Supplementary memorandum from Dr Michael
Nelson
DIET, IRON
STATUS, AND
COGNITIVE FUNCTION
Recent evidence from the US and the UK shows
that cognitive function is poor in children with poor dietary
status but no overt signs of deficiency, and that intervention
to improve diet (eg supplementation studies, breakfast interventions)
results in improved cognitive function and learning ability. Iron
deficiency is the best nutritional documented cause of poor cognitive
function in UK school children. Nutrients which enhance iron absorption
(eg vitamin C) are also important. Improved nutrient intakes related
to breakfast consumption are associated with better cognitive
function and scholastic achievements.
RECOMMENDATIONS:
1. Guidelines for school meals should be
specified with respect to both foods and nutrients
Enusuring good cognitive function requires feeding
guidelines which promote healthy eating patterns and adequate
nutrient intakes.
2. Iron status should be monitored in adolescent
girls
Girls who are in the bottom quarter of the distribution
of iron status have an increased risk of impaired cognitive function.
School catering should pay particular attention to the promotion
of good iron status. Girls with poor iron status may require iron
supplements to boost cognitive function and scholastic performance.
3. Schemes for the distribution in schools
of free or subsidised fruit and vegetables are likely to have
widespread benefits
Higher intakes of vitamin C are associated with
better iron absorption, better iron status and hence better cognitive
function. Patterns of healthy eating would be established early.
Risks of heart disease, some cancers, stroke and hypertension
in adulthood would be reduced.
4. Schools should have a "whole school"
approach to nutrition
Messages in the classroom should be reinforced
in the canteen and tuck shop through appropriate catering practices.
Programmes to improve nutritional knowledge and behaviour must
link activities which include pupils, teachers, and caterers.
The studies outlined on the following pages
provide evidence in support of these recommendations.
PRIMARY SCHOOL
CHILDREN
Tiniakos F. Iron status and cognitive function
in poor inner city primary school children. MSc thesis. University
of London. 1993.
Location: Four primary schools located
in a poor area of Central London.
Sample: 129 children aged 6.5-9.5
years.
Measurements: Food frequency questionnaires.
Finger prick blood sample.
Two tests of cognitive function (digit span and
coding).
Findings: Children with poor iron status
had poor cognitive function.
1. Four children (3 per cent) had haemoglobin
levels less than 12.0 g/dl indicating iron deficiency anaemia.
2. Nine children (7 per cent) had ferritin
levels less than 20mg/l indicating poor iron stores.
3. There was a statistically significant
correlation between digit span and total iron intake (r=0.197,
P=0.0033) which was not explained by age.
4. Table 1 shows that there were trends of
increasing cognitive function with higher haemoglobin status.
The trends failed to reach statistical significance, however.
Table 1
Mean haemoglobin (Hb), iron intake, digit
span and coding test scores in 129 children aged 6.5-9.5 years
living in a poor area of Central London
|
| Hb group
|
|
| | |
| |
| Hb (range) | <11.6 | 11.6-13.0
| >13.0 | P |
| Means | |
| | |
| Hb | 11.2 | 12.3
| 13.8 | 0.000 |
| Iron intake | 6.8 | 4.8
| 5.1 | 0.156 |
| Digit span | 4.9 | 5.9
| 5.5 | 0.391 |
| Coding | 10.3 | 10.6
| 11.3 | 0.285 |
|
Shovlin A. Iron status and cognitive and physical performance
in 7-11 year old school children. MSc thesis. London University.
1993.
Location: One primary school in a poor area of
Central London.
Sample: 77 children aged 7-11 years.
Measurements: 7-day food diaries completed parents'
or guardians' help.
Two tests of cognitive function (digit span and coding).
Findings: Children with poor iron status had poor
cognitive function.
1. There was a statistically significant correlation between
digit span and haemoglobin levels (r=0.23, P=0.018), not explained
by age.
2. There were no other significant associations observed
between cognitive function and measures of iron status or diet.
The results from these two studies suggest that in children
from poor backgrounds better iron status is weakly but positively
associated with better cognitive function (digit span). Levels
of iron deficiency anaemia are low but a substantial number of
children are likely to have poor iron stores. This may become
problematic in adolescence when the limited reserves of iron (in
conjunction with low dietary intakes of iron and vitamin C) are
not sufficient to maintain adequate growth and development, and
cognitive function is limited.
SECONDARY SCHOOL
CHILDREN
Studies published in the UK in the mid-1990s suggested that
poor iron status and iron deficiency anaemia were common (10 per
cent to 20 per cent) in apparently healthy adolescent girls[28][29]
(The rates in boys were only about 3 per cent). This appeared
to be a consequence of arriving at puberty with low iron stores,
eating a diet that was low in iron and/or vitamin C (often in
an attempt to lose weight or on becoming vegetarian), and starting
menstruation with its associated iron losses.
Recent intervention studies in adolescent girls in the Baltimore,
Md, USA and in North London suggest that:
(a) poor iron status (with or without the presence of
anaemia) limits cognitive function; and
(b) improvements in iron status amongst those with poor
initial status leads to better cognitive function within 8-10
weeks.
Bruner AB, Joffe A, Duggan AK, Casella JF, Brandt J. Randomised
study of cognitive effects of iron supplementation in non-anaemic
iron deficient adolescent girls. Lancet 1996; 348:992-996
Location: Four High Schools in Baltimore, Maryland,
USA.
