A micronutrient-fortiﬁed young-child formula improves the iron and
vitamin D status of healthy young European children: a randomized,
double-blind controlled trial
Marjolijn D Akkermans,
* Simone RBM Eussen,
Judith M van der Horst-Graat,
Ruurd M van Elburg,
Johannes B van
and Frank Brus
Department of Pediatrics, Juliana Children’s Hospital/Haga Teaching Hospital, The Hague, Netherlands;
Danone Nutricia Research, Utrecht, Netherlands;
Department of Pediatrics, Emma Children’s Hospital/Academic Medical Center, Amsterdam, Netherlands; and
Department of Pediatrics, VU University
Medical Center, Amsterdam, Netherlands
Background: Iron deﬁciency (ID) and vitamin D deﬁciency (VDD)
are common among young European children because of low di-
etary intakes and low compliance to vitamin D supplementation
policies. Milk is a common drink for young European children.
Studies evaluating the effect of milk fortiﬁcation on iron and vita-
min D status in these children are scarce.
Objective: We aimed to investigate the effect of a micronutrient-
fortiﬁed young-child formula (YCF) on the iron and vitamin D
status of young European children.
Design: In this randomized, double-blind controlled trial,
healthy German, Dutch, and English children aged 1–3 y were
allocated to receive either YCF (1.2 mg Fe/100 mL; 1.7 mg vita-
min D/100 mL) or nonfortiﬁed cow milk (CM) (0.02 mg
Fe/100 mL; no vitamin D) for 20 wk. Blood samples were taken
before and after the intervention. The primary and secondary out-
comes were change from baseline in serum ferritin (SF) and
25-hydroxyvitamin D [25(OH)D], respectively. ID was deﬁned
as SF ,12 mg/L in the absence of infection (high-sensitivity
C-reactive protein ,10 mg/L) and VDD as 25(OH)D ,50 nmol/L.
Statistical adjustments were made in intention-to-treat analyses for
sex, country, age, baseline micronutrient status, and micronutrient
intake from food and supplements (and sun exposure in the case of
vitamin D outcomes).
Results: The study sample consisted of 318 predominantly Cauca-
sian (w95%) children. The difference in the SF and 25(OH)D change
between the treatment groups was 6.6 mg/L (95% CI: 1.4, 11.7 mg/L;
P= 0.013) and 16.4 nmol/L (95% CI: 9.5, 21.4 nmol/L; P,0.001),
respectively. The probability of ID (OR 0.42; 95% CI:0.18, 0.95;
P= 0.036) and VDD (OR 0.22; 95% CI: 0.01, 0.51; P,0.001)
after the intervention was lower in the YCF group than in the
Conclusion: Micronutrient-fortiﬁed YCF use for 20 wk preserves
iron status and improves vitamin D status in healthy young children
in Western Europe. This trial was registered at www.trialregister.nl
as NTR3609. Am J Clin Nutr doi: 10.3945/ajcn.116.136143.
Keywords: iron deﬁciency, vitamin D deﬁciency, cow milk, young-
child formula, vitamin D supplements, micronutrient fortiﬁcation,
Micronutrient deﬁciency is a major public health problem that
even in industrialized countries contributes to the global burden
of disease. Iron deﬁciency (ID)
and vitamin D deﬁciency
(VDD) are among the most common micronutrient deﬁciencies
in young children worldwide (1). ID can lead to iron deﬁciency
anemia (IDA) (2), and both of these conditions are associated
with impaired neurodevelopment (3–6). It has been suggested
that vitamin D has an important role in immune system func-
tioning and in preventing cancers, whereas VDD can lead to
rickets (7, 8).
Despite national nutritional recommendations, the iron and
vitamin D intake of young children in Europe has been shown to
often be insufﬁcient in preventing ID and VDD (9–13). Further-
more, although the use of vitamin D supplements is associated
with a lower prevalence of VDD, compliance seems to be low (7,
9, 10). To increase compliance, fortiﬁcation of commonly used
food products has been suggested. Fortiﬁcation produces a more
gradual increase in serum micronutrient concentration; in ad-
dition, if consumed on a regular and frequent basis, fortiﬁed
products will maintain body stores of nutrients more efﬁciently
and effectively than intermittent supplements (1).
Several international trials have shown beneﬁcial effects of
food fortiﬁcation (e.g., milk, bread, and margarine) on iron
(14–23) and vitamin D (24–27) status in children. Milk is a
Supported by Danone Nutricia Research. This is a free access article,
distributed under terms (http://www.nutrition.org/publications/guidelines-and-
policies/license/) that permit unrestricted noncommercial use, distribution, and
reproduction in any medium, provided the original work is properly cited.
Present address: Food and Biobased Research, University of Wagenin-
gen, Wageningen, Netherlands.
*To whom correspondence should be addressed. E-mail: m.d.akkermans@
Received April 5, 2016. Accepted for publication December 5, 2016.
Abbreviations used: AE, adverse event; CM, cow milk; hsCRP, high-
sensitivity C-reactive protein; ID, iron deﬁciency; IDA, iron deﬁciency ane-
mia; ITT, intention to treat; PP, per protocol; SF, serum ferritin; VDD, vitamin
D deﬁciency; YCF, young-child formula; 25(OH)D, 25-hydroxyvitamin D.
Am J Clin Nutr doi: 10.3945/ajcn.116.136143. Printed in USA. Ó2017 American Society for Nutrition 1of9
AJCN. First published ahead of print January 4, 2017 as doi: 10.3945/ajcn.116.136143.
Copyright (C) 2017 by the American Society for Nutrition
popular source for delivering fortiﬁcation because of its wide
availability and acceptance. However, randomized controlled
trials investigating the effect of this strategy in young European
children are scarce. Existing studies differ in fortiﬁcation dosage
and outcome parameters, which hampers the comparison of
results (14, 15, 17, 23, 25). Moreover, the inﬂuence of an in-
fection [e.g., on serum ferritin (SF)] or the season (on vitamin D
status) on outcome variables is not always taken into account.
