Populations for Epidemiologic Study of the Elderly

Journal of Gerontology: MEDICAL SCIENCES 2002, Vol. 57A, No. 10, M678 M682 Copyright 2002 by The Gerontological Society of America The Interaction of Cognitive and Emotional Status on Subsequent Physical Functioning in Older Mexican Americans: Findings From the Hispanic Established Population for the Epidemiologic Study of the Elderly O Mukaila A. Raji, 1,2 Glenn V. Ostir, 1,2 Kyriakos S. Markides, 2,3 and James S. Goodwin 1,2,3 Departments of 1 Internal Medicine and 3 Preventive Medicine and Community Health, and 2 Sealy Center on Aging, University of Texas Medical Branch, Galveston. Background. Optimal mental health (cognitive and emotional functioning) is an important factor for maintaining physical function. This study investigated the effects of cognitive and emotional status on subsequent lower body function in a population-based sample of older Mexican Americans. Methods. A 2-year prospective cohort study included Mexican Americans aged 65 and older who scored 18 or higher on the Mini-Mental State Examination (MMSE) at baseline interview and for whom complete data on a summary performance measure of lower body function were available at the 2-year follow-up interview (n 2068). In-home interviews in 1993 1994 and 1995 1996 assessed sociodemographic variables, physical health conditions, cognitive function, emotional health, and lower body function. Results. In a multivariate analysis, continuous MMSE (b 0.06; SE 0.02, p.004) and Center for Epidemiological Studies Depression (CES-D) (b 0.53; SE 0.17, p.002) scores at baseline were significantly associated with a summary performance measure of lower body function 2 years later, controlling for age, gender, marital status, education, selected medical conditions, and baseline summary performance score. A significant MMSE by CES-D interaction (p.002) on summary performance score was also found after adjustments were made for age, gender, marital status, education, chronic health conditions, and baseline summary performance score. After adjusting for age, gender, marital status, education, selected medical conditions, and baseline summary performance score, subjects with low cognition (MMSE score 18 21) and high depressive symptoms (CES-D score 16) were the most likely to have poor summary performance scores 2 years later (b 0.95, SE 0.36, p.008), followed by subjects with high cognition (MMSE score 21) and high depressive symptoms (CES-D score 16) (b 0.57, SE 0.19, p.003), and those with low cognition (MMSE score 18 21) and low depressive symptoms (CES-D score 16) (b 0.47, SE 0.22, p.03), with high cognition (MMSE score 21) and low depressive symptoms (CES-D score 16) used as the reference. Conclusions. Our results confirm prior investigations showing both cognitive function and emotional health predict subsequent lower body function, and extend these findings to older Mexican Americans. In addition, our results indicate that good emotional health moderates the impact of low cognition on subsequent physical function. PTIMAL mental health (cognitive and emotional functioning) is an important factor for maintaining physical function. The deleterious effects of poor mental health in late life on functional ability have been well characterized (1 11). Several aging studies have reported on the relationship between poor cognitive functioning and increased limitations in activities of daily living (ADLs) (1 5). Moritz and colleagues, using data from the New Haven Established Populations for Epidemiologic Study of the Elderly (EPESE), showed that the odds of reporting disability in ADLs at the 3-year follow-up were about 2.5 times greater in subjects with low cognitive ability at baseline compared with those with normal cognitive ability (2). Data from the Nun Study showed older women with low-normal cognitive function on the Mini-Mental State Examination (MMSE) at baseline had twice the risk of ADL disability at follow-up relative to those with high-normal cognition (3). Studies have also reported on the association between emotional health and subsequent disability among older adults (6 11). Cronin-Stubbs and colleagues, in a study of 3434 older persons living in the community, reported