1
Intake and dietary sources of haem and non-haem iron in Flemish preschoolers
2
3
Inge Huybrechts1,2, Yi Lin1, Willem De Keyzer1,3, Christophe Matthys4,1, Linda Harvey5,6,
4
Aline Meirhaeghe7, Jean Dallongeville7, Beatriz Sarria8, Guy De Backer1, Stefaan De
5
Henauw1,3
6
7
1
Department of Public Health, Ghent University, Ghent, Belgium
8
2
Dietary Exposure Assessment group, International Agency for Research on Cancer (IARC),
9
Lyon, France
10
3
Department of Nutrition and dietetics, University College Ghent, Gent, Belgium
11
4
Clinical and experimental Endocrinology, KULeuven, Leuven, Belgium.
12
5
Institute of Food Research, Norwich Research Park, Colney, Norwich, NR4 7UA, UK
13
6
School of Medicine, Health Policy and Practice, University of East Anglia, Norwich, NR4
14
7TJ, UK
15
7
INSERM, U744; Institut Pasteur de Lille; Univ. Lille Nord de France, Lille, France
16
8
Department of Metabolism and Nutrition, Food Science and Technology and Nutrition
17
Institute (ICTAN), Spanish Council for Scientific Research (CSIC) 28040 Madrid, Spain.
18
19
Running title: iron intakes among preschoolers
20
Key words: dietary sources, iron, child
21
1
22
*Corresponding author: Inge Huybrechts, Department of Public Health, Faculty of Medicine
23
and Health Sciences, Ghent University, UZ – 2 Blok A, De Pintelaan 185, B-9000 Ghent,
24
Belgium. Tel: +32 (0)9 332 24 23, Fax: +32 (0)9 332 49 94, email: inge.huybrechts@ugent.be
25
26
Total number of words in abstract: 247
27
Total number of words in manuscript: 2990
28
Funding for this project was provided by the Belgian Nutrition Information Center.
29
30
Abbreviations
31
IOM: Institute of Medicine
32
WHO: World Health Organization
33
FAO: Food and Agriculture Organization
34
SHC: Superior Health Council
35
DRI: Dietary reference intakes
36
EAR: Estimated Average Requirement
37
EDR: Estimated Dietary Record
38
SPSS: Statistical Package for the Social Sciences
39
FBDG: Food Based Dietary Guidelines
40
41
2
42
Abstract
43
Background/Objectives: In the absence of biochemical data on iron status in preschoolers,
44
data on the adequacy of iron intake may be used to assess the possible risk of iron deficiency
45
in this population group. Therefore, this study aims to investigate iron intake and its food
46
sources in Flemish preschoolers.
47
Subjects/Methods: 661 Flemish preschoolers 2.5-6.5 years old were recruited via a random
48
cluster sampling design, using schools as primary sampling units. Three-day estimated diet
49
records were used to assess dietary intakes. The contribution to iron intake (haem and non-
50
haem) of 57 food groups was computed by summing the amount provided by the food group
51
for all individuals divided by the total intake for all individuals.
52
Results: Mean total iron intake (s.d.) was 7.4 ( 2.3) and 6.7 ( 2.8) mg/day for boys and
53
girls, respectively. 65% of the children <4 years old and 45% of those 4-6.5 years old
54
presented adequate iron intakes. The food groups with the highest mean proportional
55
contribution to total iron intake were bread, meat & meat products, breakfast cereals and
56
sweet snacks (in that order). Children from small families whose mother had a low
57
educational level had higher iron intakes.
58
Conclusions: Iron intakes were similar for boys and girls and almost half of the Flemish
59
preschoolers doenot comply with the dietary iron recommendations.
60
61
3
62
Introduction
63
Iron is an essential nutrient for humans, having a prominent position in a number of
64
physiological processes, such as oxygen transport and storage, oxidative energy production
65
and others (Centers for Disease Control and Prevention (CDC), 1998; FAO/WHO, 2004;
66
Hoge Gezondheidsraad, 2006). Iron is present in different forms (haem and non-haem) with
67
differences in characteristics and bioavailability (haem iron has a higher bioavailability than
68
non-haem iron). Long term, inadequate iron intake can lead to iron deficiency anaemia, loss
69
of appetite, lassitude, delayed psychomotor and cognitive infant and child development and
70
lower resistance to infection (Centers for Disease Control and Prevention (CDC), 1998;
71
FAO/WHO, 2004).
72
Pynaert and colleagues reported insufficient iron intakes among Belgian adolescents
73
investigated in 1997, with 99.5% of the 13-18 years old girls and 38.8% of the boys not
74
reaching Belgian dietary recommendations (Pynaert et al, 2005). When bioavailable iron was
75
considered (using absorption factors of 25% for haem iron (FAO/WHO, 2004) and 10% for
76
non-haem iron (Heath & Fairweather-Tait, 2002), 84.5% of the adolescent boys and only
77
16.5% of the girls met the age-specific requirement (Pynaert et al, 2005). However, there are
78
no data related to the intake of bioavailable iron among Belgian children. Moreover, there are
79
no biochemical data on iron status in this population group. Therefore, data on the adequacy
80
of iron intake may indicate the possible risk of iron deficiency in Belgian preschoolers.
