Aji et al.

BMC Pregnancy and Childbirth (2019) 19:183

https://doi.org/10.1186/s12884-019-2341-4

RESEARCH ARTICLE Open Access

Vitamin D deficiency status and its related

risk factors during early pregnancy: a cross-
sectional study of pregnant Minangkabau
women, Indonesia
Arif Sabta Aji1* , E. Erwinda2, Y. Yusrawati3, Safarina G. Malik4 and Nur Indrawaty Lipoeto5

Abstract
Background: Vitamin D deficiency (VDD) is a common problem in reproductive-aged women and has become a
major public health problem worldwide. The effect of VDD in pregnancy has been associated with several adverse
pregnancy outcomes. This study aims to assess the serum levels of 25-hydroxyvitamin D (25(OH)D) in the first
trimester and its associated factors (socio-demographics, pregnancy profiles, dietary intake, and maternal anthropometry
measurements) for the determination of vitamin D deficiency status in early pregnancy.
Methods: A cross-sectional study of 239 pregnant mothers in West Sumatra, Indonesia was conducted. We measured
lifestyle, socio-demographics and pregnancy profile through a structured questionnaire and interview process. A semi
quantitative-food frequency questionnaire (SQ-FFQ) was used to analyse the dietary intake of the pregnant women.
Serum 25(OH)D concentrations were measured at < 13 weeks gestation using ELISA and logistic regression models were
employed to identify the predictors of low vitamin D status.
Results: The prevalence of first-trimester maternal VDD and sufficiency were 82.8 and 17.2% respectively. The median
25(OH)D was 13.15 ng/mL (3.00–49.29 ng/mL). The significant independent predictors were no working status (OR: 0.029;
0.001–0.708) (p = 0.030); nulliparous parity status (OR: 7.634;1.550–37.608) (p = 0.012); length of outdoor activity status of
less than an hour (OR: 9.659;1.883–49.550) (p = 0.007); and no consumption of supplements before pregnancy (OR: 4.49;
1.081–18.563) (p = 0.039).
Conclusions: The prevalence of VDD is common in early pregnancy among Minangkabau women. Recommendations
and policies to detect and prevent such insufficiency during pregnancy should be developed by considering
the associated factors.
Keywords: Vitamin D, Vitamin D deficiency, Risk factors, Early pregnancy, Minangkabau women, West Sumatra

Background Vitamin D and parathyroid hormone are important in

Indonesia is a tropical country with abundant sun expo- calcium homeostasis and bone mineralisation [4]. The main
sure, as it lies within the equatorial zone. However, low source of vitamin D is obtained by direct synthesis from
vitamin D status is still found in such countries Several sunlight to the skin and stimulation of pre-vitamin D3
studies on vitamin D status in pregnant women and development. Moreover, intake from diet and supplements
women of childbearing age have been conducted and have will be synthesised as vitamin D2. However, dietary intake
shown that an average of more than 95% of individuals only provides 10% of vitamin D requirements as very few
have low vitamin D status [1–3]. foods contain a high amount of vitamin D, so vitamin D
supplements could be considered as a way of boosting such
* Correspondence: [email protected]
requirements [5]. Vitamin D2 and D3 are sequentially con-
1
Postgraduate Biomedical Science Department, Faculty of Medicine, Andalas verted in the liver and kidneys into 25-hydroxyvitamin D,
University, Padang, West Sumatra 25172, Indonesia which is a major circulating form and is used to determine
Full list of author information is available at the end of the article

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Aji et al. BMC Pregnancy and Childbirth (2019) 19:183 Page 2 of 10

vitamin D status, and 1,25-dihydroxyvitamin D, the bio- Data collection

logically active form respectively [6]. The data collected included the subject’s characteristics,
Vitamin D action during pregnancy is vital for foetal anthropometric data, socio-economic status, blood pressure
growth and development as it increases the calcium re- and dietary intake. Characteristic which included drug
quirement to develop foetal bone growth. It also plays a consumption history, maternal age, gravidity, parity, socio-
role in helping the absorption of calcium in the digestive economic status, medical health history and lifestyle, were
tract. Limited exposure to sunlight and food intake dur- collected from personal interviews on enrolment on in the
ing pregnancy may cause insufficient body deposits in study. Anthropometric data, including weight, height,
the foetus and mother. These factors affect the develop- and the mid upper arm circumferences of the pregnant
ment and mineralisation of bone in the foetus and can women, were measured by a midwife and trained enu-
have an adverse effect on the pregnancy outcomes. Sev- merators. Body weight was measured to the nearest
eral studies have reported the relationship between ma- 100 g using an electronic scale (Seca 803, Seca GmbH. Co.
ternal VDD and adverse maternal and foetal outcomes, kg, Hamburg, Germany) and height was measured to the
including gestational diabetes, pre-eclampsia, preterm nearest millimetre using a stadiometer (OneMed-Medicom
labour, low birth weight, and caesarean section [7–10]. stature meter, YF.05.05.V.A.1022, Jakarta, Indonesia)
Personal lifestyle and cultural factors are important and mid-upper arm circumference was measured by
determinants for vitamin availability because of their using a meter line and rounded up to the nearest 0.1 cm
influence on sun exposure and dietary intake [11, 12]. (Medline-OneMed Medicom, Jakarta, Indonesia) in the
Determination of vitamin D status in the first trimester more commonly used hand. Pre-pregnancy BMI was
provides an opportunity to detect early vitamin D status calculated by recalling the women’s pre-pregnancy
and may help reduce pregnancy complications. The aim weight by monitoring their longitudinal maternal and
of this study is to assess serum 25(OH)D levels during child book (KIA), classified according to World
early pregnancy and to determine the factors associated Health Organization guidelines for Asian populations
with low vitamin D status. (underweight, < 18.5 kg/m2; normal, 18.5–23.49 kg/m2;
overweight, 23.5–24.99 kg/m2; pre-obese, 25–29.99 kg/m2;
Methods obese, ≥30 kg/m2) [13]. Gestational age was determined by
Study population the date of the last menstrual period and confirmed by
This was a cross-sectional study of maternal vitamin D ultrasound reports in the first trimester.
status during early pregnancy in West Sumatra, Indonesia. A lifestyle questionnaire was given to the participants,
The research location was divided into two different parts; including questions about the duration of sun exposure,
urban/rural areas and mountainous/coastal areas. The working status, physical activity, and sunscreen use. Sun
study was conducted in five different cities between July exposure was calculated as an index of the hours per week
and September 2017, the locations selected on the basis the pregnant women spent outdoors exposed to sunlight,
that they have the highest pregnancy rates for public either during their leisure or working time. We then
health centres in the sub-districts of each district in West divided the exposure into two groups (≤1 h = inadequate
Sumatra. Pregnant women who visited a health care centre or > 1 h = adequate). A midwife took systolic and diastolic
in each location were recruited with a total number of 239 blood pressure by three measurements before taking a
women taking part. The inclusion criteria were as follows: blood sampling using an aneroid sphygmomanometer
1) having received a pregnancy examination at a health (OneMed-Medicom, Jakarta, Indonesia). Physical activity
centre in one of the research locations; 2) being in the first during the first trimester of pregnancy was measured using
trimester of pregnancy (< 13 weeks); 3) being healthy as “The Global Physical Activity Questionnaire” (GPAQ),
confirmed by a doctor (with no infections or history of developed by the World Health Organization (WHO) [14].
communicable disease); and 4) being willing to comply The WHO STEPwise method was used to calculate phy-
with the study procedures by signing an informed consent sical activity and was expressed as Metabolic Equivalent
form. The exclusion criteria comprised: 1) twin pregnancy; minutes per day (METmins/day). The participants
2) having suffered or presently having a chronic disease were classified as having “high activity” if they accu-
history (e.g. diabetes, hypertension or heart attack); 3) mulated ≥3000 METmins/week, “moderate activity” if
anemia; 4) women routinely taking calcium supplements; 3000 > MET≥600 or “low activity” if < 600 METmins/week.
5) pregnant women taking medicine which would affect
their vitamin D metabolism, such as antiepileptic agents, Dietary assessment
glucocorticoids, anti-estrogens or antiretroviral drugs; and Dietary data were obtained from a validated semi quanti-
6) hypothyroidism. The selected subjects were interviewed tative food frequency questionnaire (SQ-FFQ) developed
and had a blood sample taken to measure their biochem- by Lipoeto (2004) [15]. Mean energy and nutrient intakes
ical serum 25(OH)D level. were calculated and compared with the Recommended
Aji et al. BMC Pregnancy and Childbirth (2019) 19:183 Page 3 of 10

