Vision and hearing impairments are common and are

Transcription

1 Near Vision Impairment Predicts Cognitive Decline: Data from the Hispanic Established Populations for Epidemiologic Studies of the Elderly Carlos A. Reyes-Ortiz, MD, w Yong-Fang Kuo, PhD, wz Anthony R. DiNuzzo, PhD, w Laura A. Ray, MPA, z Mukaila A. Raji, MD, w and Kyriakos S. Markides, PhD wz OBJECTIVES: To estimate the association between sensory impairment and cognitive decline in older Mexican Americans. DESIGN: A prospective cohort study. SETTING: The Hispanic Established Populations for Epidemiologic Studies of the Elderly from five southwestern states. PARTICIPANTS: The sample consisted of 2,140 noninstitutionalized Mexican Americans aged 65 and older followed from 1993/1994 until 2000/2001. MEASUREMENTS: The outcome, cognitive function decline, was assessed using the Mini-Mental State Examination blind version (MMSE-blind) at baseline and at 2, 5, and 7 years of follow-up. Other variables were near vision, distance vision, hearing, demographics (age, sex, marital status, living arrangements, and education), depressive symptoms, hypertension, diabetes mellitus, stroke, heart attack, and functional status. A general linear mixed model was used to estimate cognitive decline at follow-up. RESULTS: In a fully adjusted model, MMSE-blind scores of subjects with near vision impairment decreased 0.62 points (standard error (SE) , P 5.03) over 2 years and decreased (slope of decline) 0.13 points (SE , P 5.045) more per year than scores of subjects with adequate near vision. Other independent predictors of cognitive decline were baseline MMSE-blind score, age, education, marital status, depressive symptoms, and number of activity of daily living limitations. From the Division of Geriatric Medicine, w Sealy Center on Aging, and z Department of Preventive Medicine and Community Health, University of Texas Medical Branch, Galveston, Texas. This study was supported by Grant P50 CA (University of Texas Medical Branch (UTMB) Center for Population Health and Health Disparities) funded by the National Institute of Health, Grant AG10939 (Hispanic Established Population for the Epidemiological Study of the Elderly), funded by the National Institute of Aging, and Grant P60 AG17231 (UTMB Claude D. Pepper Older Americans Independence Center), funded by the National Institute of Health and the National Institute of Aging. Address correspondence to Carlos A. Reyes-Ortiz, MD, Division of Geriatric Medicine, Sealy Center on Aging, University of Texas Medical Branch, Galveston, TX careyeso@utmb.edu CONCLUSION: Near vision impairment, but not distance vision or hearing impairments, was associated with cognitive decline in older Mexican Americans. J Am Geriatr Soc 53: , Key words: near vision; cognitive decline; older Mexican Americans; MMSE-blind; EPESE Vision and hearing impairments are common and are associated with poor outcomes in older people. For example, vision impairment is a risk factor for falls, fear of falling, hip fracture, disability, cognitive decline, and mortality. 1 5 Similarly, hearing impairment is associated with disability, high depressive symptomatology, and cognitive decline. 3,6,7 Optimal cognitive function depends on processing and retrieval of information acquired through the visual and auditory sensory systems, among other senses. Because agerelated changes in these systems could affect performance on mental assessment tests, older patients with poor vision and hearing are frequently excluded from research on cognitive aging, 8,9 but these exclusions fail to take into account that there may be a higher proportion of cognitive impairment or dementia cases in this group of individuals. 10,11 Alternatively, omitting vision-dependent items from the Mini-Mental State Examination (MMSE) 12 does not reduce the power of the MMSE to make a distinction between demented and nondemented individuals. 10,11 Indeed, a previous study 10 reported that neither the sensitivity nor the specificity of its version of the MMSE for the visually impaired was lower than for the full MMSE. Similarly, another study 11 reported that the MMSE blind version (MMSE-blind) was a valid instrument that permits its application in old age. Nevertheless, little is known about how different degrees of hearing and vision impairments affect changes in cognitive function over time in older people. Rates of cognitive impairment in older Mexican Americans appear to be higher than rates found in other older non-hispanic populations. 13,14 A previous analysis of the JAGS 53: , 2005 r 2005 by the American Geriatrics Society /05/$15.00

