Epidemiology of vitamin D insufﬁciency and deﬁciency in a population in a sunny
country: Geospatial meta-analysis in Brazil
e Yure Gomes dos Santos
, Gisele Queiroz Carvalho
, Djanilson Barbosa dos Santos
and Ana Marl
Center of Biological and Health Sciences, Universidade Federal do Oeste da Bahia, Barreiras; Instituto de Sa
ude Coletiva, Universidade Federal da
Bahia, Salvador, Brazil;
Center of Humanities, Universidade Federal do Oeste da Bahia, Barreiras, Brazil;
Campus Avan¸cado de Governador Valadares,
Universidade Federal de Juiz de Fora, Minas Gerais, Brazil;
Center for Health Sciences, Universidade Federal do Rec^
oncavo da Bahia, Santo Ant^
School of Nutrition, Universidade Federal da Bahia, Salvador, Brazil
Studies conducted among populations of tropical countries have reported high prevalences of vitamin D
deﬁciency and insufﬁciency. Information resulting from meta-analyses on the spatial distribution of vitamin
Ddeﬁciency and insufﬁciency in tropical countries is still rare. The aim of this review was investigated the
prevalence of vitamin D deﬁciency and insufﬁciency among the Brazilian population. Observational studies
were searched in eight electronically databases. Additionally, theses and dissertations and abstracts were
screened. Details on study design, methods, population, mean and data on serum concentrations of vitamin
D in different age groups in Brazil were extracted. Data were pooled using a random-effects model and
choropleth maps were created based on the geopolitical regions of the country. 72 published paper met the
inclusion criteria. The mean vitamin D concentration among the Brazilian population between 2000 and
2017 of 67.65 nmol/L (95% CI: 65.91, 69.38 nmol/L).The prevalences of vitamin D deﬁciency and insufﬁciency
were 28.16% (95% CI: 23.90, 32.40) and 45.26% (95% CI: 35.82, 54.71), respectively, for the Brazilian
population. The highest prevalence of deﬁciency were observed in the southern and southeastern regions
and the highest occurrence of vitamin D insufﬁciency was among the populations of the southeastern and
northeastern regions. Finally, there are high prevalence of inadequate vitamin D concentrations among the
population, regardless of age group in Brazil. The development of vitamin D food fortiﬁcation policies in
needs to be cautious and carefully planned.
Vitamin D; vitamin D
deﬁciency; systematic review;
Meta-analysis; Sunny country
Vitamin D is an essential fat-soluble vitamin for calcium
homeostasis and bone health (Palacios and Gonzalez, 2014).
An estimated 80% to 90% of vitamin D in the human body
originates from cutaneous synthesis by means of activation of
7-dehydrocholesterol through sunlight, while the remaining
20% to 10% are provided by either supplements or food
Low-latitude regions, such as Brazil, allow photosynthesis of
vitamin D at adequate concentrations during most seasons of
the year because of the availability of ultraviolet rays (UVB)
(Tsiaras and Weinstock, 2011) and the possibility that individ-
uals can be exposed to sunlight. However, paradoxically, studies
conducted among populations of tropical countries, such as
Brazil, have reported high prevalences of vitamin D deﬁciency
and insufﬁciency (Unger et al., 2010; Santos et al., 2012).
The nutritional state of vitamin D in the human body is
measured from the plasma levels of 25-hydroxyvitamin D (25
(OH)D). It was proposed from a review of the literature that 25
(OH)D vitamin concentrations below 50 nmol/l in adults
should be considered to represent vitamin deﬁciency, while
those between 50 and 80 nmol/l 25(OH)D would indicate
insufﬁciency (Holick, 2007). However, these cutoff points were
deﬁned based on data from Western countries with high
proportions of elderly individuals in their populations (Hoteit
et al., 2014).
Vitamin D deﬁciency and insufﬁciency are associated with
several chronic endocrine-metabolic diseases (Holick and
Chen, 2008). In this regard, meta-analysis of data has shown
that vitamin D deﬁciency were associated with increased risk of
cardiovascular diseases, diabetes, metabolic syndrome, obesity
and cancer (Luo et al., 2017; Qi et al., 2017; Pereira-Santos
et al., 2015; Zhang et al., 2017; Pludowski et al., 2013). The
clinical manifestations of vitamin D deﬁciency include muscu-
loskeletal disorders, such as rickets and osteoporosis, and
increased occurrence of infections, while insufﬁciency predis-
poses individuals to the risk of developing chronic diseases
with no clinical manifestations (Arabi et al., 2010).
CONTACT Marcos Pereira-Santos email@example.com Rua Professor José Seabra de Lemos, 316, Recanto dos Pássaros, CEP: 47808-021 Barreiras,
Color versions of one or more of the ﬁgures in the article can be found online at www.tandfonline.com/bfsn.
