| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
RESEARCH ARTICLE |
a Department of Medicine, Division of Gerontology, University of Maryland School of Medicine, and Geriatric Research, Education and Clinical Center (GRECC), Baltimore Veterans Affairs Medical Center
b Department of Psychology, University of Maryland, Baltimore County
Shari R. Waldstein, Department of Psychology, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250 E-mail: waldstei{at}umbc.edu.
Decision Editor: Margie E. Lachman, PhD
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
O2). Multiple regression analyses with age, education, number of antihypertensive medications, HbA1c, diastolic BP, and peak O2 as predictors revealed significant (and marginally significant) associations between lower levels of fitness (peak O2) and poorer executive functions and delayed verbal memory. Antihypertensive medications were associated with poorer attention, but better delayed verbal memory. In addition, greater levels of HbA1c were positively related to attention. These results suggest that cardiovascular risk factors are important predictors of cognitive function among middle-aged and older African Americans.
Amore pronounced decline in cognitive function with increasing age has been noted in African Americans as compared with Caucasians (Lyketsos, Chen, and Anthony 1999
; Whitfield et al. 2000). Such accelerated cognitive aging among African Americans may be explained, in part, by indices of poorer physical health. Poorer health status may also play a critical role in explaining cross-sectional cognitive performance discrepancies between African Americans and Caucasians (Whitfield et al. 1997). In this regard, health disparities between African Americans and Caucasians are well documented, with African Americans displaying a higher incidence and prevalence of numerous chronic diseases that are known to have deleterious effects on cognition (Waldstein 2000; Williams 2000). For example, a variety of cardiovascular risk factors and cardiovascular diseases are associated with lower levels of cognitive function; examples include hypertension, diabetes mellitus, high cholesterol, myocardial infarction, and lower levels of physical activity or fitness (for reviews see Dustman, Emmerson, and Shearer 1994; Waldstein and Elias 2001). Indices of pulmonary function, such as forced expiratory volume, have also been identified as significant longitudinal predictors of cognitive performance (Albert et al. 1995; Chyou et al. 1996). Interestingly, such measures are influenced greatly by smoking, which is also a major cardiovascular risk factor.
Research on health and cognition has focused predominantly on Caucasian samples, and there is little information available on the relation of physical health status to cognition among African Americans. In one of the few available studies, Whitfield and colleagues 1997 noted an association between lower levels of average peak expiratory flow and cognitive decline in older African Americans. However, in a more recent investigation, Whitfield and colleagues 2000 found an association between lower levels of peak expiratory flow and poorer cognitive function only in an older European American sample, whereas number of self-reported chronic health conditions predicted lower levels of cognition among African Americans. In contrast, Manly and colleagues 1998 did not find health-related variables such as hypertension and diabetes to account for discrepancies in cognitive function between elderly African Americans and Caucasians. However, these investigators relied on self-reported history of medical conditions, which may either underestimate or fail to identify existing associations between health status and cognition (Waldstein 2000). The goal of this preliminary investigation was to examine the influence of medically determined cardiovascular risk factors on cognitive function in a sample of middle-aged to older, nondemented, stroke-free African Americans.
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Participants were predominantly female (83%) African Americans, aged 43–82 years (M = 59; SD = 11.2) with 14 ± 2.5 years of formal education and an average of 2.8 ± 1.9 self-reported cardiovascular risk factors: 53% hypertensive, 19% diabetic, 28% obese, and 22% dyslipidemic. There were no current smokers and 17% former smokers. Eighty-three percent of the women were post-menopausal. Participants took an average of 0.6 prescription drugs, and 44% of the sample was treated with antihypertensive medications. Medical assessment of cardiovascular risk factors yielded the following means and standard deviations: systolic blood pressure (BP) = 136 (21) mm Hg, diastolic BP = 78 (11) mm Hg, total cholesterol = 210 (38) mg/dl, fasting glucose = 110 (25) mg/dl, glycosylated hemoglobin (HbA1c) = 0.5 (0.5) mg/dl, and peak O2 = 21.1 (7.1) ml/kg/min.
