|
|
| |||||||||||||||||||||||||||||||||||||||||
Background Although the prevalence of blindness and visual impairment increases with age, most surveys of ocular disease do not include nursing home residents.
Methods We conducted a population-based prevalence survey of persons 40 years of age or older residing in nursing homes in the Baltimore area. Of 738 eligible subjects in 30 nursing homes, 499 (67.6 percent) participated in the study. They had their eyes examined and their visual acuity tested and were interviewed in detail. The nonparticipants were more likely to be older, to be white, and to have lower scores on the Mini–Mental State Examination.
Results The prevalence of bilateral blindness (visual acuity <20/200) was 17.0 percent. The prevalence of visual impairment (<20/40 but >20/200) was 18.8 percent. The frequency of blindness increased from 15.2 percent among those under 60 years of age to 28.6 percent among those 90 or older. The age-adjusted prevalence of blindness was 50 percent higher among blacks than among whites (P<0.01). As compared with the noninstitutionalized population from the same communities, the rate of blindness among nursing home residents was 13.1 times higher for blacks and 15.6 times higher for whites. Cataract was the leading cause of blindness, followed by corneal opacity, macular degeneration, and glaucoma. We judged that 20 percent of the functional blindness and 37 percent of the visual impairment could be remedied by adequate refractive correction.
Conclusions Blindness and visual impairment are highly prevalent among nursing home residents. Much of this loss of vision could be treated or prevented with appropriate ophthalmologic care.
Methods
The Baltimore Nursing Home Eye Survey is a population-based prevalence survey of ocular disease and visual impairment among nursing home residents who had previously lived in the communities of East Baltimore. It was designed to complement the sample of noninstitutionalized East Baltimore residents studied in the original Baltimore Eye Survey.1 In the current study we estimated the prevalence of ocular disease among persons who would have been eligible for the Baltimore Eye Survey had they been living at home. These estimates could then be combined with the results from the community-based sample to provide a total picture of the magnitude of vision loss in a community. Data were collected from February 1988 through September 1989.
The sampling plan identified residents of nursing homes who had been admitted from an address within the sampling frame of the Baltimore Eye Survey. The admitting offices of all nursing homes in the Baltimore area were contacted to identify residents admitted from the eastern and southeastern health districts of Baltimore. People admitted to a nursing home directly from a hospital were assigned the address of the residence from which they were admitted to the hospital.
Thirty-four of the 36 nursing homes identified as having potentially eligible subjects (94.4 percent) agreed to participate. In the second stage of the study, eligible subjects were defined as those 40 years of age or older who were admitted from a residence in eastern or southeastern Baltimore. Four participating nursing homes (11.8 percent) had no patients who had been admitted from an address in the target area. The number of eligible subjects in each nursing home ranged from 1 to 89, with a total of 748.
A short version of the Mini–Mental State Examination13 was administered to all eligible subjects to determine their competence to provide informed consent. It is scored on a scale ranging from 0 to 32 points, with a score of 0 representing poor cognitive function, and a score of 32 excellent cognitive function. We conferred with the professional staff at each facility regarding the ability to provide informed consent of all subjects who scored less than 15 on the examination. In a small minority of cases, the medical staff recommended that such persons be approached directly. In all other cases, the next of kin or the legal guardian, including a state agency if necessary, was asked to provide permission for the subject to participate. All procedures were approved by the Joint Committee on Clinical Investigation of the Johns Hopkins Medical Institutions.