81
This study investigates the intake of total, haem and non-haem, iron among Flemish
82
preschoolers and the percentage of children reaching the new Belgian iron recommendations.
83
In addition, the most important food sources that contribute to children’s dietary iron intakes
84
were investigated and the potentially associated variables studied.
85
4
86
Methods
87
Survey population
88
This study used data from the Flanders preschool dietary survey (data collected from October
89
2002 until February 2003), in which usual dietary intake of Flemish preschoolers (2.5-6.5 y
90
old) was estimated from 3-day estimated dietary records (3d EDR), completed by parents.
91
The sampling design and methods have been described in detail previously, along with the
92
response rate and the representativeness of the study sample (50% response rate and 49%
93
after data-cleaning) (Huybrechts et al, 2008). In brief, a random cluster sampling design at the
94
level of schools, stratified by province and age was used (Huybrechts et al, 2008).
95
The percentage of underreporters has been described in depth previously and was shown to be
96
low (< 2% of the children when using Goldberg cut-offs (Black et al, 1991) adapted for
97
children) (Huybrechts & De Henauw, 2007). Underreporters have not been excluded from the
98
study sample that was used for the analyses here described as similar results were obtained
99
when excluding the underreporters (data not shown).
100
101
The Ethical Committee of the Ghent University Hospital (Belgium) granted ethical approval
102
for the study (file number: 2002/300). All parents of the children participating in the Flanders
103
preschool dietary survey provided informed written consent.
104
105
Assessment of iron intakes
106
Parents were asked to complete estimated dietary records about their child’s food intake on
107
three consecutive days. Details on brand name and/or food type (e.g. low fat) were reported
108
whenever available. Only diaries with three completed record days were included (n 696;
109
66% of collected diaries).
5
110
The Dutch food composition databases NEVO (NEVO, 2001) was used for calculating haem,
111
non-haem and total iron intakes. Losses during preparations were taken into account as all
112
foods were coded “as eaten” and not “as raw”.
113
114
In total 936 foods and composite dishes were encoded in the original database. All recipes
115
that were described in depth as individual ingredients in the diaries were encoded as
116
ingredients in the original database. After linking haem and non-haem iron values (g/100g) of
117
the NEVO table with the detailed food list, all food items were divided into 57 food groups of
118
similar nutrient content or consumption, based on the classification of the Flemish FBDG
119
(VIG, 2004), and the expert opinion of the investigators (see food groups listed in table 3).
120
121
Parental questionnaire about socio-demographic, economic and lifestyle factors
122
To evaluate possible determinants of food consumption habits, a general questionnaire,
123
registering additive information about the child, its parents and the family/household
124
composition was completed by the parents. The authors categorized parental work status
125
(employed or unemployed), education (lower secondary education (only the 3 first years, or less
126
of the secondary education), secondary education or post-secondary education) and smoking
127
status (yes/no); as well as family size (less than two versus two or more children); and child
128
nutritional supplement consumption (yes/no). Children’s physical activity level was estimated
129
by the parents as follows: light means sedentary, medium means that a sport is often practiced
130
and high, that one or more sports are regularly practiced, which involves training.
131
132
Statistical analyses
133
The Statistical Package for the Social Sciences for Windows version 14 (SPSS Inc., Chicago,
134
IL, USA) was used. In total, 661 children were included in the analyses (age and/or sex were
6
135
missing for 35 children what reduced our sample from 696 to 661 children in the statistical
136
analyses). Mean and median ‘usual’ intakes of the population were calculated using statistical
137
modelling to correct for day-to-day variability in the 3d EDR (Guenther et al, 1997; Nusser et
138
al, 1996). The program used to calculate usual intakes was the Software for Intake
139
Distribution Estimation (C-side) (Iowa State University, 2006).
140
Belgian Recommended Nutrient Intake (RNI) was 3.9 mg/d for children < 4 years old and 4.2
141
mg/d for children ≥ 4 years old (Gezondheidsraad, 2009),however, no Estimated Average
142
Requirement (EAR) are given. Therefore, the US EAR for preschool aged children was used:
143
3.0 mg/d for children < 4 years old and 4.1 mg/d for children ≥ 4 years old (IOM, 2001). As
144
on average 25% of haem iron (FAO/WHO, 2004) and 10% of non-haem iron (Heath &
145
Fairweather-Tait, 2002) are absorbable, bioavailable iron intake was estimated as follows
146
(Pynaert et al, 2005): bioavailable iron intake = (haem iron intake 0.25) + (non-haem iron
147
intake 0.10) (FAO/WHO, 2004). Since all these recommendations included different
148
reference values for children under four years old and children at least four years old, we
149
calculated the median intakes for these two different age groups (2.5-3 years old versus 4-6.5
150
years old) (Table 2). For determining the proportion of preschoolers who had an iron intake
151
below the recommendations the full probability approach was used (Carriquiry, 1999; Gibson
152
& Ferguson, 2008).