The primary objective of this study (NTR3609) was to in-
vestigate the effect of a micronutrient-fortiﬁed young-child
formula (YCF) given for 20 wk on ferritin concentrations of
healthy children aged 12–36 mo living in Western Europe
compared with the use of nonfortiﬁed cow milk (CM). Sec-
ondary objectives were to establish the effect of the intervention
on the prevalence of ID and IDA, serum 25-hydroxyvitamin
D [25(OH)D] concentrations, and the prevalence of VDD.
This randomized, double-blind controlled trial was performed
in Western Europe from 2012 October to 2014 September. The
participating countries were Germany, (9 private pediatric clinics
spread throughout the country), Netherlands, (Juliana Children’s
Hospital/Haga Teaching Hospital in The Hague, VU University
Medical Center in Amsterdam, and Sophia Children’s Hospital/
Erasmus Medical Center in Rotterdam) and the United Kingdom
(Royal National Orthopedic Hospital in London and St. Mary’s
Hospital in Newport, Isle of Wight). The study was approved by
the medical ethical review board of all participating sites. The
prevalence of and risk factors for ID and VDD at baseline have
previously been published (9).
Inclusion and exclusion criteria
Children aged 12–36 mo with a stable health status (i.e., without
any known chronic or recent acute disease) were eligible for this
study. The children were familiar with and currently drinking
milk products and were expected to have a study product in-
(,32 wk, or ,37 wk with a birth weight ,1800 g); known in-
fection within the last week or infection needing medical assistance
or treatment within the last 2 wk; known hemoglobinopathies; any
case of anemia treated within the last 3 mo; a blood transfusion
received within the last 6 mo; the presence of a relevant congenital
abnormality; chromosomal disorder or severe disease (such as major
congenital heart disease or Down syndrome); having a disorder
requiring a special diet (such as food intolerance or food allergy
or complaints such as reﬂux, constipation, and cramps); current use
of antiregurgitation, antireﬂux, or laxative medication; participation
in any other study involving investigational or marketed products
within 2 wk before entering the study; known allergy or intolerance
to components of the study products (e.g., milk powder, lactose, or
ﬁsh protein); and vaccination with a live or live-attenuated vaccine
received within the last 2 wk. Finally, parents had to be able to
understand the local language and read and ﬁll out questionnaires.
Subjects were recruited in 2 ways based on the local situation
at the individual sites. In the Netherlands and the United
Kingdom, parents of eligible subjects were informed about the
study during a preoperative visit before an elective nonemergency
surgical procedure (e.g., urologic surgeries, inguinal or umbilical
hernia operations, or ear-nose-throat procedures). After written
informed consent was obtained, the ﬁrst blood drawn was
combined with the placement of an intravenous catheter needed
for administering general anesthesia. The subjects from Germany
were recruited during a regular visit to their pediatrician. After
taking the blood sample, the included children were randomly
allocated to receive either micronutrient-fortiﬁed YCF (test
product) or nonfortiﬁed CM (control product) for a period of
20 wk. A computer model was used for block randomization in
which stratiﬁcation was applied for country and sex. Parents (and
their children), investigators, and treating physicians were
blinded to product allocation by coding the cans containing the
study products. Parents then answered questions about their
child’s demographic and socioeconomic characteristics, day
care center attendance, sun exposure, and medical history. Food
intake was measured by a food-frequency questionnaire that was
adapted and translated from previously published dietary ques-
tionnaires (28–30). Micronutrient intake was calculated with the
use of a Dutch nutrient databank (31). The results reﬂected the
intake 1 mo before the baseline visit.
During the intervention period, parents were asked not to
change their child’s dietary habits, including the use of sup-
plements. After 1, 5, and 15 wk, parents were contacted by
phone to discuss study product compliance and completion of
diaries. These diaries included daily study product intake, pos-
sible adverse events (AEs) and serious AEs, and the use of
medication. Diaries on stool frequency and consistency were
completed 7 d before the last 2 scheduled visits (weeks 10 and
20). Stool frequency was measured as the number of stools
passed on each day of the 7 d, and stool consistency was mea-
sured on an ordered 5-point scale with pictures (1: watery;
2: soft, pudding-like; 3: soft-formed; 4: dry-formed; 5: dry hard
pellets). Halfway through the study, parents were asked to visit
the study center to collect a new study product and to discuss
potential issues. After 20 wk in all 3 countries, a second venous
blood sample was taken while subjects visited the hospital or
clinic for the last time (Figure 1). During all 3 visits (baseline,
10 wk, and 20 wk), height or length and weight were measured.
Body weight was measured to the nearest 0.1 kg with the use
of a calibrated weighing scale. Height was measured to the
nearest 0.2 cm, standing and without wearing shoes, with the use
of a calibrated stadiometer. In those children who were not able
to stand, length was measured lying down with the use of a
length board to a precision of 0.2 cm. Weight-for-age zscores,
height- or length-for-age zscores, and BMI-for-age zscores
following WHO growth charts were calculated.
Test and control products: YCF and CM
The detailed nutrient proﬁles of both study products are shown
in Table 1. The test product was a commercially available
micronutrient-fortiﬁed YCF containing 1.2 mg Fe/100 mL and
1.7 mg vitamin D/100 mL. The control product was a non-
fortiﬁed CM that contained 0.02 mg Fe/100 mL and no vitamin D.
The energy concentrations of both products were comparable
(46 kcal/100 mL for CM compared with 50 kcal/100 mL for
YCF). Both YCF and CM were supplied in powdered form with
instructions for preparing the milk by diluting the powder with
2of9 AKKERMANS ET AL.
water. The study products were produced, provided, and coded
(for blinding purposes) by Nutricia Cuijk (commissioned by
Danone Nutricia Research).
Deﬁnitions and laboratory analyses
Blood samples were stored at Nutricia Research Analytic
Science Laboratory at 2808C before being analyzed in 4 batches.
Some parameters were analyzed at Reinier de Graaf Groep
Laboratory. SF and serum 25(OH)D were analyzed with the use
of an Abbott Architect i2000 immunology analyzer with a
chemiluminesent immunoassay and chemoluminescent micro-
particle immunoassay, respectively.
IDwasdeﬁnedasSF,12 mg/L and IDA as ID combined with a
hemoglobin concentration ,110 g/L according to the WHO (2).