Procedures
Demographic, biomedical, and psychological assessment
We collected self-reports of demographic information (e.g., age, education) and medical and pharmacological history. By using the Center for Epidemiologic Studies–Depression Scale (CES-D; Radloff 1977), we assessed depressive symptomatology. Clinical assessment of BP was completed with an automated Dinamap Vital Signs Monitor (Model #1846-SX, Critikon, Inc., Tampa, FL). We obtained and averaged three sequential resting measures for the present analyses, and drew a fasting blood sample into chilled ethylenediaminetetraacetic acid (EDTA) for later enzymatic determination of total cholesterol, plasma glucose, and glycosylated hemoglobin (HbA1c). Peak oxygen consumption (O2) was assessed by indirect calorimetry during a progressively graded treadmill test that we terminated upon maximal fatigue of the participant. By averaging the last two O2 determinations, we determined peak O2, a measure of oxygen uptake by skeletal muscle (i.e., the ability of the muscle to extract oxygen from the blood) during exercise that indexes fitness level.
Neuropsychological assessment
We administered the following cognitive tests in fixed order: (a) The Digits Forward and Digits Backward portions (total number of correct trials, respectively) of the Digit Span subscale of the Wechsler Adult Intelligence Scale–Revised (Wechsler 1981), (b) Word List Learning Test (Morris et al. 1989) assessed free recall of a 10-item word list three consecutive times and at 7-min delayed recall, (c) EXIT 25 (Royall, Mahurin, and Gray 1992), a 25-item screening of executive functioning, and (d) CLOX (Royall, Cordes, and Polk 1998), a clock drawing task that involves first drawing a clock without guidance (CLOX draw) and then copying the clock after a demonstration.
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
O2, and total cholesterol) and cognitive test performance after adjusting for age and education. We noted significant (or marginally significant) correlations between HbA1c and Digits Forward (r = .55; p = .001), peak O2 and Digits Backward (r = –.31; p = .09), and diastolic blood pressure and Digits Forward (r = .37; p = .04). Systolic BP and total cholesterol did not correlate significantly with any cognitive test scores.
Next, to examine the multivariate relation of cardiovascular risk factors to cognitive function, we computed a series of multiple regression analyses. Scores on the cognitive tests were criterion variables, and age, education, number of antihypertensive medications, HbA1c, diastolic BP, and peak O2 were predictor variables. Results are depicted in Table 1 . We noted a significant overall regression model for Digits Forward, F(6,29) = 4.85, p = .002, r2 = .50: HbA1c predicted better performance on this test ( p < .01), whereas antihypertensive medication use was associated with lower Digits Forward performance (p < .02). We found a significant overall regression model for EXIT 25 performance, F(6,29) = 3.47, p = .01, r2 = .42; HbA1c marginally predicted better performance on this test (p < .07). We also noted significant regression models for Immediate Word Recall, F(6,29) = 4.22, p = .004, r2 = .46, and Delayed Word Recall, F(6,29) = 4.00, p = .005, r2 = .45. Specifically, antihypertensive medication use was associated with higher scores on Delayed Word Recall (p < .04), and peak O2 was marginally related to poorer scores on Delayed Word Recall (p < .08). There was a trend toward a significant overall regression model for CLOX draw, F(6,29) = 2.22, p = .07, r2 = .32. Peak O2 was significantly associated with poorer performance on this test (p < .03). Finally, the overall regression model for Digits Backward was nonsignificant, F(6,29) = 1.88, p = .12, r2 = .28. The present findings were unchanged when we introduced depression (CES-D) scores or history of chronic obstructive pulmonary disease into the models.
|
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
O2 to poorer performance on select tests of verbal memory and executive function after we controlled for age, education, depression, and use of antihypertensive medications. Peak O2 independently predicted 7% to 14% of the variance in these cognitive measures. These findings are consistent with many (though not all) prior cross-sectional studies of physical fitness or physical activity and cognition (see Dustman et al. 1994) in which similar associations between these variables and measures of verbal memory and executive function were noted. Use of antihypertensive medications predicted poorer attention, but better delayed verbal memory. Although prior randomized clinical trials of antihypertensive medications generally suggest minimal to no impact on cognitive function, overall results of these studies and cross-sectional investigations have been mixed (Muldoon, Manuck, Shapiro, and Waldstein 1991). Whereas improved blood pressure control associated with antihypertensive medication use may be beneficial to certain dimensions of cognitive function, other aspects of these medications may be related to diminished performance. Next, HbA1c was related to better performance on a measure of attention and marginally associated with better executive function. These latter findings are largely inconsistent with prior research in predominantly Caucasian, diabetic samples (see Ryan 2001). However, it is important to note that relatively few of the present participants were diabetic, and this small sample may have had other unique characteristics that could have influenced these associations. Finally, the general absence of an independent relation between BP and cognitive performance is also inconsistent with prior work (Waldstein 1995).
Strengths of this preliminary investigation include direct biomedical assessment of cardiovascular risk factors and examination of an understudied population that is prone to both cognitive and physical decline with age. Study limitations include a small sample size, thus limiting statistical power. In this regard, significant individual predictors associated with the CLOX test were interpreted despite the fact that the overall regression model was only marginally significant (p < .07). These findings thus need to be viewed with caution. Nevertheless, it is important to consider the magnitude of the effect sizes (i.e., medium) associated with even the marginally significant findings. Next, several of our cognitive measures (e.g., CLOX, EXIT 25) are considered screening tools and may not have been sufficiently challenging to elicit maximum variability in test performance. In addition, BP was measured on only one day. Enhanced reliability of BP measurement across days, or an index of chronic BP elevation, may strengthen any associations with cognition (Waldstein 2000). Finally, some studies have suggested use of a lower MMSE cutoff for disability among African Americans than that employed here (Bohnstedt, Fox, and Kohatsu 1994). Our sample may therefore be particularly high functioning, which is one of many potential selection biases that limit the generalizability of these findings. For all of these stated reasons, the present study may actually underestimate the relation of cardiovascular risk factors to cognition among African Americans.
The present findings are, nonetheless, intriguing, and suggest that the relation between medically determined cardiovascular risk factors and cognitive performance should be further explored among a larger sample of African Americans by using a more extensive battery of neuropsychological tests and more challenging measures of cognitive function. Potential moderating influences of gender should also be examined. The deleterious influence of certain cardiovascular risk factors may help to explain the more pronounced cognitive aging that has been noted among African Americans (Lyketsos et al. 1999). In addition, because African Americans have a greater prevalence and severity of cardiovascular risk factors, these factors may also play an important role in explaining cross-sectional discrepancies in cognitive performance as a function of ethnicity (Whitfield et al. 1997).
|
Acknowledgments |
|---|
Received for publication October 31, 2000. Accepted for publication September 27, 2001.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  G. W. Rebok, M. C. Carlson, and J. B. S. Langbaum Training and Maintaining Memory Abilities in Healthy Older Adults: Traditional and Novel Approaches J. Gerontol. B. Psychol. Sci. Soc. Sci., June 1, 2007; 62(suppl_Special_Issue_1): 53 - 61. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Morrow, D. Clark, W. Tu, J. Wu, M. Weiner, D. Steinley, and M. D. Murray Correlates of health literacy in patients with chronic heart failure. Gerontologist, October 1, 2006; 46(5): 669 - 676. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. A. Sloan and J. Wang Disparities Among Older Adults in Measures of Cognitive Function by Race or Ethnicity J. Gerontol. B. Psychol. Sci. Soc. Sci., September 1, 2005; 60(5): P242 - P250. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Robbins, M. F. Elias, P. K. Elias, and M. M. Budge Blood Pressure and Cognitive Function in an African-American and a Caucasian-American Sample: The Maine-Syracuse Study Psychosom Med, September 1, 2005; 67(5): 707 - 714. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Birns, H. Markus, and L. Kalra Blood Pressure Reduction for Vascular Risk: Is There a Price To Be Paid? Stroke, June 1, 2005; 36(6): 1308 - 1313. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