153
The population proportion formula was used to determine the percentage contribution of each
154
of the 57 food groups to the intake of each dietary component (haem and non-haem iron).
155
This was done by summing the amount of the component provided by the food for all
156
individuals divided by the total intake of that component from all foods for the entire study
157
population (Fox et al, 2006; Krebs-Smith et al, 1989; Royo-Bordonada et al, 2003).
158
A Kolmogorov–Smirnov test was used to test for normality. To compare the means of
159
different groups, the Independent-Samples T-Test was used for normally distributed data,
7
160
otherwise the Mann–Whitney U test was used. All analyses were also executed using bio-
161
available total iron intake.
162
A generalized linear model (GLM) was used to investigate the associations of total iron,
163
haem-iron and non-haem iron intake (dependent variables) with the sociodemographic,
164
economic and lifestyle variables available in parental questionnaires (independent variables).
165
Associations were simultaneously controlled for all variables included in the models: total
166
energy intake, sex, age, physical activity level, supplement use, household size, occupational
167
status and educational level of the parents and smoking of the parents. Interactions of the
168
independent variables with age and gender were also included (only the interactions between
169
mother’s education level and age were significantly influenced total iron, and non-haem iron
170
status). Afterward, a fitted model was used by reducing the full model until only statistically
171
significant (p<0.01) variables were left. . This was done by several backward steps, so that the
172
least significant covariable is dropped except it is significant at the critical level of 0.01. The
173
reduced models are successively re-fitted applying the same rule until all remaining variables
174
are statistically significant. The Type I Wald Chi-Square test was used to determine
175
significance. A P-value of <0.01 was taken in order to reduce the probability of false-positive
176
findings.
177
178
Results
179
In Table 1 the energy intake (kcal/day), the absolute (mg/day) and energy-adjusted
180
(mg/1000 kcal) intake of total, haem and non-haem iron intake in boys and girls separately are
181
shown.
182
8
183
Insert Table 1
184
The absolute mean intake of non-haem and haem iron respectively was on average 6.9 and 0.6
185
mg/day for boys and 6.2 and 0.6 mg/day for girls, respectively. The higher intake of non-
186
haem iron compared with haem iron was consistently observed in boys and girls. Although
187
significant differences were found between genders for total iron intake and non-haem iron
188
intakes (p<0.001), no significant differences were found after correcting for energy intake.
189
190
Table 2 shows the median iron intakes and the percentage of the population that reaches iron
191
intake recommendations using the full probability approach. Also median bioavailable iron
192
intakes (mg/day) in the different age categories are shown. The median intake of total iron
193
and bioavailable iron was above the EAR in both age categories (Table 2). The percentage of
194
children meeting the iron recommendation was 65% and 45% in the youngest and oldest age
195
group respectively (table 2).
196
197
Insert Table 2
198
While the dietary reference intakes for iron differed between the two age groups, the reference
199
intakes for boys were the same as for girls in the three different dietary recommendations
200
discussed above. When stratifying for gender, the proportions of children meeting the iron
201
recommendations when using the full probability approach was very similar between boys
202
and girls and comparable to the values given for the total population in table 2 (data not
203
shown).
204
9
205
The average proportional contribution of different food groups to dietary total, haem and non-
206
haem iron intakes is shown in Table 3. Food groups with the highest mean proportional
207
contribution to total dietary iron intake in both boys and girls were in the following order:
208
bread (14.5%), meat & meat products (10.5%), breakfast cereals (10.2%) and sweet snacks
209
(9.5%). For haem iron, the main contributors were meat & meat products (69.5%), cold cuts
210
from meat products (18.5%) and poultry (5.6%). For non-haem iron, the main contributors
211
were bread (15.9%), breakfast cereals (11.2%), and sweet snacks (10.4%), followed by
212
sugared milk drinks (6.7%), and cooked vegetables (6.3%).The contributions for the total
213
study population are presented in table 3, being the contributions for both genders similar
214
(data not shown).
215
216
Insert Table 3
217
Associations of socio-demographic factors with total iron, haem and non-haem iron intakes
218
are shown in Table 4. Total energy intake and mother’s lower secondary education were
219
positively associated with total, haem and non-haem iron intake. The maternal educational
220
level was inversely associated with total, haem and non-haem iron intake (p<0.004) (Table 4).
221
Children from lower educated mothers had significantly higher iron intakes compared with
222
children from higher educated mothers. In addition, children from smaller families had higher
223
total iron and non-haem iron intakes than those from bigger families (>2 children). In
224
contrast, age was negatively associated with total and non-haem iron intakes.
225
226
Insert Table 4
227
10
228
Discussion
229
Previous research reported insufficient iron intakes among Belgian adolescents, mainly in
230
adolescent girls (13-18 years old) (Pynaert et al, 2005). When bioavailable iron was
231
considered, 84.5% of the adolescent boys and only 16.5% of the girls met the age-specific
232
requirement (Pynaert et al, 2005). However, data related to the intake of bioavailable iron
233
among Belgian children is lacking, as well as biochemical data on their iron status. Therefore,
234
we evaluated the iron intake of the children in order to assess the possible risk of iron
235
deficiency in this population group. We observed that total iron intakes were higher among
236
boys than among girls, however this difference disappeared when correcting for total energy
237
intake. When comparing total dietary intakes with the recommendations via the full
238
probability approach, 65% of the children <4 years old had total iron intakes above the iron
239
recommendation and 45% in the age group 4-6.5 years old. This means that almost half of the
240
preschool aged population in Belgium has inadequate iron intakes when comparing with the
241
recommendations.