a gradient of risk between increasing number of depressive symptoms and ADL disability, adjusting for relevant risk factors (11). Penninx and colleagues reported in a group of 1286 community-dwelling persons aged 71 years and older that an increasing number of depressive symptoms at baseline was a significant predictor of declining lower body function 4 years later, adjusting for relevant sociodemographic factors and health status (9). It is not clear, however, if cognitive ability and emotional health interact to affect physical functioning in older Mexican Americans. Thus, the purpose of this study is two-fold. First, we investigate the independent relationships of cognitive ability and emotional health on lower-body function 2 years later in a community sample of older Mexican Americans. Second, because past research has shown good emotional health to positively influence the rate of recovery 1-year post medical event (12), we plan to test the hypothesis that good M678

MENTAL HEALTH AND SUBSEQUENT PHYSICAL FUNCTION M679 emotional health will moderate the association between low cognitive ability and decline in lower body function. METHODS Sample Data are from the Hispanic Established Population for the Epidemiologic Study of the Elderly (H-EPESE). The H-EPESE is an ongoing National Institute on Aging-funded community-based study of 3050 Mexican American subjects aged 65 and older (13). The sample was designed to be generalizable to approximately 85% of older Mexican Americans living in five southwestern states including Texas, California, Colorado, Arizona, and New Mexico (13). A full description of the rationale, methods, and subject characteristics can be found elsewhere (13). The response rate at baseline interview (1993 1994) was 83%, with 2873 (94.2%) subjects interviewed in person and 177 (5.8%) by proxy. At the 2-year follow-up interview, 2167 (71.1%) subjects were reinterviewed in person and 272 (8.9%) by proxy. Only subjects who scored 18 or higher on the MMSE were asked to attempt the lower-body performance tasks. Thus, the present study includes data on subjects who scored 18 or higher on the MMSE and who also had complete data on a summary performance measure of lower body function (n 2068) at the 2-year follow-up interview (1995 1996). Measures Cognitive function. The MMSE is a 30-item measurement instrument used to assess cognitive function. It is among the most frequently used cognitive screening measures in studies of older adults (14), but it cannot be used to diagnose dementia. The English and Spanish versions of the MMSE were adopted from the Diagnostic Interview Scale (DIS) and have been used in prior community surveys (15). Scores have a potential range of 0 to 30, with lower scores indicating poorer cognitive ability. MMSE score was used as a continuous variable (range 18 30) and was dichotomized as 18 21 (n 334) and 22 30 (n 1734) (16). Emotional health. The Center for Epidemiological Studies Depression (CES-D) scale is widely used as a measure of emotional health in community studies of older people (17,18). The CES-D consists of 20 items in which subjects are asked whether they have experienced certain positive or negative feelings or symptoms in the past week. Responses are scored on a four-point scale (0 to 3). Scores for the positive items are reversed and the 20 items summed. The CES-D has a potential range of 0 to 60, with higher scores indicating increased depressive symptoms. In the analysis, CES-D score was used both as a continuous and dichotomized variable. Individuals with a score of less than 16 were classified as having low depressive symptoms (n 1619), and those with scores 16 or more were classified as having high depressive symptoms (n 432) (17). Summary performance measure of lower body function. The summary performance measure is comprised of three lower-body activities: a timed 8-foot walk, rising from a chair five times, and a hierarchical standing balance task (19,20). Using previously established criteria (19), performance on each task is classified on a scale ranging from 0 to 4. Subjects unable to perform the particular task received a 0 score, and a 1 4 score was assigned to those able to complete the task. When the