2 682 REYES-ORTIZ ET AL. APRIL 2005 VOL. 53, NO. 4 JAGS Hispanic Established Populations for Epidemiologic Studies of the Elderly (H-EPESE) data revealed that cognitive impairment was significantly associated with older age, low education, and near vision impairment, but blind subjects were excluded. 14 To extend earlier work, 14 the association between sensory impairment and cognitive decline was investigated in this population, although the present analysis differs from the earlier work 14 in three respects; a longitudinal analysis was performed by using linear mixed models over 7 years of follow-up, cognitive function was assessed by using the MMSE-blind instead of the MMSE (with inclusion of subjects who were blind or severely visually impaired), 10,11 and other potential confounders of the association between sensory impairment and cognitive decline such as baseline mental status and functional status were controlled for. 8,9 METHODS Sample Data are from the H-EPESE, a population-based study of 3,050 noninstitutionalized Mexican Americans aged 65 and older (83% response rate) residing in five southwestern states: Texas, California, New Mexico, Colorado, and Arizona. Sampling and data collection are described elsewhere. 15 Data at baseline (first wave, 1993/1994) were used to predict change and decline on cognitive status over time. Data on cognitive status were also obtained at the second (1995/1996), third (1998/1999), and fourth (2000/2001) waves on subjects aged 65 and older. One hundred ninetyseven subjects without the MMSE-blind at baseline and 178 without sensory measurements at baseline were excluded. Also excluded were 216 deceased subjects, 195 who refused to be reinterviewed or were lost to follow-up, 75 who had no MMSE-blind follow-up measures, and 49 who had missing values on other relevant variables. Subjects who died or were lost to follow-up were not significantly different in terms of sensory impairments but were more likely to have lower MMSE-blind scores than subjects who remained in the study. The final sample for analysis included 2,140 subjects. Measures Outcome The outcome was cognitive status as measured according to absolute decline of MMSE-blind scores from baseline to 2, 5, and 7 years later. 10 Subjects were administered a blind version of the MMSE (MMSE-blind) 10 from which all eight items involving image processing in the test situation (items with naming, reading, and obeying a sentence; writing a sentence; copying; and performing a three-stage command) were deleted. The number of items was reduced to a total yielding a possible score of 0 to 22, compared with a score of 0 to 30 for the full MMSE. 11 Main Independent Variables Near and distance vision and hearing were the main variables of interest in this analysis. Visual acuity was measured using a performance-based assessment in-home interview. Respondents who reported being blind were not further tested for visual acuity. Participants were encouraged to wear their glasses or contact lenses during testing. Seventy-nine percent of participants who reported wearing corrective lenses for reading wore them during near vision screening, and 75% wore them during distance vision screening. Corrected bilateral near vision acuity was measured by having subjects hold cards at least 7 inches from their eyes and asking them to read the numbers, as described previously. 16 Each card had 7-digit telephone numbers of three different type sizes: 7, 10, and 23 points. 16 Participants who could only read the 10-or 23-point type size or were unable to read the 23-point type size were considered to have near vision impairment (code 5 0), and participants who could read the 7-point were considered to have adequate near vision (code 5 1). Corrected bilateral distance vision was measured using a modified Snellen test employing directional Es at 4 m to estimate acuity from 20/40 to 20/200; if visual acuity is greater than 20/200, a subject is classified as being functionally blind. 16 Participants who could only read Es of 20/ 60 or greater were considered to have distance vision impairment (code 5 0), and participants who could read 20/40 or less were considered to have adequate distance vision (code 5 1). Hearing was assessed using the 10-item Hearing Handicap Inventory for the Elderly, Screening version, 17 which is a well-validated, cross-culturally adapted self-report measure of hearing impairment, with potential scores ranging from 0 to 40. Participants who had scores of 10 or greater were considered to have hearing impairment (code 5 0), and participants who had scores of less than 10 were considered to have adequate hearing (code 5 1). 17 Covariates Sociodemographic variables examined were age, sex (male 5 0, female 5 1), marital status (unmarried 5 0, married 5 1), living arrangement (living alone 5 0, living with others 5 1), and education (years of schooling). Medical conditions were assessed using a series of questions asking the respondents whether a doctor had ever told them that they had hypertension, diabetes mellitus, heart attack, or stroke. Depressive symptoms were measured using the Center for Epidemiologic Studies Depression Scale (CES- D). 18 This scale is widely used in older populations and consists of 20 items in which subjects are asked whether they have experienced certain feelings or symptoms during the week. Response items are scored on a 4-point scale (0 3), with an overall range of 0 to 60, with higher scores indicating greater depressive symptomatology. Persons scoring 16 or more were categorized with high depressive symptomatology. 