242
243
A comparison of iron intake in Flemish preschoolers with other preschool populations from
244
other countries is difficult to make and needs to be interpreted carefully, mainly because of
245
potential differences in methodology, study population and dietary reference values used.
246
Furthermore, only few studies reported bioavailable iron or haem and non-haem iron
247
separately. Taking this into account, our iron intake results were comparable with those from
248
France, the Netherlands and the UK (Lambert et al, 2004). This review reported that iron
249
intakes in 2- to 3-year-olds ranged from about 5 to 10 mg/d, and those of 4- to 6-year-old
250
boys and girls from around 6 to 13 mg/d (Lambert et al, 2004).
11
251
In the DONALD cohort in Germany, detailed data were collected on diet, metabolism, growth
252
and development from healthy subjects between infancy and adulthood. Food consumption
253
was assessed with 3-day weighed dietary records. From 1995–2000, mean iron intake in the
254
age category 4–8 y was 8 (SD=2.1) mg/day which was slightly higher than in our population
255
(Sichert-Hellert & Kersting, 2003).
256
In our study, bread is the main source of total iron intake (15%), followed by meat (11%) and
257
breakfast cereals (10%). Meat was the primary source of haem iron intake (32%). The
258
contribution of breakfast cereals to iron intake was much higher in US children than in
259
Flemish preschoolers. Subar et al. already reported an important contribution from fortified
260
foods to different micronutrients in US children in 1989-1991 (Subar et al, 1998).
261
Comparison with the main iron contributors among 4-year old Swedish children showed that
262
meat products had a higher contribution to total iron intakes (23%), whereas bread had a
263
much lower (<10%) (Garemo et al, 2007).
264
265
As previously indicated, there are no data available on biological iron status in our study
266
population, thus it is not clear to what extent the observed intake translates into adequate
267
overall iron status. It is known that the risk for iron deficiency (serum ferritin < 12 µg/L)) is
268
apparent, especially in adolescent girls. For instance, in Great Britain, the National Diet and
269
Nutrition Survey (1997) of young people (aged 4–18 y), demonstrated that iron status indices
270
were strongly correlated with haem iron intake, but not with total or non-haem iron intake,
271
and improved with increasing meat consumption (Thane et al, 2003). Low haemoglobin levels
272
were observed in 9% of children aged 4- y old (Thane et al, 2003).
273
From a Swedish follow-up study that began at age 6 months, Ohlund and colleagues reported
274
that prevalences of anaemia and iron deficiency were low, affecting 2 (1.8%) and 3 (2.8%)
275
children at the age of 4 y (n = 127), respectively; no child had iron deficiency anaemia
12
276
(Ohlund et al, 2008). These authors also showed that food choices had little effect on iron
277
status and that haemoglobin concentrations and mean corpuscular volume tend to track from
278
infancy into childhood. In this study, dietary iron intake was not significantly correlated with
279
hemoglobin concentrations, whereas the consumption of meat products had a positive effect
280
on serum ferritin concentrations and mean corpuscular volume in boys (P = 0.015 and 0.04,
281
respectively). In healthy, well-nourished children with a low prevalence of iron deficiency,
282
the mother's haemoglobin concentration was significantly associated with that of her child,
283
but the underlying mechanism is unclear (Ohlund et al, 2008).
284
285
When looking at possible child or family characteristics that could be associated with dietary
286
intakes, the authors found a significant influence of the educational level of the mother and
287
family size on total iron and haem-iron intake in preschool aged children (children from lower
288
educated mothers and small families had higher iron intakes). Possible explanations for these
289
associations between participant characteristics and iron intake are the fact that people with
290
different socio-economic status might have different dietary behaviours that lead to different
291
iron intakes.
292
293
Our study presents the following strengths and limitations. A large representative sample of
294
661 Flemish preschoolers participated in the study. Although, willingness to participate leads
295
to some selection bias, these data represent a more general population of preschool children in
296
Flanders compared to other food consumption surveys which are mostly restricted to local
297
areas. Nonetheless, as previously shown (Huybrechts et al, 2008), the study sample was
298
subject to certain selection bias, with lower socio-economic classes being underrepresented. It
299
is also noteworthy that like any dietary assessment methodology, diet records are prone to a
300
degree of misreporting that may have influenced our classification of compliance and non13
301
compliance with Dietary reference intakes (DRI). In addition, a 3d diet record does not
302
necessarily reflect an individual’s usual intake. Therefore, a statistical modelling method that
303
accounts for within-individual variability was used to calculate usual iron intakes (except for
304
the food group intakes reported in table 3). Since all days of the week were included in the
305
study, it was possible to adjust our data to remove the effect of the day of the week.
