Changes in Visual Acuity in a Population The Beaver Dam Eye Study

Ronald Klein, MD, Barbara E. K. Klein, MD, Kristine E. Lee, MS

Purpose: To describe the change in visual acuity over a 5-year period in persons participating in a large population-based study.

Methods: Best-corrected visual acuity was measured, after refraction, with 10gMAR charts using a modification of the Early Treatment Diabetic Retinopathy Study protocol in 3684 persons living in Beaver Dam, Wisconsin, who ranged in age from 43 to 86 years at the time of a baseline examination from 1988 to 1990, and at a follow-up examination from 1993 to 1995.

Results: The change in the number of letters read correctly over the 5-year period varied from 0.4 4.9 (mean standard deviation) in people between 43 and 54 years of age to -5.2 15.4 in people 75 years of age or older at baseline. Over the 5-year period, vision became impaired (20/40 or worse in the better eye) in 2.9% of the pop-ulation and severely impaired (20/200 or worse in the better eye) in 0.3%. The visual angle doubled in 1.7% of the population, and 2.4% had improved vision. People 75 years of age or older at baseline were 12.5 times (95% confidence interval [CI], 8.6-18.2; P < 0.001) more likely to have impaired vision, 9.7 times (95% CI, 5.9-16.0; P < 0.001) more likely to have doubling of the visual angle, and 78 times more likely (95% CI, 9.9-614.1; P < 0.001) to have severe visual impairment than people younger than 75 years of age at baseline. People 75 years of age or older who were living in nursing homes or group homes were 3.8 times more likely to have impaired vision, 3.3 times more likely to have severely impaired viSion, and 5.7 times more likely to have a doubling of the visual angle than those not residing in a nursing home or a group home.

Conclusion: These data provide precise population-based estimates of incidence of visual loss over a wide spectrum of ages and show that decreased visual acuity in people 75 years of age is a common finding, especially in those who are in nursing homes or group homes. Ophthalmology 1996; 1 03: 1169-1178

Originally received: November 27. 1995. Revision accepted: May I. 1996. From the Department of Ophthalmology and Visual Sciences. University of Wisconsin Medical School. Madison. Supported by National Institutes of Health grant EY06594 (Drs. R Klein and BEK Klein). Bethesda. Maryland. and. in part. by the Research to Prevent Blindness (Dr. R Klein) Senior Scientific Investigator Award. New York. New York. Each author states that he/she has no proprietary interest in any aspect of this work. Reprint requests to Ronald Klein. MD. MPH. Department ofOphthal-mology and Visual Sciences. University of Wisconsin-Madison. 610 North Walnut St. 460 WARF. Madison. WI 53705-2397.

Recent population-based studies have provided data on the prevalence and severity ofloss of vision in the United States. I- 3 These data show a significant increase in the prevalence of impaired vision with increasing age, es-pecially in those 75 years of age or older. However, esti-mates of the incidence of loss of vision are few and are limited by the small size or the select nature of the groups studied and by the lack of objective data.4- 7 Reliable in-cidence rates ofloss of vision have important public health implications, including the ability to project needs for ser-vices and costs, defining etiologic relations, and assessing the effect of treatment. The purpose of this report is to

1169

Ophthalmology Volume 103, Number 8, August 1996

describe the change in visual acuity over a 5-year period in persons participating in the Beaver Dam Eye Study, a population-based study of older adults.

Methods

Population Methods used to identify the population and descriptions of the population have appeared in previous reports.38.9 In brief, a private census of the population of Beaver Dam, Wisconsin, was performed from September 15, 1987, to May 4, 1988. Eligibility requirements for entry into the study included living in the city or township of Beaver Dam and being 43 to 84 years of age at the time of the census. A total of 5924 individuals, both institutionalized and noninstitutionalized, were eligible for inclusion in the study. Of these individuals, 4926 participated in the examination phase between March 1, 1988, and Septem-ber 14, 1990. Ninety-nine percent of the population was white. Of those who did not participate, 227 (3.8%) had died before the examination, 99 (1.7%) had moved out of the area, 18 (0.3%) could not be located, 276 (4.7%) permitted an interview only, and 378 (6.4%) refused to participate. Comparisons between participants and non-participants at the time of the baseline examination have appeared elsewhere.3

The members of the cohort who were eligible to par-ticipate at baseline had been divided randomly into ten groups to be examined sequentially over the course of the study. Surviving members of the cohort were invited to participate in a 5-year follow-up examination in the same order as at baseline examination. Of the 4926 people who participated at the baseline examination, 385 (7.8%) died before March 1, 1993, the beginning of the 5-year follow-up examination. Of the 4541 surviving persons who had, participated in the baseline examination, 3684 (81.1 %) participated in the follow-up examination from March 1, 1993, through June 14, 1995. One hundred seventy-one (3.8%) died after the start of the follow-up but before ex-amination. Four could not be located. Two hundred fifty-nine (5.7%) permitted an interview only (48 of whom moved out of the area) and 423 (9.3%) refused to partic-ipate (44 of whom moved out of the area). Both the mean and median times between the baseline and 5-year follow-up examinations were 4.8 years (standard deviation, 0.4 years).

