PERSONS with diabetes are at increased risk of impaired

Journal of Gerontology: MEDICAL SCIENCES 2004, Vol. 59A, No. 2, 148 153 Copyright 2004 by The Gerontological Society of America Patterns of Disability Related to Mellitus in Older Women Siobhan C. Maty, 1 Linda P. Fried, 1,2 Stefano Volpato, 3 Jeff Williamson, 2,4 Frederick L. Brancati, 1,2 and Caroline S. Blaum 2,5 Departments of 1 Epidemiology and 2 Medicine, The Johns Hopkins Medical Institutions, Baltimore, Maryland. 3 Department of Clinical and Experimental Medicine, University of Ferrara, Italy. 4 Department of Medicine, Wake Forest University, Winston-Salem, North Carolina. 5 Department of Medicine, The University of Michigan, Ann Arbor. Purpose. To identify pattern(s) of disability related to diabetes in older women and to determine the extent to which disability is mediated by selected diabetes complications. Methods. Cross-sectional analysis of data from a population-based study composed of a representative sample of 3570 noninstitutionalized women aged 65 years and older living in the Baltimore metropolitan area who agreed to be screened for the Women s Health and Aging Study. Results. 483 (13.5%) of the women reported physician-diagnosed diabetes. Compared to women without diabetes, women with diabetes were significantly more likely to report difficulty in 14 of 15 daily tasks, including walking 2 3 blocks, lifting 10 pounds, using the telephone, and bathing (range of odds ratios [OR] 1.5 2.8; all p,.01). After adjustment for age, race, and marital status, women with diabetes were about twice as likely to report difficulty in any one of four functioning groups (mobility, upper extremity, higher functioning tasks, or self-care) (OR 2.2; 95% confidence interval [CI] 1.8 2.7), and over three times as likely to report difficulty in a group combining higher functioning and self-care tasks (OR 3.2; 95% CI 2.4 4.1). Adjustment for self-reported heart disease, stroke, high blood pressure, and visual problems did not attenuate these associations. Conclusions. is strongly associated with a wide range of disabilities in older women. This association does not appear to be mediated by prevalent diabetes complications and risks, heart disease, high blood pressure, stroke, or eye disease. Other complications of diabetes (e.g., neuropathy, peripheral vascular disease) may mediate diabetes-associated disability in older adults. PERSONS with diabetes are at increased risk of impaired physical function (1). Because diabetes prevalence rises with age, the impact of diabetes on disability is likely to be particularly significant in an aging population, where disability rates are known to be higher than in a younger population (2). According to the Third National Health and Nutrition Examination Survey (NHANES III), over 18% of people aged 60 years and older have diabetes by the American Association fasting plasma glucose criterion (3), while an equal or slightly higher proportion would have diabetes using criteria based on an oral glucose tolerance test (4). is well known to be associated with atherosclerotic and microvascular complications, which in themselves can result in physical disability (5 8). In addition, diabetes is associated with other risks and conditions, including hypertension and dyslipidemia, that increase the risk of vascular diseases. Therefore, diabetes-related disability is often presumed to be related to the vascular complications of diabetes, such as heart disease, stroke, peripheral vascular disease, and visual disorders. Given the broad impact of diabetes on multiple diseases and conditions, diabetes may have a similarly broad effect on multiple functioning difficulties. Most studies of the association of diabetes and disability have studied only a few indicators of functional impairment, have not focused on older adults, or have considered diabetes as one of several chronic conditions that affect disability (9,10). Recent research has begun to evaluate whether diabetes-related disability is mediated exclusively through diabetes complications, or if an independent or unmeasured effect of the disease is responsible. We conducted a cross-sectional study of women aged 65 and older to test two hypotheses about diabetes in older women: first, that type 2 diabetes in older women is associated with a broad range of functional difficulties, or physical disability; and second, that physical disability in women with diabetes would be explained mainly by the presence of prevalent diabetes-related conditions and complications: heart disease, stroke, hypertension, and visual impairment. METHODS Setting The study population consisted of women aged 65 years and older screened for possible participation in the Women s Health and Aging Study (WHAS), an on-going, longitudinal study of risk factors for the progression of disability (11,12). This screened sample used in our analyses included women with the entire range of disability, from completely able to disabled. Sample Selection A random, population-based sample of women residing in eastern Baltimore City, Maryland, was identified using 148

DIABETES AND DISABILITY 149 Health Care Financing Administration Medicare eligibility records. This interview, successfully completed by 3841 women (72% of eligible sample), determined disability in four summary groups of functional tasks (mobility, upper extremity, basic self-care tasks (nonmobility activities of daily living [ADLs]), and higher functioning tasks (instrumental activities of daily living [IADLs]). Data on demographics, self-reported health status, physiciandiagnosed medical history, and Folstein Mini-Mental State Exam (MMSE) (13) score were also collected. Women with a MMSE score of 15 or below (n ¼ 179), with missing ethnicity data (n ¼ 18), or with missing data for any significant variables used in the final regression models (n ¼ 74) were excluded. The remaining 3570 women were the subjects of our analyses. Data Collection All information was gathered by trained personnel who conducted in-person interviews. status was obtained by self-report in response to the question, Has your physician ever told you that you have diabetes?. Similarly, the presence of comorbid medical conditions was assessed with the question, Has your physician ever told you that you have,disease.? for the following medical conditions: myocardial infarction, angina, congestive heart failure, hypertension, other heart disease, arthritis, stroke, cancer, broken or fractured hip, lung disease, hearing problems, vision problems, and Parkinson s disease. Women reporting myocardial infarction, angina, or congestive heart failure were classified as having heart disease. Finally, participants were questioned about 15 distinct physical tasks in four functional domains: mobility, upper extremity function, self-care tasks, and higher functioning skills. For each task the participant was asked, By yourself, that is without help from another person or special equipment, do you have any difficulty,performing task.?. The responses were dichotomized into yes difficulty versus no difficulty. Classification of Disability Functional groups were derived from several standardized scales [ADL (14), IADL (15), Rosow-Breslau (16,17)]. Individual tasks were grouped as follows: 1) mobility walking 2 3 blocks, climbing 10 steps, getting in or out of bed or chairs, and heavy housework; 2) upper extremity function raising arms, grasping, and lifting up to 10 pounds; 3) higher functioning tasks (IADLs) using telephone, light housekeeping, meal preparation, and personal shopping; and 4) self-care tasks (ADLs) bathing and showering, dressing, eating, and using the phone. A woman was defined as having a physical disability in a particular group if she reported difficulty in any of the tasks specific for that category. However, individuals often reported difficulty for tasks in several groups. Therefore, study participants were further differentiated into six mutually exclusive domains of disability derived from the original four groups: 1) upper extremity only, 2) mobility only, 3) both mobility and upper extremity, 4) higher functioning, 5) self-care, and 6) higher functioning and selfcare. The latter three generally included people who had mobility and/or upper extremity difficulty in addition to their higher functioning and/or self-care difficulty. These domains can be considered hierarchical because they describe women with difficulty in progressively more types of tasks considered to reflect worsening disability (18,19). Analysis (present vs absent) was the main independent variable. Any difficulties versus no self-reported difficulties for each task were compared in diabetic versus nondiabetic women using the Pearson chi-square test (20). Women who reported any difficulty in any task were classified as disabled. Then, disability was aggregated into the six mutually exclusive domains described above. To assess the independent relationship of diabetes to each of the six mutually exclusive domains, we constructed four sequential multiple logistic regression models, with each one adding additional variables. Women with no disability in any task served as the reference group in each logistic regression calculation. The first logistic regression model was unadjusted. The second was adjusted for sociodemographic variables (age, race, marital status, and education). The third was adjusted for sociodemographic variables plus conditions generally related to diabetes (heart disease, high blood pressure, stroke, and vision problems). Finally, the fourth, a fully adjusted model, included sociodemographic variables, plus diabetes-related conditions, plus comorbid conditions unrelated to diabetes (Parkinson s disease, arthritis, fractured hip, lung disease, hearing problems, cancer, and MMSE score). Because the self-reported data did not include a measure of obesity, we performed identical analyses on a subset of women randomly sampled from the same sampling frame (total n ¼ 763, with diabetes n ¼ 113) as our study sample and for whom measured height and weight were available [women in this subsample were among the participants in the WHAS I (11) and WHAS II (12) studies]. For this subset, we included body mass index in unadjusted and adjusted models as described above. All analyses were performed using Statistical Analysis System software (SAS, Inc., Cary, NC) (21). All tests of significance were two tailed. RESULTS Baseline Characteristics Of 3570 women, 483 (13.5%) reported having diabetes mellitus. Characteristics of the women by diabetes status are displayed in Table 1. Compared to women without diabetes, women with diabetes were more likely to be black and less likely to have completed high school. As expected, women with diabetes were more likely to have a history of heart disease (myocardial infarction or angina or congestive heart failure, 32.9% vs 20.5%), high blood pressure (69.4% vs 50.0%), stroke (14.9% vs 6.7%), and vision problems (81.4% vs 73.7%). Disability in Specific Tasks Compared to women without diabetes, women with diabetes were significantly more likely to report difficulty

150 MATY ET AL. with 14 of the 15 tasks (Table 2). The percentage of women with diabetes reporting some task-specific difficulty ranged from 5.0% for eating to 57.7% for doing heavy housework. Overall, women with diabetes were 1.5 to 2.8 times more likely than women without diabetes to report some difficulty with tasks. Disability in Different Domains Compared to women without diabetes, women with diabetes were more likely to report disability related to mobility only (14.7 vs 13.8%), mobility and upper extremity (11.6 vs 8.5%), higher functioning (9.7 vs 7.8 %), and higher functioning and self-care (27.3 vs 14.4%) (Table 3). The association of diabetes with upper extremity only and self-care was not significant in any models. Overall, 73.9% of diabetic women reported disability in at least one functional domain compared to only 56.5% of women without diabetes. When women without any disability were used as the reference group, the unadjusted odds of disability associated with diabetes ranged from 1.4 in self-care to 3.2 in higher functioning and self-care. The observed associations between diabetes and the disability domains were either unchanged or strengthened after adjustment for age, race, marital status, and education. (Table 3; model 2). Prevalent diabetes-related comorbidities attenuated the association of diabetes with the disability domains (model 3), but an independent association of diabetes with disability remained. Likewise, in the four disability domains significant in unadjusted models, the associations remained significant although somewhat attenuated in a final model that adjusted simultaneously for all four diabetes-related conditions as well as seven comorbid conditions unrelated to diabetes. Finally, in the subset of women for whom body mass index was available, inclusion of that variable in either the unadjusted or adjusted models did not change the relationships of diabetes to any of the six disability groupings. DISCUSSION Our research shows that diabetes is associated with a wide range of disabilities in every major task and summary grouping. These associations are independent of age, race, and education, and they are not completely explained by the excess prevalence of heart disease, high blood pressure, stroke, or vision problems in women with diabetes, or by multiple comorbid conditions unrelated to diabetes. Our use of a population-based data set with both healthy and disabled women, an excellent response rate, and a high quality survey lend confidence to our findings. Of the studies that document the cross-sectional association of diabetes with functioning limitations and disability, most include diabetes among many chronic diseases that are associated with disability. Different studies have reported increased associations of diabetes with poor physical performance measured by functional (5,10,22 24), IADL (5,10,22,23,25), and ADL limitations (5,10,22 28). However, few of these studies have simultaneously considered multiple tasks and disability categories. We evaluated a broad array of disability measures and found that diabetes is significantly associated with nearly every functional task Table 1. Selected Demographic and Clinical Characteristics of 3570 Women Aged 65 and Older by Status Characteristics* Sociodemographic data Women With (N ¼ 483) Women Without (N ¼ 3087) p Value Age 65 74 y 54.2 46.8 p,.01 75 84 y 30.0 30.9 NS 85þ y 15.7 22.3 p,.01 Race Black 40.8 22.7 p,.0001 Married currently 28.4 28.4 NS Completed high school 30.6 43.5 p,.0001 Comorbid conditions Heart disease 32.9 20.5 p,.0001 High blood pressure 69.4 50.0 p,.0001 Parkinson s disease 0.8 0.9 NS Stroke 14.9 6.7 p,.0001 Mini-Mental Status Score 23 17.4 14.6 NS Arthritis 63.8 57.6 p,.05 Fractured hip 3.9 5.7 NS Lung disease 12.0 9.5 NS Hearing problems 24.2 20.6 NS Vision problems 81.4 73.7 p,.001 Cancer 13.3 14.0 NS Notes: *All data represent percent of women with and without diabetes reporting the characteristic. Individuals with Mini-Mental Status scores,15 were excluded from this analysis. NS ¼ not significant. from the most complex to the most basic. Our strongest association was found between diabetes and difficulty in the higher functioning and self-care grouping, the most severe disability group. Full adjustment for chronic diseases and demographics somewhat increased the strength of that association. There are several hypotheses as to why diabetes has such a broad effect on physical functioning. has many associated risks, comorbidities, and complications that lead to multiple impairments and a broad impact on functioning. Another hypothesis, a subject of much recent research, is that diabetes, and the metabolic syndrome, represent a more global underlying disorder, potentially related to inflammatory system alterations. Increased inflammatory system activity is associated with the metabolic syndrome (29), appears to increase risk of type 2 diabetes development (30,31), and has been associated with type 2 diabetes (32) and acute hyperglycemia (33). -related inflammation affecting multiple physiological systems, such as vascular, nervous, and musculoskeletal systems, could potentially underlie widespread functioning problems. A key finding of our study is that the most prevalent vascular complications of diabetes (heart disease, stroke, visual impairment), and a highly prevalent co-occurring risk factor (hypertension), did not substantially reduce the association of diabetes with disability. This finding is contrary to the common clinical perception that disability associated with diabetes is mainly attributable to its vascular complications. However, Gregg and colleagues, using data from NHANES III, also found an independent cross-sectional

DIABETES AND DISABILITY 151 Table 2. Distribution of Difficulty With Selected Tasks by Status in 3570 Women Aged 65 and Older Domain Task * Status % With Any Difficulty Unadjusted Odds Ratio (95% CI) Mobility Walking 2 3 blocks Yes 51.8 1.9 No 35.4 (1.6, 2.4) Climbing 10 steps Yes 38.5 2.2 No 22.4 (1.8, 2.6) Getting in/out of bed or chairs Yes 19.9 1.5 No 14.4 (1.2, 1.9) Heavy housework Yes 57.7 2.3 No 37.2 (1.9, 2.8) Upper extremity Raising arms Yes 16.8 1.7 No 10.6 (1.3, 2.2) Grasping Yes 22.2 1.6 No 14.9 (1.3, 2.1) Lifting up to 10 lb Yes 37.8 1.8 No 25.2 (1.5, 2.2) Higher functioning Using phone Yes 7.2 1.6 No 4.6 (1.1, 2.3) Light housekeeping Yes 20.9 2.5 No 3.6 (1.9, 3.2) Meal preparation Yes 18.2 2.8 No 7.3 (2.2, 3.7) Personal shopping Yes 31.5 2.2 No 17.7 (1.7, 2.7) Self-care Bathing/showering Yes 29.9 2.0 No 17.6 (1.6, 2.5) Dressing Yes 13.5 1.9 No 7.5 (1.4, 2.6) Eating Yes 5.0 1.5 No 3.3 (1.0, 2.4) Using toilet Yes 12.9 1.8 No 7.8 (1.3, 2.4) Notes: *483 women with diabetes; 3087 women without diabetes. Odds ratio for presence of any difficulty vs no difficulty in women with diabetes vs women without diabetes. CI ¼ confidence interval. association of diabetes with three functioning measures after controlling for vascular complications of diabetes, obesity, and nondiabetes-related comorbidities (5). However, this analysis concentrated only on three specific functioning measures (walking a quarter mile, climbing steps, and housework) and did not explore the full range of physical limitations and disability related to diabetes. Volpato and colleagues evaluated severe mobility and ADL disability in older women who were already disabled, and also found that these limitations were not explained by body mass index, multiple diabetes-related impairments and conditions, and other comorbidities. Substantial decrease in the independent relationship of diabetes to the two functional outcomes studied was only obtained when multiple comorbidities and impairments were accounted for by statistical adjustment (34). Because we used self-reported data, we could not study the associations of disability with precise measures of vascular disease, or with diabetes complications and potentially related conditions and risks that are not amenable to self-report such as peripheral vascular disease, neuropathy, renal disease, or hyperlipidemia. Study participants were asked about arthritis (the type of arthritis cannot be specified from self-report); control for arthritis did not change the association of diabetes with disability in our study sample. There was no measure of obesity in our self-reported data, although in the subgroup of women (n ¼ 763) who participated in the WHAS studies, control for body mass index did not change associations of diabetes with disability measures. A reasonable hypothesis is that other unmeasured complications of diabetes, both vascular and nonvascular, mediate the association between diabetes and disability. Besides unmeasured vascular complications (peripheral neuropathy, peripheral vascular disease, and renal disease), other conditions and risks may be involved, including hyperlipidemia, depressive symptoms, and cognitive decline. While obesity is independently associated with disability (35), our subgroup analysis, consistent with other studies (5,34,36), did not suggest that obesity accounts for the relationship of diabetes to disability. In older women, it is possible that level of hyperglycemia, catabolic complications, alterations in the inflammatory system, and even geriatric syndromes (28) such as undernutrition, falls, incontinence, and weight loss may be pathways for the effects of diabetes on disability. Psychosocial variables not evaluated during the WHAS screening interview may also influence disability (9). Our study has several limitations. Our data involved older women exclusively, and therefore cannot evaluate potential effects of sex and age on the relationship between diabetes

152 MATY ET AL. Table 3. Unadjusted and Adjusted Associations of With Any Disability by Hierarchical Category in 3570 Women Aged 65 and Older Heirarchical Disability Category Women With (N ¼ 483) Women Without (N ¼ 3087) Unadjusted Adjusted (Demographics) Odds Ratio (95%) CI* Model Adjusted à (Demographics and Complications) Fully Adjusted No disability 126 (26.1) 1344 (43.5) 1.0 (ref) 1.0 (ref) 1.0 (ref) 1.0 (ref) Upper extremity only 19 (3.9) 130 (4.2) 1.6 (0.9, 2.6) 1.6 (1.0, 2.7) 1.5 (0.9, 2.5) 1.4 (0.8, 2.5) Mobility only 71 (14.7) 426 (13.8) 1.8 (1.3, 2.5) 1.8 (1.3, 2.4) 1.6 (1.1, 2.2) 1.6 (1.1, 2.1) Mobility and upper extremity 56 (11.6) 263 (8.5) 2.3 (1.6, 3.2) 2.3 (1.6, 3.2) 1.9 (1.3, 2.7) 1.8 (1.2, 2.7) Higher functioning 47 (9.7) 242 (7.8) 2.1 (1.4, 3.0) 2.2 (1.5, 3.3) 1.8 (1.0, 2.7) 1.7 (1.1, 2.6) Self-care 32 (6.6) 237 (7.7) 1.4 (1.0, 2.2) 1.6 (1.0, 2.4) 1.4 (0.9, 2.1) 1.4 (0.9, 2.3) Higher functioning and self-care 132 (27.3) 445 (14.4) 3.2 (2.4, 4.1) 4.0 (2.9, 5.3) 2.8 (2.0, 3.8) 3.4 (2.3, 4.9) Disability in any category 357 (73.9) 1743 (56.5) 2.2 (1.8, 2.7) 2.4 (1.9, 3.0) 1.9 (1.5, 2.4) 1.9 (1.5, 2.4) Notes: *Indicates odds of any disability in one or more tasks within disability categories in women with diabetes vs women without. Adjusted for age, race, marital status, and education. à Adjusted for age, race, marital status, education, and self-reported comorbidities/conditions potentially related to diabetes (heart disease, stroke, high blood pressure, vision problems). Adjusted for age, race, marital status, education, diabetes-related comorbidities, all other available self-reported conditions (Parkinson s disease, arthritis, hip fracture, lung disease, hearing problems, and cancer), and Mini-Mental scores. CI ¼ confidence interval. and disability. Like all cross-sectional studies, our study cannot provide direct support for causal inferences. Because we lacked clinical data, we were unable to classify diabetic women as to diabetes type (although one would expect that the vast majority had type 2 diabetes), level of obesity, diabetes duration or severity, or level of metabolic control. Also, we had no information on the severity of comorbid diseases. Finally, we used self-report for data on diabetes, diseases, and disability. However, Rathouz and colleagues found significant internal validity and reliability with similar selfreported functional measurements used during weekly assessments of the WHAS study cohort (37). In addition, the reliability of self-reported medical diagnoses has been high among independent elderly individuals in outcomes research related to incontinence (38,39), dependence (40), and falling (41). Finally, the self-reported prevalence of diabetes we found, as well as its associations with increased heart disease and vision problems, is consistent with published research on nationally representative sample data (3) and lends credibility to the self-report of diabetes in this survey. Our study shows that disability associated with diabetes is common, affects many common activities, and is at least partially independent of the most prevalent vascular comorbidities and risks. While causation cannot be established in our cross-sectional data, our findings support the notion that diabetes might produce disability via alternative pathways involving more-difficult-to-measure vascular complications, excess hyperglycemia, or catabolism, and that management of the most prevalent vascular complications of diabetes alone may not necessarily eliminate the burden of functional limitations and disability attributable to diabetes. More research is needed on possible pathways from diabetes to disability and the pathophysiological mechanisms responsible for these pathways in order to design interventions for type 2 diabetes that will substantially decrease the overall burden of disability of older women with diabetes. ACKNOWLEDGMENTS This work was supported by a grant from the National Institute on Aging and a Clinical Research Award from the American Association (Dr. Brancati) and National Institutes of Health grant K08-AG00749 (Dr. Blaum). The WHAS studies were supported by National Institute on Aging grant RO1 AG11703-01A1, NIH-NCRR, OPD-GCRC grant RR00722, RO1 AI41956, and NIA Contract NO1-AG12112. Dr. Maty is now an Assistant Professor of Community Health at Portland State University, Portland, Oregon; Dr. Blaum is also a Visiting Assistant Professor of Medicine, The Johns Hopkins University. The results were presented in part at the 57th Scientific Sessions of the American Association, Boston, Massachuestts, June 1997. Address correspondence to Caroline S. Blaum, MD, Division of Geriatrics, The University of Michigan Medical Center, Rm. 1111 CCGCB, 1500 East Medical Center Dr., Ann Arbor, MI 48109-0926. E-mail: cblaum@umich.edu REFERENCES 1. Nathan DM. Long-term complications of diabetes mellitus. N Engl J Med. 1993;328:1676 1685. 2. Songer T. Disability in diabetes. In: Harris M, Cowie C, Stern MP, et al., eds. in America. Bethesda, MD: National Institutes of Health, National Institute of and Digestive and Kidney Diseases; 1995:259 283. 3. Harris MI, Flegal KM, Cowie CC, et al. Prevalence of diabetes, impaired fasting glucose, and impaired glucose tolerance in U.S. adults. The Third National Health and Nutrition Examination Survey, 1988 1994. Care. 1998;21:518 524. 4. Resnick HE, Harris MI, Brock DB, Harris TB. American Association diabetes diagnostic criteria, advancing age, and cardiovascular disease risk profiles: results from the Third National Health and Nutrition Examination Survey. Care. 2000;23:176 180. 5. Gregg EW, Beckles GL, Williamson DF, et al. and physical disability among older U.S. adults. Care. 2000;23: 1272 1277. 6. Klein R, Klein B. Vision disorders in diabetes. In: Harris M, Cowie C, and Stern MP, et al., eds. in America. Bethesda, MD: National Institutes of Health, National Institute of and Digestive and Kidney Diseases; 1995:293 338. 7. Palumbo P, Melton L III. Peripheral vascular disease and diabetes. In: Harris M, Cowie C, Stern MP, et al., eds. in America. Bethesda, MD: National Institutes of Health, National Institute of and Digestive and Kidney Diseases; 1995:401 408. 8. Wingard D, Barrett-Connor E. Heart disease and diabetes. In: Harris M, Cowie C, Stern MP, et al., eds. in America. Bethesda,

DIABETES AND DISABILITY 153 MD: National Institutes of Health, National Institute of and Digestive and Kidney Diseases; 1995:429 448. 9. Fried LP, Bandeen-Roche K, Kasper JD, Guralnik JM. Association of comorbidity with disability in older women: the Women s Health and Aging Study. J Clin Epidemiol. 1999;52:27 37. 10. Guccione AA, Felson DT, Anderson JJ, et al. The effects of specific medical conditions on the functional limitations of elders in the Framingham Study. Am J Public Health. 1994;84:351 358. 11. Guralnik J, Fried L, Simonsick E, et al. The Women s Health and Aging Study: Health and Social Characteristics of Older Women with Disability. Bethesda, MD: National Institute on Aging; 1995. NIH Pub. No. 95 4009. 12. Fried LP, Bandeen-Roche K, Chaves PH, Johnson BA. Preclinical mobility disability predicts incident mobility disability in older women. J Gerontol Med Sci. 2000;55:M43 M52. 13. 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. 14. Katz S, Ford A, Moskowitz R, et al. Studies of illness in the aged: The index of ADL, a standardized measure of biological and psychosocial function. JAMA. 1963;9:179 186. 15. Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist. 1969;9:179 186. 16. Nagi SZ. An epidemiology of disability among adults in the United States. Milbank Q. 1976;54:439 467. 17. Rosow I, Breslau N. A Guttman health scale for the aged. J Gerontol. 1966;21:556 559. 18. Jette AM, Branch LG, Berlin J. Musculoskeletal impairments and physical disablement among the aged. J Gerontol. 1990;45:M203 M208. 19. Verbrugge LM, Jette AM. The disablement process. Soc Sci Med. 1994;38:1 14. 20. Agresti A. Categorical Data Analysis. New York: John Wiley and Sons; 1990. 21. SAS Institute. SAS Procedures Guide, Version 8, Volumes 1 and 2. Cary, NC: SAS Institute; 2000. 22. Boult C, Kane RL, Louis TA, et al. Chronic conditions that lead to functional limitation in the elderly. J Gerontol. 1994;49:M28 M36. 23. Miller DK, Lui LY, Perry HM III, et al. Reported and measured physical functioning in older inner-city diabetic African Americans. J Gerontol Med Sci. 1999;54:M230 M236. 24. Moritz DJ, Ostfeld AM, Blazer D II, et al. The health burden of diabetes for the elderly in four communities. Publ Hlth Rep. 1994; 109:782 790. 25. Ettinger WH, Fried LP, Harris T, et al. Self-reported causes of physical disability in older people: the Cardiovascular Health Study. CHS Collaborative Research Group. J Am Geriatr Soc. 1994;42:1035 1044. 26. Kishimoto M, Ojima T, Nakamura Y, et al. Relationship between the level of activities of daily living and chronic medical conditions among the elderly. J Epidemiol. 1998;8:272 277. 27. Macheledt JE, Vernon SW. and disability among Mexican Americans: the effect of different measures of diabetes on its association with disability. J Clin Epidemiol. 1992;45:519 528. 28. Tinetti ME, Inouye SK, Gill TM, Doucette JT. Shared risk factors for falls, incontinence, and functional dependence. Unifying the approach to geriatric syndromes. JAMA. 1995;273:1348 1353. 29. Pickup JC, Mattock MB, Chusney GD, Burt D. NIDDM as a disease of the innate immune system: association of acute-phase reactants and interleukin-6 with metabolic syndrome X. Diabetologia. 1997;40: 1286 1292. 30. Pradhan AD, Manson JE, Rifai N, et al. C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. JAMA. 2001; 286:327 334. 31. Barzilay JI, Abraham L, Heckbert SR, et al. The relation of markers of inflammation to the development of glucose disorders in the elderly: the Cardiovascular Health Study.. 2001;50:2384 2389. 32. Ford ES. Body mass index, diabetes, and C-reactive protein among U.S. adults. Care. 1999;22:1971 1977. 33. Esposito K, Nappo F, Marfella R, et al. Inflammatory cytokine concentrations are acutely increased by hyperglycemia in humans: role of oxidative stress. Circulation. 2002;106:2067 2072. 34. Volpato S, Blaum C, Resnick H, et al. Comorbidities and impairments explaining the association between diabetes and lower extremity disability: The Women s Health and Aging Study. Care. 2002;25:678 683. 35. Weil E, Wachterman M, McCarthy EP, et al. Obesity among adults with disabling conditions. JAMA. 1265;288:1265 1268. 36. Gregg EW, Mangione CM, Cauley JA, et al. and incidence of functional disability in older women. Care. 2002;25:61 67. 37. Rathouz PJ, Kasper JD, Zeger SL, et al. Short-term consistency in self-reported physical functioning among elderly women: the Women s Health and Aging Study. Am J Epidemiol. 1998;147:764 773. 38. Herzog AR, Fultz NH. Prevalence and incidence of urinary incontinence in community-dwelling populations. J Am Geriatr Soc. 1990;38:273 281. 39. Resnick NM, Beckett LA, Branch LG, et al. Short-term variability of self report of incontinence in older persons. J Am Geriatr Soc. 1994;42: 202 207. 40. Smith LA, Branch LG, Scherr PA, et al. Short-term variability of measures of physical function in older people. J Am Geriatr Soc. 1990;38:993 998. 41. Nevitt MC, Cummings SR, Kidd S, Black D. Risk factors for recurrent nonsyncopal falls. A prospective study. JAMA. 1989;261:2663 2668. Received October 25, 2002 Accepted February 13, 2003