We interviewed each participating subject or, if necessary, a proxy respondent. The interview included demographic characteristics, medical and ophthalmic history, the use of eye-care and general medical services, self-perceived problems with vision, and a variety of other variables. An ophthalmologic screening examination was conducted in the nursing home and included measurements of height, weight, blood pressure, pulse, ocular refraction, visual acuity, visual fields, and intraocular pressure, as well as stereo-fundus color photographs of the optic disk and maculae. For refraction and visual-acuity testing, subjects' eyes were first refracted with the AO Reichert SR-IV Programmed Subjective Refractor (AO Reichert Scientific Instruments, Buffalo, N.Y.). This result was refined as necessary with the use of standard subjective-refraction techniques. Retinoscopy and trial lenses were used to refract the eyes of subjects who could not use the SR-IV. Visual acuity was measured at 4 m with the use of the charts and light box described by Ferris et al.14,15 Visual acuity was measured separately for each eye and was defined according to the lowest line on the chart for which the majority of letters were read correctly. Visual acuity was measured twice, first with the subject's typical distance correction (that is, eyeglasses or contact lenses, if any) and again with the full required distance correction as determined by the study refraction data. Best corrected visual acuity was defined as the visual acuity in the better eye with full distance correction.
Because definitions of blindness and visual impairment vary widely around the world,16 we have used two alternative sets of criteria in this report. The first set is the criteria most commonly applied in the United States — a best corrected visual acuity in the better eye of 
20/200 for blindness and between 20/40 and 20/200 for visual impairment. This definition of blindness is used by the Social Security Administration to define eligibility for disability due to severe loss of vision. The cutoff for visual impairment conforms to requirements for obtaining a driver's license in most states.17 The second set of criteria is that established by the World Health Organization and used in the International Classification of Diseases — a best corrected visual acuity in the better eye of less than 20/400 for blindness and less than 20/60 but no less than 20/400 for visual impairment.18
The statistical analysis consisted of an estimation of prevalence, with 95 percent confidence intervals based on the normal approximation. The adjusted prevalence was calculated with multiple linear-regression models.
Results
A total of 748 potentially eligible subjects were identified during recruitment — 455 whites, 283 blacks, and 10 persons of other races. Because of their small number, subjects in the "other" race category were excluded from further analysis, leaving a total of 738 subjects. Five hundred fourteen (69.6 percent) of these subjects underwent at least part of the ocular examination. Of the 224 subjects whose eyes were not examined, 12 (5.4 percent) died before an examination could be scheduled, 71 (31.7 percent) declined to provide consent for any contact, 123 (54.9 percent) consented initially but ultimately declined permission when it came time for the eye examination, and 18 (8.0 percent) did not participate for a variety of other reasons, most commonly because of hospitalization.
The rate of participation in the ocular examination declined from an average of 74.6 percent among those under the age of 80 to 64.5 percent among those 80 or older (Table 1). There was no indication of a linear decline in participation with increasing age. Black subjects participated significantly more frequently than did whites (77.7 percent vs. 64.6 percent, P<0.001), mirroring the participation in the noninstitutionalized sample drawn from the same communities.1 There was no difference in the rate of participation according to sex. Nonparticipants had significantly lower scores on the Mini–Mental State Examination than participants; over 80 percent scored less than 12, as compared with approximately 60 percent of the participants.
|
Age was strongly associated with visual impairment and blindness. There was a small peak in the prevalence of blindness among subjects 40 to 59 years of age, which was followed by an increasing prevalence with older ages to a high of 28.6 percent among those 90 years of age or older — more than double the rate among those in their 60s (Table 2, Figure 1). Bilateral visual impairment was also common in this population, with rates rising from 10.9 percent in the youngest age group to 26.5 percent among those 90 years of age or older. Overall, the total prevalence of poor vision (visual impairment or blindness) was 35.8 percent according to the U.S. definition and 30.5 percent according to the criteria of the World Health Organization.
|
|
Women had higher rates of both blindness and visual impairment than did men, although these differences were not statistically significant (Table 2). The prevalence of blindness was 1.37 times greater for women. Similar differences were found for visual impairment.