Of the 3684 people examined at baseline and at follow-up, 92 were living in a nursing home or a group home at follow-up. Of these people, 24 were in a nursing home or a group home at baseline. One person was in a nursing home at baseline but moved home by follow-up.

Comparisons between participants and nonparticipants at follow-up are presented in Table 1. The 686 nonpar-ticipants who were alive at follow-up were more likely to be older than the participants (Table 1). After controlling for age at baseline, these nonparticipants were.more likely at baseline to have retired and had fewer years of education completed, a history of never drinking alcohol, lower in-

1170

come, poorer visual acuity, a history of cardiovascular disease, more packyears smoked, a higher serum choles-terollevel, and higher systolic and diastolic blood pressure than persons who participated. There were no statistically significant differences in the presence of age-related mac-ulopathy at baseline between participants and nonparti-cipants.

Procedures Similar procedures were used at both the baseline and follow-up examinations. Letters from primary care phy-sicians or from the principal investigators, if no primary care physician was identified, were sent to those who were eligible. The letters described the study and invited eligible people to participate. This was followed by a call from the study coordinator, who provided further information about the study and made an appointment for the ex-amination. People who were not interested in participating in the examination were asked to respond by telephone to the questionnaire that was administered at the time of examination.

Human experimentation committee approval was granted and informed consent was obtained from each participant at the beginning of the examination. The parts of the examination pertinent to this report consisted of a standardized refraction and measurement of the visual acuity using the Humphrey 530 refractor (Allergan Hum-phrey, San Leandro, CA). At follow-up only, before re-fraction, the participants first were asked to read the Early Treatment Diabetic Retinopathy Study chart R with their current prescription without covering either eye. The number ofletters correctly read was recorded. At baseline and follow-up, the refraction obtained using the refractor was placed in a trial lens frame and the best-corrected visual acuity was remeasured following the Early Treat-ment Diabetic Retinopathy Study protocol using chart R and modified for a 2-m distance. 3. IO If the best-corrected visual acuity was 20/40 or worse, an Early Treatment Diabetic Retinopathy Study refraction was performed and the visual acuity was measured. The interobserver vari-ation among the examiners for obtaining the refractive error or the best-corrected visual acuity was low and not clinically significant (data not shown). . At both examinations, visual acuity was obtainable and

considered reliable in both eyes in 3480 persons, in the right eye only in 41 persons, and in the left eye only in 47 persons. Visual acuity was obtainable and considered reliable in at least one eye at both visits for 17 of those in a nursing home or a group home at both visits (11 were older than 75 years of age) and 55 of those in a nursing home or a group home at follow-up only (39 were older than 75 years of age).

Definitions For each eye, visual acuity was recorded as the number of letters correctly identified (range, 0 [

Klein et al . Changes in Visual Acuity

Table 1. Distribution of Baseline Characteristics among Participants and Nonparticipants in Beaver Dam II

Nonparticipants

Participants Alive Dead

Crude Crude Age-adjusted Crude Age-adjusted Characteristic % (no.) % (no.) P' % (no.) Pt

Age at baseline (yrs) 43-54 34.7 (1277) 30.5 (209) 9.4 (16) 55-64 28.8 (1063) 24.9 (171)

Ophthalmology Volume 103, Number 8, August 1996

20/320, 20/400, 20/500, 20/640, 20/800, hand motions, light perception, and no light perception. Levels of im-pairment in visual function were defined by the best-cor-rected visual acuity in the eye, or, for a participant, in the better eye. The definitions were no impairment (better than 20/40, 41-70 letters correct), any visual impairment (20/40 or worse, 40 or fewer letters correct), and severe impairment (20/200 or worse, 5 or fewer letters correct).

Persons were at risk for vision becoming impaired if their visual acuity was better than 20/40 in one or both eyes at baseline. Similarly, persons were at risk for severe loss of vision if their visual acuity was better than 20/200 in one or both eyes at baseline. Loss of vision over the 5-year period is defined as a doubling of the visual angle, a loss of 15 letters (e.g., a change from baseline to follow-up from 60 to 45 letters read correctly, corresponding to a change in visual acuity from 20/16 to 20/32). For each person, loss of vision was defined according to these cri-teria in the better eye. Persons with visual acuity of no light perception at baseline were, therefore, not at risk to lose vision. The incidence of monocular impairment was defined as visual acuity declining from better than 20/40 visual acuity in both eyes at baseline to 20/40 or worse in one eye only at follow-up. The incidence of severe monocular impairment was defined as visual acuity de-clining from better than 20/200 in both eyes at baseline to 20/200 or worse in one eye only at follow-up. Improve-ment in vision was defined as vision improving by 15 or more letters (halving of the visual angle). People had im-provement in vision ifthere were 55 letters or more (visual acuity, 20/20 or worse) in at least one eye. Improvement was computed for visual acuity measured in the worse eye. Age was defined as the age at the time of the baseline examination.