Dietary Allowance (RDA) for pregnant women [16]. The to baseline characteristics and dietary intake between
SQ-FFQ was adapted to Minangkabau food habits; the two groups. Furthermore, analysis of associated
Minangkabau is an ethnic indigenous group in West factors (socio-demographics, pregnancy profile, lifestyle,
Sumatra, Indonesia. The validated SQ-FFQ listed food anthropometry and dietary intake) as predictors of VDD
fortified with vitamin D, natural food rich in vitamin D, status was made using logistic regression.
and dietary supplements. More than 223 general food The categorical data were analysed by a chi-squared
items containing potential sources of vitamin D in West test, and a student’s t test was used to compare the
Sumatra were included in the SQ-FFQ food list. Daily vitamin D serum levels of the two groups. Logistic re-
energy and nutritional intakes were calculated and com- gression models were used to estimate the OR’s of the
pared with pregnant women RDA [14]. Based on calcium dependent variable (vitamin D status) and the indepen-
and vitamin D intake, the women were divided into dent effects of known risk factors (e.g. lifestyle, dietary
two groups; those with an inadequate intake (< 15 μg/ intake, socio-demographics, and pregnancy profile (data
day or < 600 IU/day) and adequate intake (≥15 μg/day indicator for maternal health history, such as parity status,
or ≥ 600 μg/day) of vitamin D. Furthermore, the partici- gestational age, maternal anthropometry, blood pressure
pants were grouped based on those with an adequate intake and adverse pregnancy outcome)). All the data were
(≥1300 mg/day) and inadequate intake (< 1300 mg/day) managed and analysed descriptively using IBM SPSS,
of calcium. version 20.0, and presented as tables and Figs. A signifi-
cance level of a p value of less than 0.05 with Odd Ratio
Measurement of serum 25(OH)D (OR) and 95% CI was used to determine the relationship.
All the first trimester pregnant women enrolled in the
study had blood samples taken from their antecubital Results
vein. Non-fasting maternal blood samples were collected Population characteristics
and banked by phlebotomists at the Biomedical Labora- Table 1 shows the data characteristics of the first tri-
tory, Andalas University. 232 samples from the 239 par- mester of pregnant women, such as socio-demographics,
ticipants were taken directly in public health centres in pregnancy profile, and vitamin D intake. In total, 239
each research location. Subsequently, the samples were subjects were chosen, but only 232 successfully had their
directly transferred and stored in the biomedical labo- blood serum taken for analysis. Seven subjects failed to
ratory at the Faculty of Medicine, Andalas University, enroll in the study. The subjects completed the ques-
for serum 25(OH)D level assay. The serum samples tionnaire and had anthropometric measurement and a
were separated by centrifugation at 3500 rpm at 4 °C blood test for vitamin D taken on their first visit for
for 10 min, then stored in aliquots at − 80 °C. The antenatal care in the public health centres. Their mean
serum 25(OH)D test was assessed using ELISA from age was 29.77 ± 5.68 years, with most subjects in the > 30
Diagnostic Biochemistry Canada (DBC) 25-Hydroxyvitamin years age group (45.30%). The geographical scope of the
D (DBC, London, Ontario Canada). The inter-assay and study was divided into coastal areas (40.90%) and moun-
intra-assay coefficients of variation of total serum 25(OH)D tainous areas (59.10%). Based on the working area, the
level were 5.0 and 8.10% at 21.87 ng/mL, and 2.4 and 9.9% study population comprised 48.70 and 51.30% from urban
at 45.01 ng/mL, respectively. There is no consensus on and rural areas respectively. Maternal education levels
the optimal vitamin D level. In this study, vitamin D were 28.90% up to primary school, 40.50% to secondary
status was determined by 25-hydroxyvitamin D (25(OH)D) school, and 30.60% from tertiary school level (diploma or
levels. We used the cut-off points suggested by the Institute high school). Women who were housewives and did not
of Medicine and vitamin D levels were categorised as have an occupation were included in the no working
sufficient (25(OH)D ≥ 20 ng/mL), insufficient (25(OH)D = group, while those who had jobs were included in the
12–19 ng/mL), or deficient (25(OH)D < 12 ng/mL) [15]. working group.
25(OH)D levels were the best marker for identifying vita-
min D status and the major circulatingform of vitamin D. Dietary, anthropometric, and pregnancy profile of the
Minangkabau women
Statistical analyses The median gestational age of the subjects on recruit-
Data were presented as the mean levels of continuous ment was 10 weeks (in the range 5–12 weeks). Approxi-
variables as a mean ± SD (range), and numbers and mately, 75% of them were and 25% nulliparous. The
percentages were used for the binary logistic of the means of the maternal anthropometric values of the preg-
binary and categorical data. The unit of measurement of nant women were 55.48 ± 11.33 kg for pre-pregnancy
vitamin D concentrations was standardised to the S.I. bodyweight; 154.35 ± 6.0 cm for height; 23.45 ± 4.56 cm
unit, ng/mL for 25-hydroxyvitamin D. Data analysis was for pre-pregnancy BMI; 56.15 ± 11.87 kg for first trimester
performed to identify any differences in the data relating of bodyweight; and 27 ± 3.79 cm for mid-upper arm
Aji et al. BMC Pregnancy and Childbirth (2019) 19:183 Page 4 of 10