Supplemental data for this article can be accessed on the publisher’s website.
Systematic Review Registration: PROSPERO number CRD42017076118.
© 2018 Taylor & Francis Group, LLC
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION
The prevalence of vitamin D deﬁciency worldwide remains
uncertain, since there is a lack of data from many countries
(Palacios and Gonzalez, 2014). In 2007, an estimated one bil-
lion people worldwide presented either vitamin D insufﬁciency
or deﬁciency (James, 2008). However, one decade later, the out-
come from this estimate remains unknown. Moreover, the
occurrence and distribution of vitamin D among South Ameri-
can populations is still poorly understood (van Schoor and
Lips, 2011) and research on the nutritional state of vitamin D
among the populations of sunny countries also remains scarce
(Santos et al., 2012).
In Brazil, studies conducted over the past few years have
revealed high prevalence of vitamin D deﬁciency and insufﬁ-
ciency among different age groups and in both sexes (Arabi
et al., 2010). In a healthy population in S~
ao Paulo, aged 18–
90 years, vitamin D insufﬁciency (vitamin D serum 25
<75 ml/L) was reported to affect 77.4% (Unger et al., 2010).
Slightly lower prevalence of insufﬁciency (54.3%; 50–75 nmol/
L) was identiﬁed among girls living in the municipality of Curi-
tiba (Santos et al., 2012).
Regarding vitamin D deﬁciency (<50 nmol/l), high occur-
rences have been observed among different groups in Brazil:
36.3% among girls living in the municipality of Curitiba(Santos
et al., 2012); and 86% among elderly individuals in the southern
region of the country (Scalco et al., 2008). Therefore, in Brazil,
the results from studies on the nutritional state of vitamin D
among different age groups, whether healthy or not, support
the need to conduct a meta-analysis on vitamin D concentra-
tions in Brazil, a tropical country with high incidence of ultravi-
olet rays. Thus, the objective of the present study was to
estimate the prevalence and spatial distribution of vitamin D
deﬁciency and insufﬁciency in the Brazilian population.
The present systematic review and meta-analysis followed the
recommendations of the Meta-analysis Of Observational Stud-
ies in Epidemiology (MOOSE ) guidelines (Stroup et al., 2000).
In the supplementary ﬁle we have attached a MOOSE checklist.
The protocol for this systematic review was registered in the
PROSPERO registry prior to starting the literature search
Search strategy and eligibility criteria
The PubMed, Bireme, Scopus, Web of Science, Science Direct,
SciELO and Lilacs databases were used to identify observational
studies that estimated the serum concentrations of vitamin D
among different age groups in Brazil. Published papers regis-
tered in these databases up to May 10, 2017 were identiﬁed
using the descriptors "vitamin D”,“ergocalciferols”,“cholecal-
ciferol”,“Brazil”and “humans”. Additionally, theses and disser-
tations were identiﬁed from the thesis database of Peri
Capes, and also abstracts from Brazilian conferences in the
ﬁelds of endocrinology and nutrition. At the end of the search,
we evaluated the reference lists of the articles on vitamin D, so
as to identify any studies that were not indexed in the databases
but might be pertinent for inclusion in this review (Table S1).
The search strategy was designed and conducted by an
epidemiologist with experience in development of systematic
review and meta-analysis (MPereira-Santos).
The inclusion criteria were that the articles, theses, disserta-
tions and conference papers should report on observational
studies from research that measured the serum concentrations
of vitamin D among non-hospitalized Brazilians, and should be
original. These studies could be published in any language.
Studies with an experimental design, opinionated reviews, edi-
torials, review articles and case reports were excluded. The
PICOS (population, intervention, comparison group, outcome,
and study design) criteria was used to guide the determination
of inclusion and exclusion criteria for this review (Table S2).
The published papers were managed using Mendeley and
the inclusion and exclusion criteria were applied by indepen-
dent reviewers (M Pereira-Santos and GQ Carvalho), who
selected the eligible articles. Articles were screened and selected
for full-text review if they met the selection criteria. At the end
of the review, disagreements about article inclusion were
resolved through reaching a consensus between the two
reviewers, who discussed eligibility and came to an agreement.
In the absence of a consensus, a third reviewer evaluated
whether the study in question was eligible (AM Oliveira).
The serum level of 25(OH)D was used as an indicator for
vitamin D status, because this metabolite reﬂects the combined
effect of intake, skin synthesis, storage, blood transport protein
and catabolism (Holick, 2007). Moreover, hydroxylation of 25
(OH)D to 1.25(OH)2D3 (active vitamin D) occurs in several
tissues: the half-life of 25-OH-D is two to three weeks, while
the half-life of 1.25(OH)2D3 is approximately six hours (Mose-
The articles thus selected were read in their entirety and infor-
mation regarding publication and design, sample, age of partic-
ipants, period when the study was conducted, technique for
assaying vitamin D, place of study and variables investigated
was recorded using a form designed for gathering this informa-
tion. The selected studies presented data from the outcomes of
the study, and no contact with authors was necessary.
