Better brain, <20 BMI

[Guest guest]

Hi All,

We CRers expect, based on empirical data, believe that we will

live longer. In our elder years, what will be the

quality of our lives? The retention of our

http://en.wikipedia.org/wiki/Cognitive abilities

largely will probably help govern our

http://en.wikipedia.org/wiki/Quality_of_life in later

years. The association of CR with our lower

http://en.wikipedia.org/wiki/Body_mass_index (BMI) has

cast CRers to be

http://en.wikipedia.org/wiki/Underweight for our age

and sex. A new report appears to find that the

quintile of subjects of the study that are underweight

by the BMI <20 description of underweight to have the

best performance in cognition.

http://tinyurl.com/hkepj was the popular press report,

beginning:

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

Blog

Back of the Pack

A fitting way to age

Tuesday, October 10, 2006 | 04:53 PM ET

By Hadzipetros

If you are terrified at the thought of spending your

final days in a nursing home—unable to manage much

more than counting floor tiles—you might want to pick

up the latest issue of Neurology, the official journal

of the American Academy of Neurology.

.... One article-- Relation between body mass index and

cognitive function in healthy middle-aged men and

women -- featured a study that looked at the

connection between body mass index (BMI) and the

ability to think, learn, and remember over time. More

than 2,000 healthy men and women in France were

divided into five groups, according to their BMI in

1996. They were given a series of memory tests. Five

years later, the process was repeated.

....

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

http://tinyurl.com/fcbvq is the URL for the

not-yet-in-Medline publication abstract:

NEUROLOGY 2006;67:1208-1214

Relation between body mass index and cognitive

function in healthy middle-aged men and women

M. Cournot, MD, J. C. Marquié, PhD, D. Ansiau, PhD, C.

aud, MD, H. Fonds, MD, J. Ferrières, MD, MSc,

FESC and J. B. Ruidavets, MD

Objective: To assess whether body mass index (BMI) is

associated with cognitive function and cognitive

decline in healthy men and women.

Methods: In this prospective cohort study, we analyzed

data from 2,223 healthy workers aged 32 to 62 years at

baseline. Medical, psychosocial, and environmental

data were collected in 1996 and in 2001. We tested

cognitive functions at baseline and at follow-up with

word-list learning (four recalls), a Digit–Symbol

Substitution Test, and a selective attention test.

Results: Cross-sectionally, a higher BMI was

associated with lower cognitive scores after

adjustment for age, sex, educational level, blood

pressure, diabetes, and other psychosocial

covariables. A higher BMI at baseline was also

associated with a higher cognitive decline at

follow-up, after adjustment for the above-cited

confounding factors. This association was significant

for word-list learning. For the changes in scores at

word-list learning (delayed recall), regression

coefficients were –0.008±0.13, –0.09±0.13, –0.17±0.14,

and –0.35±0.14 (p for trend < 0.001) for the second,

third, fourth, and fifth quintiles of BMI at baseline

when compared with the first quintile. No significant

association was found between changes in BMI and

cognitive function.

Conclusions: Body mass index was independently

associated both with cognitive function (word-list

learning and Digit–Symbol Substitution Test) and

changes in word-list learning in healthy, nondemented,

middle-aged men and women.

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

The tables from the publication are:

Table 1 Characteristics of the sample according to

quintiles of BMI at baseline

=====================================================================

Quintile 1 (15–21.5), n=445 Quintile 2

(21.5–23.4), n=445 Quintile 3 (23.4 –25.2), n=444

Quintile 4 (25.2–27.7), n=445 Quintile 5 (27.7– 45),

n=444 Total, n=2,223 p^*Trend

=====================================================================

Women (%) 340 (76.2) 266 (59.6) 175 (39.6) 132 (29.7)

169 (38.1) 1,080 (48.6) <0.001

Age at baseline (%) <0.001

32 years 187 (42.2) 133 (29.8) 127 (28.5) 98 (22.0)

89 (20.0) 634 (28.5)

42 years 153 (34.3) 160 (35.9) 129 (29.2) 143

(32.1) 119 (26.8) 704 (31.7)

52 years 73 (16.4) 110 (24.7) 123 (27.6) 123 (27.6)

161 (36.3) 589 (26.5)

62 years 32 (7.2) 43 (9.6) 65 (14.7) 81 (18.2) 75

(16.9) 296 (13.3)

Educational level (%) <0.001

</=9 years 78 (17.5) 88 (19.8) 97 (21.7) 106 (23.8)

159 (35.9) 528 (23.7)

10–12 years 74 (16.3) 84 (18.9) 82 (18.3) 79 (17.8)

91 (20.3) 409 (18.4)

>12 years 293 (65.8) 273 (61.4) 265 (60.0) 260

(58.4) 194 (43.8) 1,286 (57.9)

Occupation (%) <0.001

Craftsman, manager 35 (7.9) 37 (8.3) 71 (16.0) 63

(14.1) 38 (8.6) 244 (11.0)

Intermediate 147 (33.0) 153 (34.4) 139 (31.3) 134

(30.1) 117 (26.4) 690 (31.0)

White collar 177 (39.8) 169 (38.0) 137 (31.0) 142

(32.0) 146 (32.9) 771 (34.7)

Blue collar 86 (19.3) 86 (19.3) 97 (21.7) 106

(23.8) 143 (32.1) 518 (23.3)

BMI, kg/m2 19.8 </= 1.2 22.5 </= 0.6 24.3 </= 0.5 26.3

</= 0.8 30.5 </= 2.8 24.7 </= 3.8 —

Changes in BMI (between baseline and follow-up) +0.83

</= 1.5 +0.55 </= 1.4 +0.47 </= 1.5 +0.65 </= 1.7

+0.40 </= 2.2 +0.58 </= 1.7 <0.001

Systolic BP, mm Hg 121 </= 13 126 </= 15 128 </= 14

133 </= 14 136 </= 15 129 </= 15 <0.001

Diastolic BP, mm Hg 74 </=9 76 </= 10 77 </= 10 80 </=

10 82 </= 10 77 </= 10 <0.001

Diabetes (%) 4 (0.9) 5 (1.1) 6 (1.4) 10 (2.2) 20 (4.5)

45 (2.0) 0.004

Current smoking (%) 164 (36.8) 139 (31.2) 121 (27.2)

131 (29.4) 119 (26.8) 674 (30.2) 0.005

Daily alcohol consumption (%) 74 (16.6) 122 (27.4) 144

(32.4) 158 (35.5) 142 (32.0) 640 (28.8) <0.001

Physical activity (%) <0.001

None to < twice a week 220 (49.4) 188 (42.2) 197

(44.4) 225 (50.6) 249 (56.1) 1,029 (46.3)

Twice a week or more 225 (50.6) 257 (57.8) 247 (55.6)

220 (49.4) 195 (43.9) 1,194 (53.7)

Word-list learning words

Recall 1 6.3 </= 1.9 6.2 </= 1.9 6.0 </= 1.9 5.7

</= 1.8 5.5 </= 1.9 5.9 </= 1.9 <0.001

Recall 2 8.8 </= 2.5 8.6 </= 2.4 8.4 </= 2.4 8.0

</= 2.4 7.8 </= 2.3 8.3 </= 2.4 <0.001

Recall 3 10.6 </= 2.7 10.2 </= 2.6 10.1 </= 2.5 9.5

</= 2.6 9.2 </= 2.6 9.9 </= 2.6 <0.001

Delayed recall 8.6 </= 2.8 7.8 </= 2.8 7.8 </= 2.7

7.2 </= 2.7 6.9 </= 2.7 7.6 </= 2.8 <0.001

DSST score 55.0 </= 15.1 53.2 </= 13.8 51.3 </= 13.9

49.0 </= 13.8 47.1 </= 14.5 51.2 </= 14.5 <0.001

Selective attention, seconds 201.3 </= 90 201.9 </= 87

208.0 </= 100 219.5 </= 101 226.1 </= 100 211.4 </=

96.5 <0.001

Perceived health 8 (6–9) 8 (6–9) 7.5 (5–9) 7.5 (5–9) 7

(5–9) 7 (5–9) 0.17

Nottingham Health Profile score/100

Energy 18.5 </= 28.9 15.4 </= 26.6 14.1 </= 26.9

14.7 </= 26.5 18.9 </= 21.4 16.3 </= 28.2 0.03

Emotional reactions 17.0 </= 18.5 14.5 </= 16.6

14.1 </= 17.6 13.4 </= 17.0 15.8 </= 18.8 15.0 </=

17.7 0.02

Social isolation 6.5 </= 14.8 5.7 </= 13.6 4.7 </=

13.4 5.8 </= 13.9 6.0 </= 14.5 5.7 </= 14.0 0.15

Perceived stress score/20 8.8 </= 2.9 8.2 </= 2.8 8.1

</= 2.9 8.2 </= 2.9 8.4 </= 3.0 8.4 </= 2.9 0.002

=====================================================================

BMI=body mass index; BP=blood pressure;

DSST=Digit–Symbol Substitution Test.

Table 2 Relationship between BMI and cognitive test

scores at follow-up in univariate and multivariate

analyses (linear regression)

==================================================================

-

--- -

--- Selective attention time, ß±SE

Word-list learning, delayed recall, number of

cited words, ß±SE---Digit–Symbol Substitution Test

score, ß±SE---Multivariate

Univariate Multivariate---Univariate

Multivariate---Univariate

---Men Women

==================================================================

Age^# -0.92 </= 0.05^* -0.68 </= 0.05^* -5.97 </=

0.33^* -4.37 </= 0.33^* +20.96 </= 1.67^* +20.31 </=

0.50^* +16.41 </= 2.71^*

Female sex (vs male) +0.86 </= 0.12^* +0.62 </= 0.11^*

+4.35 </= 0.71^* +2.64 </= 0.64^* +0.57 </= 3.88 — —

BMI at baseline

1st quintile (ref) — — — — — — —

2nd quintile -0.37 </= 0.18^§ -0.14 </= 0.17 -1.60

</= 1.22 -0.20 </= 0.99 -0.96 </= 5.52 -15.50 </=

10.25 1.15 </= 6.8

3rd quintile -0.84 </= 0.18^* -0.50 </= 0.17^*

-2.71 </= 1.22^* -0.53 </= 1.00 3.62 </= 5.52 -21.16

</= 9.63 4.9 </= 7.9

4th quintile -1.22 </= 0.18^* -0.74 </= 0.17^*

-6.16 </= 1.11^* -3.04 </= 1.00^§ 12.62 </= 5.52^§

-13.99 </= 9.48 16.7 </= 7.8^§

5th quintile -1.43 </= 0.18^* -0.84 </= 0.17^*

-7.90 </= 1.10^* -2.92 </= 1.00^§ 16.71 </= 5.52^§

-14.66 </= 9.71 17.8 </= 8.6^§

p for trend <0.001 <0.001 <0.001 <0.001 <0.001 0.50

0.001

BMI changes, kg/m2 +0.02 </= 0.03 — -0.05 </= 0.21 —

-0.14 </= 1.14 —

R2 — 0.23 0.17 0.17 0.17

==================================================================

In multivariate analyses, coefficients are adjusted

for educational level, diabetes, systolic blood

pressure, daily alcohol intake, physical activity,

perceived health score, perceived stress score, and

social isolation (Nottingham Health Profile). R^2

denotes the coefficient of determination.

^* p < 0.001.

^§ p < 0.05.

^# ß for trend across age classes.

Table 3 Variables associated with changes in cognitive

test scores between baseline and follow-up (linear

regression)

===================================================================

Changes in word-list learning, delayed recall,

number of cited words, ß±SE---Changes in Digit–Symbol

Substitution Test score, ß±SE---Changes in selective

attention time, ß±SE

Univariate Multivariate Univariate Multivariate

Univariate Multivariate

===================================================================

Age^# -0.45 </= 0.01^* -0.29 </= 0.04^* -0.42 </=

0.02^* -2.53 </= 0.32^* 14.03 </= 1.66^* 11.77 </=

1.69^*

Female sex (vs male) +0.32 </= 0.09^* +0.37 </= 0.09^*

-0.35 </= 0.02^* — +4.01 </= 3.88 —

BMI at baseline

1st quintile (ref) — — — — — —

2nd quintile -0.07 </= 0.13 -0.008 </= 0.13 -0.44

</= 0.94 0.28 </= 0.91 2.39 </= 5.21 -0.80 </= 5.11

3rd quintile -0.19 </= 0.14 -0.09 </= 0.13 -0.43

</= 0.94 0.61 </= 0.93 5.62 </= 5.23 2.15 </= 5.27

4th quintile -0.28 </= 0.14^§ -0.17 </= 0.14 -2.37

</= 0.94^§ -1.10 </= 0.94 6.60 </= 5.22 2.57 </= 5.38

5th quintile -0.58 </= 0.14^* -0.35 </= 0.14^§

-2.96 </= 0.94^§ -0.92 </= 0.96 10.23 </= 5.23^§ 0.84

</= 5.35

p for trend <0.001 <0.001 <0.001 0.14 <0.001 0.69

BMI changes, kg/m2 -0.02 </= 0.02 — -0.02 </= 0.18 —

2.5 </= 1.14 —

R2 — 0.27 — 0.19 — 0.32

===================================================================

All coefficients are adjusted for cognitive test

score at baseline and region of residence. In

multivariate analyses, coefficients are adjusted for

educational level, diabetes, systolic blood pressure,

daily alcohol intake, physical activity, perceived

health score, perceived stress score, and energy

(Nottingham Health Profile). R2 denotes the

coefficient of determination.

^* p < 0.001.

^§ p < 0.05.

^# ß for trend across age classes.

Table 4 Association between BMI at baseline and

cognitive decline in univariate and multivariate

analyses (logistic regression)

======================================================================

Decline in word-list learning, change <25th

percentile (-1 word) vs -25th percentile, OR (95%

CI)---Decline in DSST, change <25th percentile (-4.3)

vs -25th percentile, OR (95% CI)---Decline in

selective attention score, change >75th percentile

(+40 seconds) vs </=75th percentile, OR (95% CI)

Univariate Multivariate Univariate Multivariate

Univariate Multivariate

======================================================================

BMI at baseline

1st quintile (ref) 1 1 1 1 1 1

2nd quintile 0.82 (0.57–1.18) 0.78 (0.54–1.11) 1.09

(0.79–1.52) 1.01 (0.72–1.42) 0.96 (0.70–1.32) 0.85

(0.61–1.18)

3rd quintile 1.18 (0.83–1.67) 1.06 (0.74–1.52) 1.20

(0.87–1.68) 1.07 (0.76–1.50) 1.44 (1.06–1.97) 1.24

(0.90–1.71)

4th quintile 1.18 (0.83–1.67) 1.05 (0.72–1.53) 1.93

(1.40–2.67) 1.69 (1.20–2.37) 1.01 (0.73–1.39) 0.85

(0.60–1.20)

5th quintile 1.49 (1.05–2.11) 1.24 (0.84–1.80) 2.05

(1.47–2.83) 1.68 (1.18–2.37) 1.15 (0.83–1.59) 0.86

(0.60–1.22)

p for trend 0.006 0.09 <0.001 <0.001 0.35 0.48

All coefficients are adjusted for region of residence.

In multivariate analyses, coefficients are adjusted

for age, sex, educational level, diabetes, systolic

blood pressure, daily alcohol intake, physical

activity, perceived health score, perceived stress

score, and energy (Nottingham Health Profile).

======================================================================

OR=odds ratio;=DSST=Digit–Symbol Substitution Test.

-- Al Pater, alpater@...

-- Al Pater, PhD; email: Alpater@...

__________________________________________________