Ferritin is an acute-phase protein that may increase when an
infection is present, even in the presence of low iron stores.
Therefore, high-sensitivity C-reactive protein (hsCRP), also an
acute-phase protein, was determined in all venous blood samples,
and all children with elevated hsCRP concentrations ($10 mg/L)
were excluded from the ID and IDA analyses.
VDD was deﬁned as serum 25(OH)D ,50 nmol/L because
this concentration is the cutoff recommended by most experts (7,
10, 32). As previously described (9), mean annual vitamin D
concentrations were calculated from the single values to adjust
for seasonal variations in circulating 25(OH)D concentrations
with the use of the cosinor model of Sachs et al. (33).
Sample size calculations were based on the primary parameter
(SF) with the use of data from Szymlek-Gay et al. (20).
Assuming a difference between treatment groups in SF change
(from baseline to endpoint) of 8.1 mg/L (621 mg/L), 216
subjects (108/group) were required for a statistical power of 0.8
(a= 0.05) in a 2-sided ttest. In addition, to account for strati-
ﬁcation and dropout (w25%), 288 subjects were anticipated to
be required for inclusion in the study.
Statistical analyses, described in a statistical analysis plan
that was ﬁnalized before unblinding of the study, were performed
with the use of SPSS version 21.0 (IBM). As a ﬁrst step, the
distribution of variables was assessed with the use of histograms
and quantile-quantile plots. Categorical variables were then
summarized by frequency and percentage distributions, and
normally distributed continuous variables were summarized
by means and SDs. Nonnormally distributed continuous var-
iables were expressed as medians (IQRs) (quantiles 1 and 3).
The basic principle of our analyses was to analyze data on an
intention-to-treat (ITT) basis, in which all children for whom
there was information were analyzed in the groups to which they
were originally allocated, irrespective of whether they actually
followed the treatment regimen. Vitamin D status and hemo-
globin concentrations were analyzedinthisITTstudysample.
Analyses regarding iron status (including IDA) were then
performed in the modiﬁed ITT study sample. This sample in-
cluded all subjects from the ITT study sample in whom normal
hsCRP concentrations (,10 mg/L) were measured at both
baseline and at the end of the study.
The effect of the study products on SF and serum 25(OH)D
concentration was investigated with the use of linear regression
analyses, whereas its effect on the prevalence of both micro-
nutrient deﬁciencies was determined with the use of logistic
regression analyses. In principle, these analyses were performed
while adjusting for sex and country (stratiﬁcation factors), age,
micronutrient status at baseline, and iron or vitamin D intake
(from food and supplements) at baseline. In the case of vitamin D
analyses, we also adjusted for sun exposure of $1 h/d.
FIGURE 1 Flowchart of the study design showing the study procedure during the 20-wk intervention period and 2-wk follow-up period. C, phone contact;
IMPROVING IRON AND VITAMIN D STATUS OF CHILDREN 3of9
Finally, all previously mentioned analyses, including adjust-
ments for the predeﬁned variables, were also performed in the 2
per-protocol (PP) samples. These samples consisted of subjects
from the ITTand the modiﬁed ITT sample that demonstrated good
compliance with instructions for consuming the assigned study
product. Good compliance was deﬁned as consuming $151 mL
study product/d for $80% of the days within the last 28 d of
study product intake. All CIs are 2-sided with a conﬁdence level
of 95%. Statistical signiﬁcance was deﬁned as P,0.05.
Study sample and baseline characteristics
Because of a higher rate of dropouts than anticipated, 318
subjects were ﬁnally included in the ITT study sample: 158 in the
YCF group and 160 in the CM group (Figure 2). This ITT sample
consisted of 264 children from Germany (83.0%), 42 from the
Netherlands (13.2%), and 12 from the United Kingdom (3.8%).
Tables 2 and 3show the baseline characteristics and baseline iron
and vitamin D status of the 2 treatment groups, respectively.
These tables show a higher educational and working status of the
parents of the YCF group than the CM group, although more data
on this are missing in the CM group than the YCF group. Fur-
thermore, the CM group had a higher iron intake from milk and
higher vitamin D intake from food than the YCF group (Table 2).
Figure 2 shows the number of children included in our different
study groups and analysis sets. There were no differences in the
number of or reasons for early termination (Figure 2) or in the
percentage of children demonstrating good compliance (69.6%
compared with 71.9%; P= 0.659) between YCF and CM users.
The aforementioned observed differences in educational status,
working status, and iron and vitamin D intake between CM and
YCF users were also found in our modiﬁed ITT sample and the
PP and modiﬁed PP sample (data not shown).
Iron status and ID and IDA prevalence
In the (complete) modiﬁed ITT sample, the difference in
change from baseline in SF between the treatment groups was
6.6 mg/L (95% CI: 1.4, 11.7 mg/L; P= 0.013). The estimated mean 6
SEM change in SF concentration from baseline was 24.9 6
2.2 mg/L for the CM group and +1.7 62.4 mg/L for the YCF
group (Table 3). We then performed explorative analyses in which
the modiﬁed ITT sample was divided into 4 subgroups repre-
senting categories of most frequently consumed daily volume
within the last 4 wk (1–150, 151–300, 301–500, and .500 mL/d).
The effect sizes in these subgroups were analyzed while adjusting
for sex, country, and baseline SF concentration. In children con-
suming .500 mL/d, the group difference in change from baseline
in SF was 11.2 mg/L (95% CI: 1.8, 20.6 mg/L).
Tab le 4 shows the prevalence rates of ID and IDA before and
after the intervention. The probability of ID after the intervention
was lower in the YCF group than the CM group (OR: 0.42; 95%
CI: 0.18, 0.95; P= 0.036). The IDA prevalence rates were too low
to evaluate the effect of the intervention on IDA prevalence.
Hemoglobin concentrations and anemia
At baseline, 18.9% of the children were anemic: 23 in the YCF
group and 37 in the CM group. After the intervention, 4 YCF
users and 13 CM users were anemic (P= 0.021). In contrast, the
mean change from baseline in hemoglobin was comparable for
YCF and CM users (Table 3).