Table 1 Socio-demographic, pregnancy profile, and vitamin D Table 1 Socio-demographic, pregnancy profile, and vitamin D
intake of women (N = 232) intake of women (N = 232) (Continued)
Independent variables Percent Mean ± SD (range) / Median Independent variables Percent Mean ± SD (range) / Median
(%) (IQR, 25, 75%) (%) (IQR, 25, 75%)
Socio-demographic factors b. No 97.40
Location Spontaneous abortion
a. Padang 5.60 a. Yes 11.60
b. Padang Pariaman 18.50 b. No 88.40
c. Payakumbuh 26.30 History of preterm birth
d. Lima Puluh Kota 32.80 a. Yes 3
e. Pariaman 16.80 b. No 97
Geographical statusa Blood pressure (mmHg)
a. Coastal 40.90 a. Systolic 110.39 ± 11.32
(min = 90, max = 150)
b. Mountainous 59.10
b. Diastolic 75.41 ± 7.11
Urban statusb (min = 60, max = 90)
a. Urban 48.70 Factors related to vitamin D
b. Rural 51.30 intake

Age (year) 29.77 ± 5.68 Consuming supplement before

(min = 17, max = 44) pregnancy

Age groups a. Yes 12.90

a. ≤20 3 b. No 87.10

b. 21–25 23.70 Consuming supplement during

1st trimester
c. 26–30 28
a. Yes 38.40
d. > 30 45.30
b. No 61.60
Educational status
Calcium and vitamin D
a. Primary school 28.90 supplementsa
b. Secondary school 40.50 a. Yes 16.40
c. Tertiary school 30.60 b. No 83.60
Working status Intake of vitamin D 5.32 (3.10)
a. Working 32.80 from food (min = 0.37, max = 23.63)

b. Not working 67.20 The level of vitamin D 3.40

from food
Household income IDR 2.400 (1.400)
per month (IDR) (min = IDR 700, max = IDR 48.000) a. ≥15 mcg/day 96.60
(adequate)
Household members 3 (1) (min = 4, max = 7)
b. < 15 mcg/day (inadequate)
Pregnancy profiles
Intake of calcium from food 459.230 (367.87)
Gestational age (week) 10 (4) (min = 5, max = 12) (min = 353.27, max = 721.14)
Parity status The level of calcium from food
a. Nulliparous 25.40 a. ≥1300 mg/day 9.90
(adequate)
b. Multiparous 74.60
b. < 1300 mg/day 90.10
Pre-pregnancy weight (kg) 55.48 ± 11.33
(inadequate)
(min = 36, max = 95)
a
Geographical status divided by research location in coastal area and
Height (cm) 154.35 ± 6.0 mountainous area; b Urban status divided by location at urban area
(min = 140, max = 176) (Padang, Payakumbuh, and Pariaman) and rural area (Lima Puluh Kota and
Pre-pregnancy BMI (kg/m2) 23.45 ± 4.56 Padang Pariaman); BMI, body mass index
(min = 14.10, max = 37)
Maternal weight of 56.15 ± 11.87 circumference (MUAC). Furthermore, with regard to diet-
1st trimester (kg) (min = 31, max = 93.9)
ary consumption and factors related to vitamin D intake
Upper arm 27 ± 3.79 (min = 17, max = 38)
circumference (cm) in the study population, 80% of the women had not taken
supplements before pregnancy, 62% were not taking sup-
Smoking status
plements during pregnancy, and more than 90% had low
a. Yes 2.60
vitamin D and calcium intake status. The median of
Aji et al. BMC Pregnancy and Childbirth (2019) 19:183 Page 5 of 10

