Research

JAMA Pediatrics | Original Investigation

Effect of Higher vs Standard Dosage of Vitamin D3

Supplementation on Bone Strength and Infection
in Healthy Infants
A Randomized Clinical Trial
Jenni Rosendahl, MD; Saara Valkama, MD; Elisa Holmlund-Suila, MD, PhD; Maria Enlund-Cerullo, MD, MSc;
Helena Hauta-alus, MSc; Otto Helve, MD, PhD; Timo Hytinantti, MD, PhD; Esko Levälahti, MSc;
Eero Kajantie, MD, PhD; Heli Viljakainen, PhD; Outi Mäkitie, MD, PhD; Sture Andersson, MD, PhD

Editorial
IMPORTANCE Although guidelines for vitamin D supplementation in infants have been widely Audio
implemented, they are mostly based on studies focusing on prevention of rickets. The
optimal dose for bone strength and infection prevention in healthy infants remains unclear. JAMA Pediatrics Patient Page
and Related article

OBJECTIVE To determine whether daily supplementation with 1200 IU of vitamin D3 Supplemental content
increases bone strength or decreases incidence of infections in the first 2 years of life
compared with a dosage of 400 IU/d.

DESIGN, SETTING, AND PARTICIPANTS A randomized clinical trial involving a random sample of
975 healthy term infants at a maternity hospital in Helsinki, Finland. Study recruitment
occurred between January 14, 2013, and June 9, 2014, and the last follow-up was May 30,
2016. Data analysis was by the intention-to-treat principle.

INTERVENTIONS Randomization of 489 infants to daily oral vitamin D3 supplementation of

400 IU and 486 infants to 1200 IU from age 2 weeks to 24 months.

MAIN OUTCOMES AND MEASURES Primary outcomes were bone strength and incidence of
parent-reported infections at 24 months.

RESULTS Of the 975 infants who were randomized, 485 (49.7%) were girls and all were of
Northern European ethnicity. Eight hundred twenty-three (84.4%) completed the 24-month
follow-up. We found no differences between groups in bone strength measures, including
bone mineral content (mean difference, 0.4 mg/mm; 95% CI, −0.8 to 1.6), mineral density Author Affiliations: Children’s
(mean difference, 2.9 mg/cm3; 95% CI, −8.3 to 14.2), cross-sectional area (mean difference, Hospital, Pediatric Research Center,
University of Helsinki and Helsinki
–0.9 mm2; 95% CI, −5.0 to 3.2), or polar moment of inertia (mean difference, –66.0 mm4, University Hospital, Helsinki, Finland
95% CI, −274.3 to 142.3). Incidence rates of parent-reported infections did not differ between (Rosendahl, Valkama, Holmlund-
groups (incidence rate ratio, 1.00; 95% CI, 0.93-1.06). At birth, 914 of 955 infants (95.7%) Suila, Enlund-Cerullo, Hauta-alus,
Helve, Hytinantti, Kajantie, Mäkitie,
were vitamin D sufficient (ie, 25-hydroxyvitamin D [25(OH)D] concentration ⱖ20.03 ng/mL).
Andersson); Folkhälsan Research
At 24 months, mean 25(OH)D concentration was higher in the 1200-IU group than in the Center, Helsinki, Finland (Enlund-
400-IU group (mean difference, 12.50 ng/mL; 95% CI, 11.22-13.78). Cerullo, Viljakainen, Mäkitie);
National Institute for Health and
Welfare, Helsinki, Finland (Helve,
CONCLUSIONS AND RELEVANCE A vitamin D3 supplemental dose of up to 1200 IU in infants Levälahti, Kajantie); PEDEGO
did not lead to increased bone strength or to decreased infection incidence. Daily Research Unit, MRC Oulu, Oulu
supplementation with 400 IU vitamin D3 seems adequate in maintaining vitamin D University Hospital and University of
Oulu, Oulu, Finland (Kajantie); Center
sufficiency in children younger than 2 years.
for Molecular Medicine, Karolinska
Institutet and Clinical Genetics,
TRIAL REGISTRATION ClinicalTrials.gov Identifier: NCT01723852 Karolinska University Hospital,
Stockholm, Sweden (Mäkitie).
Corresponding Author: Sture
Andersson, MD, PhD, Children’s
Hospital, Pediatric Research Center,
University of Helsinki and Helsinki
University Hospital, PO Box 281,
JAMA Pediatr. doi:10.1001/jamapediatrics.2018.0602 FIN-00029 HUS, Helsinki, Finland
Published online May 29, 2018. (sture.andersson@hus.fi).

(Reprinted) E1
© 2018 American Medical Association. All rights reserved.