Vitamin D status and VDD prevalence
In the (complete) ITT sample, the difference in change from
baseline in 25(OH)D between the treatment groups was 16.4 nmol/L
(95% CI: 9.5, 21.4 nmol/L; P,0.001). The estimated mean 6SEM
change in 25(OH)D concentration from baseline was 27.2 6
2.5 nmol/L for the CM group and 9.2 62.8 nmol/L for the YCF
group (Table 3). We then performed explorative analyses in which
we determined the effect sizes in subgroups based on the most fre-
quently consumed daily volume within the last 4 wk while adjusting
for sex, country, and baseline 25(OH)D concentration. In children
who consumed .500 mL/d, the group difference in change from
baseline in 25(OH)D was 18.1 nmol/L (95% CI: 3.0, 33.2 nmol/L).
Table 4 shows the prevalence rates of VDD before and after the
intervention. The probability of VDD after the intervention was
lower in the YCF group than the CM group (OR: 0.22; 95% CI:
0.01, 0.51; P,0.001).
ID and VDD
At baseline, 8.2% of the YCF group and 5.6% of the CM group
were iron- and vitamin D–deﬁcient. These prevalence rates in-
creased in the CM group to 15.3% and decreased for YCF users
to 4.0% after 20 wk of study product intake.
CM and YCF content per 100 mL of prepared product
Proteins 3.5 1.1
Carbohydrates 5.2 6.6
Fats 1.7 1.9
Fibers 0.0 0.8
Sodium, mg 40.0 20.0
Potassium, mg 174.0 56.0
Chloride, mg 101.0 31.0
Calcium, mg 127.0 110.0
Phosphorus, mg 100.0 67.0
Magnesium, mg 12.0 10.0
Nonheme iron, mg 0.02 1.2
Zinc, mg 0.40 0.90
Copper, mg 2.4 59.0
Manganese, mg 0.91 16.0
Selenium, mg 0.90 2.3
Iodine, mg 9.8 17.0
Vitamin A, mg REs 13.0 65.0
,mg 0.0 1.7
a-Tocopherol (vitamin E), mg 0.0 1.3
Vitamin K, mg 0.0 5.0
Thiamin (vitamin B-1), mg 28.0 70.0
Riboﬂavin (vitamin B-2), mg 142.0 87.0
Vitamin B-6, mg 30.0 60.0
Folic acid, mg 1.6 18.0
Vitamin B-12, mg 0.24 0.13
Biotin, mg 2.0 1.7
Vitamin C, mg 0.55 14.0
CM, cow milk; RE, retinol equivalent; YCF, young-child formula.
4of9 AKKERMANS ET AL.
PP and modiﬁed PP analyses conﬁrmed the results from the
ITT and modiﬁed ITT analyses, although the effect sizes were
larger in the PP analyses (data not shown).
Safety of study products: AEs, gastrointestinal tolerance,
Overall, there were no statistically signiﬁcant differences in
the number and severity of reported AEs between the YCF and
FIGURE 2 Flowchart of the study sample. Children with elevated hsCRP concentrations ($10 mg/L) were excluded from the analyses regarding iron status to
prevent falsely elevated or normal ferritin concentrations in the case of an infection. The PP groups consisted thereafter of children that demonstrated good compliance
with instructions for consuming the assigned study product. Good compliance was deﬁned as consuming $151 mL study product/d $80% of the days within the last
28 d of study product intake. CM, cow milk; hsCRP, high-sensitivity C-reactive protein; ITT, intention to treat; PP, per protocol; YCF, young-child formula.
IMPROVING IRON AND VITAMIN D STATUS OF CHILDREN 5of9
CM groups (data not shown). Of the reported AEs (939 in 258
subjects), 33 events in 27 subjects were considered to be related
to the study product. Most of these supposedly related AEs
compromised gastrointestinal complaints. There were 30 reports
of diarrhea in 26 subjects (17%) from the YCF group and 17
reports of diarrhea in 14 subjects (9.2%) from the CM group
(P= 0.061). In both groups, most of these reports were
documented in the ﬁrst week after the start of the study product,
and the diarrhea lasted ,5 d (data not shown). The complaints
were not severe, and most of the complaints were resolved
without any medication. Furthermore, there were 9 serious AEs
reported in 8 subjects. These events were diverse and evenly
distributed over the treatment groups (data not shown). All were
considered to be unrelated to the study product.
Table 5 shows the stool characteristics (frequency and con-
sistency) recorded before each hospital or clinic visit (baseline,
10 wk, and 20 wk) by treatment group. No statistically signiﬁ-
cant differences in gastrointestinal tolerance were observed be-
tween the treatment groups. Finally, there were also no statistically
signiﬁcant differences in the anthropometric data between the 2
treatment groups during the intervention period (data not shown).
To our knowledge, this is the ﬁrst randomized, double-blind
controlled trial to describe the effect of micronutrient-fortiﬁed YCF
on both the iron and vitamin D status of healthy children aged 12–36
mo in Western Europe. The results of this study indicate that the
daily consumption of YCF for 20 wk preserves iron status and
Adjusted mean changes in iron and vitamin D status after the intervention
Serum ferritin, mg/L
Baseline 28.9 617.1 25.6 614.8
20 wk 22.0 617.5 27.9 617.4
Change from baseline 24.9 62.2
Baseline 118.5 610.7 119.8 68.8
20 wk 121.9 69.8 123.9 68.2
Change from baseline 3.5 69.1 3.1 68.9
Serum 25(OH)D, nmol/L
Baseline 70.2 626.7 69.4 627.0
20 wk 62.0 629.9 77.8 626.6
Change from baseline 27.2 62.5
Values are means 6SDs unless otherwise indicated. The change from
baseline in serum ferritin and serum 25(OH)D were analyzed while adjusting
for sex and country (stratiﬁcation factors), age, micronutrient status at base-
line, and the iron or vitamin D intake from food and supplements (and sun
exposure in the case of vitamin D). The iron analyses were performed in the
modiﬁed intention-to-treat sample in which the children with an elevated
high-sensitivity C-reactive protein were excluded to prevent falsely elevated
or normal ferritin concentrations in the case of an infection. CM, cow milk;
YCF, young-child formula; 25(OH), 25-hydroxyvitamin D.