vitamin D and calcium intake from food were 5.32 (3.1) Table 3 Lifestyle factors of first trimester pregnant women
mcg/d and 459.230 (367.87) mg/d for vitamin D and (N = 232)
calcium intake, respectively. The dietary, anthropometric, Independent variables Percent (%) Median (IQR 25, 75%)
and pregnancy profiles of the study population are sum- Sunlight exposure (minute/day) 60 (53.75
marised in Table 1. (min = 15, max = 300)
Length of outdoor activity
Serum 25(OH)D levels and Vitamin D status a. < 60 min/week 47.80
Table 2. shows that median serum 25(OH)D levels were b. ≥60 min/week 52.20
13.15 ng/mL (in the range of 3.00–49.29 ng/mL). Most
Sunscreen application
of the subjects had insufficient vitamin D status. 47%
were vitamin D deficient, with serum levels lower than a. Yes 70.30
12 ng/mL; 36.20% were vitamin D insufficient (concen- b. No 29.70
tration levels of serum between 12 and 19 ng/mL); and Occupation
17.20% had sufficient vitamin D status (concentration levels a. Indoor 75.40
of serum above or equal to 20 ng/mL). Serum 25(OH)D b. Outdoor 24.60
level categories have varying reference guidelines. In this
Dressing style
study, they were divided into three categories based on In-
stitute of Medicine recommendations, with vitamin D levels a. < 27% uncovered 11.60
categorised as sufficient (25(OH)D ≥ 20 ng/mL), insuffi- b. ≥27% uncovered 88.40
cient (25(OH)D = 12–19 ng/mL) or deficient (25(OH)D Physical activity
< 12 ng/mL) [17]. For further multivariate analysis and a. Low 39.70
meaningful statistical analysis, the subjects categorised as b. Moderate 35.80
either vitamin D insufficient or deficient were combined
c. High 24.60
into one category, with insufficiency-deficiency status
(< 20 ng/mL) and with the sufficiency category being
≥20 ng/mL. 80.80% of the subjects had an insufficient-
deficient vitamin D status, while 17.20% had a sufficient protect their skin from sun exposure. The occupation
vitamin D status. status of the subjects was as follows: 75.40% worked
indoors and 24.60% worked outdoors. The style of dressing
Lifestyle factors individuals included those who were covered, which means
Lifestyle factors in this study were successfully explored that they wore veils during their daily outdoor activities,
in the interview process at the recruitment stage and are and those who were uncovered, meaning they did not wear
shown in Table 3. The median of sunlight exposure veils or cover their entire body surface when outdoors. The
duration was 60 min (in the range of 15–300 min); results of the study population reveal that 11.60 and 88.40%
47.80% of the subjects had less than an hour of sun for < 27% uncovered and ≥ 27% uncovered group respec-
exposure during the day, while 52.20% had more than tively. The physical activity levels in the first trimester
60 min. 70.30% used sunscreen and 29.70% did not to were 39.70% low level, 35.80% moderate and 24.60%
high physical activity level (PAL).
Table 2 Serum 25(OH)D levels among first trimester pregnant
women (N = 232) Factors associated with VDD
Variables Percent (%) Median Univariate analysis of the association between potential
(IQR, 25, 75%) associated factors and vitamin D deficiency-insufficiency
Serum 25(OH)D levels (ng/mL) 13.15 (9.98, min = 3.00, status during early pregnancy are shown in Table 4.
max = 49.29) When using univariate binary logistic regression analysis,
Vitamin D status the vitamin D sufficient group (25(OH)D ≥ 20 ng/mL)
a. Deficiency (< 12 ng/mL) 46.60 was shown to be older; with a higher household income
b. Insufficiency (12–19 ng/mL) 36.20
per month; lower bodyweight in the first trimester and
before pregnancy; lower education levels; lower mid-
c. Sufficiency (≥20 ng/mL) 17.20
upper arm circumference; lower pre-pregnancy BMI
Vitamin D status status in the overweight and obese group; represent a lower
(dichotomous categorized)
proportions of nulliparous pregnant women; had a lower
a. Deficiency-insufficiency 82.80
(< 20 ng/mL)
level of sunscreen application; and more outdoor activity
hours compared to the vitamin D deficiency-insufficiency
b. Sufficiency (≥20 ng/mL) 17.20
group (25(OH)D < 20 ng/mL). A higher proportion of
Aji et al. BMC Pregnancy and Childbirth (2019) 19:183 Page 6 of 10

Table 4 Factors associated with low vitamin D status

Variables OR (95% CI) P value
Age (year) 1.020 0.862–1.208 0.814
Age group < 20 1.00
21–25 0.777 0.060–10,143 0.848
26–30 0.172 0.009–3.279 0.242
> 30 0.189 0.040–111.259 0.360
Education levels Primary 1.00
Secondary 0.568 0.182–1.775 0.330
Tertiary 0.429 0.100–1.829 0.253
Working status Working 1.00 0.030
Not Working 0.029 0.001–0.708
Household income/month (IDR) 1.00 1.00–1.00 0.681
Urban status Urban 1.00 0.882
Rural 1.109 0.285–4.312
Geographical status Coastal 1.00
Mountainous 0.424 0.121–1.486 0.180
Gestational age (week) 0.945 0.771–1.157 0.583
Parity status Multiparous Nulliparous 1.00 1.550–37.608 0.012
7.634
Bodyweight of 1st trimester (kg) 0.975 0.826–1.151 0.764
Upper arm circumference (cm) 1.118 0.882–1.416 0.358
Pre-pregnancy bodyweight (kg) 1.007 0.840–1.207 0.939
Pre-pregnancy BMI (kg/m2) 0.920 0.644–1.315 0.647
Pre-pregnancy BMI status Underweight 1.00
Normal 0.664 0.118–3.731 0.642
Overweight 0.101 0.003–2.977 0.184
Pre-obese 0.990 0.038–26.028 0.995
Obese 2.423 0.025–232.393 0.704
Blood pressure (mmHg) Systolic 1.00 0.947–1.056 0.994
Diastolic 0.971 0.882–1.070 0.552
Sunscreen application No 1.00 0.810
Yes 1.147 0.375–3/507
Type occupation Indoor 1.00 0.081
outdoor 17.713 0.704–445.491
Dressing style < 27% uncovered 1.00 0.267–5.552 0.799
≥27% uncovered 1.218
Physical activity level Low-moderate 1.00 0.731
High 0.815 0.255–2.610
Outdoor activity 0986 0.972–1.001 0.070
The length of outdoor activity ≥60 min/week 1.00 0.007
< 60 min/week 9.659 1.883–49.550
Vitamin D intake (mcg/d) 0.988 0.885–1.103 0.835
Calcium intake (mg/d) 1.00 0.999–1.000 0.269
Vitamin D intake status Inadequate 1.00 0.222–188.964 0.277
Adequate 6.483
Aji et al. BMC Pregnancy and Childbirth (2019) 19:183 Page 7 of 10

Table 4 Factors associated with low vitamin D status (Continued)