18 Functional status was assessed using self-report of seven items from a modified version of the Katz activity of daily living (ADL) scale. 19 ADLs included walking across a small room, bathing, grooming, dressing, eating, transferring from a bed to a chair, and using the toilet. Subjects were asked whether they could perform the activities without help, if they needed help, or whether they were unable to perform them (0 7). Statistical Analyses Sociodemographic and health characteristics were examined at baseline for subjects stratified by categories of near vision, distance vision, and hearing (impairment vs

3 JAGS APRIL 2005 VOL. 53, NO. 4 NEAR VISION AND COGNITIVE DECLINE 683 adequate) using the chi-square test for categorical variables and the t test or the Wilcoxon rank-sum test for continuous variables. To test whether the relationship between sensory measures (near vision, distance vision, and hearing) and cognitive function is modified across time, a general linear mixed model was fitted using the MIXED procedure in SAS (SAS Institute, Inc., Cary, NC), while adjusting for age, sex, medical conditions (hypertension, diabetes mellitus, stroke, and heart attack), depressive symptoms, and number of ADL limitations. Mixed model for analysis of the multiwave H-EPESE data set was chosen because this analytic approach accounts for unbalanced data and modeling of covariance structure. In the mixed models, associations between near vision, distance vision, hearing, and decline in cognition (MMSEblind score) over time were tested for, adjusting for baseline MMSE-blind. Interaction termsfnear vision by time, distance vision by time, hearing by time, baseline MMSE-blind by timefwere also introduced in the mixed model to assess the association between near vision (impairment vs adequate), distance vision (impairment vs adequate), hearing (impairment vs adequate), and slope of cognitive decline over time. There were two linear mixed models. Model 1 included sensory measures (near vision, distance vision, and hearing), adjusting for baseline MMSE-blind and time, and their corresponding interactions with time. In Model 2, sociodemographics, medical conditions (hypertension, stroke, diabetes mellitus, and heart attack), depressive symptoms, and number of ADL limitations were added. All analyses were performed using SAS for Windows, version 8.2. RESULTS The distribution of subjects according to near- or distance vision categories was as follows: 7.3% (n 5 157) had nearand distance vision impairment, 14.2% (n 5 304) had only near vision impairment, 5.0% (n 5 106) had only distance vision impairment, and 73.5% (n 5 1,573) had neither near nor distance vision impairment. Of the visually impaired subjects, 137 were severe visually impaired or functionally blind (e.g., could only read more than 23-point type size for near vision or more than 20/200 for distance vision or were self-reported blind). Table 1 presents baseline characteristics of subjects by vision and hearing categories (impairment vs adequate). Subjects with near vision, distance vision, and hearing impairment were more likely to be older, to have low baseline MMSE-blind scores, and to report high depressive symptoms and a high number of ADL limitations. In addition, subjects with near and distance vision impairment were more likely to have self-reported diabetes mellitus and stroke than subjects with adequate vision. Figure 1 presents the unadjusted absolute decline in MMSE-blind scores at 2, 5, and 7 years of follow-up, by vision and hearing categories. Subjects with near vision impairment had significantly lower levels of MMSE-blind than subjects with adequate near vision at 2, 5, and 7 years of follow-up; the slopes of decline were steadily deeper for Table 1. Baseline Characteristics of Older Mexican Americans by Vision and Hearing Categories (N 5 2,140) Variable Near Vision Impairment (n 5 461) Adequate Near Vision (n 5 1,679) Distance Vision Impairment (n 5 263) Adequate Distance Vision (n 5 1,877) Hearing Impairment (n 5 220) Adequate Hearing (n 5 1,920) Age, mean SD z z z Male, % Unmarried, % z Living alone, % Education, years, z z mean SD Hypertension, % Stroke, % 7.2 w z Diabetes mellitus, % 25.8 w w Heart attack, % Center for Epidemiologic 26.7 z w z 19.1 Studies Depression Scale score 16, % Baseline Mini-Mental State z z z Examination Blind score, mean SD k Number of activity of daily living limitations, mean SD z z z z Po.05; w.01; z.001. P-values were calculated using the chi-square test for categorical variables and the t test or the Wilcoxon rank-sum test for continuous variables; comparisons are between impairment and adequate categories. Range 0 60, higher score indicates greater depression. k Range z Range 0 7, higher score indicates greater impairment. SD 5 standard deviation.