Estimated mean 6SEM (all such values).
The group difference in the change from baseline in serum ferritin and
serum 25(OH)D between treatment groups was 6.6 mg/L (95% CI: 1.4, 11.7 mg/L;
P= 0.013) and 16.4 nmol/L (95% CI: 9.5, 21.4 nmol/L; P,0.001), respectively.
Baseline characteristics of the intention-to-treat study sample
Demographic and general characteristics
Male, n(%) 91 (56.9) 89 (56.3)
Caucasian, n(%) 151 (94.4) 152 (96.2)
Age, mo 20.5 67.72 20.8 67.3
Gestational age, wk 39.0 61.9 39.3 61.4
Birth weight, g 3238 6553 3400 6513
Educational status of either parent,
None 1 (0.6) 0 (0)
Primary school 30 (18.8) 21 (13.3)
High school or trade school 66 (41.2) 80 (50.6)
University 30 (18.8) 35 (22.2)
Unknown 33 (20.6) 22 (13.9)
Professional status of parents, n(%)
$1 working 118 (73.8) 126 (79.7)
None working 3 (1.9) 8 (5.1)
Unknown 39 (24.3) 24 (15.2)
Daycare attendance, n(%)
Yes 75 (46.9) 66 (41.8)
No 84 (52.5) 91 (57.6)
Unknown 1 (0.6) 1 (0.6)
$1 h spent outside/d, n(%)
Yes 135 (84.4) 124 (78.5)
No 25 (15.6) 34 (21.5)
Use of sunscreen or protective
Yes 37 (23.1) 51 (32.3)
No 117 (73.1) 103 (65.2)
Unknown 6 (3.8) 4 (2.5)
Characteristics at baseline
Weight-for-age zscore 0.15 60.98 0.28 60.92
Height- or length-for-age zscore 0.19 60.99 0.11 61.00
BMI-for-age zscore 0.3 61.1 0.3 61.0
Milk intake during previous month
CM, n(%) 68 (42.5) 73 (46.2)
YCF, n(%) 85 (53.1) 79 (50.0)
Other, n(%) 7 (4.4) 6 (3.8)
Amount per day, mL 517 6223 512 6230
Use of supplements containing
Yes 2 (1.3) 3 (1.9)
No 152 (95.0) 151 (95.6)
Unknown 6 (3.7) 4 (2.5)
Use of supplements containing
vitamin D, n(%)
Yes 51 (31.8) 43 (27.2)
No 103 (64.4) 111 (70.3)
Unknown 6 (3.8) 4 (2.5)
Iron intake at baseline,
3.1 (0.0–5.2) 2.3 (0.0–4.8)
From food 6.8 (5.0–9.9) 6.7 (4.3–10.1)
Vitamin D intake at baseline,
4.4 (0.0–7.0) 4.4 (0.0–6.3)
From food 5.3 (1.1–7.7) 2.0 (0.8–7.0)
Values are n(%) or means 6SDs unless otherwise indicated. CM,
cow milk; YCF, young-child formula.
Medians (IQRs) because of no normal distribution.
Includes YCF and CM.
6of9 AKKERMANS ET AL.
improves vitamin D status in young European children. Further-
more, neither study product was related to the incidence of serious
AEs. The use of YCF may therefore be an effective and practical
strategy for preventing ID and VDD in young European children.
We observed a modest increase in SF among the children who
consumed YCF. Explorative analyses based on the type of milk
before the start of the study (formula or CM) showed a higher
increase in SF in original CM users than in original formula users
(data not shown). Therefore, young children who consumed CM
would probably beneﬁt the most from micronutrient-fortiﬁed
YCF. One would normally expect a decrease of SF over time
because blood volume expands rapidly during growth, requiring
increasing erythropoiesis with the use of stored iron and sub-
sequently decreasing SF concentrations (14, 15, 20, 23). Because
the SF concentration increased modestly in the YCF group, we
suggest that the use of micronutrient-fortiﬁed YCF preserves iron
stores in young European children.
Four European studies have also reported on the effect of
fortiﬁed formula on iron status (14, 15, 17, 23), although only 2
(14, 17) used formula with a comparable iron content of
1.2 mg/100 mL. First, Daly et al. (14) investigated the hema-
tologic effects of a follow-on formula in a group of inner-city
toddlers whose mothers had already switched to pasteurized
CM by 6 mo of age. SF concentrations in formula users remained
stable but decreased signiﬁcantly in the toddlers that continued on
CM. The second study that used the same iron dosage focused on
the mental and psychomotor developmental indexes at the age of
18 mo after 9 mo of consuming fortiﬁed formula. The authors
reported signiﬁcantly higher SF concentrations at 18 mo in the
fortiﬁed formula users than the nonfortiﬁed formula and CM
users. Unfortunately, they did not report on SF concentrations at
baseline (17). Therefore, the results of our study are only
comparable to Daly et al. (14). However, they included younger
children, had a longer intervention period, and did not specify
details on the ethnicity and socioeconomic status of the par-
ticipating children. Furthermore, they did not take into account
the inﬂuence of a possible infection on SF concentrations. These
differences in study design make it difﬁcult to compare results.
Vitamin D status
Our observed increase in serum 25(OH)D concentrations in the
YCF group is conﬁrmed by a study from Hower et al. (25) among
German children. In contrast, Madsen et al. (26) found a decrease
in serum 25(OH)D concentration in both formula and CM users in
Denmark. In the latter study, a lower fortiﬁcation dosage of only
0.38 mg vitamin D/100 mL was used compared with 1.7 mg/100 mL
in our YCF. This lower fortiﬁcation dosage may not be sufﬁcient for
maintaining adequate serum 25(OH)D concentrations.
The VDD prevalence in our study decreased in the YCF group
but increased in the CM group (to 33.3%). In Hower et al. (25),
higher prevalence rates #79.2% were found in CM users. In this
study, the inﬂuence of vitamin D–fortiﬁed formula (2.85 mg
vitamin D/100 mL) on vitamin D status was investigated in
children aged 2–6 y. Vitamin D status was determined before
and after winter. It is known that the risk for VDD increases
during the winter (7, 10), which may explain why VDD preva-
lence rates were higher than in our study.