Variables OR (95% CI) P value
Calcium intake status Inadequate 1.00 0.086–9.904 0.946
Adequate 0.921
Consuming supplement Yes 1.00 1.081–18.563 0.039
before pregnancy
No 4.49
Consuming supplement during No 1.00 0.172–1.443 0.199
1st trimester
Yes 0.499
Consuming vitamin D and calcium No 1.00 0.259–4.289 0.940
supplement
Yes 1.056
OR odds ratio, CI confidence interval, Ref reference category
Bold number is indicated P < 0.05
Serum 25(OH)D < 20 ng/mL (n = 192) and serum ≥20 ng/mL (n = 40)
1.00 as a reference group

pregnant women who lived in coastal and rural areas had group; those between 30 and 44 ng/mL were in the suffi-
vitamin D sufficiency status. However, there were no sig- ciency group; and those above 100 ng/mL were in the
nificant differences between those variables and vitamin D toxicity group [18]. Using the Endocrine Society cut-offs
status during early pregnancy in the Minangkabau in our study population would change the prevalence of
pregnant women. VDD to 36.20% for severe deficiency, 46.60% for defi-
A significant association was shown between all the ciency and 14.70% for insufficiency, with only 2.60%
risk factor variables and low vitamin D status after per- having vitamin D sufficiency status.
forming a binary logistic regression. The results show Indonesia has a prevalence of VDD status among
that four risk factors were significantly associated with women of childbearing age, children, and pregnant
low vitamin D status; non working status (OR: 0.029, CI women. Recent studies in North Sumatra have shown
95% 0.001–0.708) (p value = 0.030); Nulliparous parity sta- that 70% of women of childbearing age (n = 100) had
tus (OR: 7.634, CI 95% 1.550–37.608) (p value = 0.012); deficient status, while 29 and 1% had insufficient and suffi-
length of outdoor activity status of less than an hour (OR: cient status, respectively [19]. A survey of children aged
9.659, CI 95% 1.883–49.550) (p value = 0.007); and not from 0.5 to 12 years in the South East Asian Nutrition
taking supplements before pregnancy (OR: 4.49, CI 95% Surveys (SEANUTS), based on a total of 16.744 children
1.081–18.563) (p value = 0.039). No significant asso- from Indonesia, Malaysia, Thailand, and Vietnam, re-
ciations were observed between any other factors and vealed that the percentage of children with adequate
vitamin D deficiency-insufficiency status. 25(OH)D (≥75 nmol/L equal to ≥30 ng/mL) was as low as
5% (in Indonesia) and up to 20% (in Vietnam). Vitamin D
Discussion insufficiency (< 50 nmol/L equal to < 20 ng/mL) was
This study has reported the prevalence of low maternal observed in between 40 to 50% of the children in all these
vitamin D status in early pregnancy of 10 weeks (in the countries [20]. Bardosono et al., [2] reported that among
range of 5–12 weeks). The subjects were obtained by 143 healthy pregnant women on their first visit to mater-
approaching singleton pregnant women at Public Health nal clinics in Jakarta, Indonesia, 25.20% were found to be
Centres in the six districts of West Sumatra Province, deficient in calcium and 90.20% vitamin D. This high
Indonesia. Only healthy pregnant women were included, prevalence of VDD was due to gender, childhood con-
ones Who did not have any pre-existing hypertension, dia- sequences, type of occupation, physical activity, vitamin D
betes or other medical issues that could have increased intake, body fat percentage, duration of sun exposure,
their risk of developing pregnancy complications or living in an urban area, region of residence within the
VDD. The majority of our study population were either country, and religion, which all significantly increased the
vitamin D deficient or insufficient (82.80%, < 20 ng/mL), odds of being vitamin D insufficient [2, 19, 20].
compare to those who were vitamin D sufficient (17.20%, Low serum 25(OH)D levels in our study population
≥20 ng/mL) according to the criteria published by the are most likely due to the subjects being in their first
Institute of Medicine [17]. On the other hands, the trimester of pregnancy, therefore a greater proportion of
Endocrine Society published different cut-offs for vita- them would be suffering from morning sickness, resulting
min D status, and reported that women who had less in limited outdoor physical activity, and very few women
than 10 ng/mL were in the severe deficiency group; take multivitamin supplements. Bukhary et al., [21] re-
those who had less than 20 ng/mL were in the deficiency ported that 90% of first trimester pregnant women studied
Aji et al. BMC Pregnancy and Childbirth (2019) 19:183 Page 8 of 10

were suffering from hypovitaminosis D. Jan Mohammed to sunlight. However, sources of vitamin D from food
et al., [22] conducted a study in different terms of preg- are lower and only contributes 10% of total needs. The
nancy, and found that the prevalence of low vitamin D main source of vitamin D is exposure to UVB sunlight,
status (< 30 ng/mL) among second and third trimester which contributes to 90% of vitamin D requirements
women was 60 and 37%, respectively. A study measuring [30]. Therefore, the major cause of VDD is inadequate
the vitamin D level of 200 women in India also mentioned sunlight exposure [31]. Indonesia is a tropical country
similar findings, where the prevalence of VDD was with only two season per year and we conducted this
significantly higher in the third trimester, than in the study in the dry season at that time. In this current
first and second trimester [23]. The reduced preva- study, we did the data collection during July–September
lence of VDD during pregnancy might be due to a 2017 in West Sumatra with the daily hours of sunshine
significant increase in dietary intake and supplied multi- averagely about 5 h [32]. This makes Sumatra has longer
vitamins during pregnancy [22]. sunshine hours compared to other months and those
In this study, few subjects took supplements before preg- data might influence our result of the study. A positive
nancy. There was an association between low consumption correlation was found between sun exposure and vita-
of supplements and vitamin D status (p = 0.039). Pregnant min D status. Most of the women who had less exposure
women who did not take supplements before pregnancy to sunlight were vitamin D deficient. Pregnant women
had a 4.5 times increased risk of VDD (OR: 4.49, CI 95% who are less physically active or have a sedentary indoor
1.081–18.563) (p value = 0.039). Having a proper nutritional lifestyle have less exposure to sunlight [23, 33]. In this
status through dietary supplementation before pregnancy study it was revealed that women who had an outdoor
enhances prenatal health, prepares nutrition reserves in the activity status of less than an hour had a tenfold increase
preconception period, and prevents the risk of adverse in the risk of developing VDD (OR: 9.659, CI 95%
pregnancy and foetal outcomes associated with nutrient 1.883–49.550) (p value = 0.007). Our previous study on
deficiencies [24]. Furthermore, our subjects also had a the third trimester pregnant women also reported that
lower vitamin D and calcium intake (5.23 mcg/d (3.10, those who living in mountainous areas, worked indoors,
range 0.37–23.63 mcg/d) vs 459.230 mg/d (367.87, range or who were engaged in low-middle physical activity had
353.27–721.14). The results with regard to vitamin D intake a greater risk of developing VDD [11]. Applying
in our study population shows that the level is almost half sunscreen with a high sun protection factor could reduce
the Recommended Intake of 10 mcg/day proposed by the vitamin D synthesis in the skin by more than 95%. Many
Institute of Medicine (IOM) and Indonesia Recommended factors influence the vitamin D status of individuals and
Dietary Allowance (RDA) [16, 17]. Even though diet and populations, including latitude, season, time spent
supplement intake only provide low levels of vitamin D, outdoors, clothing habitually worn, sunscreen use,
low vitamin D intake status is considered to be one of the weight status, skin colour, and some medications and
risk factors in VDD. A study by Bukhary et al. showed that medical conditions [34].
dietary intake of vitamin D was a significant predictive Pregnant women with nulliparous parity status tend to
factor, together with ethnic group, educational status and have higher serum 25(OH)D levels than multiparous
sun protection [21]. The first trimester of pregnancy is women [35]. However, different studies related to serum
associated with multiple instances of morning sickness. 25(OH)D levels and parity have presented conflicting
This condition leads to poor dietary intake during this results [36–38]. Our study population, who were nullipa-
period and worsens if it means women are unable to rous women, were likely to have lower 25(OH)D serum
achieve a balanced nutrition, which is an important factor levels than their multiparous counterparts with levels of
in fulfilling nutrient requirements. Increasing dietary 12.70 ng/mL and 14.45 ng/mL respectively. After adjusting
consumption of vitamin D, such as milk and other dairy for multivariate logistic regression, parity status was in-
products, and the use of multivitamins and dietary supple- cluded as one of the risk factors in VDD. Pregnant women
ments may be needed, and these solutions might help to with nulliparous parity status had an eight times higher
maintain a sufficient vitamin D level [25]. Recent clinical risk of developing VDD (OR: 7.634, CI 95% 1.550–37.608)
guidelines report that taking more than 600 IU/day of (p value = 0.012). This finding is similar to that of Perez et
vitamin supplements may increase and maintain vitamin D al., who examined first trimester vitamin D status and the
levels at levels higher than 30 ng/mL (75 nmol/L), but factors associated with the lower 25(OH)D levels.
further research is needed to determine the appropriate Nulliparous women (OR: 2.47;p = 0.002), those of non-
dosage of supplements [26–29]. Caucasian ethnicity (OR: 36.29;p = 0.001), and the season
There are several possible reasons for decreasing (OR: 10.93;p = 0.001) at the time of sampling were factors
25(OH)D serum levels; not only low consumption of related to deficient 25(OH)D levels [36]. Different results
vitamin D-rich food, but also certain external factors were obtained by Choi et al., and Ates et al., who reported
that contribute to preventing bodies from being exposed that parity status was not associated with serum 25(OH)D
Aji et al. BMC Pregnancy and Childbirth (2019) 19:183 Page 9 of 10

levels during the first trimester of pregnancy [39, 40]. How- Acknowledgments
ever, nulliparous pregnant women have a greater risk of The authors would like to acknowledge the participation of the volunteers,
nutrition students, research assistants, biomedical laboratory assistant, and
certain pregnancy complications such as pre-eclampsia and field data enumerator for their support in this study. We appreciate all the
preterm delivery compared to multiparous women [25]. subjects for their co-operation and support of this study. Special thanks are
The limitations of this study include possible bias due also due to all the midwives at the maternal clinics in Payakumbuh, Padang,
Lima Puluh Kota, Pariaman, and Padang Pariaman. All authors read and
to the small sample size and the self-selection of study approved the final manuscript.
subjects, and the fact that a gold-standard method for
measuring 25(OH)D levels was not used. In addition, we Funding
This study was funded by Indonesian Danone Institute Foundation and the
could not completely evaluate underlying factors such as Ministry of Research Technology (Grant No: 007/ROG-D/IDIF/X2016)
pigmentation, physical activity levels, macronutrient and Directorate General of Higher Education Ministry of National Education
values, sun protection score, and the length of sun (KEMENRISTEK DIKTI), Indonesia, with project name The Research of Master
Program Leading to Doctoral Degree for Excellent Students (PMDSU Batch-2)
exposure which is during the dry season, Sumatra has in the year of 2018 (Grant No: 050/SP2HL/LT/DRPM/2018). The views
more sunshine hours compared to other months. These expressed herein are those of the individual authors and do not necessarily
data might influence our result of study. Recall bias from reflect those of Indonesian Danone Institute Foundation.

questionnaires is a common problem when conducting Availability of data and materials

an observational study, as the data are obtained through Data from this study will not be shared because additional results from the
an interview session. However, we tried to reduce bias study are yet to be published.

by recruiting a trained interviewer who was a nutrition- Author’s contributions

ist to collect the dietary assessments and other question- ASA carried out statistical analysis, data collection, data interpretation, and
naires. The information about the background of the drafted the manuscript, NIL, YY and SGM conceiving, supervising, designing,
providing directions, performing data analysis, and helped revise the manuscript.
study population (socioeconomic and sociodemographic) EE involve as a research assistant for collecting data, monitoring and evaluation
used in the study was collected from accurate and com- subject, and project administration. NIL revised the final manuscript. All authors
prehensive databases from the maternal care units in the read and approved the final manuscript.

public health centres in each district. The prevalence of Ethics approval and consent to participate
VDD status in the first trimester may not be compared This study was conducted in accordance with the declaration of Helsinki. All
to other population studies with different geographical procedures involving human subjects were approved by the Ethics Committee
of Faculty of Medicine, Andalas University (No. 262/KEP/FK/2016). All subjects
characterisation. However, the main strength of this provided written consent for their participation in this study.
study is the involvement of biochemical and dietary in-
take measurements of vitamin D in the West Sumatran Consent for publication
Not applicable.
population. External factors were examined to determine
predictive factors in developing VDD status, such as life- Competing interests
style, factors related to vitamin D intake, and pregnancy The authors declare that they have no competing interests.

profiles. A larger sample is needed in future studies,

especially from the Indonesian population, and it would Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in
be better if subjects from different areas of the country published maps and institutional affiliations.
were recruited, in order to form a better picture of
vitamin D status during pregnancy in Indonesia. Author details
1
Postgraduate Biomedical Science Department, Faculty of Medicine, Andalas
University, Padang, West Sumatra 25172, Indonesia. 2Postgraduate Public
Health Department, Faculty of Medicine, Andalas University, Padang, West
Conclusions Sumatra 25172, Indonesia. 3Department of Obstetrics and Gynecology,
Despite living in a country in Southeast Asia with high Faculty of Medicine, Andalas University, Padang, West Sumatra 25172,
sun exposure, pregnant Minangkabau women reported a Indonesia. 4Eijkman Institute for Molecular Biology, Ministry of Research,
Technology and Higher Education, Jakarta 10430, Indonesia. 5Nutrition
high prevalence of first trimester VDD, which was related Science Department, Faculty of Medicine, Andalas University, Padang, West
to not working, nulliparous parity, partaking in outdoor Sumatra 25172, Indonesia.
activities for less than one hour per day, and not con-
Received: 11 December 2018 Accepted: 14 May 2019
suming vitamin D supplements before pregnancy.
Recommendations and policies to detect and prevent
insufficiency of vitamin D during pregnancy should be References
1. Wibowo N, Bardosono S, Irwinda R, Syafitri I, Putri AS, Prameswari N.
developed with consideration of the associated factors. Assessment of the nutrient intake and micronutrient status in the first
trimester of pregnant women in Jakarta. Med J Indonesia. 2017;26:109–15.
Abbreviations https://doi.org/10.13181/mji.v26i2.1617.
25(OH)D: 25-hydroxyvitamin D; BMI: Body mass index; CI: Confidence interval; 2. Bardosono S. Maternal micronutrient deficiency during the first trimester
ELISA: Enzyme-linked immunosorbent assay; IOM: Institute of Medicine; among Indonesian pregnant women living in Jakarta. JKI. 2016;4:76–81.
MUAC: Mid-upper arm circumference; PAL: Physical activity levels; https://doi.org/10.23886/ejki.4.6281.76-81.
RDA: Recommended dietary allowance; SQ-FFQ: Semi quantitative food 3. Sari DK, Damanik HA, Lipoeto NI, Lubis Z. Is micro evolution in tropical
frequency questionnaire; VDD: Vitamin D deficiency country women resulting low 25(OH)D level?: a cross sectional study in
Aji et al. BMC Pregnancy and Childbirth (2019) 19:183 Page 10 of 10

