Knudtson MD, Klein BEK, Klein R, Cruickshanks KJ, Lee KE. Age-Related Eye Disease, Quality of Life, and Functional Activity. Arch Ophthalmol. 2005;123(6):807-814. doi:10.1001/archopht.123.6.807
To examine the associations of measures of quality of life (Medical Outcomes Study Short Form Health Survey) and functional activities (activities of daily living, instrumental activities of daily living, and visual function) in persons with and without age-related eye diseases.
Two thousand, six hundred seventy persons participated in the 1998 through 2000 examinations of both the Beaver Dam Eye Study and the Epidemiology of Hearing Loss Study. Age-related eye disease (age-related maculopathy, cataract, diabetic retinopathy, glaucoma, macula edema, occlusions, amblyopia, and macular holes) were assessed by fundus, slitlamp, and retroilluminated photographs and self-reported ocular history. Also administered was a standard interview that included the Medical Outcomes Study Short Form Health Survey, activities of daily living, instrumental activities of daily living, and visual function questionnaires and information on other medical conditions.
After controlling for age and sex, we found that persons with an age-related eye disease had decreased scores in almost all the domains of the Medical Outcomes Study Short Form Health Survey, and persons with eye disease in both eyes had poorer scores than persons with eye disease in only 1 eye. Stratifying by age-related maculopathy and central cataract yielded similar results. Further adjustment for current visual acuity and the number of comorbid conditions explained most associations. Several of the mental scales were still marginally significantly lower (P<.10) in persons with age-related maculopathy after adjustment. Persons with an age-related eye disease were not more likely to have impaired activities of daily living or instrumental activities of daily living. After adjustment for current visual acuity and number of comorbidities, persons who had trouble reading small print or recognizing people across the street were more likely to have an age-related eye disease. Otherwise, there were no significant associations with the visual function questions and any of the specific ocular conditions.
Many measures of general quality of life and functional activities were related to age-related eye diseases, but few associations remained significant after adjustments for vision and other comorbidities. Our data are compatible with the notion that decreased visual function, irrespective of the pathologic reason for the decrease, is associated with diminished quality of life and functional activities of living.
The personal impact of disease, in general, is its implications for impaired function. For some diseases, the functional impairment is directly related to the body part affected, such as an amputation where a lower extremity is affected by trauma or vascular occlusion. However, some sensory illnesses have far-reaching effects on function. Ocular problems, due to a variety of eye diseases, have been suggested as playing a role in decreased self-assessed visual function and on other general activities of living independently.1- 3 While many researchers are interested in the pathology of the specific eye diseases and strategies to prevent them, the functional impact is often of most concern to patients. Diminished vision has economic and psychological impact on people’s lives.4,5 Further, certain eye diseases have been shown to be associated with declines in quality-of-life measures independent of vision.6 Thus, in addition to age itself, certain eye diseases are among the most feared chronic disabilities among the US public.7 In the current report, we assess quality of life, as measured by the Medical Outcomes Study Short Form Health Survey (SF-36), activities of daily living (ADL), and a visual function (VF) questionnaire in persons with and without age-related eye diseases in a large Midwestern population-based cohort study.
Study subjects included persons participating in both the Beaver Dam Eye Study (BDES) and the Epidemiology of Hearing Loss Study (EHLS) during the 1998-2000 examination phase (n = 2670). Detailed descriptions of the design of both studies appear in other reports.8,9 A private census was conducted in the city and township of Beaver Dam, Wisconsin, in 1987 and 1988 to identify and recruit subjects for the BDES. Of the 5924 eligible persons, 4926 (83%) participated in the baseline examination, with ages ranging from 43 to 86 years.9 The EHLS began at the 5-year follow-up examination of the BDES.8 The data for the current study were collected from 1998-2000 at the 10-year follow-up examination of the BDES and the 5-year follow-up of the EHLS, when the questionnaires concerning quality of life and activity levels were first administered to the entire cohort. Twenty-five percent of the population completed both studies on the same day, and over 95% of the cohort completed both studies within 5 months of each other. Differences between participants and nonparticipants at examinations have been presented for both the BDES9- 11 and the EHLS.8,12 In general, live nonparticipants in 1 or both studies were older and sicker than persons who completed both studies. All data were collected with institutional review board approval in conformity with all federal and state laws, and the study was in adherence to the tenets of the Declaration of Helsinki.