4 684 REYES-ORTIZ ET AL. APRIL 2005 VOL. 53, NO. 4 JAGS - vision. Neither distance vision impairment nor hearing impairment predicted cognitive decline. In Model 2, MMSEblind scores of subjects with near vision impairment decreased 0.62 points (SE , P 5.03) over 2 years and decreased (slope of decline) 0.13 points (SE , P 5.045) more per year than scores of subjects with adequate near vision. Other independent predictors of cognitive decline were time (estimate (b) , SE , Po.001), baseline MMSE-blind (b , SE , Po.001; slope of decline each year b , SE , P 5.001), age (b , SE , Po.001), education (b , SE , Po.001), being unmarried (b , SE , P 5.028), high depressive symptoms (CES-D score 5 16, b , SE , P 5. 04), and number of ADL limitations (b , SE , Po.001). = - - Figure 1. Unadjusted absolute decline with 95% confidence intervals in Mini-Mental State Examination-blind (MMSEB) scores at 2, 5, and 7 years of follow-up by vision and hearing categories (N 5 2,140). subjects with near vision impairment after 2 and 5 years of follow-up. Subjects with distance vision impairment had significantly lower levels of MMSE-blind than subjects with adequate distance vision at 2, 5, and 7 years of follow-up; the slopes of decline went down between 2 and 5 years but went up after 5 years. There were no differences in levels of MMSE-blind between subjects with hearing impairment and those with adequate hearing. Table 2 shows the adjusted mixed models predicting cognitive decline in MMSE-blind scores over the 7-year follow-up as a function of baseline sensory impairments and other characteristics. In Model 1, MMSE-blind scores of subjects with near vision impairment decreased 0.87 (standard error (SE) , P 5.002) over 2 years and decreased (slope of decline) 0.13 points (SE , P 5.04) more per year than scores of subjects with adequate near DISCUSSION This cohort study of community-dwelling older Mexican Americans found that near vision impairment at baseline was associated with low cognitive function and deeper cognitive decline over 7 years of follow-up. This association remained significant after adjustment for possible confounders such as other sensory impairments (distance vision and hearing impairments), demographics, comorbidities, functional status, and baseline MMSE-blind scores. To the authors knowledge, this is the first study of cognitive impairment in older Mexican Americans that included visually impaired subjects. Near vision tasks include craft and leisure activities, eating, personal care and hygiene, some work tasks, and reading. Although the exact mechanism is unknown, near vision impairments have been postulated to affect cognitive performance by reducing level of participation in intellectually stimulating activities and subsequently leading to decreasing brain reserve. 8,20 For example, in animal and human studies, sensory impairment has a deleterious effect on brain function by producing cerebral atrophy and intellectual decline. 21 In contrast, sensorial stimulation has a crucial role in neuronal activation across the life course. Such stimulation is reported to have beneficial effects on neuronal development (e.g., new synapses), function (e.g., deoxyribonucleic acid repair), and survival. 21 Sensory stimuli, in the form of social and leisure activities, are also associated with a lower risk of developing dementia in older people. 8,20 Indeed, older cognitively impaired adults have improved cognitive function after cataract surgery, probably reflecting better vision and enhanced brain activity as sensory stimulation increased. 22 Nevertheless, because ophthalmological examination was not performed, which ocular pathology (e.g., cataracts) or refractive error accounted for the impaired vision could not be identified. In addition, associations between near vision disorders and brain functions may be indirect. For example, near vision modulates somatosensory function (e.g., interaction between vision and touch), helps to construct brain representations of peripersonal space (e.g., parietal cortex), and establishes relationships between sensorimotor or physical (e.g., grip strength and visual acuity) and cognitive performance. 23,24