Statistical Methods Comparisons of participants and nonparticipants were done using analysis of variance and the Cochran-Mantel-Haenszel test of independence to adjust for age groups with continuous (i.e., blood pressure) and categorical (i.e., visual acuity) characteristics, respectively. Student's t test and analysis of variance were used to compare the change in the number ofletters read between eyes and age groups, respectively. Linear regression analyses were used to compute estimates of age-adjusted (continuous and qua-dratic) and sex-adjusted change. Chi-square and Cochran-Mantel-Haenszel tests were used to compute unadjusted and age-group-adjusted comparisons of incidence, re-spectively. Unless noted, all results were unadjusted for any confounders. SAS was used for all analyses. I 1.12

Results

The mean age of the participants at baseline was 60.4 years, and 56.8% were women (Table 1). The mean num-ber of years of school completed was 12, and the median income was $25,000. Other characteristics of the partic-ipants at baseline are presented in Table 1.

1172

The decrease in the number of letters read correctly (mean standard deviation) over the 5-year period was small--0.9 8.6 letters in the right eye and -1.3 8.9 letters in the left eye. There was a significant inverse re-lation between the change in the number of letters read correctly between examinations and increasing age. In the right eye, the change in the number ofletters read correctly varied from 0.4 4.9 in people 43 to 54 years of age to -5.2 15.4 in people 75 years of age or older at baseline (Fig 1). Similar relations between age and change in the number of letters read correctly between examinations were found in the left eye. Age-adjusted changes in visual acuity scores (mean standard error) over the 5-year pe-riod were slightly higher in right eyes of men (-0.7 0.3) versus women (0.0 0.2). Regarding changes in visual acuity in left eyes, there was no difference between men (-0.8 0.3) and women (-0.8 0.2). After controlling for sex, the amount of change in visual acuity score (logMAR) increased quadratically with age. In people 55 years of age at baseline, there was no change in visual acuity score over the 5-year period (letters gained in right eyes, 0.3 0.2; letters lost in left eyes 0.1 0.2). In persons 65 years of age at baseline, the visual acuity score dimin-ished by 1.0 0.2 letter in right eyes and 1.6 0.2 letters in left eyes. In persons 75 years of age at baseline, the visual acuity score diminished by 3.5 0.3 letters in right eyes and 4.1 0.3 letters in left eyes.

The incidence of visual impairment (right eye versus left eye, 4.9% versus 5.4%; P = 0.35) or doubling of the visual angle (right eye versus left eye, 2.8% versus 3.0%; P = 0.53) was similar for right and left eyes. The 5-year incidence of the development of monocular visual im-pairment (either eye) was 5.6%; for severe monocular vi-sual impairment it was 1.0%; and for doubling of the visual angle it was 4.1 % (Table 2). The 5-year incidence of im-provement in vision in one eye only (either eye) (defined by an increase of 15 or more letters read correctly at fol-low-up compared with baseline) was 5.1%. The 5-year incidence of monocular impairment, severe monocular visual impairment, and doubling of the visual angle in one eye only increased with increasing age, whereas im-provement of visual acuity in one eye only did not change

~ 0 III -1

"0 m

-2 .c E " Z .5 III

-4 Cl c III

.J:: -5 0

-6 43-54 55-64 65-74 75+

Age (years) Figure 1. Five-year change in the number of letters read correctly in the right eye by age and sex in the Beaver Dam Eye Study.

.....

..... ~

Tabl

e 2.

Five

-yea

r Inc

iden

ce o

f Mon

ocul

ar C

hang

es i

n V

isio

n by

Age

an

d Se

x in

the

Bea

ver D

am E

ye S

tudy

Dou

blin

g o

f V

isua

l A

ngle

V

isua

l Im

pair

men

t Se

vere

Im

pair

men

t Im

prov

emen

t

No. o

f Par

ticip

ants

No

. of P

artic

ipan

ts

No. o

f Par

ticip

ants

No

. of P

artic

ipan

ts

Age

(yrs)

a

t Ri

sk

(%)

Pt

at

Risk

(%

) Pt

a

t Ri

sk

(%)

Pt

at

Risk

(%

) Pt

Fe

mal

e 43

-54

650

(0.8)

626

(1.3)

648

(0.0)

153

(4.6)

55-6

4 53

7 (2.

8) 50

7 (3.

0) 53

3 (0.

6) 19

7 (4.

6) 65

-74

521

(6.0)