Health-related quality of life was ascertained using the SF-36.13 The SF-36 assessed function, sense of well-being, disability, and personal evaluation of both physical and mental health. The individual questions were grouped into 8 subscales to measure the following health domains: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. The scores calculated for each domain ranged from 0 to 100, with higher scores indicating fewer limitations or disabilities. The 8 scores were also summarized into a mental component score and a physical component score, which have been shown to be accurate summary measures of the 8 subscales.14 All scores were calculated according to the recommended guidelines and algorithms.14,15
Information on ADL and instrumental activities of daily living (IADL) was collected. The ADL portion included 7 questions about activities, such as walking across a small room, bathing, and using a toilet. The IADL portion included 12 questions about activities, such as preparing meals, using a telephone, and doing laundry. More detailed information on these questionnaires is contained in other reports.16,17
We also administered a VF questionnaire,18 which included questions about how vision limited the participants in (1) reading small print as found in the telephone book or the classified ads, (2) reading regular print as found in newspapers, magazines, recipes, menus, or numbers on the telephone, (3) reading road signs or counting pins at the end of a bowling alley, and (4) recognizing people or objects across the street. Possible answers to these questions were none, little, some, most, all, or do not know. Participants were further asked whether they drove at night and, if so, how much they were limited by their vision. If participants did not drive at night, they were asked whether it was because of their vision. Participants were also asked to rate their best-corrected vision as excellent, very good, good, fair, or poor.
The presence of age-related maculopathy (ARM), diabetic retinopathy, and other retinal conditions were evaluated from stereoscopic 30° color fundus photographs centered on the disc (Diabetic Retinopathy Study standard field 1) and macula (Diabetic Retinopathy Study standard field 2) and a nonstereoscopic color fundus photograph temporal to but including the fovea.19 Grading of ARM was performed in a masked fashion using a standardized protocol, the Wisconsin Age-Related Maculopathy Grading scheme described elsewhere.20 For purposes of this article, an eye was considered to have ARM if soft indistinct drusen or any drusen plus pigmentary abnormalities (increased retinal pigment or retinal pigment epithelial depigmentation) or signs of late-stage ARM (exudative ARM or geographic atrophy) were present within an area of a circle with a radius of 3000 μm centered on the fovea. There were not enough persons with late ARM who were free of other eye disease (eg, cataract) to evaluate those with late ARM as a separate group (18 persons with unilateral and 16 persons with bilateral late ARM). Exclusion of these persons did not significantly alter results.
Slitlamp and retroilluminated photographs of the lens were taken with specifically modified cameras to evaluate presence and type of cataract.21 Procedures for grading cataract (lens opacities) appear elsewhere.21,22 An eye was defined as having any cataract if cortical cataract covered more than 5% of the lens surface, nuclear sclerotic cataract was graded as level 4 or 5 on a 5-step scale, or posterior subcapsular cataract covered more than 5% of any subfield of the lens measurement grid.21 Central cataract was defined as the presence of a nuclear sclerotic cataract or a posterior subcapsular cataract or cortical cataract involving 25% or more of the central circle of the lens measurement grid. People who underwent cataract surgery were not included in the cataract analysis.
During the eye examination and subsequent grading of photographs, a variety of other eye conditions were assessed, including central or branch artery/vein occlusions, macular edema, diabetic retinopathy, glaucoma (defined as taking glaucoma drops or a self-reported history of glaucoma), amblyopia (lazy eye), trauma, and congenital diseases. The number of persons in each of these subgroups was too small to analyze separately.