5 JAGS APRIL 2005 VOL. 53, NO. 4 NEAR VISION AND COGNITIVE DECLINE 685 Table 2. General Linear Mixed Models Estimates of Absolute Decline in MMSE-B Scores from 2 to 7 Years as a Function of Vision or Hearing Impairment and Other Characteristics of Subjects at Baseline (N 5 2,140) Predictor Variable Model 1 Estimate SE P-value Model 2 Estimate SE P-value Intercept o o.001 Time o o.001 Near vision impairment Interaction between near vision impairment and time Distance vision impairment Interaction between distance vision impairment and time Hearing impairment Interaction between hearing impairment and time Baseline MMSE-B score o o.001 Interaction between baseline MMSE-B score and time Age o.001 Male Unmarried Living alone Education, years o.001 Center for Epidemiologic Studies Depression Scale score 16 w Hypertension Stroke Diabetes Heart attack Number of activity of daily living limitations z o.001 Note: Variable time is testing the slope of decline on Mini-Mental State Examination-blind (MMSEB) scores over time. The term for the interaction between the predictor variable and time represents the longitudinal effect of the baseline measure of that variable on the annual rate of decline in performance of the MMSEB. Range w Range 0 60, higher score indicates greater depression. z Range 0 7, higher score indicates greater impairment. Estimate 5 estimated change in MMSEB score; SE 5 standard error. An association was not found between distance vision impairment or hearing impairment and cognitive decline. An Australian longitudinal population study found an association between distance vision impairment, but not hearing impairment, and memory decline. 25 Studies with dementia patients found that near and distance vision impairments were associated with cognitive decline. 4,5,26 The finding of the current study that high depressive symptoms were a predictor of cognitive decline is consistent with other reports in the literature. For example, one study 27 found that older women without dementia but with depressive symptoms have worse cognitive function and exhibit greater cognitive decline than women with few or no depressive symptoms after 4 years of follow-up. Another study 28 reported that depression was associated with risk of Alzheimer s disease and cognitive decline in highly educated individuals. The current finding that the number of ADL limitations was a predictor of cognitive decline should be interpreted with caution. Cognitive decline and ADL impairment are usually correlated because the progression of cognitive decline is associated with ADL impairment. 8,9,20 One study suggested a bidirectional effect of ADL and cognitive impairment because both impairments appear to influence the development of one another. 29 Because both are markers of frailty, the causal direction of ADL impairments need further investigation. The finding of association between high education and slower cognitive decline is consistent with other studies linking high education with lower odds of future dementia, where education was considered a marker of brain reserve to maintain cognitive function in later life. 8,20,21 For example, it was reported that highly educated subjects were less likely to develop dementia than less-educated subjects and that subjects with college education reported greater participation in doing crossword puzzles and playing musical instruments, activities found to be significantly associated with lower risk of developing dementia after 5 years of follow-up, than subjects with high school education or less. 8 One limitation of this study was its reliance on selfreport data for comorbidities, but self-report of medical conditions has been used in many studies with communitydwelling older people. 1,2,6 Another limitation is that information was not available in this population about the causes of near vision impairment, each of which may have different effects on sensory performance. In addition, data on sensory measures after the baseline were not included because the same measures of vision and hearing were not used for all subjects. In addition, information about whether corrections of near or distance vision problems (e.g., cataract surgery) could have salutary effects on cognitive function was not available. Nevertheless, an important contribution of this study was the inclusion of severely visually impaired individuals, a population understudied in cognitive aging. 10,11 In conclusion, these results showed that near vision impairment was predictive of cognitive decline in older

