Professional Documents
Culture Documents
, Bustamante, E. E., Manning, A. F. (2012). The relationship between physical activity and cognition in
Staffileno,
older Latinos. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 10.1093/geronb/gbr137
67(5), 525–534, doi:10.1093/geronb/gbr137. Advance Access published on February 9, 2012
1Women, Children, and Family Nursing, College of Nursing, Rush University, Chicago, Illinois.
Objectives. The purpose of this study was to explore the relationship between minutes spent participating in light and
moderate/vigorous-intensity physical activity (PA) and cognition in older Latinos, controlling for demographics, chronic
health problems, and acculturation.
Method. A cross-sectional study design was used. Participants were self-identified Latinos, without disability, who
had a score less than 14 on a 21-point Mini-Mental State Examination. Participants were recruited from predominantly
Latino communities in Chicago at health fairs, senior centers, and community centers. PA was measured with an acceler-
ometer, worn for 7 days. Episodic memory and executive function (inference control, inattention, and word fluency) were
measured with validated cognitive tests.
Results. Participants were 174 Latino men (n = 46) and women (n = 128) aged 50–84 years (M = 66 years). After
adjusting for control variables (demographics, chronic health problems) and other cognitive measures, regression analy-
ses revealed that minutes per day of light-intensity PA (r = −.51), moderate/vigorous PA (r = −.56), and counts per minute
(r=−.62) were negatively associated with lower word fluency.
Discussion. Findings suggest that the cognitive benefits of both light-intensity PA and moderate/vigorous PA may be
domain-specific.
Closer scrutiny of these 10 studies revealed selected alternative to self-report (Hawkins et al., 2009). They are
measurement issues that may have particular bearing for small and unobtrusive and can be worn throughout the
Latinos. First, all relied on self-report measures that focused day without interfering with everyday movement. Another
heavily on leisure-time or structured PA obtained through unique feature is their ability to capture light-intensity PAs,
exercise and sport. According to 2009 National Health In- such as household cleaning, which is difficult to recall on
terview Survey, however, 44% of adult Latinos versus 28% self-report questionnaires. Although it is well-established
of non-Latino White adults reported no light, moderate, or that moderate-intensity PA reduces the risk of cardiovascu-
vigorous leisure PA (Roger et al., 2011). Furthermore, data lar disease and diabetes (U.S. Department of Health and
from Latinos indicate that the prevalence of leisure-time PA Human Services, 2008), recent work has shown that light-
is lowest among immigrants, and, although it increases over intensity PA is positively associated with physical health
and obeying a command, writing a sentence, and copying a was taken with the Omron HEM-907XL automated blood
design. To date 1,451 older people in the Rush Memory pressure machine (Omron Health Care Inc., Vernon Hills,
and Aging Project have completed the standard 30-point IL). It has reliability and reproducibility of mercury sphyg-
MMSE as part of their baseline evaluation (Bennett et al., momanometer measurements (El Assaad, Topouchian, Darne,
2005). The Pearson correlation between the standard version & Asmar, 2002). Blood pressure measurements followed
and the 21-point MMSE is 0.968 (p < .001). Previous studies the American Heart Association Recommendations for
on cognition and aging using the 30-point MMSE with par- human blood pressure determination (Pickering et al.,
ticipants of low-educational attainment have identified one- 2005), and the average of three measurements was recorded.
third incorrect as a cutpoint for impaired/poor cognition (Raji, Consistent with National Health and Nutrition Examination
Al Snih, Ostir, Markides, & Ottenbacher, 2010); therefore, Survey (NHANES), a participant was identified as having
began data collection. The cognitive function tests are The ActiGraph monitor provides a valid assessment of PA
administered in the following fixed order: East Boston in adult men and women during treadmill walking/running
Memory Test (immediate recall), Stroop Neuropsychological and daily activity (Hendelman, Miller, Baggett, Debold, &
Screening Test, East Boston Memory Test (delayed recall), Freedson, 2000). It records vertical accelerations as “counts.”
numbers comparison, and category fluency. Participants were instructed to wear the monitor on their
hips for seven consecutive days. Participants were given an
East Boston Memory Test. Episodic memory was mea- accelerometer log on which they recorded the hours they
sured with the East Boston Memory Test, which involves wore it. The monitor was initialized and set to collect count
immediate and delayed recall of a brief story (Albert et al., values at 1-min sampling intervals (epochs). To be included
1991). The story, which has 12 key elements, was read in the analyses, ≥10 hr of data or wear time had to be available
Table 1. Background Characteristics correlation analyses to determine how the dependent vari-
n % ables were related to the predictors.
Age
50–59 54 31.0
60–69 52 29.9
Results
70–79 51 29.3
≥80 17 9.8 Background Characteristics
Education The mean age of the participants was 66 years (SD =
Did not complete high school 99 56.9 9.12; range 50–84 years; Table 1). The majority of the
Completed high school or GED 29 16.7
Some college or vocational school 26 14.9 participants were women (74%), unmarried (59%), and not
tests that remained. At the end of the second visit, partici- Correlations Among Background Characteristics, PA, and
pants were given the oral and written results of their blood Cognitive Function
pressure, weight, and height information and PA informa- Age was significantly negatively correlated with light-
tion from the American Heart Association in Spanish or intensity PA, minutes of moderate/vigorous-intensity PA,
English. Participants were compensated $10 at the end of and activity counts (Table 2). Correlations of small magni-
the first appointment and received another $10 at the end of tude were found between chronic health problems and
the second appointment. minutes of light-intensity PA a day (r = −.221). There was a
correlation between sex and moderate/vigorous-intensity
Analyses PA, with men having more minutes in moderate/vigorous-
All analyses were conducted using SPSS V16.0. The intensity PA than women.
bivariate correlations between background/control measures Correlations were assessed between the PA measures and
and PA with the cognitive measures were conducted to cognitive measures. In addition, we looked at the composite
determine if further multiple regression analysis was justified. scores for episodic memory (Boston immediate and Boston
Next, a multivariate multiple regression was done with delay) and PA measures. The pattern of correlations for
episodic memory, interference control, inattention, and the the component and summary/composite scores for episodic
two-word fluency measures (animals and fruits and vegeta- memory was nearly identical across all three PA measures.
bles) as the dependent variables and the background and PA Consequently, the subsequent analyses examined only the
variables as predictors. This was followed by a canonical episodic composite.
Table 2. Correlations Among Background Characteristics, Physical Activity (PA), and Cognition
M (SD) 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.
Background characteristics
1. Age
2. Sexa −.012
3. Educationb −.268*** .000
4. Chronic health problems .296*** −.203* −.117
5. Acculturation −.020 −.010 .277*** .176*
PA
6. Lightc 259.4 (73.3) −.438*** .096 .139 −.221** .087
7. Moderate/vigorousc 31.2 (30.5) −.318*** −.255** −.036 −.093 .036 .321***
8. Counts per minute 267.0 (181.9) −.280*** −.112 −.022 −.151 .041 .368*** .865***
Cognition
9. Boston immediate 8.8 (2.0) −.000 .165* .054 −.121 .078 .085 .013 .014
10. Bostondelay 8.6 (2.0) −.023 .216** .076 −.117 .112 .091 .056 .067 .771***
6530 WILBUR
WILBUR
11. Episodic memory 8.7 (2.0) −.012 .203* .069 −.126 .101 .093 .037 .043 .939*** .943***
12. Interference control 14.8 (9.5) −.486*** .026 .258** −.140 .192* .234** .150 .137 .295*** .319*** .327***
13. Inattention 16.7 (7.3) −.311*** .002 .494*** −.086 .244** .119 .062 .058 .197* .139 .178* .495***
14. Number of animals 17.2 (4.5) −.247** .009 .244** −.040 .028 .253** .231** .240** .276** .277** .294*** .389*** .401***
15. Number of fruits and 17.1 (5.2) −.211** .306*** .204* −.164* .029 .246** .079 .114 .204* .263** .249** .334*** .341*** .574***
vegetables
16. Word fluency 34.3 (8.7) −.260** .165* .254** −.109 .032 .281*** .181* .205* .273*** .305*** .308*** .410*** .421*** .868*** .905***
Notes. n = 151.
a 1. = men; 2. = women.
b 1. = no high school degree; 2. = high school degree or higher.
c Minutes per day at named intensity.
Table 3. The Correlations Between Background and Physical Table 3 also displays the extent to which the four func-
Activity (PA) Predictor Variables and Canonical Functions tions are correlated with cognitive functioning. Functions
Canonical functions 1 and 3 both predict interference control, and inattention
Predictor variables 1 2 3 4 Function 2 is most correlated with episodic memory and
Background word fluency (fruits and vegetables recalled). Function 4 is
Age .81 −.04 .54 .05 most highly correlated with the two measures of word flu-
Sex .05 −.92 −.12 .04
Education −.72 −.13 .63 −.10
ency, the number of fruit and vegetables recalled (r = −.44)
Chronic health problems .19 −.41 .18 .15 and the number of animals recalled (r =−.83). Function 4 is
Acculturation −.38 −.07 .21 .40 composed of the three measures of PA and one measure of
PA acculturation. From these results, it is clear that there is a
We examined correlations between all study variables to provide cognitive benefits for word fluency as does moder-
identify congruence with prior studies on age, education, ate/vigorous PA. This could be particularly important for
sex, PA, and cognition. As expected, age was negatively older adults who, like participants in our study, tend to
correlated and education positively correlated with the spend more time participating in light-intensity than more
majority of the dimensions of cognitive function (Plassman vigorous activities. We must keep in mind that our estimates
et al., 2007). The lack of a correlation between age and epi- between moderate PA and cognition may be somewhat mis-
sodic memory was surprising. There is evidence suggesting leading because of the floor effects. It may be that increases
that frequent participation in cognitively stimulating activ- in moderate PA will lead to substantial increases in cogni-
ities is associated with reduced risk of dementia (Wilson tive functioning, but we were not able to find this relation-
et al., 2005). Therefore, we speculate that the rich tradition ship due to the limited range of moderate/vigorous PA in
Albert, M. S., Smith, L. A., Scherr, P. A., Taylor, J. O., Evans, D. A., & Greenberg, S. (2009). A profile of older Americans: 2009. Retrieved from
Funkenstein, H. H. (1991). Use of brief cognitive tests to identify http://www.aoa.gov/AoAroot/Aging_Statistics/Profile/2009/docs/
individuals in the community with clinically diagnosed Alzheimer’s 2009profile_508.pdf
disease. International Journal of Neuroscience, 57, 167–178. Haan, M. N., Mungas, D. M., Gonzalez, H. M., Ortiz, T. A., Acharya, A., &
doi:10.3109/00207459109150691 Jagust, W. J. (2003). Prevalence of dementia in older Latinos: The
Alzheimer’s Association. (2010). 2010 Alzheimer’s disease facts and influence of type 2 diabetes mellitus, stroke and genetic factors. Journal
figures. Alzheimer’s and Dementia, 6, 158–194. doi:10.1016/ of American Geriatrics Society, 51, 169–177. doi:10.1046/j.1532-
j.jalz.2010.01.009 5415.2003.51054.x
Bachman, D. L., Wolf, P. A., Linn, R., Knoefel, J. E., Cobb, J., Belanger, A., Ham, S. A., & Ainsworth, B. E. (2010). Disparities in data on Healthy
. . . White, L. R. (1992). Prevalence of dementia and probable People 2010 physical activity objectives collected by accelerometry
senile dementia of the Alzheimer type in the Framingham Study. and self-report. American Journal of Public Health, 100, S263–S268.
the United States: The aging, demographics, and memory study. Trenerry, M., Crosson, B., DeBoe, J., & Leber, W. (1989). Stroop Neuro-
Neuroepidemiology, 29, 125–132. doi:10.1159/000109998 psychological Screening Test manual. Adessa, FL: Psychological
Raji, M. A., Al Snih, S., Ostir, G. V., Markides, K. S., & Ottenbacher, K. J. Assessment Resources (PAR).
(2010). Cognitive status and future risk of frailty in older Mexican Troiano, R. P., Berrigan, D., Dodd, K. W., Masse, L. C., Tilert, T., &
Americans. The Journals of Gerontology, Series A: Biological McDowell, M. (2008). Physical activity in the United States measured
Sciences and Medical Sciences, 65, 1228–1234. doi:10.1093/gerona/ by accelerometer. Medicine and Science in Sports and Exercise, 40,
glq121 181–188. doi:10.1249/mss.0b013e31815a51b3
Richards, M., Hardy, R., & Wadsworth, M. E. (2003). Does active leisure Tryon, W. W., & Williams, R. S. (1996). Fully proportional actigraphy:
protect cognition? Evidence from a national birth cohort. Social A new instrument. Behavior Research Methods, Instruments, and
Science and Medicine, 56, 785–792. doi:S0277953602000758 Computers, 28, 392–403. doi:10.3758/BF03200519
Roger, V. L., Go, A. S., Lloyd-Jones, D. M., Adams, R. J., Berry, J. D., U.S. Department of Health and Human Services. (2008). 2008 Physical