Only a minority of the children (w30%) in our study received
vitamin D supplements (mean content: 10.7 mg/d), although
policies regarding vitamin D supplementation exist in all 3 par-
ticipating countries. This emphasizes the need for new strategies,
such as the use of fortiﬁed food products.
Advantage of fortiﬁcation with iron and vitamin D (and
Most of the previously mentioned randomized controlled trials
studied the effect of single iron- or single vitamin D–fortiﬁed
food products. However, a comprehensive review by Best et al.
Gastrointestinal tolerance: stool frequency and consistency
CM (n= 153) YCF (n= 153)
Baseline 2 (1–2)
10 wk 1 (1–2) 1 (1–2)
20 wk 2 (1–2) 1 (1–2)
Baseline Soft-formed (54.6) Soft-formed (55.3)
10 wk Soft-formed (52.8) Soft-formed (45.0)
20 wk Soft-formed (57.8) Soft-formed (47.1)
Analyses were performed in all children from the intention-to-treat
sample that actually drank any study product. There were no statistically
signiﬁcant differences between the 2 treatment groups. CM, cow milk; YCF,
Values at baseline were recorded as a single integer; values at 10 and
20 wk were derived from 7 daily frequency values.
Median; IQR in parentheses (all such values).
Presented on an ordered scale as the most frequently recorded stool
consistency. The options were watery; soft, pudding-like; soft-formed; dry-
formed; and dry hard pellets.
Iron and vitamin D deﬁciency before and after the intervention
Iron deﬁciency, n(%) 0.42
Baseline 17 (11.9) 21 (14.3)
20 wk 29 (29.6) 14 (13.9)
Iron deﬁciency anemia, n(%) —
Baseline 8 (5.6) 4 (2.7)
20 wk 4 (4.0) 0 (0.0)
Vitamin D deﬁciency, n(%) 0.22
Baseline 35 (21.9) 40 (25.3)
20 wk 37 (33.3) 15 (13.5)
Iron deﬁciency was deﬁned as serum ferritin ,12 mg/L in children
without an elevated high-sensitivity C-reactive protein. Iron deﬁciency
anemia was deﬁned as iron deﬁciency combined with a hemoglobin concen-
tration ,110 g/L. Vitamin D deﬁciency was deﬁned as serum 25-hydroxy-
vitamin D ,50 nmol/L. OR column shows the odds of having iron deﬁciency
calculated while adjusting for sex and country (stratiﬁcation factors), age,
micronutrient status at baseline, and the iron or vitamin D intake from food
and supplements (and sun exposure in the case of vitamin D). *P,0.05.
CM, cow milk; YCF, young-child formula.
IMPROVING IRON AND VITAMIN D STATUS OF CHILDREN 7of9
(34) showed that multimicronutrient fortiﬁcation, such as our
YCF, results in more positive effects on biochemical indicators
of micronutrient status. In general, it is believed that micro-
nutrients can interact with each other (e.g., by competing for the
same transporter) and hereby lead to a different absorption of
other micronutrients (34, 35).
For example, ID and VDD seem to inﬂuence each other in a
negative way, but the precise pathogenesis is unclear (36–39).
Vitamin D has been suggested to increase the storage and re-
tention of iron by reducing the activity of proinﬂammatory cy-
tokines that inhibit iron absorption. On the other hand, it is known
that ID impairs the intestinal absorption of fat and the fat-soluble
vitamin A and therefore maybe also the absorption of fat-soluble
vitamin D. Moreover, iron is a cofactor for the enzyme 1a-
hydroxylase, which is responsible for the hydroxylation of 25(OH)D
D (40). Combined fortiﬁcation of iron and vitamin D
may therefore have a synergistic effect on iron and vitamin D
status. On the other hand, the bioavailability of iron also depends
on the composition of the diet. Food products containing heme
iron (e.g., meat) are better absorbed than those containing non-
heme iron (e.g., vegetables, milk). Furthermore, several factors
enhance (e.g., vitamin C) or inhibit (e.g., calcium) iron absorp-
tion. The amount of calcium is lower and the amount of vitamin C
is higher in our YCF than in our CM, and this could have also
inﬂuenced the found effect of our YCF on the change in iron
status. Another impact of multimicronutrient fortiﬁcation is that,
in addition to iron, several other micronutrients can also inﬂuence
hemoglobin concentrations (34).
Safety of micronutrient-fortiﬁed YCF
Iron could theoretically increase pro-oxidant stress with po-
tential adverse effects, including infection risk, and possibly
affect stool pattern. However, consistent with previous reports
(41), we observed no difference either in the frequency and
severity of AEs (15) nor in the stool characteristics between YCF
and CM users. Consistent with 2 other studies, we also did not
ﬁnd differences in anthropometric variables between YCF and
CM users (14, 17).
Strengths and limitations
The strength of our study is that it was a randomized, double-
blind controlled trial in a well-deﬁned sample of healthy young
Caucasian children in Western Europe. Furthermore, we took into
account the inﬂuence of infections and the season on iron and
vitamin D status variables, respectively.
Most of our study sample consisted of German children, and
almost all children were Caucasians. This lack of diversity may
hamper generalizing our results to other parts of the world.
However, although country and race may inﬂuence baseline
micronutrient status, we do not believe that it will change the
observed effect of our intervention. Another limitation of our
study is the use of an adapted food-frequency questionnaire that
was not validated for determining iron and vitamin D intake in
young children. However, these kind of questionnaires have been
found suitable for determining iron and vitamin D intake in
infants and preschoolers (42). Finally, the percentage of dropouts
(mostly because of nonacceptance of the study product), although
similar for both treatment groups, was higher than expected.
Approximately 40% of the children consumed CM before the
start of the study. During the intervention period, these children
were exposed to milk with a different consistency and possibly a
different taste. These differences can explain the refusal of some
children to drink the study products. Future studies should
therefore investigate the best form and taste of fortiﬁed YCF.
In conclusion, the daily use of micronutrient-fortiﬁed YCF
instead of nonfortiﬁed CM for 20 wk preserves iron status and
improves vitamin D status in children aged 12–36 mo in Western
Europe. The current recommendations state that CM is accept-
able after the age of 1 y, although the iron and vitamin D intake
in these children, including the use of vitamin D supplements, is
insufﬁcient for preventing ID and VDD. YCF, as part of a tod-
dler’s diet, could play a role in ensuring sufﬁcient intake of
certain micronutrients. The long-term beneﬁts of fortiﬁed YCF
on neurodevelopment and overall health remain to be elucidated.
We thank the following pediatricians for their contribution to this study:
¨rfer, Gerhard Bleckmann, Klaus Helm, Eivy Franke-
Beckmann, Peter A Soemantri, Thomas Adelt, Manfred Praun, Michael
Horn, Franziskus Schuhboeck, Hugo Heij, Koen Joosten, Benjamin Jacobs,
and Carina Venter.
The authors’ responsibilities were as follows—MDA: analyzed the data
and had primary responsibility for the ﬁnal content; and all authors: designed
and conducted the research, wrote the manuscript, and read and approved the
ﬁnal manuscript. JBvG is a member of the Dutch National Breastfeeding
Council, European Society for Paediatric Gastroenterology Hepatology and
Nutrition, Health Council of the Netherlands, and neonatal nutrition section
of the Dutch Pediatric Association and director of the Dutch National Donor
Human Milk Bank; he has received honoraria for presentations and consul-
tations from Danone, Nutricia, Mead Johnson Nutrition, Nestl´
Institute, Hipp, Prolacta, and Nutrinia in the past 3 y. SRBME and RMvE are
employees and JMvdH-G a former employee of Danone Nutricia Research.
Danone Nutricia Research was involved in the study design and implemen-
tation. The statistical analyses and interpretation of the data were performed
independently. None of the remaining authors reported a conﬂict of interest
related to the study.
1. Allen L, de Benoist B, Omar D, Hurrell R, editors. Guidelines on food
fortiﬁcation with micronutrients [Internet]. [cited 2016 Jan 2].
Available from: http://www.who.int/nutrition/publications/guide_
2. WHO. Iron deﬁciency anaemia: assessment, prevention and control
[Internet] [cited 2016 Jan 12]. Available from: http://www.who.int/
3. Algarin C, Nelson CA, Peirano P, Westerlund A, Reyes S, Lozoff B.
Iron-deﬁciency anemia in infancy and poorer cognitive inhibitory
control at age 10 years. Dev Med Child Neurol 2013;55:453–8.
4. Lozoff B. Iron deﬁciency and child development. Food Nutr Bull 2007;
5. Lozoff B, Clark KM, Jing Y, Armony-Sivan R, Angelilli ML,
Jacobson SW. Dose-response relationships between iron deﬁciency
with or without anemia and infant social-emotional behavior. J Pediatr
6. Lukowski AF, Koss M, Burden MJ, Jonides J, Nelson CA, Kaciroti N,
Jimenez E, Lozoff B. Iron deﬁciency in infancy and neurocognitive
functioning at 19 years: evidence of long-term deﬁcits in executive
function and recognition memory. Nutr Neurosci 2010;13:54–70.
7. Misra M, Pacaud D, Petryk A, Collett-Solberg PF, Kappy M. Vitamin
D deﬁciency in children and its management: review of current
knowledge and recommendations. Pediatrics 2008;122:398–417.
8. Thacher TD, Fischer PR, Strand MA, Pettifor JM. Nutritional rickets
around the world: causes and future directions. Ann Trop Paediatr
8of9 AKKERMANS ET AL.
9. Akkermans MD, van der Horst-Graat J, Eussen S, van Goudoever J,
Brus F. Iron and vitamin D deﬁciency in healthy young children in
Western Europe despite current nutritional recommendations. J Pediatr
Gastroenterol Nutr 2016;62:635–42.
10. Braegger C, Campoy C, Colomb V, Decsi T, Domellof M, Fewtrell M,
Hojsak I, Mihatsch W, Molgaard C, Shamir R, et al. Vitamin D in the
healthy European paediatric population. J Pediatr Gastroenterol Nutr
11. European Food Safety Authority. Scientiﬁc opinion on nutrient re-
quirements and dietary intakes of infants and young children in the
European Union [Internet]. [cited 2016 Jan 2]. Available from: http://
12. Eussen S, Alles M, Uijterschout L, Brus F, van der Horst-Graat J. Iron
intake and status of children aged 6–36 months in Europe: a systematic
review. Ann Nutr Metab 2015;66:80–92.
13. Ocke MC, van Rossum CTM, Fransen HP, Buurma EJM, de Boer EJ,
Brants HAM. Niekerk EM, van der Laan JD, Drijvers JJMM,
Ghameshlou Z. Dutch national food consumption survey—young
children 2005/2006 [Internet]. [cited 2016 Jan 2]. Available from:
14. Daly A, MacDonald A, Aukett A, Williams J, Wolf A, Davidson J,
Booth IW. Prevention of anaemia in inner city toddlers by an iron
supplemented cows’ milk formula. Arch Dis Child 1996;75:9–16.
15. Gill DG, Vincent S, Segal DS. Follow-on formula in the prevention of
iron deﬁciency: a multicentre study. Acta Paediatr 1997;86:683–9.
16. Gondolf UH, Tetens I, Michaelsen KF, Trolle E. Iron supplementation
is positively associated with increased serum ferritin levels in 9-month-
old Danish infants. Br J Nutr 2013;109:103–10.
17. Morley R, Abbott R, Fairweather-Tait S, MacFadyen U, Stephenson T,
Lucas A. Iron fortiﬁed follow on formula from 9 to 18 months im-
proves iron status but not development or growth: a randomised trial.
Arch Dis Child 1999;81:247–52.
18. Sazawal S, Dhingra U, Dhingra P, Hiremath G, Sarkar A, Dutta A,
Menon VP, Black RE. Micronutrient fortiﬁed milk improves iron sta-
tus, anemia and growth among children 1–4 years: a double masked,
randomized, controlled trial. PLoS One 2010;5:e12167.
19. Serdula MK, Lundeen E, Nichols EK, Imanalieva C, Minbaev M,
Mamyrbaeva T, Timmer A, Aburto NJ. Effects of a large-scale mi-
cronutrient powder and young child feeding education program on the
micronutrient status of children 6–24 months of age in the Kyrgyz
Republic. Eur J Clin Nutr 2013;67:703–7.
20. Szymlek-Gay EA, Ferguson EL, Heath AL, Gray AR, Gibson RS.
Food-based strategies improve iron status in toddlers: a randomized
controlled trial. Am J Clin Nutr 2009;90:1541–51.
21. Thankachan P, Selvam S, Surendran D, Chellan S, Pauline M,
Abrams SA, Kurpad AV. Efﬁcacy of a multi micronutrient-fortiﬁed
drink in improving iron and micronutrient status among school-
children with low iron stores in India: a randomised, double-masked
placebo-controlled trial. Eur J Clin Nutr 2013;67:36–41.
22. Villalpando S, Shamah T, Rivera JA, Lara Y, Monterrubio E. Fortifying
milk with ferrous gluconate and zinc oxide in a public nutrition program
reduced the prevalence of anemia in toddlers. J Nutr 2006;136:2633–7.
23. Virtanen MA, Svahn CJ, Viinikka LU, Raiha NC, Siimes MA,
Axelsson IE. Iron-fortiﬁed and unfortiﬁed cow’s milk: effects on iron
intakes and iron status in young children. Acta Paediatr 2001;90:724–31.
24. Houghton LA, Gray AR, Szymlek-Gay EA, Heath AL, Ferguson EL.
Vitamin D-fortiﬁed milk achieves the targeted serum 25-hydroxyvitamin
D concentration without affecting that of parathyroid hormone in New
Zealand toddlers. J Nutr 2011;141:1840–6.
25. Hower J, Knoll A, Ritzenthaler KL, Steiner C, Berwind R. Vitamin D for-
tiﬁcation of growing up milk prevents decrease of serum 25-hydroxyvitamin
D concentrations during winter: a clinical intervention study in Germany. Eur
J Pediatr 2013;172:1597–605.
26. Madsen KH, Rasmussen LB, Andersen R, Molgaard C, Jakobsen J,
Bjerrum P, Andersen EW, Mejborn H, Tetens I. Randomized
controlled trial of the effects of vitamin D-fortiﬁed milk and bread
on serum 25-hydroxyvitamin D concentrations in families in
Denmark during winter: the VitmaD study. Am J Clin Nutr 2013;
27. Piirainen T, Laitinen K, Isolauri E. Impact of national fortiﬁcation of
ﬂuid milks and margarines with vitamin D on dietary intake and serum
25-hydroxyvitamin D concentration in 4-year-old children. Eur J Clin
28. Davies DSC, Jewell T, McBride M, Burns H. Vitamin D—advice on
supplements for at risk groups [Internet]. [cited 2016 Jan 2]. Available
29. Geleijnse JM, Giltay EJ, Schouten EG, de Goede J, Oude Griep LM,
Teitsma-Jansen AM, Katan MB, Kromhout D. Effect of low doses of
n-3 fatty acids on cardiovascular diseases in 4,837 post-myocardial
infarction patients: design and baseline characteristics of the Alpha
Omega trial. Am Heart J 2010;159:539–46.e2.
30. Uijterschout L, Vloemans J, Vos R, Teunisse PP, Hudig C, Bubbers S,
Verbruggen S, Veldhorst M, de Leeuw T, van Goudoever JB, et al.
Prevalence and risk factors of iron deﬁciency in healthy young children
in the southwestern Netherlands. J Pediatr Gastroenterol Nutr 2014;58:
31. Netherlands National Institute for Public Health and the Environment.
Dutch nutrient composition table [Internet]. [cited 2016 Jan 2].
Available from: http://www.rivm.nl/Onderwerpen/N/Nederlands_
32. Holick MF. Vitamin D deﬁciency. N Engl J Med 2007;357:266–81.
33. Sachs MC, Shoben A, Levin GP, Robinson-Cohen C, Hoofnagle AN,
Swords-Jenny N, Ix JH, Budoff M, Lutsey PL, Siscovick DS, et al.
Estimating mean annual 25-hydroxyvitamin D concentrations from
single measurements: the Multi-Ethnic Study of Atherosclerosis. Am J
Clin Nutr 2013;97:1243–51.
34. Best C, Neuﬁngerl N, Del Rosso JM, Transler C, van den Briel T,
Osendarp S. Can multi-micronutrient food fortiﬁcation improve the
micronutrient status, growth, health, and cognition of schoolchildren?
A systematic review. Nutr Rev 2011;69:186–204.
35. Fischer Walker C, Kordas K, Stoltzfus RJ, Black RE. Interactive effects
of iron and zinc on biochemical and functional outcomes in supple-
mentation trials. Am J Clin Nutr 2005;82:5–12.
36. Jin HJ, Lee JH, Kim MK. The prevalence of vitamin D deﬁciency in
iron-deﬁcient and normal children under the age of 24 months. Blood
tamin D levels are associated with both iron deﬁciency and anemia
in children and adolescents. Pediatr Hematol Oncol 2015;32:99–
38. Sharma S, Jain R, Dabla PK. The role of 25-hydroxy vitamin D de-
ﬁciency in iron deﬁcient children of North India. Indian J Clin Bio-
39. Yoon JW, Kim SW, Yoo EG, Kim MK. Prevalence and risk factors for
vitamin D deﬁciency in children with iron deﬁciency anemia. Korean J
40. Takeyama K, Kitanaka S, Sato T, Kobori M, Yanagisawa J, Kato S. 25-
Hydroxyvitamin D3 1alpha-hydroxylase and vitamin D synthesis.
41. Singhal A, Morley R, Abbott R, Fairweather-Tait S, Stephenson T,
Lucas A. Clinical safety of iron-fortiﬁed formulas. Pediatrics 2000;
42. Roman-Vinas B, Ortiz-Andrellucchi A, Mendez M, Sanchez-Villegas A,
Pena Quintana L, Aznar LAM, Hermosos M, Serra-Majem L. Is the
food frequency questionnaire suitable to assess micronutrient intake
adequacy for infants, children and adolescents? Matern Child Nutr
IMPROVING IRON AND VITAMIN D STATUS OF CHILDREN 9of9