, gender 27. Angle J, Wissman DA. The epidemiology of myopia. and refractive error related differences in the normal Am J Epidemiol. 1980; 11: optic disc. Arch Ophthalmol. 1994; 112: Taylor HR. Racial variations in vision. Am J Epidemiol. 15. Katz J, Tielsch JM. Visual function and visual acuity 1981;113: in an adult urban population. Journal of Visual Impairment and Blindness. 1996;90: Ophthalmol. 1991; 29. arssinen O. Astigmatism and school myopia. Ada 69: