Hostname: page-component-77f85d65b8-hzqq2 Total loading time: 0 Render date: 2026-03-29T02:19:06.035Z Has data issue: false hasContentIssue false
  • English
  • Français

Dietary glycaemic index and cognitive function: associations in adults aged 53y of the 1946 British birth cohort

Published online by Cambridge University Press:  03 February 2017

E. Philippou ,
G.K. Pot ,
A. Heraclides ,
R. Bendayan  and
M. Richards
E. Philippou
Affiliation:
Department of Life and Health Sciences, University of Nicosia, Cyprus Diabetes and Nutritional Sciences Division, King's College London, London, UK
G.K. Pot
Affiliation:
Diabetes and Nutritional Sciences Division, King's College London, London, UK Section Health and Life, Vrije Universiteit Amsterdam, the Netherlands
A. Heraclides
Affiliation:
Centre for Primary Care and Population Health,Medical School, University of Nicosia, Cyprus
R. Bendayan
Affiliation:
MRC Unit for Lifelong Health and Ageing at UCL, London, UK
M. Richards
Affiliation:
MRC Unit for Lifelong Health and Ageing at UCL, London, UK
Rights & Permissions [Opens in a new window]

Abstract

An abstract is not available for this content. As you have access to this content, full HTML content is provided on this page. A PDF of this content is also available in through the ‘Save PDF’ action button.

The brain relies on glucose for its function(Reference Amiel1) obtained mainly from carbohydrate (CHO)-containing foods. Several, mostly single meal studies, investigated whether the rate of glucose release would affect cognitive functioning using the established CHO classification of the glycaemic index (GI), with inconsistent results(Reference Philippou and Constantinou2). We studied associations between usual diet GI and cognitive function at age 53y in members of the MRC National Survey of Health and Development (1946 British birth cohort)(Reference Wadsworth, Kuh, Richards and Hardy3). We hypothesised that consumption of a low GI diet would be associated with a better cognitive function since it is associated with a more constant postprandial blood glucose concentration(Reference Ludwig4).

Analysis included data from 939 (54·5 % female) cohort members with complete diet data (based on 5-day estimated food diaries) at age 53, cognition data at both ages 53y and 63y and potential confounders. Diet GI was calculated by assigning a glycaemic load (GL) value for each food item, then summing the GL values for the day and dividing this by the total daily CHO in grams(Reference Wolever and Jenkins5). Cognitive function was assessed by a 15-item word-learning task (short term episodic memory), and a letter search test requiring mental speed, visual scanning and focused concentration(Reference Richards, Shipley, Fuhrer and Wadsworth6). Multivariable linear or logistic regression models were used to examine associations between quartiles of diet GI, with the lowest GI quartile used as reference, and results of cognitive function tests presented as continuous outcome variables or in tertiles (in case of highly skewed variables) respectively. Analyses were adjusted for potential confounders as shown in the Table.

There was a significant trend for lower memory score, lower letter search number of hits, letter search speed and higher speed-accuracy trade-off (calculated as speed/accuracy) with increasing diet GI quartile. Only the associations for speed-accuracy trade-off remained significant in the fully adjusted model. Significant GI x sex interactions were found for letter search number of hits and speed-accuracy trade-off. No difference in letter search accuracy by diet GI quartile was found.

In conclusion, the study findings indicate a robust association between dietary GI and speed-accuracy trade-off. The attenuation of the association between letter search speed and cognition when adjusted for adolescent cognition and overall education strongly suggests reverse causality. These initial cross-sectional findings will be further explored in longitudinal analysis.

Table 1. Associations between diet GI quartiles and cognitive function test results at age 53 adjusting for potential confounders 1

1. Model 1: unadjusted; Model 6: adjusted for sex, cognition at age 15, educational attainment, social class, BMI, waist circumference, smoking status, physical activity, blood pressure, serum triglyceride concentration, HDL cholesterol, energy intake, intake of:  % fat,  % saturated fat,  % alcohol,  % protein,  % carbohydrates,  % sugars, non-starch polysaccharides and energy intake: estimated energy requirements.

2. Letter search speed was analyzed using linear regression in its continuous form. Letter search speed-accuracy trade-off was highly skewed so it was categorized in tertiles and analyzed using ordinal logistic regression.

References

1. Amiel, SA (1994) Proc Nutr Soc 53(2):401–5.CrossRefGoogle Scholar
2. Philippou, E, Constantinou, M (2014) Adv Nutr 5(2):119–30.CrossRefGoogle Scholar
3. Wadsworth, M, Kuh, D, Richards, M, Hardy, R (2006) Int J Epidemiol 35:49–54.CrossRefGoogle Scholar
4. Ludwig, DS (2002) JAMA 287(18):2414–23.CrossRefGoogle Scholar
5. Wolever, TM, Jenkins, DJ (1986) Am J Clin Nutr 43(1):167–172 CrossRefGoogle Scholar
6. Richards, M, Shipley, B, Fuhrer, R, Wadsworth, MEJ (2004) BMJ 328, 552–4.CrossRefGoogle Scholar
Figure 0

Table 1. Associations between diet GI quartiles and cognitive function test results at age 53 adjusting for potential confounders1