306
Unfortunately, it was impossible to correct for seasonal variations, because our fieldwork was
307
only conducted during autumn and wintertime. No data were found relating to potential
308
seasonal influences on nutrient intakes in this population group in Belgium. However, from
309
the National food consumption survey in 2004, it could be concluded that seasonal variations
310
were only limited at the nutrient level (De Vriese et al, 2006). These low seasonal variations
311
could be due to the widespread availability of most foods all year round.
312
In addition, it should be noted that food composition data, used for calculating nutrient intakes
313
might also introduce some error in dietary surveys reporting nutrient intakes. Food
314
composition data is essential for calculating nutrient intakes from consumption data.
315
However, most of the available food composition tables do not include detailed data on haem
316
and non-haem iron, therefore the authors were not able to use the Belgian food composition
317
table (NUBEL). The food composition table from The Netherlands (NEVO) was the only
318
FCT available from a neighbouring country with similar food consumption habits that
319
included haem and non-haem iron data. Therefore this FCT was used instead of our Belgian
320
Nubel table. However, using a food composition table from another country might have some
321
limitations since differences in food composition might exist between different countries for
322
similar food items. Furthermore, not all Belgian food items were included in the NEVO table
323
what required extra calculations via recipes/ingredients or what forced us to use information
324
from a similar food item that was available in the NEVO table. Therefore, the authors would
14
325
like to emphasize the growing requirement for robust food composition data as described by
326
Westenbrink and colleagues (Westenbrink et al, 2009).
327
328
The data presented in this paper do not include the use of iron supplements as this information
329
was not included in the three-day dietary records. However, from the food frequency
330
questionnaire used in the Flanders preschool dietary survey, it could be concluded that only
331
1.3% of the children were using iron supplements, limiting the impact of the absence of this
332
variable in our study (Huybrechts et al, 2010).
333
334
Conclusion
335
Iron intakes were similar for boys and girls (when adjusting for energy intake) and the mean
336
iron intake of the majority of Flemish preschoolers complies with several (inter)national
337
dietary iron recommendations. However, almost half of the children seem to have inadequate
338
iron intakes when comparing with the recommendations via the full probability approach. The
339
educational level of the mother and family size were the only socio-economic factors that had
340
a significant influence on total iron and haem-iron intake in preschool aged children (children
341
from lower educated mothers and small families had higher iron intakes).
342
343
344
Acknowledgement
345
We thank all the parents and teachers who participated into this project and generously
346
volunteered their time and knowledge. We also acknowledge Mia Bellemans and Mieke De
347
Maeyer, the dieticians of our team, who were responsible for the data input. In addition we
348
would like to thank Dr. Ilse Pynaert for her assistance in the data linking procedures to
15
349
calculate the haem and non-haem iron intakes. Funding for this project was provided by the
350
Belgian Nutrition Information Center.
351
352
Inge Huybrechts was responsible for the study design, fieldwork, analyses and the writing of
353
the manuscript. Christophe Matthys, Guy De Backer and Stefaan De Henauw contributed in
354
the conceptualisation of the study design and the development of the questionnaires. Yi Lin
355
and Willem De Keyzer assisted in the statistical analyses. Linda Harvey, Aline Meirhaeghe
356
and Jean Dalongeville and Beatriz Sarria assisted in the interpretation of the results and in the
357
writing of the manuscript. All authors assisted in the writing of the manuscript.
358
359
Conflict of Interest
360
There is no conflict of interest to report
361
362
16
363
Table 1 – Description of energy intake (kcal/day), absolute (mg/day) and energy-adjusted
364
(mg/1000 kcal) intake of total, haem and non-haem iron and the ratio of non-haem/haem iron
365
intake (mean (s.d.)) in boys (n=338) and girls (n=323).
Total group
Mean
SD
Boys (n=338)
Energy(kcal/d)
1509.4 287.5
Total iron (mg/d)
7.4
2.3
Haem iron (mg/d)
0.6
0.4
Non-haem iron (mg/d)
6.9
2.2
Total iron (mg/1000kcal)
4.9
1.3
Haem iron (mg/1000kcal)
Non-haem iron
(mg/1000kcal)
0.4
0.3
4.6
1.2
Girls (n=323)
Energy(kcal/d)
1397.6 288.3
Total iron (mg/d)
6.7
2.2
Haem iron (mg/d)
0.6
0.4
Non-haem iron (mg/d)
6.2
2.1
Total iron (mg/1000kcal)
4.8
1.2
Haem iron (mg/1000kcal)
Non-haem iron
(mg/1000kcal)
0.4
0.3
4.4
1.2
366
367
17
368
Table 2 - Median iron intake (mg/day) and proportions of preschoolers who met the
369
recommendation in children younger than 4 years old (n=197) and children aged 4-6 years old
370
using the full probability approach (n=465).