Downloaded From: on 05/29/2018

 Research Original Investigation
F
or decades, vitamin D deficiency has been a worldwide
health concern.1 Many countries have implemented
guidelines for vitamin D supplementation and food for-
tification, but preventive measures have not been adequate.2
In the United States, vitamin D deficiency, defined as serum
25-hydroxyvitamin D (25[OH]D) concentration less than 20.03
ng/mL (to convert to nanomoles per liter, multiply by 2.496),
was reported in 1988 to 2010 in 26% to 32% of the population.3
Likewise, a 2016 study involving more than 50 000 Euro-
pean adults and children estimated 40% to be vitamin D
deficient.4
Vitamin D deficiency in infants can lead to impaired bone
mineralization and rickets.5 Since the 1920s, vitamin D has been
recognized as effective in preventing and treating rickets, but
optimal supplementation for bone health is still unclear, with
data on the skeletal effects of vitamin D in infants being
scarce.6-8
Vitamin D is a potent modulator of both innate and adap-
tive immunity.9 Some observational studies report an asso-
ciation between vitamin D deficiency and increased infec-
tious diseases. 10-12 Randomized trials, however, show
conflicting results and few involve infants.13-15 Because infec-
tions are a major cause of early childhood morbidity, the ef-
fect of vitamin D supplementation is of great interest.
We conducted a randomized clinical trial comparing daily
vitamin D3 supplementation of 400 IU and 1200 IU in healthy
infants aged 2 weeks to 24 months. Our aim was to evaluate
effects of vitamin D supplementation on bone strength and in-
cidence of infections. We hypothesized that higher dosages of
vitamin D supplementation would lead to increased bone
strength and to decreased frequency of infections in early
childhood.
Methods
Study Design and Participants
The Vitamin D Intervention in Infants (VIDI) study was a ran-
domized, double-blind, 24-month clinical trial of daily vita-
min D3 supplementation of 400 IU or 1200 IU administered
to infants. Between January 14, 2013, and June 9, 2014, fami-
lies were recruited at Kätilöopisto Helsinki Maternity Hospi-
tal, Helsinki, Finland, 1 to 2 days after delivery. All follow-up
was completed on May 30, 2016. Mothers took no regular medi-
cation and had a singleton pregnancy. Ethnicity was re-
stricted to Northern European to exclude the effect of skin pig-
mentation on vitamin D status.16 Eligible infants were born at
term (37 weeks and 0 days’ to 42 weeks and 0 days’ gesta-
tion), with a birth weight within 2 SDs of the mean for gesta-
tional age.17
Infants excluded were those requiring intravenous glu-
cose, antibiotics, nasal continuous positive airway pressure
treatment for more than 1 day, phototherapy for more than 3
days, or nasogastric tube feeding for more than 1 day and in-
fants with seizures.
The study was conducted in accordance with the Decla-
ration of Helsinki, and the Research Ethics Committee of the
Hospital District of Helsinki and Uusimaa approved the study.
E2 JAMA Pediatrics Published online May 29, 2018 (Reprinted)
© 2018 American Medical Association. All rights reserved.
Downloaded From: on 05/29/2018
Dosage of Vitamin D3 and Bone Strength and Infections in Infants
Key Points
Question Does a higher dose (1200 IU) of supplemental vitamin
D3 administered to healthy infants increase bone strength or
decrease incidence of infections compared with the standard dose
(400 IU) ?
Findings This randomized clinical trial of 975 infants found no
difference in bone strength or incidence of infections between
intervention groups at 24 months of age.
Meaning In healthy infants, daily supplementation with 1200 IU
of vitamin D3 provides no additional benefits compared with
supplementation with 400 IU for bone strength or incidence of
infections in early childhood.
The project protocol is provided in Supplement 1 and has been
described in a previously published article.18 Parents gave
written informed consent at recruitment.
Randomization and Blinding
Infants were randomized (1:1) to receive 400 IU or 1200 IU of
vitamin D3 daily from age 2 weeks to 24 months. To ensure fair
distribution across the year, a pharmacist at Helsinki Univer-
sity Hospital with no relation to the study performed random-
ization in blocks of 50. Both study preparations, manufac-
tured by Orion Pharmaceuticals, contained vitamin D 3
dissolved in medium-chain triglyceride oil and were identi-
cal in appearance. Participants and investigators were masked
to group assignment, and no changes to the methods were
made after trial commencement. An external steering group
monitored the study.
Procedures
Vitamin D supplements were administered orally once daily
in a volume of 5 drops for both concentrations. No other vita-
min D supplements were allowed concurrently. The families
recorded daily vitamin D supplementation and all their child’s
infections in study diaries. Completed diaries were collected
and reviewed at follow-up visits arranged at ages 6, 12, and 24
months. Adherence to treatment was calculated from the dia-
ries as the percentage of days of supplement administration
compared with the total number of days of follow-up.
We analyzed 25(OH)D concentration (in nanograms per
milliliter; to convert to nanomoles per liter, multiply by 2.496)
at birth (cord blood) and at ages 12 and 24 months and intact
parathyroid hormone concentration at ages 12 and 24 months
with a fully automated immunoassay (IDS-iSYS; Immunodi-
agnostic System Ltd). Details on data collection and labora-
tory analyses are provided in eAppendix 1 and eAppendix 2 in
Supplement 2.
Outcome Measures
The primary outcomes were bone strength and incidence of
parent-reported infections at 24 months. We used 4 different
bone measurements for total and cortical bone to assess bone
strength because none of these alone determines fracture
resistance.19 Bone mineral content reflects bone mineral quan-
tity (in milligrams) per millimeter, bone mineral density re-
flects bone mineral quantity (in milligrams) per cubic centi-
jamapediatrics.com
 Dosage of Vitamin D3 and Bone Strength and Infections in Infants
meter, cross-sectional area of the bone reflects bone area in
square millimeters, and polar moment of inertia reflects bone
resistance to torsion (measured in quartic millimeters). These
bone strength measurements were assessed at age 24 months
with peripheral quantitative computed tomography (pQCT)
(Stratec XCT 2000 L Research+; Stratec Medizintechnik GmbH)
of the left tibia. The length of the tibia was measured from the
medial malleolus to the medial condyle, and bone strength was
measured at 20% distal proximal length. We graded each pQCT
scan quality according to movement artifacts from 1 to 5 and
further as good (grades 1-2), moderate (grades 3-4), or poor
(grade 5).20 If major movement artifacts or incorrect leg posi-
tioning occurred, the scan was excluded.
Infections were assessed from study diaries in which
parents prospectively recorded all their child’s infections. Par-
ents reported the date of the infection (month/year), infec-
tion type, symptoms or specific diagnosis, duration, medica-
tion, physician visit, or hospitalization. The cumulative number
of infection episodes was calculated for the 24-month study.
If several symptoms or diagnosed infections were reported for
the same period, these were considered as belonging to the
same episode.
For characterization, the parent-reported infections
were divided post hoc into 7 subtypes: (1) upper respiratory
tract infection, defined as presence of rhinitis, cough, sore
throat, nasal congestion, or sneezing, with or without fever
(temperature, >38.0°C); (2) acute otitis media diagnosed by a
physician; (3) pneumonia diagnosed by a physician; (4) con-
junctivitis, defined as reddish eye, discharge from the eye, or
both; (5) gastroenteritis, defined as vomiting and/or diar-
rhea; (6) nonspecified viral infection, defined as fever (tem-
perature, >38.0°C) with or without skin manifestations; and
(7) other bacterial infection including miscellaneous bacte-
rial infections, such as of the urinary tract or skin. For out-
come analyses, the infections were further grouped into 3
main types: (1) respiratory infections, including upper respi-
ratory tract infections, acute otitis media, and pneumonia;
(2) gastroenteritis; and (3) other bacterial and viral infec-
tions, including conjunctivitis.
Safety Monitoring
Trial safety was ensured by measurement of the ionized cal-
cium concentration at follow-up visits. In case of severe hy-
percalcemia, defined as an ionized calcium concentration ex-
ceeding the upper age-specific reference limit by 10% or more
(ie, 6.12 mg/dL at ages 6 and 12 months and 5.92 mg/dL at age
24 months; to convert to millimoles per liter, multiply by 0.25),
calcium and 25(OH)D concentrations were to be remeasured,
symptoms indicative of hypercalcemia investigated, and con-
tinuation of vitamin D supplementation reevaluated. In case
of any adverse effects or severe hypercalcemia, an external
monitor was to be informed.
Statistical Analysis
The trial was designed with a planned sample of 1000 study
participants. This total allowed an estimated 20% dropout rate,
leaving 800 participants. To reach the statistical power to de-
tect a 0.2-SD difference in bone mineral content or bone cross-
jamapediatrics.com
© 2018 American Medical Association. All rights reserved.
Downloaded From: on 05/29/2018
Original Investigation Research
sectional area, the estimate was 210 and 297 pQCT scans, re-
spectively, in each group.21 For infections, assuming an average
annual infection rate of 6 for a child younger than 2 years, we
estimated that detection of a decrease from 12 to 9 infections
during the 24-month study period required a sample size of
220 per group to achieve a statistical power of 90%.22,23
Comparisons of baseline and follow-up characteristics and
of biochemical measures were analyzed with the 2-tailed in-
dependent sample unpaired t test, Mann-Whitney test, or Pear-
son χ2 as applicable. Logarithmic transformation was used for
variables with nonnormal distribution.
We assessed differences in bone strength between groups
with a multivariate analysis of covariance. We used a crude
model and an adjusted model, the latter using as covariates sex,
age, weight, and quality of scans. Analyses were performed
with IBM SPSS, version 22 (IBM).
To evaluate the effect of group on infection outcome mea-
sures, we applied a negative binomial model. Incidence was
estimated as the proportion of follow-up time in person-
months, which allowed use of all available infection data, in-
cluding data from incomplete study diaries. A complemen-
tary log-logistic model served to model the probability of at
least 1 infection episode.
For group-season interaction analyses of infections, data
were split into 1-month intervals. Random-effects models
served for estimation to account for dependence of indi-
vidual data on separate intervals (range, 3-25 months per par-
ticipant). Analyses were performed with Stata statistical soft-
ware, version 14 (StataCorp).
In the analyses, we applied the intention-to-treat prin-
ciple. Per-protocol analyses included participants with treat-
ment adherence of at least 80%. Two-sided P < .05 was con-
sidered statistically significant.
Results
Baseline Characteristics
Of the 4980 participants screened, 987 were found eligible and
consented to participate. Twelve infants were excluded after
randomization for not meeting eligibility criteria, leaving 975
infants, of whom 489 were randomly assigned to the 400-IU
group and 486 to the 1200-IU group (Figure 1). Of the 975 in-
fants who were randomized, 485 (49.7%) were girls and all were
of Northern European ethnicity. Baseline characteristics be-
tween groups did not differ (Table 1).
Adherence to Treatment
A total of 823 children (84.4%) completed the 24-month study;
865 (88.7%) attended the 12-month and 892 (91.5%) the
6-month follow-up (eTable 1 in Supplement 2). The cumula-
tive number of participants lost to follow-up was 83 at 6
months, 110 at 12 months, and 152 at 24 months and did not
differ between groups. The mean adherence to vitamin D
supplementation during 24 months was 88% and was similar
in both groups (eTable 2 in Supplement 2). At age 24 months,
the proportion of children with at least 80% adherence was
83.5% (eTable 2 in Supplement 2).
(Reprinted) JAMA Pediatrics Published online May 29, 2018 E3
 Research Original Investigation Dosage of Vitamin D3 and Bone Strength and Infections in Infants

Figure 1. Participant Flow of the Vitamin D Intervention in Infants Study

4980 Infants assessed for eligibility

3993 Excluded
1572 Did not meet inclusion criteria
2421 Declined to participate

987 Randomized

495 Randomized to receive vitamin D3, 400 IU/d 492 Randomized to receive vitamin D3, 1200 IU/d
495 Received as randomized 492 Received as randomized

84 Lost to follow-up 71 Lost to follow-up

28 Social reasons 28 Social reasons
6 Abdominal symptoms 7 Abdominal symptoms
6 Illness 5 Illness
1 Preference for commercial 10 Preference for commercial
vitamin D supplement vitamin D supplement
43 Other 21 Other

8 Discontinued intervention but completed 9 Discontinued intervention but completed

follow-up follow-up
2 Social reasons 8 Preferred commercial vitamin D supplement
4 Preferred commercial vitamin D supplement 1 Other
2 Other

6 Excluded from analysis 6 Excluded from analysis

(did not meet inclusion criteria) (did not meet inclusion criteria)

489 Included in intention-to-treat analysis 486 Included in intention-to-treat analysis

Biochemical Effects A per-protocol analysis was applied to all biochemical effects

Mean (SD) 25(OH)D concentration at birth was 32.73 (11.14)
ng/mL in the 400-IU group and 32.57 (9.62) ng/mL in the
1200-IU group; at 12 months, mean concentration was 33.13
(7.93) ng/mL in the 400-IU group and 46.07 (11.10) ng/mL in
the 1200-IU group; and at 24 months, 34.70 (7.85) ng/mL in
the 400-IU group and 47.16 (10.46) ng/mL in the 1200-IU group
(mean difference, 12.50 ng/mL; 95% CI, 11.22-13.78) (Figure 2
and eTable 3 in Supplement 2). At birth, 914 of 955 infants (95.7
%) were vitamin D sufficient (25[OH]D concentration ≥20.03
ng/mL), with no difference between groups (eTable 4 in
Supplement 2). At age 24 months, the 25(OH)D concentration
in the 404 children evaluated in the 400-IU group was 30.05
ng/mL or greater in 285 children (70.5%) and 50.08 ng/mL or
greater in 13 (3.2%). Among the 410 children in the 1200-IU
group, 391 (95.4%) had a 25(OH)D concentration of 30.05
ng/mL or greater, and 159 (38.8%) had a concentration of 50.08
ng/mL or greater. None of the infants had a 25(OH)D concen-
tration greater than 100.16 ng/mL (eTable 4 in Supplement 2).
Mean ionized calcium concentrations did not differ be-
tween groups (Figure 2 and eTable 3 in Supplement 2). At age
12 months, the ionized calcium concentrations were within age-
specific reference limits in 838 (98.1%) of the 854 children
evaluated, and at age 24 months concentrations were within
age-specific limits in 667 (91.9%) of the 726 children evalu-
ated (eTable 5 in Supplement 2). No participants developed se-
vere hypercalcemia. At ages 12 and 24 months, the serum para-
thyroid hormone level was lower in the 1200-IU group than
in the 400-IU group (Figure 2 and eTable 3 in Supplement 2).
with consistent results.
Primary Outcomes
We performed pQCT bone scans of the left tibia in 783 of the
823 children (95.1%) attending the age 24-month follow-up.
Owing to motion artifacts, 79 (10.1%) of the scans failed and
were excluded. A total of 704 (89.9%) scans were included in
the analyses. Of these, scan quality was assessed as good in
165 (48.1%) of the 400-IU group and 193 (53.5%) of the 1200-IU
group participants, moderate in 124 (36.2%) of the 400-IU
group and 133 (36.8%) of the 1200-IU group, and poor in 54
(15.7%) of the 400-IU group and 35 (9.7%) of the 1200-IU group.
Bone strength measurements for total bone and cortical bone
did not differ between groups (Table 2). The findings re-
mained unaltered in per-protocol analysis and after adjust-
ment for sex, age, weight, and quality of scans.
Infection data from study diaries were obtained from 449
participants in the 400-IU group and 448 in the 1200-IU group;
for 78 participants, data were completely missing. At the tri-
al’s end, 8458 parent-reported infection episodes had oc-
curred, amounting to a mean 9.18 (95% CI, 8.73-9.63) in the
400-IU group and 9.14 (95% CI, 8.68-9.60) in the 1200-IU
group, with no differences in incidence rates of infection be-
tween groups (incidence rate ratio [IRR], 1.00; 95% CI, 0.93-
1.06); (Table 3 and eTable 6 in Supplement 2). Infection char-
acteristics were similar; the majority were respiratory infections
(eTable 7 in Supplement 2). The only difference between groups
was a higher incidence rate of antibiotic treatment in the

E4 JAMA Pediatrics Published online May 29, 2018 (Reprinted) jamapediatrics.com

© 2018 American Medical Association. All rights reserved.

Downloaded From: on 05/29/2018

 Dosage of Vitamin D3 and Bone Strength and Infections in Infants Original Investigation Research

Table 1. Characteristics of 975 Participants in the Vitamin D Intervention in Infants Studya

All 400-IU Group 1200-IU Group
Characteristic (N = 975) (n = 489) (n = 486)
Child
Sex, female, No. (%) 485 (49.7) 242 (49.5) 243 (50.0)
Birth weight, mean (SD), g 3539 (395) 3514 (379) 3565 (410)
Birth length, mean (SD), cm 50.3 (1.7) 50.3 (1.7) 50.4 (1.8)
Apgar score at 1 min, mean (SD) 9 (1) 9 (1) 9 (1)
Gestational age, mean (SD), wk 40.2 (1.1) 40.1 (1.1) 40.2 (1.1)
Mode of birth, vaginal, No. (%) 911 (93.4) 456 (93.3) 455 (93.6)
Season of birth, No. (%)
Winter 189 (19.4) 100 (20.4) 89 (18.3)
Spring 400 (41.0) 197 (40.3) 203 (41.8)
Summer 217 (22.3) 108 (22.1) 109 (22.4)
Autumn 169 (17.3) 84 (17.2) 85 (17.5)
Older siblings (n = 892), No. (%) 331 (37.1) 153 (34.5) 178 (39.7)
Breastfeeding (n = 854), No. (%)
0-3 mo 72 (8.4) 39 (9.1) 33 (7.7)
3.1-6 mo 108 (12.6) 59 (13.8) 49 (11.5)
>6 mo 674 (78.9) 330 (77.1) 344 (80.8)
Daily dietary intake of vitamin D at 248 (148) 252 (144) 240 (148)
age 12 mo (n = 730), mean (SD), IU
Daily dietary intake of calcium at 613 (308) 627 (310) 599 (307)
age 12 mo (n = 730), mean (SD), mg
Day care attendance, No. (%)
At age 12 mo (n = 866) 38 (4.4) 15 (3.4) 23 (5.3)
At age 24 mo (n = 796) 491 (61.7) 243 (61.4) 248 (62.0)
Age at baseline, mean (SD), mo 16.3 (3.4) 16.3 (3.5) 16.3 (3.3)
Influenza vaccination (n = 889), No. (%) 353 (39.7) 190 (42.7) 163 (36.7)
Mother
Age at delivery, mean (SD), y 31.5 (4) 31.2 (4) 31.8 (5)
BMI before pregnancy (n = 891), mean (SD) 23.2 (3.7) 23.1 (3.7) 23.3 (3.7)
Use of vitamin D supplements during 813 (94.2) 407 (94.7) 406 (43.8)
pregnancy (n = 863), No. (%)
Daily vitamin D supplemental dose during 620 (640) 660 (760) 580 (520)
pregnancy (n = 863), IU, mean (SD)
Socioeconomic
Maternal educational level (n = 883), No. (%)b
Low 226 (25.6) 119 (27.2) 107 (24.0)
High 657 (74.4) 318 (72.8) 657 (76.0)
Paternal educational level (n = 870), No. (%)b Abbreviation: BMI, body mass index
Low 328 (37.7) 164 (37.8) 164 (37.6) (calculated as weight in kilograms
High 542 (62.3) 270 (62.2) 272 (62.4) divided by height in meters squared).
a
Household annual income (n = 750), No. (%)c Infants were randomized to receive
vitamin D3, 400 IU/d or 1200 IU/d.
Low 135 (18.0) 64 (16.8) 71 (19.2) b
For educational level, low indicates
Medium 544 (72.5) 281 (73.9) 263 (71.1) less than a bachelor’s degree; high,
High 71 (9.7) 35 (9.2) 36 (9.7) at least a bachelor’s degree.
c
Maternal smoking before pregnancy (n = 878), No. (%) 130 (14.8) 64 (14.8) 66 (14.8) For annual income, low indicates
less than €40 000 (<$48 316);
Maternal smoking after delivery (n = 869), No. (%) 32 (3.7) 17 (4.0) 15 (3.4)
medium, €40 000 to €109 999
Mother or father smoking after delivery (n = 837), 137 (16.4) 65 (15.7) 72 (17.0) ($48 316-$132 868); and high,
No. (%) €110 000 ($132 869) or more.

1200-IU group (IRR 1.17; 95% CI, 1.00-1.36). Results did not
change in per-protocol analysis or after adjustment for poten-
tial confounding factors (ie, parental educational level, exis-
tence of older siblings, parental smoking, day care atten-
dance, season of birth, or duration of breastfeeding).
In interaction analyses, the only differences were for in-
fection duration, indicating that, in winter and spring, dura-
tion was shorter in the 1200-IU group (test for 2 interactions:
P = .01; winter: IRR, 0.89; 95% CI, 0.81-0.97; spring: IRR, 0.89;
95% CI, 0.81-0.99), and the likelihood of contracting gastro-

jamapediatrics.com (Reprinted) JAMA Pediatrics Published online May 29, 2018 E5

© 2018 American Medical Association. All rights reserved.

Downloaded From: on 05/29/2018

 Research Original Investigation Dosage of Vitamin D3 and Bone Strength and Infections in Infants

Figure 2. Serum 25-Hydroxyvitamin D, Ionized Calcium, and Intact Parathyroid Hormone Concentrations
During the Vitamin D Intervention in Infants Study

A 25(OH)D concentration throughout study B 25(OH)D concentration at 24 mo

52.08 100.16
Mean 25(OH)D Concentration, ng/mL

Group1200 P < .001

25(OH)D Concentration, ng/mL

Group400
80.13
44.07

60.10
36.06
40.06

28.04
20.03

0 0
0 12 24 Group400 Group1200
Follow-up, mo Daily Vitamin D3 Supplementation Group A, Mean 25-hydroxyvitamin D
(25[OH]D) concentrations with 95%
C Ionized calcium concentration at 24 mo D Parathyroid hormone level at 24 mo
CIs at baseline (cord blood), age 12
months, and age 24 months in infants
6.00 80 randomized to vitamin D, 400 IU/d
Ionized Calcium Concentration, ng/mL

P = .49 P = .004
Parathyroid Hormone Level, ng/mL

and 1200 IU/d. B through D, Crude

values at 24-month follow-up by
5.60 60
intervention group. Horizontal lines
in panels B, C, and D represent
median values. P values refer to
5.20 40
differences between intervention
groups as tested by the 2-tailed,
4.80 20 unpaired, independent-samples
t test. To convert serum 25(OH)D
to nanomoles per liter, multiply by
0 0 2.496; ionized calcium to millimoles
Group400 Group1200 Group400 Group1200 per liter, multiply by 0.25; and
Daily Vitamin D3 Supplementation Group Daily Vitamin D3 Supplementation Group parathyroid hormone to nanograms
per liter, multiply by 1.0.

Table 2. Total and Cortical Bone Measurements Assessed by pQCT of the Left Distal Tibia at Age 24 Months
by Intervention Groupa

Bone Measurement 400-IU Group 1200-IU Group Mean Difference (95% CI)
Total bone
Scans, No. (%) 343 (48.7) 361 (51.3) NA
BMC, mean (95% CI), mg/mm 54.2 (53.4 to 55.1) 54.7 (53.8 to 55.5) 0.4 (−0.8 to 1.6)
BMD, mean (95% CI), mg/cm3 375.8 (367.7 to 383.8) 378.7 (370.9 to 386.5) 2.9 (−8.3 to 14.2)
CSA, mean (95% CI), mm2 148.6 (145.7 to 151.5) 147.7 (144.9 to 150.6) −0.9 (−5.0 to 3.2)
Polar moment of inertia, mean 3759 (3610 to 3908) 3693 (3548 to 3838) −66.0 (−274.3 to 142.3)
(95% CI), mm4 Abbreviations: BMC, bone mineral
Cortical bone content; BMD, bone mineral density;
CSA, cross-sectional area; NA, not
Scans, No. (%) 343 (48.7) 361 (51.3) NA applicable; pQCT, peripheral
BMC, mean (95% CI), mg/mm 42.6 (41.6 to 43.6) 43.6 (42.7 to 44.6) 1.0 (−0.4 to 2.4) quantitative computed tomography.
a
BMD, mean (95% CI), mg/cm3 723.8 (717.2 to 730.3) 729.7 (723.3 to 736.1) 6.0 (−3.2 to 15.2) Infants were randomized to receive
vitamin D3, 400 IU/d or 1200 IU/d.
CSA, mean (95% CI), mm2 58.4 (57.4 to 59.4) 59.4 (58.4 to 60.4) 1.0 (−0.4 to 2.4)
Values represent means and mean
Polar moment of inertia, mean 2055 (1999 to 2112) 2065 (2010 to 2120) 9.9 (−69.1 to 88.9) differences with 95% CIs from
(95% CI), mm4
multivariate analysis of covariance.

enteritis in winter was also lower in the 1200-IU group (test observed no differences in bone strength measurements or in
for interaction: P = .04; IRR, 0.76; 95% CI, 0.60-0.97). incidence of parent-reported infections between the 400-IU
and 1200-IU intervention groups. Our study had adequate
power to detect or exclude any but small differences between
groups.
Discussion At birth, 914 of 955 (95.7%) of the infants were vitamin D
This study involving 975 healthy children is, to our knowl- sufficient (had a 25[OH]D concentration ≥20.03 ng/mL), re-
edge, the first large randomized clinical trial evaluating vita- flecting adequate maternal vitamin D intake.24 At age 24
min D supplementation from infancy to early childhood. We months, 809 of 814 (99.4%) of the study participants were vi-

E6 JAMA Pediatrics Published online May 29, 2018 (Reprinted) jamapediatrics.com

© 2018 American Medical Association. All rights reserved.

Downloaded From: on 05/29/2018

 Dosage of Vitamin D3 and Bone Strength and Infections in Infants Original Investigation Research

Table 3. Effect of Vitamin D Supplementation on Infection Outcome Measures During the 24-Month Vitamin D Intervention in Infants Studya

400-IU Group 1200-IU Group

Participants, Events, Incidence Participants, Events, Incidence
Outcome Measure No. No. Rateb No. No. Ratec IRR (95% CI)d
Infection episodes 449 4244 0.41 448 4214 0.41 1.00 (0.93-1.06)
Respiratory infections 449 3321 0.32 448 3303 0.32 1.00 (0.93-1.07)
Gastroenteritis 449 379 0.04 448 349 0.03 0.92 (0.79-1.08)
Other infections 449 544 0.05 448 562 0.06 1.04 (0.91-1.19)
Antibiotic treatments 441 795 0.08 435 922 0.09 1.17 (1.00-1.36)
Physician visits 443 1421 0.14 435 1502 0.15 1.07 (0.94-1.21)
Hospitalizations 443 33 0.00 435 38 0.00 1.16 (0.71-1.89)
Days infected, No.e 437 26 314 0.08 430 25 541 0.08 0.97 (0.88-1.07)
Duration per infection 437 3272 0.01 430 3100 0.01 0.95 (0.89-1.01)
episode, df
Abbreviation: IRR, incidence rate ratio. number of person-months (10 204) except for duration per infection episode
a
Infants were randomized to receive vitamin D3, 400 IU/d or 1200 IU/d. and number of days infected, which were divided by the number of
b
person-days (308 675).
Incidence rate was calculated from the total number of events divided by the
d
number of person-months (10 237) except for duration per infection episode Calculated using negative binomial regression.
e
and number of days infected, which were divided by the number of Total number of days of illness during the 24-month follow-up.
person-days (309 657). f
Duration of illness in days per infection episode.
c
Incidence rate was calculated from the total number of events divided by the

tamin D sufficient. None had a 25(OH)D concentration greater
than 100.16 ng/mL or severe hypercalcemia, which are indi-
cators of vitamin D toxicity.5 These findings imply that a daily
dose of 1200 IU of vitamin D3 in this age group is safe, but even
400 IU will maintain vitamin D sufficiency in most children.
We found no significant differences between the groups
in any of the bone strength measurements, in line with ran-
domized trials involving older vitamin D–sufficient children
(aged 8-17 years) in which vitamin D dosages ranging from 132
IU/d to 14 000 IU/wk did not improve bone strength as mea-
sured by dual-energy x-ray absorptiometry or pQCT.25 How-
ever, in an earlier study in 3-month-old infants, those with
higher-dose vitamin D supplementation (1600 IU vs 400 IU)
had larger tibial bone area compared with those receiving the
lower dose.26 The divergent results in the present study may
result from this study’s longer duration and larger cohort. In
vitamin D–sufficient children, factors other than 25(OH)D con-
centration, such as motor competence, lean mass, or calcium
intake, may have a greater influence on bone strength.21,27,28
We observed no differences in the incidence of parent-
reported infections between the groups, with identical
results in the per-protocol analyses. One recent large
meta-analysis13 concluded that vitamin D supplementation is
effective in preventing acute respiratory tract infections, but
it comprised heterogeneous studies including children older
than 2 years and adults with varying baseline 25(OH)D con-
centrations; in subgroup analysis, the protective effect of
supplementation was stronger in those with a baseline 25
(OH)D concentration less than 10.02 ng/mL. Because our study
participants were mostly ([914 of 955] [95.7%]) vitamin D suf-
ficient, our results support the findings that, in vitamin D–suf-
ficient children, additional vitamin D supplementation pro-
vides no further benefit in resisting infections.29
In interaction analyses, however, we did observe the
duration of infections in winter and spring in the 1200-IU
group to be shorter than in the 400-IU group. This finding
suggests that higher dosages of vitamin D may shorten the
duration of illness during winter months, when infections
are more frequent and when cutaneous synthesis of vitamin
D is limited. However, this difference was marginal, was
based on a post hoc analysis, and demands confirmation in
future studies.
We hypothesized that higher dosages of vitamin D supple-
mentation improve bone strength and reduce infections in early
childhood in a northern European population with limited sun-
light exposure. We expected a greater proportion of children
to be vitamin D deficient at birth because studies conducted
in 2007 and 2010 showed that approximately half the healthy
newborns studied were vitamin D deficient.26,30 However,
since 2010, Finnish public health authorities have improved
vitamin D intake at the population level by food fortification
and promotion of vitamin D supplementation. Currently, liq-
uid milk products are fortified with 40 IU of vitamin D3 per 100
mL and fat spreads with 800 IU per 100 g. Guidelines for vi-
tamin D supplementation recommend 400 IU daily for preg-
nant and breastfeeding women and for children younger than
2 years and 300 IU daily for persons aged 2 to 17 years.31 These
public health actions have recently improved vitamin D sta-
tus in Finland.31,32 The absence of vitamin D–deficient in-
fants may explain the lack of any effect of higher vitamin D
supplementation on our primary outcomes. Moreover, the
higher dose of 1200 IU is unlikely to have been too small
because, for 676 of the 814 (83.0%) participants at study end,
25(OH)D concentrations exceeded 30.05 ng/mL.
Limitations
The quality of bone scans varied, and movement artifacts were
common owing to technical challenges in pQCT measure-
ment in young children. However, 615 of the 704 scans (87.4%)
were of good or moderate quality, with no differences be-
tween groups. We adjusted the statistical model by scan qual-
ity, with consistent results. On the other hand, pQCT gives more

jamapediatrics.com (Reprinted) JAMA Pediatrics Published online May 29, 2018 E7

© 2018 American Medical Association. All rights reserved.

Downloaded From: on 05/29/2018

 Research Original Investigation Dosage of Vitamin D3 and Bone Strength and Infections in Infants

detailed information about bone characteristics, and we re-
gard its use, instead of dual-energy x-ray absorptiometry, as a
significant strength of our study. Assessment of infections was
based on parents’ report without clinical evaluation or labo-
ratory confirmation, which may have led to underestimation
or overestimation of infection prevalence. Definitions of in-
fection may also vary. However, because the study was double-
blinded and controlled, any recall and observation bias is likely
to be equally distributed between groups, although such bias
could increase random error and thus reduce the likelihood
of detecting differences. That the frequency and seasonal
distribution of infections in our cohort corresponds to earlier
findings in a similar age group further supports the validity of
our data.33,34
Conclusions
In vitamin D–sufficient healthy infants, daily supplementa-
tion with 1200 IU vitamin D3 compared with 400 IU provides
no additional benefits for bone strength or for parent-
reported incidence of infections during the first 2 years of life.
In a country where sunlight exposure is limited but food for-
tification with vitamin D is common, supplementation with
400 IU of vitamin D3 daily seems adequate to ensure vitamin
D sufficiency in children younger than 2 years.

ARTICLE INFORMATION
Accepted for Publication: February 15, 2918.
Published Online: May 29, 2018.
doi:10.1001/jamapediatrics.2018.0602
Author Contributions: Drs Rosendahl, Valkama,
Mäkitie, and Andersson contributed equally to the
study. Dr Andersson had full access to all the data in
the study and takes responsibility for the integrity
of the data and the accuracy of the data analysis.
Study concept and design: Rosendahl, Valkama,
Holmlund-Suila, Helve, Hytinantti, Levälahti,
Kajantie, Viljakainen, Mäkitie, Andersson.
Acquisition, analysis, or interpretation of data:
Rosendahl, Valkama, Holmlund-Suila,
Enlund-Cerullo, Hauta-alus, Helve, Hytinantti,
Levälahti, Viljakainen, Mäkitie, Andersson.
Drafting of the manuscript: Rosendahl, Valkama,
Holmlund-Suila, Hauta-alus, Helve, Hytinantti,
Levälahti, Viljakainen, Mäkitie, Andersson.
Critical revision of the manuscript for important
intellectual content: All authors.
Statistical analysis: Rosendahl, Valkama,
Holmlund-Suila, Hauta-alus, Levälahti, Viljakainen,
Mäkitie.
Obtained funding: Enlund-Cerullo, Helve, Kajantie,
Mäkitie, Andersson.
Administrative, technical, or material support:
Helve, Hytinantti, Kajantie, Viljakainen, Mäkitie,
Andersson.
Study supervision: Holmlund-Suila, Viljakainen,
Mäkitie, Andersson.
Conflict of Interest Disclosures: None reported.
Funding/Support: The study was financially
supported by the Foundation for Pediatric
Research, Finska Läkaresällskapet, the Finnish
Medical Foundation, Governmental Subsidy for
Clinical Research, the Päivikki and Sakari Sohlberg
Foundation, Stiftelsen Dorothea Olivia, Karl Walter
och Jarl Walter Perkléns minne, the Academy of
Finland, the Sigrid Jusélius Foundation, the
Folkhälsan Research Foundation, the Novo Nordisk
Foundation, the Orion Research Foundation, and
Barncancerfonden. Orion Pharmaceuticals is
acknowledged for providing the vitamin D
supplements gratis for the study.
Role of the Funder/Sponsor: The funders had no
role in the design and conduct of the study;
collection, management, analysis, and
interpretation of the data; preparation, review, or
approval of the manuscript; and decision to submit
the manuscript for publication.
Additional Contributions: We are most grateful to
all the children and families for their commitment
to the study. Research nurses Sirpa Nolvi, Rhea
Paajanen, Nea Boman, and Päivi Turunen, Children’s
Hospital, University of Helsinki and Helsinki
University Hospital, medical student Jesse
Karppinen, University of Tartu, Estonia, and
graduate students Emma Kynkäänniemi and Sonja
Kaijanen, University of Helsinki, assisted with data
collection; laboratory technician Sari Lindén,
Children’s Hospital, University of Helsinki and
Helsinki University Hospital, and midwives and
laboratory staff at the Kätilöopisto Helsinki
Maternity Hospital made valuable contributions
to the work; Anna-Liisa Järvenpää, MD, PhD,
Children’s Hospital, University of Helsinki and
Helsinki University Hospital, performed external
monitoring; Alex Ireland, PhD, School of Healthcare
Science, Manchester Metropolitan University,
United Kingdom, assisted with peripheral
quantitative computed tomography; and Carol
Norris, PhD, University of Helsinki, performed
linguistic editing of the manuscript. All except
Drs Järvenpää and Ireland were compensated for
their work.
REFERENCES
1. Holick MF. Vitamin D deficiency. N Engl J Med.
2007;357(3):266-281.
2. Spiro A, Buttriss JL. Vitamin D: an overview of
vitamin D status and intake in Europe. Nutr Bull.
2014;39(4):322-350.
3. Schleicher RL, Sternberg MR, Lacher DA, et al.
The vitamin D status of the US population from
1988 to 2010 using standardized serum
concentrations of 25-hydroxyvitamin D shows
recent modest increases. Am J Clin Nutr. 2016;104
(2):454-461.
4. Cashman KD, Dowling KG, Škrabáková Z, et al.
Vitamin D deficiency in Europe: pandemic? Am J
Clin Nutr. 2016;103(4):1033-1044.
5. Munns CF, Shaw N, Kiely M, et al. Global
consensus recommendations on prevention and
management of nutritional rickets. J Clin Endocrinol
Metab. 2016;101(2):394-415. doi:10.1210/jc.2015-2175
6. Holick MF. Resurrection of vitamin D deficiency
and rickets. J Clin Invest. 2006;116(8):2062-2072.
doi:10.1172/JCI29449
7. Sayers A, Fraser WD, Lawlor DA, Tobias JH.
25-Hydroxyvitamin-D3 levels are positively related
to subsequent cortical bone development in
childhood: findings from a large prospective cohort
study. Osteoporos Int. 2012;23(8):2117-2128. doi:10
.1007/s00198-011-1813-9
8. Viljakainen HT, Korhonen T, Hytinantti T, et al.
Maternal vitamin D status affects bone growth in
early childhood—a prospective cohort study.
Osteoporos Int. 2011;22(3):883-891.
9. Hewison M. An update on vitamin D and human
immunity. Clin Endocrinol (Oxf). 2012;76(3):315-325.
10. Science M, Maguire JL, Russell ML, Smieja M,
Walter SD, Loeb M. Low serum 25-hydroxyvitamin
D level and risk of upper respiratory tract infection
in children and adolescents. Clin Infect Dis. 2013;57
(3):392-397.
11. Wang JW, Hogan PG, Hunstad DA, Fritz SA.
Vitamin D sufficiency and Staphylococcus aureus
infection in children. Pediatr Infect Dis J. 2015;34
(5):544-545.
12. Ginde AA, Mansbach JM, Camargo CA Jr.
Association between serum 25-hydroxyvitamin D
level and upper respiratory tract infection in the
Third National Health and Nutrition Examination
Survey. Arch Intern Med. 2009;169(4):384-390.
13. Martineau AR, Jolliffe DA, Hooper RL, et al.
Vitamin D supplementation to prevent acute
respiratory tract infections: systematic review and
meta-analysis of individual participant data. BMJ.
2017;356:i6583.
14. Xiao L, Xing C, Yang Z, et al. Vitamin D
supplementation for the prevention of childhood
acute respiratory infections: a systematic review of
randomised controlled trials. Br J Nutr. 2015;114
(7):1026-1034.
15. Vuichard Gysin D, Dao D, Gysin CM, Lytvyn L,
Loeb M. Effect of vitamin D3 supplementation on
respiratory tract infections in healthy individuals:
a systematic review and meta-analysis of
randomized controlled trials. PLoS One. 2016;11(9):
e0162996.
16. Clemens TL, Adams JS, Henderson SL, Holick
MF. Increased skin pigment reduces the capacity of
skin to synthesise vitamin D3. Lancet. 1982;1(8263):
74-76.
17. Saari A, Sankilampi U, Hannila ML, Kiviniemi V,
Kesseli K, Dunkel L. New Finnish growth references
for children and adolescents aged 0 to 20 years:
length/height-for-age, weight-for-length/height,
and body mass index-for-age. Ann Med. 2011;43(3):
235-248.
18. Helve O, Viljakainen H, Holmlund-Suila E, et al.
Towards evidence-based vitamin D
supplementation in infants: vitamin D intervention
in infants (VIDI)—study design and methods of a
randomised controlled double-blinded intervention
study. BMC Pediatr. 2017;17(1):91. doi:10.1186
/s12887-017-0845-5

E8 JAMA Pediatrics Published online May 29, 2018 (Reprinted) jamapediatrics.com

© 2018 American Medical Association. All rights reserved.

Downloaded From: on 05/29/2018

 Dosage of Vitamin D3 and Bone Strength and Infections in Infants
19. Ammann P, Rizzoli R. Bone strength and its
determinants. Osteoporos Int. 2003;14(suppl 3):
S13-S18.
20. Blew RM, Lee VR, Farr JN, Schiferl DJ,
Going SB. Standardizing evaluation of pQCT image
quality in the presence of subject movement:
qualitative versus quantitative assessment. Calcif
Tissue Int. 2014;94(2):202-211.
21. Ireland A, Rittweger J, Schönau E,
Lamberg-Allardt C, Viljakainen H. Time since onset
of walking predicts tibial bone strength in early
childhood. Bone. 2014;68:76-84. doi:10.1016/j
.bone.2014.08.003
22. Wald ER, Guerra N, Byers C. Frequency and
severity of infections in day care: three-year
follow-up. J Pediatr. 1991;118(4, pt 1):509-514.
23. Denny FW, Collier AM, Henderson FW. Acute
respiratory infections in day care. Rev Infect Dis.
1986;8(4):527-532.
24. Hauta-Alus HH, Holmlund-Suila EM, Rita HJ,
et al. Season, dietary factors, and physical activity
modify 25-hydroxyvitamin D concentration during
pregnancy [published online March 2, 2017]. Eur J
Nutr. doi:10.1007/s00394-017-1417-z
25. Winzenberg T, Powell S, Shaw KA, Jones G.
Effects of vitamin D supplementation on bone
jamapediatrics.com
© 2018 American Medical Association. All rights reserved.
Downloaded From: on 05/29/2018
density in healthy children: systematic review and
meta-analysis. BMJ. 2011;342:c7254.
26. Holmlund-Suila E, Viljakainen H, Hytinantti T,
Lamberg-Allardt C, Andersson S, Mäkitie O.
High-dose vitamin D intervention in infants—effects
on vitamin D status, calcium homeostasis, and bone
strength. J Clin Endocrinol Metab. 2012;97(11):4139-
4147.
27. Jeddi M, Dabbaghmanesh MH,
Ranjbar Omrani G, Ayatollahi SM, Bagheri Z,
Bakhshayeshkaram M. Relative importance of lean
and fat mass on bone mineral density in Iranian
children and adolescents. Int J Endocrinol Metab.
2015;13(3):e25542. doi:10.5812/ijem.25542v2
28. Umaretiya PJ, Thacher TD, Fischer PR, Cha SS,
Pettifor JM. Bone mineral density in Nigerian
children after discontinuation of calcium
supplementation. Bone. 2013;55(1):64-68. doi:10
.1016/j.bone.2013.03.017
29. Aglipay M, Birken CS, Parkin PC, et al; TARGet
Kids! Collaboration. Effect of high-dose vs
standard-dose wintertime vitamin D
supplementation on viral upper respiratory tract
infections in young healthy children. JAMA. 2017;
318(3):245-254.
30. Viljakainen HT, Saarnio E, Hytinantti T, et al.
Maternal vitamin D status determines bone
(Reprinted) JAMA Pediatrics Published online May 29, 2018
Original Investigation Research
variables in the newborn. J Clin Endocrinol Metab.
2010;95(4):1749-1757.
31. Raulio S, Erlund I, Männistö S, et al. Successful
nutrition policy: improvement of vitamin D intake
and status in Finnish adults over the last decade.
Eur J Public Health. 2017;27(2):268-273.
32. Jääskeläinen T, Itkonen ST, Lundqvist A, et al.
The positive impact of general vitamin D food
fortification policy on vitamin D status in a
representative adult Finnish population: evidence
from an 11-y follow-up based on standardized
25-hydroxyvitamin D data. Am J Clin Nutr. 2017;105
(6):1512-1520. doi:10.3945/ajcn.116.151415
33. Toivonen L, Schuez-Havupalo L, Karppinen S,
et al. Rhinovirus infections in the first 2 years of life.
Pediatrics. 2016;138(3):e20161309. doi:10.1542/peds
.2016-1309
34. Taipale TJ, Pienihäkkinen K, Isolauri E,
Jokela JT, Söderling EM. Bifidobacterium animalis
subsp. lactis BB-12 in reducing the risk of infections
in early childhood. Pediatr Res. 2016;79(1-1):65-69.
doi:10.1038/pr.2015.174
E9