Means, standard deviations and prevalence data regarding
vitamin D insufﬁciency and deﬁciency among the samples of
the studies selected were gathered. All 25(OH)D values in
nanograms per milliliter were converted to nanomoles per liter
by multiplying by 2.496, as necessary.
The 25(OH)D serum concentrations were deﬁned as the
response variable in the present study. This measurement was
considered in its continuous form (mean vitamin D level) and
in categories, which were deﬁned as either deﬁcient (<
50 nmol/L) or insufﬁcient (50 nmol/L to 80 nmol/L). Normal
concentrations of vitamin D were not used in the present study.
Evaluation of the methodological quality of the studies
Two researchers (M Pereira-Santos, DB Santos) independently
scored the quality of the observation studies included in the
meta-analyses using an adapted version of the instrument pro-
posed by Loney et al. (1998) for critical evaluations on preva-
2 M. PEREIRA-SANTOS ET AL.
This instrument contains eight criteria for evaluating the
methodological quality of studies. However, one criterion
regarding impartial assessment made by trained evaluators was
excluded because it was not considered pertinent to the nature
of data collection, which in this case was blood sampling. Thus,
the following criteria were used: 1) probabilistic or census sam-
pling; 2) adequate source of sampling (ofﬁcial census or school
census, among others); 3) previously calculated sample size; 4)
adequate method for measuring vitamin D; 5) adequate
response rate (>70.0%) and description of refusals; 6) presen-
tation of the conﬁdence intervals and analysis of the subgroups
of interest; and 7) well-described study subjects that are similar
to those of the research question (Loney et al., 1998).
For each criterion met, one point was attributed to the study
analyzed. High-quality studies were considered to be those that
reached 6 to 7 points; moderate quality between 4 and 5 points;
and low quality between 0 and 3 points (Loney et al., 1998).
The evaluation of methodological quality and risk of bias
was performed in relation to dissertations, theses and
articles. Conference abstracts were not assessed because the
information regarding the methodological quality of this
type of study was insufﬁcient. Thus, abstracts were used to
obtain vitamin D mean, prevalence and insufﬁciency data
from studies that were not published in the form of articles
or theses and dissertations. To do so, the Lattes curriculum
of the authors of the abstracts was accessed to certify
absence of publications regarding vitamin D.
The mean and prevalence of vitamin D deﬁciency and insufﬁ-
ciency were used as the meta-analysis summary measurements,
according to the geopolitical regions of Brazil: north, northeast,
center-west, southeast and south.
Mean vitamin D concentrations among the populations of
each region in Brazil and their respective conﬁdence intervals
(95% CI) were obtained following either the ﬁxed or the ran-
dom effects model, depending on the heterogeneity among the
studies. The heterogeneity and inconsistency of measurements
was identiﬁed through Cochran’s Q test. If heterogeneity was
conﬁrmed (p <0.05; I
>50%), the random effects model was
applied with inverse variance and weights according to the
results of individual studies (Higgins and Thompson, 2002).
Statistical analyses were performed using the Stata 12 software
(Stata Corp, College Station, TX, USA).
Geospatial analysis was used in the meta-analysis to obtain the
mean values for the prevalences of vitamin D deﬁciency and insuf-
ﬁciency among the Brazilian population, which were calculated
according to the ﬁxed effects model. Choropleth maps were cre-
ated based on the regions of the country (north, northeast, center-
west, south and southeast), for the period of 2000 to 2017.
A quantiﬁcation map was ﬁrst created using the method of
proportional symbols. To do so, occurrences of studies in each
of the catalogued municipalities were quantitatively represented
(Archela and Th
ery, 2008). This map also included the regional
classiﬁcation of Brazil (north, northeast, center-west, southeast
and south) using the chorochromatic method, with colors as
the visual variable for establishing zones. Thus, each region of
the country received a speciﬁc color, which allowed them to be
Maps of vitamin D deﬁciency, insufﬁciency and mean concen-
tration were created using the mean vitamin D levels and preva-
lences of vitamin D deﬁciency and insufﬁciency from the meta-
analysis. Thus, the data from each region were georeferenced such
that higher prevalences of vitamin D deﬁciency and insufﬁciency
were represented on a scale of increasing color intensity.
The spatial representation of data was processed using the
ArcGISÒ10.4 software, to produce thematic maps. The carto-
graphic base used to produce the thematic maps of the present
study was obtained from the Brazilian Institute for Geography
and Statistics (IBGE, 2015).
After screening (title and abstract), 108 studies were analyzed
regarding eligibility and 36 were excluded because they did not
meet the inclusion criteria. The reasons for excluding articles
were that they did not meet the criteria for the type of design
that had been previously deﬁned; involved hospitalized partici-
pants; were not conducted within the study period deﬁned; or
reported on the development of quantiﬁcation methods for
vitamin D (Figure 1 and Table S2). In total, 72 studies were eli-
gible for inclusion in the systematic review and meta-analysis
(Figure 1, Table S3).
The studies thus analyzed (Table S4) involved a total of
340,476 Brazilians in different age groups. Most of them were
cross-sectional studies (93.0%) and analyzed adults and elderly
adults (Tables 1 and 2). Most investigations diagnosed the vita-
min D situation between 2006 and 2011 (37.5%) and between
2012 and 2017 (43.1%).
Quantiﬁcation of serum 25(OH)D among the evaluated indi-
viduals was carried out using different techniques. Chemilumi-
nescence was the most commonly used technique (38.15%),
followed by radioimmunoassay (21.4%). Regarding methodo-
logical quality, 52.3% of the studies were classiﬁed as presenting
moderate and high levels of quality (33.5%) (Table 1).
Most of the studies were conducted in the southeastern region
(58.67%) and northeastern region (20.00%), while lower pro-
portions were conducted in the central-western and northern
regions of Brazil (Figure 2. Graph A). The mean vitamin D con-
centration for the Brazilian population between the years 2000
and 2017, from the meta-analysis, was 67.65 nmol/L (95% CI:
65.91, 69.38 nmol/L; Figure 2. Graph B). The highest level was
observed among the population in the northeastern region
The prevalence of vitamin D deﬁciency was 28.16% (95% CI:
23.90,32.40) for the Brazilian population (Figure 4, Graph A).
The highest levels of deﬁciency were observed in the southern
and southeastern regions (Figure 3- Graph A). In turn, the rate
of vitamin D insufﬁciency (Figure 3- Graph B) was 45.26%
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 3
(95% CI: 35.82,54.71), and the highest occurrence was among
the populations of the southeastern and northeastern regions
(Figure 3- Graph B).
The situation of vitamin D deﬁciency and insufﬁciency did
not change according to the year in which the research was
conducted, nor was it inﬂuenced by the age group involved in
the study (Table 2). However, the highest occurrence of vitamin
Ddeﬁciency was reported among the elderly population
The present study was the ﬁrst meta-analysis to estimate the
prevalence of vitamin D deﬁciency and insufﬁciency for the
Figure 1. Study selection ﬂowchart.
Table 1. Main characteristics of selected studies on vitamin D concentration among the Brazilian population.
Year when study was published
2005–2009 13 18.3
2010–2014 23 26.7
2015–2017 36 55.0
Cross-sectional 67 93.0
Prospective cohort 5 7.0
CLIA –chemiluminescence 27 38.5
HPLC –high-sensitivity liquid chromatography 14 20.0
RIA –radioimmunoassay 15 21.4
ECLIA –chemiluminescence immunoassay 4 5.7
ELISA –enzyme-linked immunosorbent assay 1 1.4
IRMA –immunoradiometric assay 3 4.5
Not reported 6 8.5
Methodological quality- 63 studies
High quality - 6 to 7 points 21 33.5
Moderate quality - 4 to 5 points 33 52.3
Low quality - 0 to 3 points 9 14.2
4 M. PEREIRA-SANTOS ET AL.
Brazilian population according to the country’s geopolitical
regions. The mean serum concentration of vitamin D for the
population, which was obtained from observational studies,
was characterized as insufﬁcient and did not seem to differ sig-
niﬁcantly according to the age groups investigated. Thus,
despite the high solar incidence in Brazil, the prevalence of vita-
min D deﬁciency and insufﬁciency in this country was similar
to that of nations with reduced solar availability, regardless of
the age group investigated (Hilger et al., 2014).
The evidence available suggests that vitamin D deﬁciency
can be considered to be a public health issue, since it affects all
phases of life in populations on different continents. In Euro-
pean countries, for example, the prevalence of vitamin D deﬁ-
ciency among the population was reported to be 40.4%
(Cashman et al., 2016). European data collected since 1913
(Mellaby, 1919) had already reported that vitamin D deﬁciency
was more common among the population living in countries
with higher solar incidence, such as Italy, Spain and Greece,
than among the population of countries where solar exposure
was considered inadequate (Mellaby, 1919; Hilger et al., 2014).
This paradox can be explained by the concern for adopting
actions to prevent the risks of exposure to high levels of solar
incidence, independent of the concentration of solar irradiation
in these countries. This may lead to a decrease in individuals
contact with solar irradiation, thus raising the prevalence of
inadequate vitamin D levels in populations in tropical coun-
tries. This preventive action may also explain why there is a
high prevalence of vitamin D deﬁciency in countries with low
High prevalence of vitamin D insufﬁciency and deﬁciency
were identiﬁed in different age groups in the Brazilian
population. A similar result was observed in a meta-analysis
study that estimated occurrences of vitamin D deﬁciency and
insufﬁciency in different populations around the world (Hilger
et al., 2014).
Most of the studies included in the present meta-analysis
evaluated sample populations of elderly individuals or meno-
pausal women, which are populations that are vulnerable to
vitamin deﬁciency and insufﬁciency due to their reduced
capacity for cutaneous activation of 7-dehydrocholesterol, the
precursor for vitamin D. This condition may have contributed
towards increasing the prevalence of inadequate vitamin D
concentrations, but does not decrease the epidemiological
importance of the event. The studies selected also involved
Table 2. Serum vitamin D concentration and prevalence of deﬁciency and insufﬁciency of vitamin D, according to blood sampling and age group [95% conﬁdence
Variable Number of studies % Mean of vitamina D Vitamin D deﬁciency Vitamin D insufﬁciency
Year of data colleted
2000–2005 14 19.4 59.90 (49.47, 70.34) 39.25(20.41, 58.10) 38.24(27.03 ,49.44)
2006–2011 27 37.5 61.64(55.67, 67.617) 30.11 (18.18, 42.05) 55.54(45.48,65.60)
2012–2017 31 43.1 66.37(56.16, 76.59) 32.99 (24.28, 41.69) 46.65(40.06,53.25)
Pregnant women 5 6.6 65.0 (62.38, 67.70) 33.10 (8.84, 57.76) 48.91(8.81, 89.00)
Children 5 6.6 66.68 (35.51, 97.86) 22.95(10.00, 35.89) 44.04 (22.28,65.79)
Adolescents 11 15.7 72.44 (69.81, 75.08) 14.50 (1.80, 27.19) 57.93 (49.09, 66.76)
Adults 19 27.1 61.93 (51.18,72.68) 35.73(26.41, 45.04) 46.35 (35.78,56.92)
Elderly 20 28.5 52.859(45.01, 60.70) 41.53 (27.62, 55.44) 45.85 (36.21, 55.50)
Multiples groups 10 15.5 —— —
Figure 2. Occurrences of studies in each of the catalogued municipalities (A) and mean of vitamin D levels (B) among the Brazilian population.
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 5
adults and adolescents. Children and pregnant women were the
two least investigated groups.
Another possible cause of reduced synthesis of vitamin D
is intensive use of sunscreen, which is greatly encouraged by
dermatologists to prevent skin diseases. Use of sunscreen
has been correlated with vitamin D insufﬁciency among the
Brazilian population. The physiological mechanism that
explains this process is the possible blocking of cutaneous
activation of provitamin D due to the sun protection factor.
Thus, sun protection factor (SPF) 8 can decrease vitamin D
photoproduction capacity by 90%, while SPF 30 decreases it
by 99% (Holick, 2007; Tsiaras and Weinstock, 2011). By
analogy, this mechanism can also explain how use of sun-
screen while performing daily activities, while at work, or
when practicing physical activities in enclosed environ-
ments, with low solar exposure, can decrease exposure to
UVB rays and thus how it may represent a risk factor for
vitamin D deﬁciency and insufﬁciency.
Moreover, during winter, people use more layers of clothes
and decrease their time spent outdoors. Air pollution and
working in closed environments, which limit cutaneous synthe-
sis of vitamin D, are other potential factor relating to vitamin D
insufﬁciency and deﬁciency (Holick, 2007; Tsiaras and Wein-
The factors associated with vitamin D deﬁciency in the Bra-
zilian studies are similar to those reported in other Western
countries and include extremes of age, female sex, winter sea-
son, dark skin pigmentation, lack of sun exposure, a covered
clothing style and obesity (Arabi et al., 2010). Conversely,
younger age, practicing physical activities outdoors, the spring
and summer seasons, living by the sea in sunny locations and
lower latitudes are factors that seem to favor higher serum vita-
min D concentrations among the Brazilian population (Maeda
et al., 2014). However, Brazil is a continent-sized country, rang-
ing in latitude from 5Nto33
S. Thus, the intensity of UVB
light also varies signiﬁcantly across the different regions of
Figure 3. Spatial representation for the prevalences of vitamin D deﬁciency ( A) and insufﬁciency (B) among the Brazilian population.
Figure 4. Meta-analysis of prevalence of vitamin D deﬁciency (A) and insufﬁciency (B) among the Brazilian population.
6 M. PEREIRA-SANTOS ET AL.
Brazil, which can promote differences in occurrences of vitamin
D insufﬁciency and deﬁciency in the populations of the various
states of this country.
In lower-latitude regions, closer to the equator, cutaneous
synthesis may be high due to higher temperatures and intensity
of UVB rays (Palacios and Gonzalez., 2014). This explains the
lower prevalence of vitamin D deﬁciency among the popula-
tions of the northern and northeastern regions of Brazil, where
the availability of sunshine and intensity of ultraviolet rays are
greater. Moreover, the habits and lifestyle of these populations
also favor solar exposure, especially in coastal cities, which
increases the possibility of vitamin D synthesis. Higher preva-
lences of deﬁciency were observed in the southern and south-
eastern regions, where the incidence of ultraviolet rays is lower.
However, regarding vitamin D insufﬁciency, the highest occur-
rence was observed among the populations of states in the
northeastern region. This indicates that availability of sunshine
and the presence of tropical coastal cities did not seem to inﬂu-
ence the occurrence of this problem, even if the population was
protected from greater degrees of inadequacy.
The highest prevalences of vitamin D deﬁciency were
reported from populations in the central-western, southeastern
and southern regions. Environmental factors may inﬂuence the
vitamin D levels in the populations of these regions. High levels
of atmospheric pollution are among the environmental factors
that have been reported to occur in the states of these regions
(Vormittag et al., 2014). Air pollution due to particulate matter
), sulfur dioxide (SO
), ozone (O
), carbon monoxide
(CO) and nitrogen oxides (NOx), among others, block ultravio-
let rays and decrease the possibility that individuals will be able
to photosynthesize vitamin D (Feizabad et al., 2017).
In Brazil, vitamin D intake through dietary sources is low.
The small bioavailable quantities of this vitamin in food are
insufﬁcient for the physiological needs of the human body
(Peters et al. 2009). Conversely, the availability of vitamin D in
foods in Brazil is mostly unknown. Not all food products
include information regarding vitamin D in the tables of per-
centage composition. This hinders studies evaluating vitamin
D intake and the repercussions of consuming dietary sources of
vitamin D on the serum concentrations of this compound in
any given population. Thus, quantiﬁcation of vitamin D in Bra-
zilian foods and compilation of food composition tables should
be targeted in future studies.
In a prospective study on vitamin D levels in the adult popu-
lation of a Finnish cohort, it was reported that food fortiﬁca-
tion, especially of ﬂuid milk products, gave rise to adequate
vitamin D status when vitamin D intake was based on nutri-
tional recommendations [25(OH)D 50 nmol/L]. In such sit-
uations, supplementation would generally not be needed
ainen et al., 2017).
Vitamin D and 25(OH)D concentrations were observed to
increase after public policy actions aimed towards vitamin D
fortiﬁcation in foods that are frequently consumed by the pop-
ulation (Black et al., 2012;J
ainen et al., 2017). Current
evidence does not support the recommendation of generalized
supplementation of vitamin D through medications, for popu-
lations (Peters et al., 2009). However, the results from the pres-
ent study and other studies support the recommendation that
policies towards fortiﬁcation of foods with vitamin D for the
Brazilian population should be implemented, considering that
the main source of vitamin D in Brazil consists of occasional
exposure to sunlight.
The development of vitamin D food fortiﬁcation policies in
Brazil needs to be cautious and carefully planned. The interests
of stakeholders within the pharmaceutical industry are pre-
dominantly focused towards implementation and commerciali-
zation of oral supplements and administration of mega-doses
of vitamin D among individuals in different age groups, who
may or may not be healthy, instead of promoting consumption
of foods that have been enriched with vitamin D.
Regarding the methodological quality of the studies ana-
lyzed, most presented moderate quality and moderate risk of
bias. The main problems of the studies related to absence of
sample calculations and use of non-probabilistic sampling.
Therefore, it is evident that epidemiological studies with repre-
sentative samples should be conducted on the repercussions of
vitamin D on the population’s health, considering the scarcity
of prevalence assessments on 25(OH)D
levels among speciﬁc
population groups, such as pregnant women and children. The
cutoff points available for evaluating vitamin D concentrations
are another issue to be addressed, since these limits do not take
into account climatic particularities and the physiological needs
of each age group. The most common methods used to deter-
mine vitamin D concentrations were competitive assays based
on speciﬁc antibodies and non-radioactive markers. However,
high-performance liquid chromatography (HPLC), which
requires methodological certiﬁcation through DEQAS (Inter-
national Vitamin D External Quality Assessment Scheme), was
also used. Proﬁciency in determining vitamin D concentration
was not observed in the methodologies of the published papers
that were evaluated. Thus, there may be variations in the results
regarding vitamin D concentrations in studies that do not
adopt this parameter.
The present study provides contributions towards the ﬁeld
of public health through identifying the epidemiological situa-
tion of vitamin D deﬁciency and insufﬁciency among 340,476
Brazilians, including children, adolescents, pregnant women,
adults and the elderly. Moreover, these results relating to vita-
min D deﬁciency and insufﬁciency were spatially distributed
across the country’s geopolitical regions. However, most studies
included in this meta-analysis were prevalence studies, which
did not allow evaluation of causality relationships between vita-
min D deﬁciency and insufﬁciency and associated factors. In
addition, the northern and central-western regions were only
represented by a few studies on the vitamin D situation among
their populations, which hindered precise estimation of vitamin
D occurrences and deﬁciencies.
The results from the present study indicate that there is a
need for intervention actions towards controlling vitamin D
deﬁciency. Public policies towards vitamin D fortiﬁcation in
foods consumed by the general public, except for speciﬁc
groups, should be developed. Supplementation can be reserved
for clinical practice, to assist individuals in situations of vulner-
ability to vitamin D deﬁciency, such as elderly people, pregnant
women or people for whom supplementation would have a sig-
niﬁcant impact, towards restoring and maintaining health. The
standardization of cutoff points for 25(OH)D should also be
further investigated, considering the physiological needs of
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 7
each group. According to the evidence of this study, we recom-
mend monitoring of vitamin D concentrations and comple-
mentation of daily needs through oral supplementation, when
necessary, among groups that are vulnerable to vitamin D deﬁ-
ciency (Maeda et al., 2014). Vitamin D deﬁciency and insufﬁ-
ciency should be considered to be a worldwide public health
issue, and Brazil forms part of this epidemiological scenario,
with high prevalences of deﬁcient and insufﬁcient vitamin D
levels among the country’s population.
Disclosure of potential conﬂicts of interest
No potential conﬂicts of interest were disclosed.
M. Pereira-Santos was supported by the Brazilian National Research
Marcos Pereira-Santos http://orcid.org/0000-0003-3766-2502
Arabi, A., R. El Rassi, and F. G. El-Hajj. 2010. Hypovitaminosis D in devel-
oping countries—prevalence, risk factors and outcomes. Nature
Reviews Endocrinology 6 (10):550–61. doi:10.1038/nrendo.2010.146.
Archela, R. S., and H. Th
ery. 2008. Orientation m
ethodologique pour la
construction et la lecture de cartes th
ematiques. Conﬁns 3:1–21.
Black, L. J., K. M. Seamans, K. D. Cashman, and M. Kiely. 2012. An
updated systematic review and meta-analysis of the efﬁcacy of vitamin
D food fortiﬁcation. Journal of Nutrition 142 (6):1102–8. doi:10.3945/
Cashman, K. D., K. G. Dowling, and Z. Skrabakova. 2016. Vitamin D deﬁ-
ciency in Europe: pandemic?. The American Journal of Clinical Nutri-
tion 103 (4):1033–44. doi:10.3945/ajcn.115.120873.
Feizabad, E., A. Hossein-Nezhad, Z. Maghbooli, M. Ramezani, R. Hashe-
mian, and S. Moattari. 2017. Impact of air pollution on vitamin D deﬁ-
ciency and bone health in adolescents. Archives of Osteoporosis 12
Higgins, J. P. T., and S. G. Thompson. 2002. Quantifying heterogeneity in a
meta-analysis. Statistics in Medicine 21 (11):1539–58. doi:10.1002/
Hilger, J., A. Friedel, and R. Herr. 2014. A systematic review of vitamin D
status in populations worldwide. British Journal of Nutrition 111
Holick, M. F. 2004. Sunlight and vitamin D for bone health and prevention
of autoimmune diseases, cancers, and cardiovascular disease. The
American Journal of Clinical Nutrition 80 (6 Suppl):1678S–88S.
Holick, M. F. 2007. Vitamin D deﬁciency. The New England Journal of
Medicine (357):266–81. doi:10.1056/NEJMra070553.
Holick, M. F., and T. C. Chen. 2008. Vitamin D defciency: A worldwide
problem with health consequences. The American Journal of Clinical
Nutrition 87 (4):1080S–6S.
Hoteit, M., L. Al-Shaar, C. Yazbeck, S. M. Bou, T. Ghalayini, and F. G. El-
Hajj. 2014. Hypovitaminosis D in a sunny country: Time trends, pre-
dictors, and implications for practice guidelines. Metabolism 63
Instituto Brasileiro de GeograﬁaeEstat
ıstica. 2015. Base cartogr
do brasil base cartogr
ınua do brasil –1:250000. Brasilia: IBGE.
ainen, T., S. T. Itkonen, and A. Lundqvist. 2017. 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. The American
Journal of Clinical Nutrition 105 (6):ajcn151415. doi:10.3945/ajcn.
James, W. P. T. 2008. 22nd marabou symposium: The changing faces of
vitamin D. Nutrition Reviews 66 (5):286–90. doi:10.1111/j.1753-
Loney, P. L., L. W. Chambers, K. J. Bennett, J. G. Roberts, and P. W. Strat-
ford. 1998. Critical appraisal of the health research literature: preva-
lence or incidence of a health problem. Chronic Diseases in Canada 19
Luo, B. A., F. Gao, and L. L. Qin. 2017. The association between vitamin D
deﬁciency and diabetic retinopathy in type 2 diabetes: A meta-analysis
of observational studies. Nutrients 9 (3):307. doi:10.3390/nu9030307.
Maeda, S. S., V. Z. C. Borba, M. B. R. Camargo, et al. 2014. Recommenda-
tions of the Brazilian Society of Endocrinology and Metabology
(SBEM) for the diagnosis and treatment of hypovitaminosis D. Arqui-
vos Brasileiros de Endocrinologia & Metabologia 58 (5):411–33.
Mellaby, E. 1919. An experimental investigation on rickets. The Lancet 193
Mosekilde, L. 2005. Vitamin D and the elderly. The Journal of Clinical
Endocrinology and Metabolism 62 (3):265–81.
Palacios, C., and L. Gonzalez. 2014. Is vitamin D deﬁciency a major global
public health problem? The Journal of Steroid Biochemistry and Molec-
ular Biology 144:138–45. doi:10.1016/j.jsbmb.2013.11.003.
Pereira-Santos, M., P. R. F. Costa, A. M. O. Assis, C. A. S. T. Santos,
and D. B. Santos. 2015. Obesity and vitamin D deﬁciency: A sys-
tematic review and meta-analysis. Obesity Reviews 16 (4). doi:10.
Peters, B. S. E., L. C. dos Santos, M. Fisberg, R. J. Wood, and L. A. Martini.
2009. Prevalence of vitamin D insufﬁciency in Brazilian adolescents.
Annals of Nutrition and Metabolism 54 (1):15–21. doi:10.1159/000
Pludowski, P., M. F. Holick, and S. Pilz. 2013. Vitamin D effects on muscu-
loskeletal health, immunity, autoimmunity, cardiovascular disease, can-
cer, fertility, pregnancy, dementia and mortality-A review of recent
evidence. Autoimmunity Reviews 12 (10):976–89. doi:10.1016/j.autrev.
Qi, D., X. Nie, S. Wu, and J. Cai. 2017. Vitamin D and hypertension: Pro-
spective study and meta-analysis. Slominski A. T., editor. PLOS One 12
Spritzer.2012.VitaminDdeﬁciency in girls from South Brazil: A cross-
sectional study on prevalence and association with vitamin D receptor
gene variants. BMC Pediatrics 12:62. doi:10.1186/1471-2431-12-62.
Scalc, R., M. O. Premaor, P. E. Froehlich, and T. W. Furlanetto. 2008. High
prevalence of hypovitaminosis D and secondary hyperparathyroidism
in elders living in nonproﬁt homes in South Brazil. Endocrine 33
Stroup, D. F., J. A. Berlin, and S. C. Morton. 2000. Meta-analysis of obser-
vational studies in Epidemiology. A Proposal for Reporting. JAMA 283
Tsiaras, W., and M. Weinstock. 2011. Factors inﬂuencing vitamin D status.
Acta Dermato-Venereologica 91 (2):115–24. doi:10.2340/00015555-
Unger, M. D., L. Cuppari, and S. M. Titan. 2010. Vitamin D status in a
sunny country: Where has the sun gone? The American Journal of Clin-
ical Nutrition 29 (6):784–8. doi:10.1016/j.clnu.2010.06.009.
van Schoor, N. M., and P. Lips. 2011. Worldwide vitamin D status. Best
Practice & Research Clinical Endocrinology & Metabolism 25 (4):671–
Vormittag, E. M. P. A., R. R. Costa, A. A. Braga, M. Miranda, N. C. Nasci-
mento, and P. H. N. Saldiva. 2014. Monitoramento da qualidade do ar
no Brasil. Inst Sa
ude e Sustentabilidade. 99.
Zhang, R., B. Li, and X. Gao. 2017. Serum 25-hydroxyvitamin D and the
risk of cardiovascular disease: dose-response meta-analysis of prospec-
tive studies. The American Journal of Clinical Nutrition 105 (4):810–9.
8 M. PEREIRA-SANTOS ET AL.