6 686 REYES-ORTIZ ET AL. APRIL 2005 VOL. 53, NO. 4 JAGS Mexican Americans independent of other health factors. An important step toward preventing cognition-related disability is better identification and treatment of factors mediating vision decline in older Mexican Americans, one of the fastest-growing segments of the U.S. population. More studies are needed to investigate the effect of corrections of low vision on subsequent cognitive function in older people. REFERENCES 1. Klein BE, Moss SE, Klein R et al. Associations of visual function with physical outcomes and limitations 5 years later in an older population: The Beaver Dam Eye Study. Ophthalmology 2003;110: DiNuzzo AR, Black SA, Lichtenstein MJ et al. Prevalence of functional blindness, visual impairment, and related functional deficits among elderly Mexican Americans. J Gerontol A Biol Sci Med Sci 2001;56A:M548 M Reuben DB, Mui S, Damesyn M et al. The prognostic value of sensory impairment in older persons. J Am Geriatr Soc 1999;47: Cronin-Golomb A, Corkin S, Growdon JH. Visual dysfunction predicts cognitive decline in Alzheimer s disease. Optom Vis Sci 1995;72: Uhlmann RF, Larson EB, Koepsell TD et al. Visual impairment and cognitive dysfunction in Alzheimer s disease. J Gen Intern Med 1991;6: Davanipour Z, Lu NM, Lichtenstein M et al. Hearing problems in Mexican American elderly. Am J Otol 2000;21: Uhlmann RF, Larson EB, Rees TS et al. Relationship of hearing impairment to dementia and cognitive dysfunction in older adults. JAMA 1989;261: Verghese J, Lipton RB, Katz MJ et al. Leisure activities and the risk of dementia in the elderly. N Engl J Med 2003;348: Korten AE, Henderson AS, Christensen H et al. A prospective study of cognitive function in the elderly. Psychol Med 1997;27: Reischies FM, Geiselmann B. Age-related cognitive decline and vision impairment affecting the detection of dementia syndrome in old age. Br J Psychiatry 1997;171: Busse A, Sonntag A, Bischkopf J et al. Adaptation of dementia screening for vision-impaired older persons. Administration of the Mini-Mental State Examination (MMSE). J Clin Epidemiol 2002;55: Folstein M, Folstein S, McHugh P. Mini-mental state. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12: Black SA, Espino DV, Mahurin R et al. The influence of non-cognitive factors on the Mini-Mental State Examination in older Mexican Americans: Findings from the Hispanic EPESE. J Clin Epidemiol 1999;52: Nguyen HT, Black SA, Ray LA et al. Predictors of decline in MMSE scores among older Mexican Americans. J Gerontol A Biol Sci Med Sci 2002;57A: M181 M Markides K, Rudkin L, Angel RJ et al. Health status of Hispanic elderly in the United States. In: Martin L, Soldo B, eds. Racial and Ethnic Differences in Late Life Health in the United States. Washington, DC: National Academy Press, 1997, pp Salive ME, Guralnik J, Christen W et al. Functional blindness and visual impairment in older adults from three communities. Ophthalmology 1992;99: Lichtenstein MJ, Bess FH, Logan SA. Validation of screening tools for identifying hearing-impaired elderly in primary care. JAMA 1988;259: Radloff LS. The CES-D scale: A self-report depression scale for research in the general population. Appl Psychol Meas 1977;1: Katz SC, Ford AB, Moskowitz RW et al. Studies of illness in the aged: The index of ADL, a standardized measure of biological and psychosocial function. JAMA 1963;185: Wilson RS, Mendes de Leon CF, Barnes LL et al. Participation in cognitively stimulating activities and risk of incident Alzheimer disease. JAMA 2002;287: Swaab DF, Dubelaar EJ, Hofman MA et al. Brain aging and Alzheimer s disease: Use it or lose it. Prog Brain Res 2002;138: Tamura H, Tsukamoto H, Mukai S et al. Improvement in cognitive impairment after cataract surgery in elderly patients. J Cataract Refract Surg 2004; 30: Pears S, Jackson SR. Cognitive neuroscience: Vision and touch are constant companions. Curr Biol 2004;14:R349 R Li KZH, Lindenberger U. Relations between aging sensory/sensorimotor and cognitive functions. Neurosci Biobehav Rev 2002;26: Anstey KJ, Luszcz MA, Sanchez L. Two-year decline in vision but not hearing is associated with memory decline in very old adults in a population-based sample. Gerontology 2001;47: Rizzo M, Anderson SW, Dawson J et al. Vision and cognition in Alzheimer s disease. Neuropsychologia 2000;38: Yaffe K, Blackwell T, Gore R et al. Depressive symptoms and cognitive decline in nondemented elderly women: A prospective study. Arch Gen Psychiatry 1999;56: Geerlings MI, Schoevers RA, Beekman ATF et al. Depression and risk of cognitive decline and Alzheimer s disease. Br J Psychiatry 2000;176: Black SA, Rush RD. Cognitive and functional decline in adults aged 75 and older. J Am Geriatr Soc 2002;50: