Abstract
OBJECTIVES
To assess how elevated body mass index (BMI) affects cognitive function in elderly people.
DESIGN
Cross-sectional study.
SETTING
Data for this cross-sectional study were taken from a multicenter randomized controlled trial, the Advanced Cognitive Training for Independent and Vital Elderly trial.
PARTICIPANTS
The analytic sample included 2,684 normal-weight, overweight, or obese subjects aged 65 to 94.
MEASUREMENTS
Evaluation of cognitive abilities was performed in several domains: global cognition, memory, reasoning, and speed of processing. Cross-sectional association between body weight status and cognitive functions was analyzed using multiple linear regression.
RESULTS
Overweight subjects had better performance on a reasoning task (β = 0.23, standard error (SE) = 0.11, P = .04) and the Useful Field of View (UFOV) measure (β = −39.46, SE = 12.95, P = .002), a test of visuospatial speed of processing, after controlling for age, sex, race, years of education, intervention group, study site, and cardiovascular risk factors. Subjects with class I (BMI 30.0–34.9 kg/m2) and class II (BMI>35.0 kg/m2) obesity had better UFOV measure scores (β = −38.98, SE = 14.77, P = .008; β = −35.75, SE = 17.65, and P = .04, respectively) in the multivariate model than normal-weight subjects. The relationships between BMI and individual cognitive domains were nonlinear.
CONCLUSION
Overweight participants had better cognitive performance in terms of reasoning and visuospatial speed of processing than normal-weight participants. Obesity was associated with better performance in visuospatial speed of processing than normal weight. The relationship between BMI and cognitive function should be studied prospectively.
Keywords: cognitive function, body mass index
Overweight and obesity are growing in prevalence among older adults.1 The prevalence of overweight and obesity is more than 50% in U.S. adults, with the highest prevalence observed in adults aged 50 and older.1 In association with various cardiovascular risk factors such as hypertension, diabetes mellitus, and heart disease, overweight and obesity are related to a host of adverse health outcomes in older adults and place a severe burden on the U.S. healthcare system.2
The relationship between body weight and selected geriatric syndromes has recently attracted significant research interest. Investigators have shown a positive association between body mass index (BMI) and depressive symptoms in elderly people.3 Studies have demonstrated that, in the elderly population, overweight and obesity were related to self-reported functional limitations,4 performance-based functional disability,5 urinary incontinence,6 and chronic osteoarthritic pain, but controversies still exist regarding the relationship between BMI and cognitive ability in older people. A positive relationship between BMI and cognition has been demonstrated,7 although some other reports have showed an inverse association between obesity and cognitive performance.8 Longitudinally, the relationship between BMI and dementia has not been consistent. One study followed 392 nondemented Swedish adults for 18 years and found that overweight was a risk factor for dementia in women,9 but the Personnes Agees QUID Study followed 3,646 individuals for 8 years and demonstrated that subjects with a BMI less than 21 had a greater risk of developing dementia than subjects whose BMI was between 23 and 26. The relationship was no longer significant after excluding individuals who developed dementia early during the follow-up.10
Therefore, the overarching goal of this study was to elucidate the relationship between body weight and cognitive function in late life by examining data from the Advanced Cognitive Training for Independent and Vital Elderly (ACTIVE) trial.
METHODS
Study Design and Participants
The ACTIVE trial, a randomized, controlled study, was designed to evaluate the effectiveness and durability of three cognitive training interventions (memory training, reasoning training, and speed-of-processing training) on mental abilities and daily functioning in older, independent-living adults. The trial is described in greater detail elsewhere.11 Briefly, persons aged 65 to 94 were enrolled across six field sites in the United States using various sampling frames and recruitment strategies (state driver’s license and identification card registries, medical clinic rosters, senior center and community organization rosters, senior housing sites, local churches, and roster of assistance and service programs for low-income elderly persons) from March 1998 to October 1999. Persons were excluded if they were cognitively impaired (score <22 on the Mini-Mental State Examination (MMSE)); had a known diagnosis of Alzheimer’s disease; were functionally impaired (requiring assistance in dressing, personal hygiene, or bathing three or more times in the previous 7 days); were medically unstable, predisposing them to imminent functional decline or death; participated currently or recently in other cognitive training; were unavailable during the study; or were impaired in vision, hearing, or communicative ability, making participation in the study impossible. All participants were randomized into three cognitive intervention groups and a no-contact control group. The cohort consists of 2,802 independent-living, otherwise normal older adults.
The current study analyzed the baseline association between body weight and cognitive function in the ACTIVE participants.
Classification of Body Weight
Body weight status was categorized based on BMI, calculated from baseline objective height and weight measurement, as weight in kilograms divided by height in square meters (kg/m2). BMI was categorized according to the National Institutes of Health obesity standards: BMI less than 18.5, underweight; BMI 18.5–24.9, normal weight, BMI 25.0–29.9, overweight; BMI 30.0–34.9, class I obesity; and BMI equal to and greater than 35.0, class II obesity.12 Of the 2,802 subjects in the ACTIVE cohort, there were 30 subjects who were underweight, 609 normal weight, 1,067 overweight, 642 class I obesity, 366 class II obesity, and 88 missing BMI. Because there were small numbers of underweight subjects (n = 30) and subjects with missing BMI (n = 88), they were excluded from analysis.
Cognitive Function Assessment
Evaluation of cognitive abilities was performed in several domains: global cognition, memory, reasoning, and speed of processing. Global cognitive function was measured using the MMSE. For memory, a composite score was created from three test scores that were standardized to a common mean of zero and summed: the Hopkins Verbal Learning Test Related Word Lists,13 the Rey Auditory-Verbal Learning Test Unrelated Word Lists,14 and the Rivermead Behavioral Memory Test Paragraph Recall task.15 These measures include an immediate and delayed learning condition, critical for the detection of anterograde memory disorders. These properties are also associated with loss of newly presented information.16 A higher memory composite score indicates better memory function.
For the reasoning domain, three outcomes were assessed: Word Series, Letter Series, and Letter Sets. A reasoning composite was similarly created from scores of the three tests. A higher reasoning composite score indicates better performance. Word Series and Letter Series tasks consisted of a list of months or letters forming increasingly complex repetitive patterns, requiring the subject to scan the list, generate and reject hypotheses regarding the nature of the pattern, and then select the next stimulus in a list of possible answers. Letter Set task required subjects to find a rule common to four of five letter sets, marking the set inconsistent with the rule. These reasoning tests require strategy formation, behavioral spontaneity, and retrieval ability from long-term memory.16 It is critical that these tasks do not substantially require the learning of new information (as is the case in the tests of memory) but rather such cognitive abilities as concept formation, response generation, and cognitive flexibility.
For speed of processing, the ACTIVE battery included the Useful Field of View (UFOV)17 measure and the Digit Symbol Substitution (DSS) test. The computer-administered UFOV was a visual divided-attention measure requiring speeded visual processing and focused on measures of visual search skills. The DSS test focused on the ability to identify and locate visual information quickly in a divided-attention format. Lower scores on the UFOV and higher scores on the DSS test indicate better performance. These tasks measure visual and motor speed of processing.
Other Covariates
The ACTIVE protocol included assessments of smoking status and history of diabetes mellitus, stroke, myocardial infarction, and hypercholesterolemia by self-report. A trained research assistant assessed blood pressure after 5 minutes of quiet sitting twice, with 2 minutes between each measure. The average systolic blood pressure and diastolic blood pressure were obtained. Blood pressure was categorized into four groups according to the Seventh Report of the Joint National Committee (normotensive, prehypertension, stage 1 hypertension, and stage 2 hypertension18) and was adjusted as a categorical variable. Other covariates included age, sex, race or ethnicity, study site, and years of education.
Analysis
The analytic sample consisted of 2,684 normal-weight, overweight, or obese ACTIVE participants. Distributions of memory composites, reasoning composites, UFOV scores, and DSS scores were normal. Therefore, normality assumption was tenable for linear regression analyses with these cognitive performances as dependent variables. Distribution of MMSE was only mildly left-skewed. The association between BMI and MMSE was unchanged whether transformed or untransformed MMSE scores were used. Therefore, untransformed MMSE scores were used in the analyses to have interpretable beta coefficients.
Cross-sectional association between body weight status and cognitive functioning was examined by using multiple linear regression while controlling for the influence of other covariates. Indicator variables were created for overweight subjects and those with class I and class II obesity, whereas normal-weight participants were categorized as the reference group. In Model 1, age, sex, race, years of education, and study site were controlled for. In Model 2, all covariates in Model 1 plus smoking status; blood pressure; and history of diabetes mellitus, myocardial infarction, hypercholesterolemia, and stroke were controlled for. The standardized effect sizes in different models were also calculated based on the standard deviation of the cognitive performance of the reference group (normal-weight subjects). In considering the issue of multiple comparisons, P = .05 was allocated across 15 comparisons (3 hypotheses in 5 models) using Bonferroni correction, resulting in a corrected alpha level of 0.0033.
Interaction effects between sex, blood pressure, diabetes mellitus, and categories of body weight were separately assessed in the multivariate model. The linear trend between body weight and cognitive functions was also tested by performing a trend test and plotting the adjusted mean cognitive scores against body weight status.
RESULTS
The baseline characteristics of the ACTIVE participants based on body weight status are provided in Table 1. The majority of the ACTIVE participants were female and white.
Table 1.
Baseline Characteristics of Participants (N = 2,684)*
| Characteristic | Normal Weight (n = 609) | Overweight (n = 1,067) | Class I Obesity (n = 642) | Class II Obesity (n = 366) |
|---|---|---|---|---|
| Age (range), mean ± SD | (65–93) 74.7 ± 6.3 | (65–94) 74.0 ± 5.9 | (65–91) 72.9 ± 5.3 | (65–90) 71.3 ± 4.8 |
| Women, n (%) | 470 (77.2) | 749 (81.3) | 500 (77.9) | 319 (87.2) |
| Race/ethnicity, n (%) | ||||
| White | 492 (80.8) | 779 (73.0) | 431 (67.1) | 219 (59.8) |
| Black | 102 (16.7) | 263 (24.7) | 201 (31.3) | 138 (37.7) |
| All other† | 15 (2.5) | 25 (2.3) | 10 (1.6) | 9 (2.5) |
| Years of education (range), mean ± SD | (5–20) 14.1 ± 2.8 | (4–20) 13.6 ± 2.7 | (6–20) 13.2 ± 2.6 | (5–20) 12.9 ± 2.6 |
| Blood pressure status, n (%)‡ | ||||
| Normal | 195 (32.0) | 218 (20.4) | 94 (14.6) | 56 (15.3) |
| Prehypertension | 227 (37.3) | 415 (38.9) | 219 (34.1) | 105 (28.7) |
| Stage 1 hypertension | 126 (20.7) | 293 (27.5) | 199 (31.0) | 121 (33.1) |
| Stage 2 hypertension | 53 (8.7) | 115 (10.8) | 120 (18.7) | 67 (18.3) |
| Unknown | 8 (1.3) | 26 (2.4) | 10 (1.6) | 17 (4.6) |
| Diabetes mellitus, n (%) | 39 (6.4) | 110 (18.9) | 94 (14.6) | 97 (26.5) |
| Current smoker, n (%) | 58 (9.5) | 89 (4.8) | 36 (5.6) | 12 (3.3) |
| History of high cholesterol, n (%) | 238 (39.1) | 478 (47.3) | 302 (47.0) | 175 (47.8) |
| History of myocardial infarction, n (%) | 57 (9.4) | 115 (12.1) | 79 (11.8) | 43 (11.7) |
| History of stroke, n (%) | 30 (4.9) | 76 (7.3) | 48 (7.5) | 26 (7.1) |
| Cognitive performance (range) mean ± SD§ | ||||
| MMSE score | (23–30) 27.4 ± 1.9 | (23–30) 27.3 ± 2.0 | (23–30) 27.4 ± 2.0 | (23–30) 27.1 ± 2.2 |
| Memory composite | (−7.78–7.44) −0.02 ± 2.71 | (−8.16–8.38) 0.04 ± 2.48 | (−6.67–6.30) 0.06 ± 2.29 | (−6.7–7.0) 0.17 ± 2.41 |
| Reasoning composite | (−6.63–8.77) 0.04 ± 2.73 | (−7.23–10.55) 0.07 ± 2.72 | (−7.71–6.78) −0.08 ± 2.48 | (−6.63–9.36) −0.06 ± 2.78 |
| DSS | (4–77) 40.66 ± 11.65 | (2–79) 40.64 ± 10.92 | (6–92) 40.11 ± 10.74 | (10–77) 40.19 ± 11.06 |
| UFOV | (331–2,000) 975 ± 294 | (338–2,000) 935 ± 285 | (312–2,000) 929 ± 288 | (355–2,000) 911 ± 271 |
Body weight status was categorized according to the following body mass index: normal-weight, 18.5–24.9 kg/m2; overweight, 25.0–29.9 kg/m2; class I obese, 30.0–34.9 kg/m2; and class II obese, ≥35.0 kg/m2.12
Hispanic, Asian, native Hawaiian, Pacific Islander, American Indian, and native Alaskan.
Blood pressure classification is according to the seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure.18
Memory composites, reasoning composites, Useful Field of View (UFOV) scores, and the Digit Symbol Substitution (DSS) scores were normally distributed. Distribution of Mini-Mental State Examination (MMSE) was mildly left-skewed.
SD = standard deviation.
After controlling for age, sex, race, years of education, intervention group, and study site, overweight subjects showed better performance on memory tests (β = 0.24, P = .03, effect size = 0.09) than normal-weight subjects (Table 2), although cardiovascular risk factors confounded the positive relationship between overweight and memory, and the association vanished after cardiovascular risk factors were added to the multivariate model. Memory performance of obese subjects was not statistically different from that of normal-weight subjects.
Table 2.
Cross-Sectional Analysis of Body Weight on Cognitive Functions
| Model 1 |
Model 2 |
|||||||
|---|---|---|---|---|---|---|---|---|
| Variable | β (SE)* | Effect Size† | P-value | P-value for Trend | β (SE)* | Effect Size† | P-value | P-value for Trend |
| Cognitive performance Global cognition (Mini-Mental State Examination score) | ||||||||
| OW | 0.08 (0.09) | 0.04 | .38 | .72 | 0.11 (0.10) | 0.06 | .26 | .90 |
| OB I | 0.17 (0.11) | 0.09 | .10 | 0.20 (0.11) | 0.11 | .06 | ||
| OB II | −0.14 (0.12) | 0.08 | .25 | −0.08 (0.13) | 0.04 | .55 | ||
| Memory composite | ||||||||
| OW | 0.24 (0.11) | 0.09 | .03 | .55 | 0.21 (0.11) | 0.08 | .07 | .50 |
| OB I | 0.18 (0.13) | 0.07 | .16 | 0.16 (0.13) | 0.06 | .22 | ||
| OB II | 0.09 (0.15) | 0.03 | .55 | 0.11 (0.16) | 0.04 | .49 | ||
| Reasoning composite | ||||||||
| OW | 0.24 (0.11) | 0.09 | .03 | .22 | 0.23 (0.11) | 0.08 | .04 | .20 |
| OB I | 0.22 (0.12) | 0.08 | .08 | 0.24 (0.13) | 0.09 | .06 | ||
| OB II | 0.17 (0.15) | 0.06 | .24 | 0.17 (0.15) | 0.06 | .26 | ||
| Speed of processing Digit symbol substitution | ||||||||
| OW | 0.75 (0.48) | 0.06 | .12 | .33 | 0.68 (0.49) | 0.06 | .16 | .38 |
| OB I | 0.12 (0.55) | 0.01 | .83 | 0.11 (0.55) | 0.01 | .84 | ||
| OB II | −0.50 (0.64) | 0.04 | .44 | −0.46 (0.66) | 0.04 | .49 | ||
| Useful Field of View score‡ | ||||||||
| OW | −37.81 (12.77) | 0.13 | .003 | .14 | −39.46 (12.95) | 0.13 | .002 | .04 |
| OB I | −33.28 (14.41) | 0.11 | .02 | −38.98 (14.77) | 0.13 | .008 | ||
| OB II | −24.35 (17.02) | 0.08 | .15 | −35.75 (17.65) | 0.12 | .04 | ||
Note: Regression coefficients in Model 1 were adjusted for age, sex, race, years of education, and study site. Regression coefficients in Model 2 were adjusted for all covariates in Model 1 plus smoking status; blood pressure; and history of diabetes mellitus, myocardial infarction, hypercholesterolemia, and stroke.
Body weight status was divided into four categories: normal weight (reference group: body mass index (BMI) = 18.5–24.9 kg/m2), overweight (OW: BMI = 25.0–29.9 kg/m2), class I obesity (OB I: BMI = 30.0–34.9 kg/m2), and class II obesity (OB II: BMI ≥35 kg/m2).
Parameter estimates (β) can be interpreted as differences in mean scores of cognitive performance for OW, OB I, and OB II subjects versus normal-weight subjects.
Effect size was based on the standard deviation of the reference group (normal-weight subjects).
Smaller values reflect better performance.
SE = standard error.
Overweight subjects demonstrated better performance on the reasoning tests than normal-weight participants after multivariate adjustment for age, sex, race, years of education, intervention group, study site, and cardiovascular risk factors (β = 0.23, P = .04, effect size = 0.08). This association remained marginally significant for comparison between class I obesity and normal-weight subjects (β = 0.24, P = .06, effect size = 0.09). In addition, overweight and obese subjects exhibited better performance on the visuospatial speed of processing (UFOV) than normal-weight subjects (β = −39.46, P = .002, effect size = 0.13; β = −38.98, P = .008, effect size = 0.13, respectively). There were no significant differences in global cognition between overweight/obese and normal-weight subjects. Although the effect sizes were small, in keeping with Cohen’s taxonomy, overweight subjects demonstrated a significantly better score on the UFOV measure than normal-weight subjects when the corrected alpha level (0.0033) was taken into consideration.
The interactions between sex, diabetes mellitus, blood pressure, and body weight were then examined, but no evidence was found for statistically significant interactions between these risk factors. To examine effects of restricted range of measurement using conventional scoring of the MMSE, ability scores derived from MMSE item responses were examined using an item response theory scoring method (one-parameter normal ogive): substantively equivalent results were obtained regarding the relationship between BMI and cognitive function.
There was no linear relationship between body weight categories and global cognition, memory, reasoning task, and the DSS test score, as reflected by nonsignificant P-values in trend test (Table 2) and inverted-U relationship between cognitive scores and weight status (Figure 1). Although the P-value of trend test between weight status and UFOV measure was 0.04, the graph of UFOV and BMI categories demonstrated a J-shaped, not linear, relationship.
Figure 1.
Mean cognitive scores with 95% confidence intervals were obtained according to body weight status after controlling for age, sex, race, years of education, intervention group, study site, and cardiovascular risk factors. A nonlinear relationship between body weight and cognitive performance was noted. Y-axis indicates scores of cognitive tests, and x-axis indicates weight status. NW = normal-weight; OW = overweight; OB I = class I obesity; OB II = class II obesity.
DISCUSSION
The ACTIVE study offers several advantages for the investigation of the relationship between BMI and cognitive function, including a large and diverse sample of healthy older adults, objective measures of height and weight to determine BMI, tests of several cognitive abilities, and measures of several covariates that might confound the relationship between BMI and cognitive function. The results indicate a nonlinear relationship between cognitive performance and BMI in normal-weight, overweight, and obese older adults. Overweight subjects demonstrated better cognitive performance on tests of reasoning and visuospatial speed of processing than normal-weight participants. Obese subjects, unlike their overweight counterparts, did not demonstrate compelling cognitive superiority in memory and reasoning tasks than normal-weight subjects, yet similar to overweight subjects, obese subjects demonstrated better performance in visuospatial speed of processing than normal-weight subjects. There was no interactive effect of sex, blood pressure, diabetes mellitus, or body weight on cognitive performance. The findings support a recent longitudinal study that found that subjects with a BMI greater than 23 kg/m2 had lower risk of developing cognitive impairment after 5 years of follow-up.19 In addition, the researchers described a U-shaped relationship between BMI and risk of impairment in instrumental activities of daily living, which are cognition-dependent abilities.
The following mechanisms can explain these findings. First, in older people, BMI is a strong predictor of skeletal muscle mass,20 which seems positively related to cognitive abilities.21 Second, greater BMI is associated with better cardiac output and stroke volume and may thus improve cerebral blood flow and cognition.22 Finally, BMI has an inverted-U relationship with serum levels of insulin-like growth factor (IGF-1), with highest IGF-1 levels observed in those BMI between 25 kg/m2 and 27 kg/m2.23 Higher IGF-1 levels exert beneficial cognitive effects in terms of better frontal-executive functions (set shifting and psychomotor speed)24 and less cognitive decline25 in elderly people. Such evidence supports the findings of this study regarding the nonlinear inverted-U relationship between body weight status and cognition.
These results have potential public health implications. U-shape or J-shape phenomena have been widely discussed in elderly people, including the relationships between blood pressure and cognition,26 blood pressure and mortality,27 and cholesterol levels and quality of life.28 This nonlinearity could also account for inconsistent findings in different studies regarding the relationship between body weight and cognitive ability.7–10 Based on the results of the current study, there is some evidence to suggest that even normal BMI may not represent a desirable state for cognition in older adults.
In addition to the cross-sectional nature of the analysis, which precludes establishing a causal relationship between BMI and cognitive abilities, this study has the following limitations that deserve comment. First, the exclusion of subjects with BMIs less than 18.5 kg/m2 precludes any conclusions regarding the relationship between low BMI and cognitive function. Second, the measures of health conditions as adjusted variables in the analysis were limited to self-report and were not validated through medical record review. Third, the ACTIVE study used a volunteer, not a population-based, sample recruited for cognition-enhancing interventions. The cognition screening scores of the study population represented neither the general population nor a truly cognitively intact sample. Last, although the findings suggest some potential, although small, benefit of elevated BMI in specific cognitive function, it could certainly be due to random effect, and the clinical importance of these findings should be explored prospectively.
In conclusion, being overweight was associated with better cognitive performance in terms of reasoning tasks and visuospatial speed of processing than being normal weight. Being obese was associated with better performance in visuospatial speed of processing than being normal weight. This investigation provides a good rationale to examine the longitudinal effect of BMI on cognition.
Acknowledgments
The ACTIVE study is a multisite collaborative cognitive intervention trial. The National Institute on Aging Scientific Coordinator at the time of award was Jared Jobe and currently is Jeffrey Elias. The National Institute on Nursing Research Scientific Coordinator at the time of award was Taylor Harden and is now Kathy Mann-Koepke. Sharon Tennstedt is the principal investigator at the coordinating center, New England Research Institutes, Watertown, Massachusetts (AG14282). The principal investigators and field sites include Karlene Ball, University of Alabama at Birmingham (AG14289); Michael Marsiske, Institute on Aging, University of Florida, Gainesville (AG14276); John Morris, Hebrew Rehabilitation Center for Aged Research and Training Institute, Boston (NR04507); George Rebok, Johns Hopkins University, Bloomberg School of Public Health (AG14260); and Sherry Willis, Penn State University, Gerontology Center (AG14263). David Smith was the principal investigator at Indiana University School of Medicine, Regenstrief Institute, Indianapolis (NR04508), at the time of initial award; currently, Frederick Unverzagt is the principal investigator.
Financial Disclosure(s): Dr. Kuo is supported by the Men’s Associates of the Hebrew Rehabilitation Center for Aged. Dr. Kuo’s work on this project was supported, in part, by National Institutes of Health (NIH) Grant AG04390, and Dr. Jones’ work was supported in part by the Hebrew Rehabilitation Center for Aged (HRCA) ACTIVE program and NIH Grant AG08812. This research was supported in part by Department of Veterans Affairs (VA) Medical Research Service VA Merit Review Awards to William A. Milberg. John N. Morris holds the HRCA Alfred A. and Gilda Slifka Chair in Social Gerontological Research, and Lewis A. Lipsitz holds the HRCA Irving and Edyth S. Usen and Family Chair in Medical Research.
Footnotes
Author Contributions: study concept and design (Kuo, Jones, Milberg, and Lipsitz), acquisition of subjects (Tennstedt, Talbot, and Morris), analysis and interpretation of data (Kuo and Jones), preparation of manuscript (Kuo and Jones), critical revision (Jones, Milberg, Tennstedt, Talbot, Morris, and Lipsitz).
References
- 1.Flegal KM, Carroll MD, Kuczmarski RJ, et al. Overweight and obesity in the United States: Prevalence and trends, 1960 to 1994. Int J Obes Relat Metab Disord. 1998;22:39–47. doi: 10.1038/sj.ijo.0800541. [DOI] [PubMed] [Google Scholar]
- 2.Roman GC. Vascular dementia Distinguishing characteristics, treatment, and prevention. J Am Geriatr Soc. 2003;51:S296–S304. doi: 10.1046/j.1532-5415.5155.x. [DOI] [PubMed] [Google Scholar]
- 3.Palinkas LA, Wingard DL, Barrett-Connor E. Depressive symptoms in overweight and obese older adults: A test of the ‘jolly fat’ hypothesis. J Psychosom Res. 1996;40:59–66. doi: 10.1016/0022-3999(95)00542-0. [DOI] [PubMed] [Google Scholar]
- 4.Jensen GL, Friedmann JM. Obesity is associated with functional decline in community-dwelling rural older persons. J Am Geriatr Soc. 2002;50:918–923. doi: 10.1046/j.1532-5415.2002.50220.x. [DOI] [PubMed] [Google Scholar]
- 5.Davis JW, Ross PD, Preston SD, et al. Strength, physical activity, and body mass index: Relationship to performance-based measures and activities of daily living among older Japanese women in Hawaii. J Am Geriatr Soc. 1998;46:274–279. doi: 10.1111/j.1532-5415.1998.tb01037.x. [DOI] [PubMed] [Google Scholar]
- 6.Hannestad YS, Rortveit G, Daltveit AK, et al. Are smoking and other lifestyle factors associated with female urinary incontinence? The Norwegian EPIN-CONT Study. BJOG. 2003;110:247–254. [PubMed] [Google Scholar]
- 7.Cattin L, Bordin P, Fonda M, et al. Factors associated with cognitive impairment among older Italian inpatients Gruppo Italiano di Farmacovigilanza nell’Anziano (G.I.F.A) J Am Geriatr Soc. 1997;45:1324–1330. doi: 10.1111/j.1532-5415.1997.tb02931.x. [DOI] [PubMed] [Google Scholar]
- 8.Geroldi C, Ferrucci L, Bandinelli S, et al. Mild cognitive deterioration with subcortical features. Prevalence, clinical characteristics, and association with cardiovascular risk factors in community-dwelling older persons (The InCHI-ANTI Study) J Am Geriatr Soc. 2003;51:1064–1071. doi: 10.1046/j.1532-5415.2003.51353.x. [DOI] [PubMed] [Google Scholar]
- 9.Gustafson D, Rothenberg E, Blennow K, et al. An 18-year follow-up of overweight and risk of Alzheimer disease. Arch Intern Med. 2003;163:1524–1528. doi: 10.1001/archinte.163.13.1524. [DOI] [PubMed] [Google Scholar]
- 10.Nourhashemi F, Deschamps V, Larrieu S, et al. Body mass index and incidence of dementia. The PAQUID Study. Neurology. 2003;60:117–119. doi: 10.1212/01.wnl.0000038910.46217.aa. [DOI] [PubMed] [Google Scholar]
- 11.Ball K, Berch DB, Helmers KF, et al. Effects of cognitive training interventions with older adults: A randomized controlled trial. JAMA. 2002;288:2271–2281. doi: 10.1001/jama.288.18.2271. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Obesity Education Initiative: The Evidence Report. Bethesda, MD: National Heart Lung and Blood Institute, National Institutes of Health; 1998. [Google Scholar]
- 13.Brandt J. The Hopkins Verbal Learning Test. Development of a new memory test with six equivalent forms. Clin Neuropsychol. 1991;5:125–142. [Google Scholar]
- 14.Rey A. L’examen psychologique dans les cas d’encephalopathie tramatique. Arch Psychol (Frankf) 1941;28:21. [Google Scholar]
- 15.Wilson BA, Cockburn J, Baddeley A. The Rivermead Behavioral Memory Test. Reading, England: Thames Valley Test Co; 1985. [Google Scholar]
- 16.Kolb B, Whishaw IQ. Fundamentals of Human Neuropsychology. New York: WH Freeman; 1996. [Google Scholar]
- 17.Owsley C, Ball K, McGwin G, et al. Visual processing impairment and risk of motor vehicle crash among older adults. JAMA. 1998;279:1083–1088. doi: 10.1001/jama.279.14.1083. [DOI] [PubMed] [Google Scholar]
- 18.Chobanian AV, Bakris GL, Black HR, et al. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. The JNC 7 Report. JAMA. 2003;289:2560–2572. doi: 10.1001/jama.289.19.2560. [DOI] [PubMed] [Google Scholar]
- 19.Deschamps V, Astier X, Ferry M, et al. Nutritional status of healthy elderly persons living in Dordogne, France, and relation with mortality and cognitive or functional decline. Eur J Clin Nutr. 2002;56:305–312. doi: 10.1038/sj.ejcn.1601311. [DOI] [PubMed] [Google Scholar]
- 20.Iannuzzi-Sucich M, Prestwood KM, Kenny AM. Prevalence of sarcopenia and predictors of skeletal muscle mass in healthy, older men and women. J Gerontol A Biol Sci Med Sci. 2002;57A:M772–M777. doi: 10.1093/gerona/57.12.m772. [DOI] [PubMed] [Google Scholar]
- 21.Nourhashemi F, Andrieu S, Gillette-Guyonnet S, et al. Is there a relationship between fat-free soft tissue mass and low cognitive function? Results from a study of 7,105 women. J Am Geriatr Soc. 2002;50:1796–1801. doi: 10.1046/j.1532-5415.2002.50507.x. [DOI] [PubMed] [Google Scholar]
- 22.Jern S, Bergbrant A, Bjorntorp P, et al. Relation of central hemodynamics to obesity and body fat distribution. Hypertension. 1992;19:520–527. doi: 10.1161/01.hyp.19.6.520. [DOI] [PubMed] [Google Scholar]
- 23.Yamamoto H, Kato Y. Relationship between plasma insulin-like growth factor I (IGF-I) levels and body mass index (BMI) in adults. Endocr J. 1993;40:41–45. doi: 10.1507/endocrj.40.41. [DOI] [PubMed] [Google Scholar]
- 24.Aleman A, Verhaar HJ, De Haan EH, et al. Insulin-like growth factor-I and cognitive function in healthy older men. J Clin Endocrinol Metab. 1999;84:471–475. doi: 10.1210/jcem.84.2.5455. [DOI] [PubMed] [Google Scholar]
- 25.Kalmijn S, Janssen JA, Pols HA, et al. A prospective study on circulating insulin-like growth factor I (IGF-I), IGF-binding proteins, and cognitive function in the elderly. J Clin Endocrinol Metab. 2000;85:4551–4555. doi: 10.1210/jcem.85.12.7033. [DOI] [PubMed] [Google Scholar]
- 26.Paran E, Anson O, Reuveni H. Blood pressure and cognitive functioning among independent elderly. Am J Hypertens. 2003;16:818–826. doi: 10.1016/s0895-7061(03)01005-7. [DOI] [PubMed] [Google Scholar]
- 27.Okumiya K, Matsubayashi K, Wada T, et al. A U-shaped association between home systolic blood pressure and four-year mortality in community-dwelling older men. J Am Geriatr Soc. 1999;47:1415–1421. doi: 10.1111/j.1532-5415.1999.tb01559.x. [DOI] [PubMed] [Google Scholar]
- 28.Shibata H. Nutritional factors on longevity and quality of life in Japan. J Nutr Health Aging. 2001;5:97–102. [PubMed] [Google Scholar]