three tasks were summed, an overall summary performance measure score ranging from 0 to 12 was created, where higher scores represented better functioning. Covariates Baseline sociodemographic variables included age, gender, marital status, education, and selected health conditions. Subjects were asked if they ever had a physician diagnosis of heart attack, stroke, arthritis, or diabetes. Statistical Analyses We examined selected baseline variables for all subjects who scored 18 or higher on the MMSE at baseline by summary performance category using contingency tables ( 2 analysis using a two-tailed p value.05). Four multivariate General Linear Models (GLMs) assessed the independent relationship of continuous MMSE and CES-D score at baseline on summary performance score 2 years later. The first model included the continuous MMSE and CES-D scores, and summary performance score; the second model added age, gender, marital status, and education; and the third model added selected medical conditions including heart attack, stroke, diabetes, and arthritis. The fourth GLM model added the continuous MMSE by CES-D interaction term. Using GLM models, we next examined the association of high (score 21) and low (score 21) MMSE score by high (score 16) and low (score 16) CES-D score on summary performance score 2 years later. The first model adjusted for baseline summary performance score, age, gender, marital status, and education, and the second model added selected health conditions including heart attack, stroke, diabetes, and arthritis. RESULTS Table 1 presents baseline sociodemographic characteristics, selected diseases, CES-D and MMSE scores, by summary performance category 2 years later. Subjects aged 65 74 were more likely to have higher summary performance scores 2 years later, while those aged 85 and older were more likely to have lower summary performance scores 2 years later. Women, the unmarried, and those with less education, CES-D 16, and lower MMSE scores were more likely to be in the lowest summary performance category at follow up. Figure 1 shows the percent change in summary performance score from baseline to the 2-year follow-up by CES-D ( 16 and 16) and MMSE category (18 21 and 22). Percent change was categorized as a decline, no change, or improvement. For the four categories, approximately one third of the subjects declined, one-half improved, and the remainder showed no change in summary performance score. Nonsignificant differences were found between the CES-D category of 16 and 16 (p.56) and the MMSE category of 18 21 and 22 (p.21).

M680 RAJI ET AL. Table 1. Selected Baseline Characteristics by Summary Performance Category 2 Years Later (n 2068) Baseline Characteristics 0 3 Summary Performance Category 4 6 7 9 10 12 Age 65 74 162 (48.4) 250 (61.1) 597 (71.2) 387 (79.7).001 75 84 126 (37.6) 136 (33.3) 216 (25.8) 92 (18.9) 85 47 (14.0) 23 (5.6) 25 (3.0) 7 (1.4) Gender Men 105 (31.3) 148 (36.2) 322 (38.4) 271 (55.8).001 Women 230 (68.7) 261 (63.8) 516 (61.6) 215 (44.2) Marital Status Married 160 (47.8) 218 (53.3) 477 (56.9) 319 (65.6).001 Unmarried 175 (52.2) 191 (46.7) 361 (43.1) 167 (34.4) Education 0 6 251 (75.6) 310 (77.5) 608 (73.4) 313 (64.5).001 7 11 55 (16.6) 69 (17.3) 140 (16.9) 97 (20.0) 12 26 (7.8) 21 (5.2) 80 (9.7) 75 (15.5) Chronic Disease (yes) Heart attack 105 (14.7) 50 (7.9) 78 (7.5) 45 (54.9).0001 Stroke 106 (14.5) 29 (4.6) 42 (4.0) 13 (43.4).0001 Diabetes 236 (33.3) 127 (22.1) 226 (22.2) 101 (1.7).0001 Arthritis 380 (52.1) 226 (35.3) 421 (39.5) 181 (30.5).0001 CES-D 0 15 204 (62.2) 301 (74.1) 679 (81.6) 435 (89.7).001 16 124 (37.8) 105 (25.9) 153 (18.4) 50 (10.3) MMSE 18 21 93 (27.8) 62 (15.2) 114 (13.6) 65 (13.4).001 22 242 (72.2) 347 (84.8) 724 (86.4) 421 (86.6) Notes: CES-D Center for Epidemiological Studies Depression; MMSE Mini-Mental State Examination. All subjects had an MMSE score 18. Table 2 presents three multivariate linear regression analyses predicting summary performance score at the 2-year follow-up as a function of baseline continuous MMSE and CES-D scores. In Model 1, after adjusting for baseline summary performance score, the regression coefficients for MMSE (b 0.10, SE.02, p.0001) and CES-D (b 0.56, SE.17, p.0009) were significantly