Visual acuity was assessed in 2 ways. First, before refraction, the participants were first asked to read the Early Treatment of Diabetic Retinopathy Study chart R, modified for a 2-m distance, with their current prescription using both eyes. The number of letters correctly read was recorded. On average, 40 letters represents a Snellen equivalent of 20/40 and 55 letters a Snellen equivalent of 20/20. Refraction was performed and best-corrected visual acuity was then measured for each eye using a modification of the Early Treatment of Diabetic Retinopathy Study protocol.11
Age was defined as the participant’s age at the time of the 1998-2000 eye exam. During the eye examination, participants were asked whether they had a history of any of the following conditions: cardiovascular disease (angina, myocardial infarction, or stroke), arthritis, Parkinson disease, Alzheimer disease, gout, cancer, and arthritis. Blood pressure was measured with a random-zero sphygmomanometer according to the Hypertension Detection and Follow-up Program protocol, and the average of 2 measurements was used for analysis.23 Hypertension status was defined as systolic blood pressure greater than or equal to 140 mm Hg or diastolic blood pressure greater than or equal to 90 mm Hg or current use of antihypertensive medications. Diabetes mellitus was defined as a history of diabetes or being newly diagnosed to have diabetes (no previous medical history of diabetes mellitus in the presence of elevated glycosylated hemoglobin and a random blood glucose ≥200 mg/dL).24 A comorbidity index was created by summing the number of the following conditions: diabetes, arthritis, hypertension, Parkinson disease, Alzheimer disease, history of cardiovascular disease, nonskin cancer, gout, and asthma. This is in keeping with the index developed by Linn et al,25 who summed a number of similar conditions.
All statistical analyses were performed using version 8.1 of SAS.26 We first performed analyses comparing various demographic and medical conditions among persons with no eye disease, persons with unilateral disease, and persons with bilateral disease (Table 1). Any eye disease was defined as present if the eye had 1 or more of the following conditions: central cataract, ARM, diabetic retinopathy, glaucoma, macula edema, central or branch artery/vein occlusion, amblyopia, or macular holes. Significance between groups was assessed by linear regression for continuous variables (eg, blood pressure) and Mantel-Haenszel procedures for categorical variables (eg, sex). Means and percentages were adjusted for age using 4 groups: 53 to 64 years, 65 to 74 years, 75 to 84 years, and 85 or more years.
We next assessed differences in quality of life (SF-36) and functional activity (ADL, IADL, and VF) measures in these persons. Different statistical models were used with each quality-of-life outcome because their distributions differed. From the SF-36 questionnaire, the physical component score, physical functioning, general health, mental component score, vitality, and mental health scales were relatively continuous but skewed to the right (ie, higher scores were more likely to occur than lower scores). Squaring the scores on their original scale yielded Box-Cox normalizing transformations.27 Using the transformed variables as the dependent variable, we performed simple linear regression. The other scales (role-physical, bodily pain, social functioning, and role-emotional) were more discrete in nature and did not lend themselves to normalizing transformations. Therefore, we categorized these variables into 3 levels and performed ordinal logistic regression.28 For ease of interpretation, results are reported in the original untransformed scale, but the P values are from the previously described models.
Activities of daily living, IADL, individual questions from the SF-36, and the VF questions were dichotomized to represent the functional activity being impaired or not. Activities of daily living were considered impaired if a participant needed help from a person or special equipment in any of the 7 ADL questions. Similarly, IADLs were considered impaired if they reported at least some difficulty in any of the 12 IADL questions. The “Personal grooming, like brushing hair, brushing teeth, or washing face” item from the ADL was considered impaired if they reported at least some difficulty with this. The “I am as healthy as anybody I know” question from the SF-36 was considered impaired if they reported mostly or definitely false. The “I expect my health to get worse” question from the SF-36 was considered impaired if they reported mostly or definitely true. The self-reported vision assessment was considered impaired if they answered poor. Night driving was considered impaired if the participants had stopped driving because of vision or if they had at least some limitation from their vision if they did drive at night. The other 4 areas of the VF questionnaire (see earlier for further description of the questions) were considered impaired if the participant answered little, some, most or all. These binary outcomes were analyzed using logistic regression techniques.27
Models were first adjusted for age and sex and then further adjusted for current visual acuity (number of letters read correctly using their current prescription, if any) and a comorbidity index. Similar analyses were run with best-corrected visual acuity in the better eye, and results were similar.
In addition to assessing any eye disease in neither, 1, or both eyes, we evaluated the associations of the scales to ARM and central cataract. For analyses of these individual diseases, persons with other ocular conditions were excluded. For example, persons with ARM were not included in any of the cataract analyses. This was done so that we could specifically investigate particular ocular traits individually. In addition, we eliminated from evaluation the persons who did not have any “known” disease but whose best-corrected visual acuity was 20/40 or worse (n = 159 persons). In a subset of these people without “known” eye disease but with best-corrected visual acuity of 20/200 or worse, one of us (R.K.) assigned a reason for their poor vision based on their fundus and lens photographs, self-reported ocular history, and examiner assessment of the participant. The majority of these people either had eye trauma, a recent stroke, or a congenital eye disease (R.K., unpublished data, 2000).
Table 1 presents characteristics of the population by having an age-related eye disease in neither, 1, or both eyes. Persons with an eye disease were older (P<.001). Further, persons with an eye disease in both eyes were older than persons with an eye disease in only 1 eye. After we adjusted for age, there was no difference by sex (P = .30). After we adjusted for age, persons with an age-related eye disease in 1 or both eyes had higher mean systolic blood pressure, were more likely hypertensive, had poorer vision, and were less educated than persons without an eye disease. There was no difference in mean diastolic blood pressure, number of comorbid conditions, having a history of cardiovascular disease, or diabetes status. Results were largely similar when restricting to specific eye conditions (results not shown).
Table 2 shows the SF-36 scales by age-related eye disease status. Table 3 shows the ADL, IADL, and VF questions by age-related eye disease status. After we adjusted for age and sex, all scores from the SF-36 scales, except bodily pain and role-emotional, were significantly lower in persons with an age-related eye disease (Table 2). Scores for 2 scales (role-physical and mental health) remained significant (P<.05) after further adjustment for current visual acuity and number of comorbid conditions. Scores for 4 of the scales (general health, mental component score, vitality, and social functioning) remained marginally significant (P<.10). In scales where a difference was found, persons with an age-related eye disease in both eyes had consistently lower scores than persons with an age-related eye disease in only 1 eye. After multivariate adjustment, persons with an age-related eye disease were more likely to be impaired in ADL, grooming oneself, general vision, reading newspaper print, recognizing people across the street, and reading small print (Table 3). Also, persons with an age-related eye disease were less likely to state they were as healthy as others.
We next stratified the data into subgroups of ARM (Table 4 and Table 5) and central cataract (Table 6 and Table 7), excluding persons with multiple ocular conditions. After we adjusted for age and sex, all SF-36 scale scores were significantly lower in persons with ARM, and in addition, persons with bilateral disease had poorer scores than persons with unilateral disease. Scores for 4 scales (bodily pain, mental component score, vitality, and social functioning) remained significant after multivariate adjustment (Table 4). After we adjusted for age and sex, the VF measures were more likely impaired in persons with ARM, but no associations remained significant after further adjustment for current visual acuity and number of comorbid conditions (Table 5).
Persons with central cataract had significantly lower scores, after we adjusted for age and sex, in 5 scales (physical component score, physical functioning, role-physical, mental component score, and mental health; Table 6). None of these relationships remained significant after further adjustment. Persons with central cataract were more likely to state they were not as healthy as other persons than persons without central cataract (Table 7). Results were largely similar using a broader definition of any cataract (results not shown).
In this study, we found that persons with an age-related eye disease had decreased quality of life as measured by the SF-36, ADL, and IADL instruments and a VF questionnaire. This study extends the findings from our investigation in 1993 of quality of well-being and self-reported ocular and other medical conditions.29 In our previous investigation, we found persons with self-reported cataract or macular degeneration generally had lower quality-of-life measures than persons free of the condition. However, at the time of that study, we did not have a measurement of visual function or ocular photographs to confirm self-reported diagnosis. Our current study adds the advantages of having ocular photographs and measurement of visual acuity. After adjusting for vision and comorbidities and using the more reliable definitions of eye disease, we found few significant associations between quality-of-life measures and age-related eye disease.
It is common practice in ophthalmology to infer the impact of vision on quality of life based on measurements in the clinic of visual acuity, near vision, or contrast sensitivity. However, these measures may not provide information on day-to-day functions requiring vision, such as reading, driving, and performing specific tasks.30 In this study, we evaluated quality-of-life and functional abilities both in a broad sense (SF-36, ADL, and IADL) and in a narrow sense (VF questions). Visual function questions (eg, National Eye Institute Visual Function Questionnaire 25) have been shown to be more sensitive than the other generic scales at assessing vision-related quality of life but do not give an overall sense of health-related quality of life.31,32
We chose to present analyses adjusted for current binocular visual acuity because this was thought to best represent their ambient vision. Rubin and collaborators33 have shown that other psychophysical measures of visual impairment (eg, contrast sensitivity) are just as related to quality of life as visual acuity. There was not a change to our results when we further adjusted the multivariate models for contrast sensitivity (results not shown).
