Dietary Glycaemic Load Associated with Cognitive Performance in Elderly Subjects

European Journal of Nutrition

Dietary glycaemic load associated with cognitive performance in elderly subjects

Susan E. Power 1,2, Eibhlís M. O’Connor 2,4, R. Paul Ross 2,3, Catherine Stanton 2,3, Paul W. O’Toole 1,2, Gerald F. Fitzgerald 1,2* and Ian B. Jeffery 1,2

1School of Microbiology, 2Alimentary Pharmabiotic Centre, University College Cork, 3Teagasc Food Research Centre, Moorepark, Fermoy, Co. Cork, Ireland, 4Department of Life Sciences, University of Limerick, Limerick, Ireland

*Corresponding author: Prof. Gerald F. Fitzgerald, tel. +353 21 4902730, fax +353 21 4903101, email

Supplementary Information

Table S1Unadjusted individual Poisson regression tests of associations between individual variables and no. of errors in the MMSE

Parameters / eRC / 95% CI / P-value / Pseudo R2 / LRT
Chi-squared Value / LRT
P-value
Glycaemic Load (Continous) / 1.01 / 1.01-1.02 / <0.001 / 4.50 / 24.15 / <0.001
Glycaemic Load (Q2) / 1.01 / 0.77-1.32 / 0.900 / 3.58 / 19.01 / <0.001
Glycaemic Load (Q3) / 1.38 / 1.08-1.77 / 0.011
Glycaemic Load (Q4) / 1.54 / 1.21-1.96 / <0.001
Age (yrs) / 1.03 / 1.02-1.04 / <0.001 / 4.39 / 25.88 / <0.001
Age (Q1) / 1.37 / 1.05-1.80 / 0.022 / 6.7 / 35.98 / <0.001
Age (Q2) / 1.62 / 1.24-2.12 / <0.001
Age (Q3) / 2.09 / 1.62-2.71 / <0.001
Gender (female) / 0.96 / 0.81-1.14 / 0.700 / 0.04 / 0.02 / 0.880
Diabetes / 1.72 / 1.35-2.17 / <0.001 / 3.43 / 25.69 / <0.001
Hypertension / 1.39 / 1.16-1.65 / <0.001 / 2.61 / 17.10 / <0.001
Adequate nutritional status
(MNA ≥24) / 0.69 / 0.56-0.86 / <0.001 / 2.08 / 16.22 / <0.001
HFD score / 0.13 / 0.05-0.32 / <0.001 / 3.79 / 21.22 / <0.001
HFD (Q2) / 0.97 / 0.78-1.20 / 0.800 / 6.93 / 36.84 / <0.001
HFD (Q3) / 0.64 / 0.50-0.81 / <0.001
HFD (Q4) / 0.53 / 0.41-0.68 / <0.001
RPP (Q1) / 1.42 / 1.11-1.83 / 0.006 / 2.33 / 12.01 / 0.007
RPP (Q2) / 1.04 / 0.79-1.37 / 0.800 / -
RPP (Q3) / 1.34 / 1.05-1.73 / 0.021 / -
Cardiovascular Meds
(1 medication) / 1.59 / 1.28-1.98 / <0.001 / 3.95 / 23.84 / <0.001
Cardiovascular Meds
(2 medications) / 1.28 / 0.96-1.70 / 0.090 / -
Cardiovascular Meds
(3+ medications) / 1.51 / 1.19-1.92 / <0.001 / -
Energy intake (1000 kcal) / 1.14 / 0.99-1.32 / 0.060 / 0.70 / 3.38 / 0.066
BMI (kg/m2) / 1.01 / 0.99-1.03 / 0.290 / 0.21 / 0.23 / 0.630

Glycaemic Load was adjusted to 200kcal value;CI, confidence interval; eRC, exponentiated regression coefficient;. HFD, healthy food diversity; MNA, Mini Nutritional Assessment; Pseudo R2, is based on the reduction of residual deviance (obtained after reﬁtting the models with the predictor under investigation being removed); RPP, residential property price

Table S2Chi-squared testsof associations between individual variables and mild cognitive impairment (MMSE score ≤24)

Parameters / X-squared / Degrees of freedom / P-value
Glycaemic Load / 6.82 / 1 / 0.009
Gender (female) / 0.00 / 1 / 1.000
Diabetes / 3.71 / 1 / 0.054
Adequate nutritional status (MNA ≥24) / 0.84 / 1 / 0.360
RPP / 1.38 / 3 / 0.710
Hypertension / 5.81 / 1 / 0.016
Cardiovascular Meds / 8.56 / 3 / 0.036

Glycaemic Load was adjusted to 200kcal value;MNA, Mini Nutritional Assessment; RPP, residential property price

TableS3 Right-censored Poisson regression model (controlling for medication associated with MMSE) to test the association between dietary glycaemic load (per 2000kcal) and the no. of errors in the MMSE, adjusted for age, gender, diabetes, nutritional status, HFD score, residential property price (RPP) quartiles, hypertension, no. of cardiovascular medications, hypertension*cardiovascular medications and total energy intake

Parameters / eRC / 95% CI / P-value
Glycaemic Load / 1.01 / 1.01-1.02 / <0.001
Age (yrs) / 1.03 / 1.01-1.04 / <0.001
Gender (female) / 1.25 / 1.04-1.51 / 0.018
Diabetes / 1.47 / 1.16-1.87 / 0.002
Normal nutritional status (MNA ≥24)a / 0.74 / 0.60-0.91 / 0.005
HFD score / 0.29 / 0.11-0.76 / 0.011
RPP (Q1)b / 1.30 / 1.02-1.67 / 0.040
RPP (Q2)b / 0.95 / 0.72-1.25 / 0.700
RPP (Q3)b / 1.25 / 0.98-1.59 / 0.070
Hypertension / 1.82 / 1.15-2.87 / 0.011
Cardiovascular Meds (1 medication)c / 1.33 / 0.97-1.82 / 0.080
Cardiovascular Meds (2 medications)c / 1.46 / 1.03-2.07 / 0.040
Cardiovascular Meds (3+ medications)c / 1.13 / 0.79-1.61 / 0.500
Hypertension*Cardiovascular Meds (1 medication) / 0.60 / 0.34-1.07 / 0.080
Hypertension*Cardiovascular Meds (2 medications) / 0.36 / 0.19-0.70 / 0.003
Hypertension*Cardiovascular Meds (3+ medications) / 0.60 / 0.33-1.06 / 0.080
Energy intake (1000 kcal) / 1.15 / 1.00-1.33 / 0.050

a, at risk of malnutrition (MNA<24) used as reference group; b, highest quartile used as reference group; c, no medications used as reference group; CI, confidence interval; eRC, exponentiated regression coefficient; HFD, healthy food diversity; MNA, Mini Nutritional Assessment

Table S4 Right-censored Poisson regression model (controlling for medication associated with MMSE) to test the association between dietary glycaemic load (crude) and the no. of errors in the MMSE, adjusted for age, gender, diabetes, nutritional status, HFD score, residential property price (RPP) quartiles, hypertension no. of cardiovascularmedications, hypertension*cardiovascular medications

Parameters / eRC / 95% CI / P-value
Glycaemic Load / 1.00 / 1.00-1.00 / 0.017
Age (yrs) / 1.02 / 1.01-1.04 / <0.001
Gender (female) / 1.22 / 1.01-1.46 / 0.040
Diabetes / 1.49 / 1.17-1.89 / 0.001
Normal nutritional status (MNA ≥24)a / 0.77 / 0.62-0.95 / 0.013
HFD score / 0.23 / 0.09-0.60 / 0.003
Property value (Q1)b / 1.36 / 1.06-1.74 / 0.014
Property value (Q2)b / 1.02 / 0.78-1.33 / 0.900
Property value (Q3)b / 1.32 / 1.04-1.68 / 0.025
Hypertension / 1.92 / 1.22-3.03 / 0.005
Cardiovascular Meds (1 medication)c / 1.33 / 0.98-1.83 / 0.070
Cardiovascular Meds (2 medications)c / 1.42 / 1.00-2.01 / 0.050
Cardiovascular Meds (3+ medications)c / 1.28 / 0.90-1.81 / 0.170
Hypertension*Cardiovascular Meds (1 medication) / 0.61 / 0.34-1.07 / 0.080
Hypertension*Cardiovascular Meds (2 medications) / 0.37 / 0.19-0.72 / 0.004
Hypertension*Cardiovascular Meds (3+ medications) / 0.52 / 0.29-0.92 / 0.025

Table S5 Poisson regression model to test the association between dietary glycaemic load (crude) and the no. of errors in the MMSE, adjusted for age, gender, diabetes, nutritional status, HFD score, residential property price (RPP)quartiles, hypertension, no. of cardiovascular medications and hypertension*cardiovascular medications

Parameters / eRC / 95% CI / P-value / Pseudo R2
Glycaemic Load / 1.00 / 1.00-1.01 / 0.004 / 1.88
Age (yrs) / 1.02 / 1.01-1.04 / 0.003 / 2.13
Gender (female) / 1.17 / 0.97-1.41 / 0.110 / 0.60
Diabetes / 1.51 / 1.17-1.93 / 0.001 / 2.28
Normal nutritional status (MNA ≥24)a / 0.75 / 0.60-0.95 / 0.015 / 1.35
HFD score / 0.25 / 0.09-0.69 / 0.007 / 1.71
RPP (Q1)b / 1.32 / 1.02-1.72 / 0.040 / 2.36
RPP (Q2)b / 0.97 / 0.73-1.28 / 0.800
RPP (Q3)b / 1.31 / 1.02-1.70 / 0.040
Hypertension / 2.17 / 1.29-3.49 / 0.002 / 3.32 d
d
Cardiovascular Meds (1 medications)c / 1.29 / 0.93-1.76 / 0.120 / 3.69 d
d
Cardiovascular Meds (2 medications)c / 1.43 / 0.99-2.04 / 0.050 / -
Cardiovascular Meds (3+ medications)c / 1.19 / 0.82-1.69 / 0.400 / -
Hypertension*Cardiovascular Meds (1 medication) / 0.62 / 0.35-1.15 / 0.110 / -
Hypertension*Cardiovascular Meds (2 medication) / 0.31 / 0.15-0.64 / 0.001 / -
Hypertension*Cardiovascular Meds (3+ medication) / 0.51 / 0.28-0.96 / 0.030 / -

a, at risk of malnutrition (MNA<24) used as reference group; b, highest quartile used as reference group; c, no medications used as reference group; d, interaction term was removed with variable; CI, confidence interval; eRC, exponentiated regression coefficient; HFD, healthy food diversity; MNA, Mini Nutritional Assessment; Pseudo R2, the loglikelihood R2, is based on the reduction of residual deviance (obtained after reﬁtting the models with the predictor under investigation being removed);

Figure S1 Comparison of MMSE score between glycaemic load quartiles.

Figure S2Comparison of dietary glycaemic load (per 2000kcal) between subjects with normal cognitive function (MMSE ≥27), mild cognitive impairment (MMSE ≤24) and those inan intermediate state between normal cognition and mild cognitive impairment (MMSE 25-26)

Figure S3 Rate of mild cognitive impairment across glycaemic load quartiles.