Indonesia. J Nutr Food Sci. 2013;4:225–32. https://doi.org/10.4172/2155- gene polymorphism. Innovations in Nutrition, Food Sciences and Public
9600.1000246. Health, JAPAN. J Nutr Food Sci. 2017;7:4(Suppl). https://doi.org/10.4172/
4. Triunfo S, Lanzone A. Potential impact of maternal vitamin D status on 2155-9600-C1-044.
obstetric well-being. J Endocrinol Investig. 2016;39:37–44. https://doi.org/10. 27. Mazahery H, von Hurst PR. Factors affecting 25-Hydroxyvitamin D
1007/s40618-015-0330-7. concentration in response to Vitamin D supplementation. Nutrients. 2015;7:
5. Binkley N, Ramamurthy R, Krueger D. Low Vitamin D status: definition, 5111–42. https://doi.org/10.3390/nu7075111.
prevalence, consequences and correction. Endocrinol Metab Clin N Am. 28. Mithal A, Kalra S. Vitamin D supplementation in pregnancy. Indian J
2010;39:287–contents. https://doi.org/10.1016/j.ecl.2010.02.008. Endocrinol Metab. 2014;18:593–6. https://doi.org/10.4103/2230-8210.139204.
6. Bikle D, Vitamin D. Production, Metabolism, and Mechanisms of Action. In: 29. Harvey NC, Holroyd C, Ntani G, Javaid K, Cooper P, Moon R, et al. Vitamin D
De Groot LJ, Chrousos G, Dungan K, Feingold KR, Grossman A, Hershman supplementation in pregnancy: a systematic review. Health Technol Assess.
JM, et al., editors. Endotext. South Dartmouth (MA): MDText.com, Inc; 2000. 2014;18:1–190. https://doi.org/10.3310/hta18450.
7. Bodnar LM, Simhan HN, Catov JM, Roberts JM, Platt RW, Diesel JC, et al. 30. Björn LO, Wang T. Vitamin D in an ecological context. Int J Circumpolar
Maternal vitamin D status and the risk of mild and severe preeclampsia. Health. 2000;59:26–32.
Epidemiology. 2014;25:207–14. https://doi.org/10.1097/EDE. 31. Holick MF, Chen TC. Vitamin D deficiency: a worldwide problem with health
0000000000000039. consequences. Am J Clin Nutr. 2008;87:1080S–6S.
8. Dror DK. Vitamin D status during pregnancy: maternal, fetal, and postnatal 32. BMKG Indonesia. Prakiraan Musim Kemarau 2017 di Indonesia | BMKG.
outcomes. Curr Opin Obstet Gynecol. 2011;23:422–6. https://doi.org/10. BMKG | Badan Meteorologi, Klimatologi, dan Geofisika. 2017. https://www.
1097/GCO.0b013e32834cb791. bmkg.go.id/iklim/prakiraan-musim.bmkg?p=prakiraan-musim-kemarau-2017-
9. Scholl TO, Chen X, Stein P. Maternal Vitamin D status and delivery by di-indonesia&tag=prakiraan-musim&lang=ID. (Accessed 9 Aug 2018).
cesarean. Nutrients. 2012;4:319–30. https://doi.org/10.3390/nu4040319. 33. Sari DK, Damanik HA, Lipoeto NI, Lubis Z. Is micro evolution in tropical
10. Khalessi N, Kalani M, Araghi M, Farahani Z. The relationship between country women resulting low 25(OH)D level?: a cross sectional study in
maternal Vitamin D deficiency and low birth weight neonates. J Family Indonesia. J Nutr Food Sci. 2013;4. https://doi.org/10.4172/2155-9600.
Reprod Health. 2015;9:113–7. 1000246.
11. Aji AS, Desmawati D, Yerizel E, Lipoeto NI. The association between 34. Rockwell M, Kraak V, Hulver M, Epling J. Clinical Management of low Vitamin
lifestyle and maternal vitamin D levels during pregnancy in West D: a scoping review of physicians’ practices. Nutrients. 2018;10. https://doi.
Sumatra. Asia Pac J Clin Nutr. 2018;27:1286–93. https://doi.org/10.6133/ org/10.3390/nu10040493.
apjcn.201811_27(6).0016. 35. Zhao X, Fang R, Yu R, Chen D, Zhao J, Xiao J. Maternal Vitamin D status in
12. Nimitphong H, Holick MF. Vitamin D status and sun exposure in Southeast the late second trimester and the risk of severe preeclampsia in
Asia. Dermatoendocrinol. 2013;5:34–7. https://doi.org/10.4161/derm.24054. southeastern China. Nutrients. 2017;9. https://doi.org/10.3390/nu9020138.
13. Expert Consultation WHO. Appropriate body-mass index for Asian 36. Pérez-López FR, Fernández-Alonso AM, Ferrando-Marco P, González-
populations and its implications for policy and intervention strategies. Salmerón MD, Dionis-Sánchez EC, Fiol-Ruiz G, et al. First trimester serum 25-
Lancet. 2004;363:157–63. https://doi.org/10.1016/S0140-6736(03)15268-3. hydroxyvitamin D status and factors related to lower levels in gravids living
14. Armstrong T, Bull F. Development of the world health organization global in the Spanish Mediterranean coast. Reprod Sci. 2011;18:730–6. https://doi.
physical activity questionnaire (GPAQ). J Public Health. 2006;14:66–70. org/10.1177/1933719110396720.
https://doi.org/10.1007/s10389-006-0024-x. 37. Makgoba M, Nelson SM, Savvidou M, Messow C-M, Nicolaides K, Sattar N.
15. Lipoeto NI, Agus Z, Oenzil F, Wahlqvist M, Wattanapenpaiboon N. Dietary First-trimester circulating 25-Hydroxyvitamin D levels and development of
intake and the risk of coronary heart disease among the coconut- gestational diabetes mellitus. Diabetes Care. 2011;34:1091–3. https://doi.org/
consuming Minangkabau in West Sumatra. Indonesia Asia Pac J Clin Nutr. 10.2337/dc10-2264.
2004;13:377–84. 38. Andersen LB, Abrahamsen B, Dalgård C, Kyhl HB, Beck-Nielsen SS, Frost-
16. Ministry of Health Republic of Indonesia. Dietary intake reference in Nielsen M, et al. Parity and tanned white skin as novel predictors of vitamin
Indonesia. Jakarta: Ministry of Health republic of Indonesia; 2013. D status in early pregnancy: a population-based cohort study. Clin
17. Institute of Medicine. Dietary reference intakes for calcium and Vitamin D. Endocrinol. 2013;79:333–41. https://doi.org/10.1111/cen.12147.
Washington, DC: The National Academia Press; 2011. 39. Choi R, Kim S, Yoo H, Cho YY, Kim SW, Chung JH, et al. High prevalence of
Vitamin D deficiency in pregnant Korean women: the first trimester and the
18. Holick MF, Binkley NC, Bischoff-Ferrari HA, Gordon CM, Hanley DA, Heaney
winter season as risk factors for Vitamin D deficiency. Nutrients. 2015;7:
RP, et al. Evaluation, treatment, and prevention of vitamin D deficiency: an
3427–48. https://doi.org/10.3390/nu7053427.
Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2011;
40. Ates S, Sevket O, Ozcan P, Ozkal F, Kaya MO, Dane B. Vitamin D status in the
96:1911–30. https://doi.org/10.1210/jc.2011-0385.
first-trimester: effects of Vitamin D deficiency on pregnancy outcomes. Afr
19. Sari DK, Damanik HA, Lipoeto NI, Lubis Z. Low serum 25(OH)D levels are
Health Sci. 2016;16:36–43. https://doi.org/10.4314/ahs.v16i1.5.
associated with single nucleotide polymorphisms of the vitamin D receptor
gene and lifestyle factors, especially in women with higher body fat
percentage. Obes Res Clin Pract. 2013;7:12–3. https://doi.org/10.1016/j.orcp.
2013.08.046.
20. Poh BK, Rojroongwasinkul N, Nguyen BKL, null S, Ruzita AT, Yamborisut U, et
al. 25-hydroxy-vitamin D demography and the risk of vitamin D insufficiency
in the south east Asian nutrition surveys (SEANUTS). Asia Pac J Clin Nutr.
2016;25:538–48.
21. Bukhary NBI, Isa ZM, Shamsuddin K, Lin KG, Mahdy ZA, Hassan H, et al. Risk
factors for antenatal hypovitaminosis D in an urban district in Malaysia. BMC
Pregnancy Childbirth. 2016;16. https://doi.org/10.1186/s12884-016-0939-3.
22. Jan Mohamed HJ, Rowan A, Fong B, Loy S-L. Maternal serum and breast Milk
Vitamin D levels: findings from the Universiti Sains Malaysia pregnancy cohort
study. PLoS One. 2014;9. https://doi.org/10.1371/journal.pone.0100705.
23. Singh D. Vitamin D status in pregnant women of Udaipur. Biochem Anal
Biochem. 2016;5:1–3. https://doi.org/10.4172/2161-1009.1000280.
24. World Health Organization. WHO recommendations on antenatal Care for a
Positive Pregnancy Experience. Geneva: World Health Organization; 2016.
25. Flood-Nichols SK, Tinnemore D, Huang RR, Napolitano PG, Ippolito DL.
Vitamin D deficiency in early pregnancy. PLoS One. 2015;10. https://doi.org/
10.1371/journal.pone.0123763.
26. Sari DK. Vitamin D supplementation increased 25(OH)D serum levels but did
not reach normal range in north Sumatera women with vitamin D receptor