Persons with ARM generally had decreased SF-36 scores and were more likely impaired in ADL, IADL, and visual function. Not all relationships were explained by vision. Differences in several of the mental scores from the SF-36 remained at least marginally significant after adjustment for current visual acuity and comorbidities. This may be explained by ARM’s association with high rates of depression.34,35 This was in contrast with findings of Mangione et al,36 who did not find ARM severity to correlate with any of the SF-36 scales. That study had a smaller sample and used different definitions, so their results cannot be directly compared with ours. Measuring quality of life with other instruments (eg, quality of well-being index, National Eye Institute Visual Function Questionnaire, activities of daily vision scale), other researchers have shown declines in quality of life for persons with ARM.37- 39 Consistent measures of depression and visual function might permit a better understanding of the relationship of ARM to self-perceived quality of life.
The relationships we found with cataract were not significant after adjustment for current visual acuity and other comorbidities. This may suggest that the only negative effect of cataract on quality of life is visual impairment. This is consistent with findings from the Proyecto VER study that found an association of cataract with quality of life as measured by the National Eye Institute Visual Function Questionnaire, which was no longer significant after multivariate adjustment.6 Without adjustment for vision, the Age-Related Eye Disease Study found lower National Eye Institute Visual Function Questionnaire scores in persons with nuclear but not cortical cataract.37 They did not present results adjusting for vision.
Persons who have undergone cataract surgery were not included in the cataract analysis. Many recent studies have shown improvements in quality-of-life and functional measures after cataract surgery.40- 43 Because we did not evaluate any of the quality-of-life outcomes before surgery, we cannot properly address associations with cataract surgery. In our study, persons who have undergone cataract surgery were older, had more comorbidities, and had poorer SF-36 scores than persons who had cataract or no known eye diseases (data not shown).
Other eye symptoms such as itching or dryness were not assessed in this analysis. Symptoms of dry eye and age-related eye disease were not related in Beaver Dam (M.D.K., unpublished data, 2004). Even though these symptoms have been shown to be associated with a decrease in quality of life, they have no impact on the results of our study.44,45
This study provided a unique opportunity to evaluate ocular conditions in a large population not selected for eye disease. Because of detailed grading procedures, comparisons can easily be made between diseased and nondiseased individuals. This study, however, has limitations. The data are cross-sectional, so we cannot evaluate changes in quality of life after an eye disease develops. Further, decreased quality-of-life measures may be attributable to other facets of the visual system not available to us (eg, itching). Also, many comparisons were evaluated in this study with no adjustment for multiple comparisons. While results should be interpreted with caution, they probably suggest the relative importance of age-related ocular disorders on quality of life and functional activities.
In summary, there is little evidence that having an age-related eye disease without visual impairment has a measurable impact on quality of life. It may actually have an effect at diagnosis (which we cannot measure), but humans are resilient, and if the eye disease does not affect function, they cope. Our data are compatible with the notion that decreased visual function, irrespective of the pathologic reason for the decrease, is associated with diminished quality of life and functional activities of living.
Correspondence: Michael D. Knudtson, MS, Department of Ophthalmology and Visual Sciences, 610 N Walnut St, Room 450 WARF, Madison, WI 53726-2336 (firstname.lastname@example.org).
Submitted for Publication: January 16, 2004; final revision received August 31, 2004; accepted September 30, 2004.
Financial Disclosure: None.
Funding/Support: This study was supported by grant EYO6594 (Drs R. Klein and B. Klein) and grant AG11099 (Dr Cruickshanks) from the National Institutes of Health, Bethesda, Md; the Lew R. Wasserman Award from Research to Prevent Blindness, New York, NY (Dr Cruickshanks); and the Research to Prevent Blindness Senior Scientific Investigator Award (Dr R. Klein).