associated with summary performance score 2 years later. With the addition of age, gender, marital status, and education (Model 2), both MMSE and CES-D remained significant predictors of summary performance score. Both variables continued to be significant with the further addition of selected medical conditions (Model 3). Using MMSE and CES-D as continuous variables and adjusting for the variables in Model 3, we then investigated the relationship between an MMSE by CES-D interaction term on 2-year summary performance score. A significant MMSE by CES-D interaction (p.002) was observed. Table 3 presents two multivariate GLM models predicting summary performance score 2 years later by a four-level hierarchical MMSE by CES-D variable. In Model 1, after adjusting for baseline summary performance score, age, gender, marital status, and education, the results indicated a gradient of risk for the MMSE by CES-D variable. Subjects in the low MMSE (score 18 21) and high CES-D (score 16) category declined by 0.99 points on the summary performance measure 2 years later (average score 6.20, SD 3.91), followed by a 0.63-point decline for subjects p Figure 1. Percent of subjects who improved, remained the same, or declined in summary performance score over 2 years. CES-D Center for Epidemiological Studies Depression; MMSE Mini-Mental State Examination. in the high MMSE (score 21) and high CES-D (score 16) category (average score 6.94, SD 3.80), and a 0.42-point decline for those in the low MMSE (score 18 21) and low CES-D (score 16) category (average score 6.20, SD 3.40). The average summary performance score for those in the reference category of high MMSE (score 21) and low CES-D (score 16) was 8.46 (SD 3.06). The results were similar in Model 2 with the addition of heart attack, stroke, diabetes, and arthritis. DISCUSSION Our findings can be summarized as follows. There is a significant relationship between higher baseline MMSE score and physical performance score 2 years later among older Mexican Americans, after adjusting for relevant sociodemographic characteristics, selected medical conditions, and summary performance score at baseline. Additionally, we found a significant inverse relationship between baseline CES-D score and physical performance score 2 years later. These findings are consistent with previous studies (1 11). We also found a previously unreported interaction between cognition and emotional status. Subjects with poor cognitive and CES-D scores were significantly more likely to have poorer summary performance scores 2 years later compared to those with poor cognition and nondepressed affect. The interaction between MMSE and CES-D score on summary performance suggests that the presence of subsyndromal depressive symptoms may exacerbate the deleterious effect of impaired cognition on physical performance in older adults. Previous research has demonstrated an association between depression and impaired physical function among older adults with dementia (21 24). These studies suggest that clinical depression is a major contributor to excess disability. Past studies in older adults have also shown that subsyndromal depressive symptoms are significantly associated with incident disability and subsequent physical decline in older blacks and whites (9 11,25). For instance, Penninx and colleagues, using a summary performance measure, showed depressive symptoms were predictive of subsequent physical decline in community-living older adults (9). Our results extend these studies by showing the

MENTAL HEALTH AND SUBSEQUENT PHYSICAL FUNCTION M681 Table 2. Multivariate Regression Analysis Assessing the Independent Relationship Between Continuous MMSE and CES-D Scores With Summary Performance Score 2 Years Later, Adjusting for Relevant Risk Factors Model 1 (n 2034) Model 2 (n 2013) Model 3 (n 1873) Baseline Characteristics b SE p b SE p b SE p MMSE (18 30) 0.10 0.02.0001 0.06 0.02.005 0.06 0.02.004 CES-D (0 60) 0.56 0.17.0009 0.59 0.17.0003 0.53 0.17.002 Baseline Summary Performance Measure 0.45 0.02.0001 0.39 0.02.0001 0.38 0.02.0001 Age (continuous) 0.11 0.01.0001 0.12 0.01.0001 Female vs Male 0.41 0.15.005 0.36 0.14.04 Married vs Unmarried 0.11 0.15.46 0.09 0.14.51 Education (continuous) 0.01 0.02.44 0.02 0.02.37 Heart Attack 0.48 0.25.05 Stroke 0.78 0.31.01 Diabetes 0.46 0.17.006 Arthritis 0.49 0.14.0006 Notes: MMSE Mini-Mental State Examination; CES-D Center for Epidemiological Studies Depression. All subjects had an MMSE score 18. independent and interactive effect of cognitive and emotional status on subsequent lower body functioning in a cohort of community-dwelling older Mexican Americans. There are several hypotheses on how emotion might interact with cognition to influence subsequent physical function. For instance, the increased risk of disability associated with depressive symptoms has been shown to be significantly associated with reduced physical activity and restricted social contacts (25). Because depression or depressive symptoms are associated with lack of energy, fatigue, and decreased vitality, individuals with high depressive symptoms may be less motivated to engage in mobility-related activities. Conversely, cognitively impaired individuals with good affect may be more motivated to engage in physical activities and other healthy lifestyles. Thus, good emotional health may moderate the adverse consequences of impaired cognition on physical health. Ostir and colleagues showed that a high level of positive emotion was significantly associated with less ADL disability 2 years later (26). Further, specific cognitive deficits may interfere with tests of physical performances such as task-sequencing impairments (apraxia). Hence, persons with low MMSE scores may be less able to engage in physical and social activities. This disengagement may also stem from decreased motivation associated with depressed mood as well as the disabling effect of specific cognitive deficits on mobility tasks. The co-occurrence of cognitive and affective impairments further compromises the adaptive and coping capability of older persons to environmental and medical stressors. This might translate into less physical exercise, poor adherence to treatment regimen, and low motivation to engage in other healthy lifestyles, factors that may moderate subsequent physical decline and scores on lower body performance measures. There are other possible explanations for how poor cognitive and emotional function at baseline may result in poorer physical performance. An unrecognized medical condition, such as a minor stroke, may cause depressive symptoms or impairment of cognition, which in turn may be reflected in poorer mobility performance. In this scenario, cognitive and affective deficits might manifest much earlier, before demonstrable deficits in physical performance measures are recognized. Previous studies have demonstrated a significant association between low cognitive function and an increased risk of subsequent stroke (27,28). We have no direct evidence in our study that incident limitations in physical performance at follow-up are due to stroke or any specific medical condition known to be associated with depression and impaired cognition. This study has some limitations. The first is the generalizability to other populations. For instance, some items on the CES-D, such as the concepts of happiness, hopefulness, life enjoyment, and feeling good, may be interpreted differently in different cultures. However, the relationship between Table 3. Regression Analysis Assessing the Relationship Between Baseline MMSE and CES-D Categories on a Summary Performance Measure 2 Years Later, Adjusting for Selected Risk Factors Model 1 (n 2013) Model 2 (n 1873) Baseline Characteristics b SE p b SE p MMSE (18 21) and CES-D ( 16) 0.99 0.35.005 0.95 0.36.008 MMSE ( 21) and CES-D ( 16) 0.63 0.18.0006 0.57 0.19.003 MMSE (18 21) and CES-D ( 16) 0.42 0.21.05 0.47 0.22.03 MMSE ( 21) and CES-D ( 16) reference reference Baseline Summary Performance Score 0.40 0.02.0001 0.38 0.02.0001 Age (continuous) 0.11 0.01.0001 0.12 0.01.0001 Female vs Male 0.40 0.15.006 0.35 0.15.02 Married vs Unmarried 0.10 0.15.51 0.08 0.15.60 Education (continuous) 0.03 0.02.09 0.03 0.02.05 Heart Attack 0.47 0.25.06 Stroke 0.76 0.31.02 Diabetes 0.46 0.17.006 Arthritis 0.49 0.14.0007 Notes: MMSE Mini-Mental State Examination; CES-D Center for Epidemiological Studies Depression. All subjects had an MMSE score 18.

M682 RAJI ET AL. emotional health and subsequent physical functioning has not been shown to be different by race or ethnicity (9 11,25). For example, past studies have reported on the association between high depressive symptoms and increased risk of subsequent physical decline among older blacks and whites (9 11). A second limitation is the reliance on selfreported CES-D measures in cognitively impaired subjects. However, all subjects included in the analysis had to score 18 or higher on the MMSE, which would remove some of the potential bias. Our study has several strengths including its large community-based sample, its prospective design, and its use of objective measures of lower body function. Objective measures of lower body function have a number of advantages over self-reports, including increased validity and reproducibility and less bias from variations in culture, language, mood, cognition, personality, and education level (19,20,29,30). In conclusion, the present findings showed the independent and interactive effects of cognitive and emotional status on subsequent physical disability 2 years later in a large community sample of older Hispanics. Multiple approaches are needed to maintain function in cognitively impaired elders. One approach is the early recognition and treatment of depression. Further, community-based programs to promote increased physical activities and intellectual engagements in the older population, particularly among the cognitively impaired, merit further study. The results of this study and others may lay the groundwork for the development of diagnostic and therapeutic strategies to prevent or delay the onset of cognitive and physical disability in older Mexican Americans, one of the fastest growing ethnic groups in the United States. Acknowledgments This study was supported by National Institute on Aging Grant AG10939. Glenn V. Ostir is supported by a fellowship award from the Canadian Institutes of Health Research. Address correspondence to Mukaila A. Raji, MD, Sealy Center on Aging, University of Texas Medical Branch, Galveston, TX 77555-0460. E-mail: muraji@utmb.edu References 1. Aguero-Torres H, von Strauss E, Viitanen M, Winblad B, Fratiglioni L. Institutionalisation in the elderly: the role of chronic diseases and dementia. Cross-sectional and longitudinal data from a populationbased study. J Clinical Epidemiol. 2001;54:795 801. 2. Moritz DJ, Kasl SV, Berkman LF. Cognitive functioning and the incidence of limitations in activities of daily living in an elderly community sample. Am J Epidemiol. 1995;141:41 49. 3. Greiner PA, Snowdon DA, Schmitt FA. The loss of independence in activities of daily living: the role of low cognitive function in elderly nuns. Am J Public Health. 1996;86:62 66. 4. Gill TM, Williams CS, Richardson ED, Tinetti ME. Impairments in physical performance and cognitive status as predisposing factors for functional dependence among nondisabled older persons. J Gerontol Med Sci. 1996;51A:M283 M288. 5. Leveille SG, Guralnik JM, Ferrucci L, et al. Black/white differences in the relationship between MMSE scores and disability: the Women s Health and Aging Study. J Gerontol Psychol Sci. 1998;53B:P201 P208. 6. Roose SP, Glassman AH, Seidman SN. Relationship between depression and other medical illnesses. JAMA. 2001;286:1687 1690. 7. Lenze EJ, Rogers JC, Martire LM, et al. The association of late-life depression and anxiety with physical disability: a review of the literature and prospectus for future research. Am J Geriatr Psychiatry. 2001; 9(2):113 135. 8. Koenig HG, Shelp F, Goli V, et al. Survival and healthcare utilization in elderly inpatients with major depression. J Am Geriatr Soc. 1989; 37:599 606. 9. Penninx BW, Guralnik JM, Ferrucci L, et al. Depressive symptoms and physical decline in community-dwelling older persons. JAMA. 1998;279:1720 1726. 10. Bruce ML, Seeman TE, Merill SS, Blazer DG. The impact of depressive symptomatology on physical disability: MacArthur Studies on Successful Aging. Am J Public Health. 1994;84:1796 1799. 11. Cronin-Stubbs D, de Leon CF, Beckett LA, et al. Six-year effect of depressive symptoms on the course of physical disability in communityliving older adults. Arch Intern Med. 2000;160:3074 3080. 12. Ostir GV, Goodwin JS, Markides KS, Ottenbacher KJ, Balfour J, Guralnik JM. Differential effects of pre-morbid physical and emotional health on recovery from acute illness. J Am Geriatr Soc. 2002;50:713 718. 13. Markides KS, Rudkin L, Angel RJ, Espino DV. Health status of Hispanic elderly. In: Martin LJ, Soldo B, eds. Racial and Ethnic Differences in the Health of Older Americans. Washington, DC: National Academy Press; 1997:285 300. 14. Folstein MF, Folstein SE, McHugh PR. Mini-mental state : a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189 198. 15. Bird HR, Canino G, Rubio-Stipec M, Shrout P. Use of the Mini-Mental State Examination in a probability sample of a Hispanic population. J Nerv Ment Dis. 1987;175:731 737. 16. Uhlmann RF, Larson EB. Effect of education on the Mini-Mental State Examination as a screening test for dementia. J Am Geriatr Soc. 1991;39:876 880. 17. Miller TQ, Markides KS, Black SA. The factor structure of the CES-D in two surveys of elderly Mexican Americans. J Gerontol Soc Sci. 1997;52B:S259 S269. 18. Sheehan TJ, Fifield J, Reisine S, Tennen H. The measurement structure of the Center for Epidemiologic Studies Depression Scale. J Pers Assess. 1995;64:507 521. 19. Guralnik JM, Ferrucci L, Simonsick EM, Salive ME, Wallace RB. Lower-extremity function in persons over the age of 70 years as a predictor of subsequent disability. N Engl J Med. 1995;332:556 561. 20. Ostir GV, Markides KS, Black SA, Goodwin JS. Lower body functioning as a predictor of subsequent disability among older Mexican Americans. J Gerontol Med Sci. 1998;53A:M491 M495. 21. Pearson JL, Teri L, Reifler BV, Raskind MA. Functional status and cognitive impairment in Alzheimer s patients with and without depression. J Am Geriatr Soc. 1989;37:1117 1121. 22. Fitz AG, Teri L. Depression, cognition and functional ability in patients with Alzheimer s disease. J Am Geriatr Soc. 1994;42:186 191. 23. Lyketsos CG, Steele C, Baker L, et al. Major and minor depression in Alzheimer s disease: prevalence and impact. J Neuropsychiatry Clin Neurosci. 1997;9(4):556 561. 24. Hargrave R, Reed B, Mungas D. Depressive syndromes and functional disability in dementia. J Geriatr Psychiatry Neurol. 2000;13(2):72 77. 25. Penninx BW, Leveille S, Ferrucci L, van Eijk JT, Guralnik JM. Exploring the effect of depression on physical disability: longitudinal evidence from the Established Populations for the Epidemiologic Studies of the Elderly. Am J Public Health. 1999;89:1346 1352. 26. Ostir GV, Markides KS, Black SA, Goodwin JS. Emotional wellbeing predicts subsequent independent functioning and survival. J Am Geriatr Soc. 2000;48:473 478. 27. Ferrucci L, Guralnik JM, Salive ME, et al. Cognitive impairment and risk of stroke in the older population. J Am Geriatr Soc. 1996;44:237 241. 28. Zhu L, Pratiglioni L, Guo Z, Winbald B, Viitanen M. Incidence of stroke in relation to cognitive function and dementia in the Kungsholmen Project. Neurology. 2000;54(11):2103 2107. 29. Cress ME, Schechtman KB, Mulrow CD, Fiatarone MA, Gerety MB, Buchner DM. Relationship between physical performance and selfperceived physical function. J Am Geriatr Soc. 1995;43:93 101. 30. Kempen R, Steverink N, Ormel J, Deeg DJH. The assessment of ADL among frail elderly in an interview survey: self-report versus performance-based tests and determinants of discrepancies. J Gerontol Psychol Sci. 1996;51B:P254 P260. Received March 8, 2002 Accepted April 30, 2002