As compared with community-dwelling persons of the same age from the same base population, nursing home residents had a dramatically higher prevalence of blindness (Table 3). Overall, prevalence rates for the nursing home residents were 15.6 times higher for whites and 13.1 times higher for blacks. For both racial groups, the difference in prevalence between nursing home residents and community residents was greatest in the younger groups, with a continuous decline with increasing age. This decline in relative prevalence was probably due to a ceiling effect, given the high rates in the older groups. Even among people 80 years of age and older, however, the prevalence of blindness was 2.4 to 3.7 times higher in the nursing home population than among those living in the community.
|
Cataract was the leading cause of blindness, accounting for blindness in 46 of the 170 eyes (27.1 percent) among bilaterally blind persons (Table 4). Cataract also occurred in combination with macular degeneration or glaucoma in another five eyes, bringing the total proportion of blind eyes with cataract as a contributory cause to 30 percent. The importance of cataract was not limited to subjects with poor cognitive function. Whereas cataract alone accounted for 58.2 percent (39 of 67) of blind eyes among persons with scores of 10 or lower on the Mini–Mental State Examination, it accounted for 20 percent (2 of 10) and 62.5 percent (5 of 8) of blind eyes for those with scores of 11 to 20 and 21 or higher, respectively. Other leading causes of blindness included corneal opacity or phthisis bulbi (shrinkage and disorganization of the eyeball) in 21 eyes (12.4 percent), glaucoma in 17 (10.0 percent), age-related macular degeneration in 20 (11.8 percent), diabetic retinopathy in 11 (6.4 percent), and cortical damage associated with stroke in 10 (5.9 percent).
|
There was a strong association of cognitive function with visual acuity. The mean Mini–Mental State score was 11.4 among those with good vision (20/40 or better), as compared with 7.8 for those with visual impairment (P<0.01) and 4.8 for those who were bilaterally blind (P<0.01). Adjusting for age, sex, and race did not materially affect this association.
Discussion
Our results show that one in six nursing home residents in whom visual acuity could be measured or estimated was bilaterally blind. This rate was 13 to 15 times higher than among a community-dwelling sample of similar age from the same locale. The differences between the study participants and nonparticipants in age and in cognitive-function scores suggest that the prevalence rates we observed are biased and that the true rates were even higher. When these results are combined with population-based estimates from other studies, it is clear that nursing home residents have significantly higher rates of blindness than people living in the community and that these excess rates are not explained solely by their advanced age.
In the Beaver Dam Eye Study, a small sample of 45 nursing home residents had a blindness rate of 8.3 percent based on examination methods comparable to those used in our survey.2 The researchers involved in the Established Populations for the Epidemiologic Studies of the Elderly used a questionnaire and measured distance visual acuity among the survivors of population-based cohorts from three communities.19 Among the 242 subjects who resided in nursing homes at the time of the six-year follow-up, 17 percent were functionally blind, and residence in a nursing home was associated with a sixfold increase in the prevalence of blindness.19 Woodruff and Pack reported a blindness rate of 9.7 percent from a survey of 43 nursing homes in three counties of Ontario, Canada.20 The variation in prevalence reported in these studies is relatively small considering the potentially large differences in definitions of blindness, the techniques used to measure visual function, and the different population characteristics.
As expected, even within a group of institutionalized elderly people, the prevalence of blindness and visual impairment was strongly associated with age. The rate of blindness was more than twice as high among those 90 years of age or older as among those in their 60s. This is consistent with rates reported by other studies of nursing home populations and by studies of community-dwelling elderly subjects.1,2,3,4,19,20,21 The high prevalence among members of the youngest age group (40 to 59 years of age) may reflect a greater importance of visual loss as a cause for admission to a nursing home in that age group.
Black participants were 50 percent more likely to be blind than white participants. This difference was greater after adjustment for age and sex. In spite of the well-documented difference in the amount of nursing home use between blacks and whites,21,22 it seems unlikely that the observed difference in the prevalence of blindness was due to differential selection according to race for admission to nursing homes. There was a similar relative difference in the prevalence of blindness according to race in the sample of community residents from the same neighborhoods.1,23 There was some evidence that women had higher rates of blindness than men, which is consistent with data from other studies.19,20,21
As in our previous report on the community-based sample from East Baltimore,1 uncorrected refractive error was an important source of visual impairment in this institutionalized group. Adequate refractive correction reduced the prevalence of blindness by 20 percent and of visual impairment by 37 percent. These results may underestimate the degree of functional loss caused by refractive error because they are based on distance acuity. Adequate near vision is important for many of the activities of elderly people, and almost all people over the age of 60 need near-vision correction.
In our previous report on the causes of blindness in a community-based sample from the same population, over 40 percent of the blindness was treatable or preventable.23 Among blacks in East Baltimore, cataract was the leading cause of blindness, indicating an inappropriately low use of cataract surgery. Our current results suggest that inadequate eye-care services may be a problem in nursing homes as well. Cataract was the leading cause of blindness for both blacks and whites, with corneal blindness second for blacks and third for whites. Even among those who had better cognitive-function scores, cataract remained the leading cause of blindness. As expected, blacks had higher rates of blindness due to glaucoma, whereas blindness caused by age-related macular degeneration was more common among whites. Although a substantial proportion of blind eyes in this population were not treatable (e.g., because of cortical blindness due to stroke), over 40 percent were treatable or preventable with appropriate intervention, the most important intervention being cataract surgery.
During the recruitment phase of this study, we interviewed all nursing home medical and nursing directors regarding the frequency of patient admission to their facilities due to serious visual impairment. None thought that patients were admitted because of a vision-related inability to care for themselves at home. Incontinence and dementia were consistently identified as the precipitating conditions associated with admission. Severe vision loss may be an unrecognized contributor to the decline in patients' cognitive function and in their ability to care for themselves, which increases the likelihood of admission to a nursing home. There was also wide variation among nursing homes in their awareness of the visual status of their patients. The high prevalence of treatable disorders suggests that policies for the testing and referral of patients need to be implemented more uniformly in these facilities.
Finally, it is clear from these data that poorer cognitive function is strongly associated with poorer visual acuity. Although we attempted to minimize the effect of subjects' cognitive status on the measurement of their visual acuity, it was impossible to eliminate this factor completely. Whether persons who have both a treatable loss of vision and cognitive impairment should receive ophthalmologic intervention will depend on the cause and severity of the cognitive deficit. Few data are available in the literature to guide a physician's choice of clinical management strategies in this situation. For patients with mild-to-moderate dementia, restoration of vision may improve their quality of life and reduce the intensity of nursing care required. The relative risks and benefits of such interventions need to be evaluated objectively.
Supported by grants (EY-05091, EY-09672, and RR-04060) from the National Institutes of Health.
Source Information
From the Department of International Health, Johns Hopkins University School of Hygiene and Public Health, Baltimore (J.M.T., J.K.); the Dana Center for Preventive Ophthalmology, Johns Hopkins University Schools of Medicine and Public Health, Baltimore (J.M.T., J.K., A.S.); the Worthen Center, Department of Ophthalmology, Georgetown University School of Medicine, Washington, D.C. (J.C.J.); the Department of Ophthalmology, University of California, Los Angeles (A.C.); and the Office of the Dean, School of Hygiene and Public Health, Johns Hopkins University, Baltimore (A.S.).
Address reprint requests to Dr. Tielsch at the Department of International Health, Johns Hopkins School of Hygiene and Public Health, 615 N. Wolfe St., Room 5515, Baltimore, MD 21205-2103.
References
| |||||||||||||||||||||||||||||||||||||||||
Related Letters:
Blindness among Nursing Home Residents
Karlawish J., Schainholz D. C., Tielsch J. M., Katz J., Sommer A., Klein B. E.K., Klein R.
Extract |
Full Text
N Engl J Med 1995;
333:879-880, Sep 28, 1995.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