371
Dietary
Age (y)
recommendation
Referen
ce
values
(mg/d)
Median
(SE)
(mg/d)
Percentage
>EAR (full probability
approach)
65.0 (p<0.01)*
<4
3.0
6.47 (0.19)
≥4
4.1
6.87 (0.10)
<4
0.46
0.74 (0.02)
≥4
0.50
0.78 (0.01)
IOM
FAO/WHO
Bioavailable iron
45.0
372
373
The Institute of Medicine (IOM) (IOM, 2001)
374
FAO/WHO 2004: required intake of bioavailable iron
375
* Pearson Chi-Square to test for significance level when comparing the two age categories
18
Table 3 - Mean proportional contribution of different food groups to dietary iron intake among Flemish preschoolers.
Food intake
(g/d)
Food
group
Subgroup
Beverages (incl. juices but no drinks from
restgroup*)
Water
Mean
(SD)
486.2
Total Iron
orde
%
r
7.1
Haem Iron
%
order
2.4
Non-Haem
Iron
%
7.6
224.2 (226.4)
0.0
0.0
0.0
23.1 (90.1)
0.1
0.0
0.1
8.2 (43.5)
0.1
0.0
0.1
172.8 (209.3)
4.2
0.0
4.6
Vegetable juice
0.2 (6.0)
0.0
0.0
0.0
Soup / bouillon
57.7 (101.7)
2.7
26.1
2.4
Light beverages
Tea and coffee without sugar
Fruit juice
Bread and cereals
Bread / rolls / crackers / rice cakes
86.7
70.3 (46.8)
14.5
Sugared bread
7.5 (22.5)
1.4
Breakfast cereals (ready-to-eat / hot)
8.9 (20.0)
10.2
6.7
Potatoes and grains
Pasta / noodles
Rice
Potatoes
Vegetables
Cooked vegetables
87.2
8
0.0
1
3
order
5
2.8
28.6
0.0
15.9
0.0
1.5
0.0
11.2
7.3
0.0
8
1
2
15.4 (41.0)
1.0
0.0
1.1
6.3 (25.5)
0.4
0.0
0.4
65.0 (69.3)
5.3
6.4
7
0.0
5.8
6
5.7
6
6.3
5
66.5
53.7 (60.1)
0.0
7.0
0.0
19
Food intake
(g/d)
Food
group
Subgroup
Raw vegetables
Fruit (sweetened / unsweetened)
Mean
(SD)
12.8 (38.3)
109.9
Total Iron
orde
%
r
0.7
3.4
Haem Iron
%
order
Non-Haem
Iron
%
0.0
0.0
0.8
3.8
Fresh fruit
94.0 (102.7)
2.7
0.0
2.9
Canned fruit
15.4 (45.4)
0.6
0.0
0.7
0.4 (3.7)
0.1
0.0
0.1
0.1 (1.5)
0.0
0.0
0.0
Dried fruit
Olives
Milk, milk products and calcium enriched
soy milk
Milka
Sugared milk drinks (e.g. Fristi,
chocolate milk, …)
439.9
10.4
179.0 (218.5)
0.8
188.3 (226.8)
6.2
4.5 (25.3)
Sugared or flavoured yoghurt
0.0
0.9
0.0
6.7
0.0
0.0
0.0
14.2 (46.9)
0.4
0.0
0.4
Soy drinks
15.7 (82.5)
0.9
0.0
1.0
Milk desserts
19.9 (56.2)
0.5
0.0
0.5
Milk desserts based on soy
2.3 (19.1)
0.1
0.0
0.1
Probiotics (e.g. Actimel, Yakult, …)
0.7 (7.4)
0.0
0.0
0.0
15.3 (43.3)
1.6
0.0
1.7
Yoghurt
White (fresh) cheese
Chees
e
14.5
Hard cheeseb
11.8 (22.6)
0.3
0.2
10
11.4
0.0
5
order
0.0
4
0.4
0.0
0.3
20
Food intake
(g/d)
Food
group
Subgroup
Mean
Cheese spread
Fat &
oilc
(SD)
2.7 (8.8)
Total Iron
orde
%
r
%
0.1
order
%
0.0
0.1
8.6
Haem Iron
Non-Haem
Iron
0.1
0.0
0.1
Butter / margarine
8.3 (9.5)
0.1
0.0
0.1
Oil
0.3 (1.4)
0.0
0.0
0.0
Frying oil
0.0 (0.6)
0.0
19.3
0.0
96.8
0.0
12.1
Meat / poultry / fish / egg / meat alternates
90.3
69.5
1
5.0
1.6
5.6
3
1.3
8.5 (28.7)
0.9
2.7
4
0.8
Cold cuts (from meat products)
20.7 (30.2)
3.9
18.5
2
2.5
Cold cuts (from fish products)
0.9 (6.8)
0.2
0.5
6
0.1
Eggsd
Meat substitutes (e.g. tofu, tempe,
…)
5.1 (18.2)
1.5
0.0
1.7
1.7 (11.6)
0.6
0.0
0.6
Nuts and seeds
0.3 (3.4)
0.1
19.5
0.0
0.1
21.3
0.0
0.6
0.0
10.4
Meat, game and meat products
37.2 (46.1)
10.5
Chicken / turkey
15.9 (34.7)
Fish / shellfish
Restgroup (snacks & desserts)
Brioches
Sweet snacks (e.g. candy bars,
candies)
Salty snacks (e.g. chips, salty
biscuits)
201.8
3.5 (17.0)
0.6
43.6 (43.5)
9.5
2.1 (9.8)
0.6
2
10
0.2
4
order
0.2
8
7
3
0.7
21
Food intake
(g/d)
Food
group
Subgroup
Mean
Tea and coffee with sugar
Soft drinks
Salty sauces (e.g. béarnaise, cream
sauces)
(SD)
Total Iron
orde
%
r
Haem Iron
%
order
Non-Haem
Iron
%
3.2 (26.6)
0.0
0.0
0.0
97.7 (169.4)
0.3
0.0
0.4
12.5 (24.9)
0.0
0.8
0.0
0.0
Cream
Sweet sauces (e.g. chocolate or
caramel sauce)
0.3 (2.6)
0.8
0.0
0.1 (2.5)
0.0
0.0
0.0
Chocolate
3.1 (9.5)
0.8
0.0
0.9
Chocolate spread
Other sweet spread (e.g. jam, honey,
…)
9.4 (13.9)
4.0
0.0
4.4
5.3 (11.6)
0.7
0.0
0.7
Sugar
0.1 (0.9)
0.0
0.0
0.0
Fried snacks (e.g. churros)
0.1 (2.6)
0.0
0.0
0.0
14.6 (37.7)
1.8
0.1
6.2 (23.2)
0.4
0.6
0.0
French fries / croquettes
Sweet desserts (e.g. ice cream,
tiramisu, …)
Miscellaneous
4.2
9
10
order
9
2.0
0.4
0.6
0.5
Pizza & quiches
2.2 (17.8)
0.3
0.4
7
0.2
Other miscellaneouse
2.0 (21.3)
0.3
0.2
9
0.2
The contributions of each food group are expressed in percentage of daily iron intakes.a Includes cow's milk and goat's milk
22
b
Excludes cream cheese
c
Includes lard / animal fats and regular / low-fat / fat-free versions of cream cheese / sour cream / cream / cream substitutes / half-skimmed
products
d
includes only eggs reported separately and eggs included in disaggregated food mixtures
e
includes foods or components with negligible contributions to total nutrient intakes that could not be categorized in the above food groups (e.g.
herbs and spices / monosodium glutamate / starch / plain gelatin / artificial sweeteners / pectin / cocoa powder / etc.)
* The restgroup includes low nutritious, high energy-dense foods that are not recommended on a daily basis
23
Table 4 - Associations of iron intakes with socio-demographic characteristics of participants.
Dependent variable:
Coefficients
β
95% Confidence
Interval
Lower
Upper
Bound
Bound
value
P
β
SE
Intercept
-0.053
0.737
-1.498
1.392
0.943
Energy (kcal)
0.006
0.000
0.005
0.006
<0.001
Family size (<2 children) ‡
2.210
0.746
0.748
3.671
0.003
Lower secondary maternal education‡
6.272
2.533
1.307
11.236
0.013
Secondary maternal education‡
-0.886
0.643
-2.147
0.374
0.168
Lower secondary maternal education* age
-1.182
0.506
-2.174
-0.190
0.020
Secondary maternal education* age
0.142
0.140
-0.131
0.415
0.309
Energy * family size (≤2children) ‡
-0.002
0.000
-0.003
-0.001
0.001
Energy (kcal)
0.000
0.000
0.000
0.001
<0.001
Lower secondary maternal education‡
0.205
0.071
0.066
0.344
0.004
Secondary maternal education‡
0.006
0.028
-0.049
0.061
0.832
Energy * family size (≤2children) ‡
0.000
0.000
0.000
0.000
0.001
Energy (kcal)
0.005
0.000
0.004
0.006
<0.001
Age (years)
-0.217
0.092
-0.397
-0.037
0.018
Lower secondary maternal education‡
8.797
2.451
3.993
13.601
<0.001
Secondary maternal education‡
-0.869
0.651
-2.145
0.406
0.181
Family size (≤2children) ‡
Lower secondary maternal education *
age
1.922
0.752
0.448
3.397
0.011
-1.718
0.494
-2.685
-0.750
0.001
Secondary maternal education* age
0.146
0.141
-0.131
0.422
0.302
Family size *energy ‡
-0.001
0.000
-0.002
0.000
0.005
Total Iron intake (mg/d)
Haem iron (mg/d)
Non-haem iron (mg/d)
24
‡ Mother’s education level (lower secondary education (only the 3 first years, or less of the secondary
education), secondary education or higher education) and family size (≤2 children and > 2 children)
using higher educated mothers and > 2 children as a reference
25
Reference List
Black AE, Goldberg GR, Jebb SA, Livingstone MB, Cole TJ and Prentice AM (1991):
Critical evaluation of energy intake data using fundamental principles of energy
physiology: 2. Evaluating the results of published surveys. Eur.J.Clin.Nutr. 45, 583599
Carriquiry AL (1999): Assessing the prevalence of nutrient inadequacy. Public Health Nutr.
2, 23-33
Centers for Disease Control and Prevention (CDC) (1998): Recommendations to prevent and
control iron deficiency in the United States. Centers for Disease Control and
Prevention. MMWR Recomm.Rep. 47, 1-29
De Vriese, S., Huybrechts, I., Moreau, M. and Van Oyen, H. (2006): The Belgian Food
Consumption Survey 1 - 2004: Report (Enquête de consommation alimentaire Belge 1
- 2004 : Rapport). D/2006/2505/16
FAO/WHO (2004): Vitamin and mineral requirements in human nutrition. Rome: World
Health Organization and Food and Agriculture Organization of the United Nations
Fox MK, Reidy K, Novak T and Ziegler P (2006): Sources of energy and nutrients in the diets
of infants and toddlers. J.Am.Diet.Assoc. 106, S28-S42
Garemo M, Lenner RA and Strandvik B (2007): Swedish pre-school children eat too much
junk food and sucrose. Acta Paediatr. 96, 266-272
Gibson RS and Ferguson, E. L. (2008): An interactive 24-hour recall for assessing the
adequacy of iron and zinc intakes in developing countries. Washington, DC and Cali:
International Food Policy Research Institute (IFPRI) and International Center for
Tropical Agriculture (CIAT) 8
Guenther PM, Kott PS and Carriquiry AL (1997): Development of an approach for estimating
usual nutrient intake distributions at the population level. J Nutr. 127, 1106-1112
Heath AL and Fairweather-Tait SJ (2002): Clinical implications of changes in the modern
diet: iron intake, absorption and status. Best.Pract.Res.Clin Haematol. 15, 225-241
Hoge Gezondheidsraad. Dietary recommendations for Belgium: revised version november
2006 (Voedingsaanbevelingen voor België: herziene versie 2006). nr. 7145-2. 2006.
Brussel.
Hoge Gezondheidsraad. Dietary recommendations for Belgium: revised version november
2009 (Voedingsaanbevelingen voor België: herziene versie 2009). nr. 8309. 2009.
Brussel.
26
Huybrechts I and De Henauw S (2007): Energy and nutrient intakes by pre-school children in
Flanders-Belgium. Br J Nutr 98, 600-610
Huybrechts I, Maes L, Vereecken C, De Keyzer W., De Bacquer D., De Backer G. et al.
(2010): High dietary supplement intakes among Flemish preschoolers. Appetite. 54,
340-345
Huybrechts I, Matthys C, Pynaert I, De Maeyer M, Bellemans M, De Geeter H et al. (2008):
Flanders preschool dietary survey: rationale, aims, design, methodology and
population characteristics. The Archives of Public Health 66, 5-25
IOM. Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium,
Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc.
Food and Nutrition Board. 2001. Washington, DC, National Academy Press.
Iowa State University. C-side. http://www.cssm.iastate.edu/software/cside.html . 2006.
Krebs-Smith SM, Kott PS and Guenther PM (1989): Mean proportion and population
proportion: two answers to the same question? J.Am.Diet.Assoc. 89, 671-676
Lambert J, Agostoni C, Elmadfa I, Hulshof K, Krause E, Livingstone B et al. (2004): Dietary
intake and nutritional status of children and adolescents in Europe. Br J Nutr 92,
S147-S211
NEVO (2001): NEVO-Table, Dutch Food Composition Table 2001. Zeist. NEVO Foundation
(in Dutch)
Nusser SM, Carriquiry AL, Dodd KW and Fuller WA (1996): A semiparametric
transformation approach to estimating usual daily intake distributions.
J.Am.Stat.Assoc. 91, 1440-1449
Ohlund I, Lind T, Hornell A and Hernell O (2008): Predictors of iron status in well-nourished
4-y-old children. Am J Clin.Nutr. 87, 839-845
Pynaert I, Matthys C, Bellemans M, De Maeyer M, De Henauw S and De Backer G (2005):
Iron intake and dietary sources of iron in Flemish adolescents. Eur.J.Clin.Nutr. 59,
826-834
Royo-Bordonada MA, Gorgojo L, de Oya M, Garces C, Rodriguez-Artalejo F, Rubio R et al.
(2003): Food sources of nutrients in the diet of Spanish children: the Four Provinces
Study. Br J Nutr 89, 105-114
Sichert-Hellert W and Kersting M (2003): Impact of Fortified Breakfast Cereals on Iron
Intake in German Children and Adolescents. [Miscellaneous Article]. J Pediatr
Gastroenterol Nutr 36, 149-153
Subar AF, Krebs-Smith SM, Cook A and Kahle LL (1998): Dietary sources of nutrients
among US children, 1989-1991. Pediatrics. 102, 913-923
Thane CW, Bates CJ and Prentice A (2003): Risk factors for low iron intake and poor iron
status in a national sample of British young people aged 4-18 years. Public Health
Nutr. 6, 485-496
27
VIG. De voedingsdriehoek: een praktische voedingsgids. Vlaams Instituut voor
Gezondheidspromotie (VIG) . 2004.
Westenbrink S, Oseredczuk M, Castanheira I and Roe M (2009): Food composition
databases: The EuroFIR approach to develop tools to assure the quality of the data
compilation process. Food chemistry 113, 759-767
28