Sample: 716 girls 13-18 years old screened for
non-anaemic iron deficiency (normal haemoglobin, low ferritin).
98 enrolled in intervention trial.
Measurements: Haemoglobin (g/dl), ferritin (mg/l)
in verrous blood.
Four cognitive function tests.
Findings: Ferritin increased by 18.2 mg/l in
the intervention group compared with 3.5 mg/l in the control group.
After eight weeks of iron supplementation, Hopkins Verbal Learning
test scores increased significantly more in the intervention group
than in the control group.
The authors conclude that improvements in iron status are
associated with improved learning ability.
Ash R and Nelson M. Iron status and cognitive function in UK
adolescent girls: and intervention study. (in preparation 1999).
Location: One single sex and two mixed comprehensive
schools in North London.
Sample: 537 girls 11.5-15.5 years old (school
years 7-10) in screening study to determine iron status. 131 girls
11.5-15.5 years old in placebo controlled iron supplementation
intervention trial.
Measurements: Finger prick blood sample for measuring
Haemoglobin (Hb: g/dl), packed cell volume (PCV: %), mean corpuscular
haemoglobin concentration (MCHC: g/dl) and zinc protoporphyrin(ZPP:
mg/dl).
Venous blood sample used to confirm iron status in girls in
intervention trial.
British Ability Scale tests for verbal and non-verbal intelligence
(IQ).
Intervention: Supervised administration of 3-5
iron tablets or placebo per week for 10 weeks.
Findings: Screening study. The lowest level
of iron status occurred at the peak of the growth spurt (age 12-13
years). Using multiple iron status indicators, 16 per cent were
classified as iron deficient anaemic (IDA), a further 11 per cent
as iron deficient (ID), and 73 per cent as iron replete (IR).
Twelve per cent of White girls, 15 per cent of Afro-Caribbean
and 29 per cent of Asian girls were classified IDA. Prevalence
of IDA was 21 per cent in girls from families with no earner,
and 31 per cent in girls who had been vegetarian for less than
one year.
Intervention trial, 131 girls were recruited into
a placebo controlled iron supplementation intervention study.
Fifteen IDA girls were age matched with 48 ID and 68 IR girls.
Iron status was confirmed using venous blood samples.
Mean IQ scores at baseline were statistically significantly
lower for the IDA girls (102) compared with ID (109) and IR (111)
girls. These differences were partly but not wholly explained
by differences in adiposity, maternal and paternal age, and socio-economic
group. A computer based test which measured the time taken to
tap specified keys (measure of motor-neural function) showed that
the IDA group was significantly slower than the ID and IR groups.
Figure 1 shows the changes in IQ scores pre- and post-intervention.
The IDA girls given an iron supplement improved the most; IDA
girls given placebo did not improve. Increases in IQ in other
groups were consistent with known levels of change on repeat testing.
These results suggest:
1. IDA girls who remained anaemic after having been given
the placebo were less able to learn than those who were iron deficient
or iron replete.
2. Girls who were given iron supplements or who were
iron replete initially were best able to learn.
BREAKFAST INTERVENTION
STUDIES
Pollitt E and Mathews R. Breakfast and cognition: an integrative
summary. American Journal of Clinical Nutrition 1998; 67 (suppl):804S-813S.
Papers presented at an International Symposium on Breakfast
and Performance in 1995 are summarised and integrated with data
published since that time. "The pooled data suggest that
omitting breakfast interferes with cognition and learning, an
effect that is more pronounced in nutritionally at-risk children
than in well-nourished children. At the very least, breakfast
consumption improves school attendance and enhances the quality
of the students' diets."
The paper suggests that missing breakfast is associated with
poorer performance on short-term memory, visual discrimination
of competing stimuli, verbal fluency, tasks of arithmetic, and
stimulus discrimination. Some of the improvement in performance
is likely to be due to higher levels of blood glucose. There is
evidence that eating breakfast is associated with higher intake
of nutrients and better nutritional status.
Participation in Breakfast Programmes was associated with
improved performance in scholastic tests and better school attendance.
It was not possible to tell from the design of the studies if
the scholastic improvement was due to improved nutritional status
or to better school attendance and longer exposure to the learning
environment.
CONCLUSIONS
Good nutrition in school is likely to be a correlate of good
educational achievement. There is strong evidence for a link between
iron status and cognitive function in adolescent girls, and limited
evidence in primary school children. To minimise the risks of
iron deficiency and iron deficiency anaemia at ages 11-12, it
is important that children arrive at the adolescent growth spurt
with good nutritional status, particularly with regard to iron.
Lack of breakfast is associated with poorer cognitive function
and scholastic performance. Participation in School Breakfast
Programmes is associated with better school attendance and better
scholastic performance. Whether the improved performance is due
to improved nutritional status or longer exposure to the school
environment is not clear.
Dr M Nelson
November 1999
28
Nelson M, White J, Rhodes C. Haemoglobin, ferritin and iron intakes
in British children aged 12-14 years: a preliminary study. British
Journal of Nutrition 1993; 70:147-155. Back
29
Nelson M, Bakaliou F, Tivedi A. Iron deficiency anaemia and physical
performance in adolescent girls from different ethnic backgrounds.
British Journal of Nutrition 1994; 72:427-433. Back
| 
