Zhu et al.

BMC Public Health (2021) 21:218

https://doi.org/10.1186/s12889-021-10268-z

RESEARCH ARTICLE Open Access

Anemia and associated factors among

adolescent girls and boys at 10–14 years in
rural western China
Zhonghai Zhu1,2, Christopher R. Sudfeld2, Yue Cheng3, Qi Qi1, Shaoru Li1, Mohamed Elhoumed1,4, Wenfang Yang5,
Suying Chang6, Michael J. Dibley7, Lingxia Zeng1,8* and Wafaie W. Fawzi2

Abstract
Background: Evidence on anemia and associated factors among young adolescent girls and boys in rural western
China is limited.
Methods: We used data from a follow-up study of adolescents (10–14 years) born to women who participated in a
randomized trial of antenatal micronutrient supplementation in western China. Anemia was defined by World
Health Organization standards. Logistic regression was used to examine the factors associated with anemia.
Results: The overall prevalence of anemia was 11.7% (178/1517). Female adolescents were 1.73 (95% CI 1.21, 2.48)
times more likely to have anemia as compared to males. Adolescents whose mothers had completed high school
were 0.35 (95% CI 0.13, 0.93) times less likely to be anemic, compared to those of whom had < 3 years of formal
education. Household wealth was also inversely associated with anemia. The association of puberty status with
anemia was modified by adolescent sex (P-value for interaction was 0.04); males with greater than mild pubertal
development had reduced odds (OR 0.35, 95% CI 0.15, 0.83) of anemia while there was no association among
females (OR 0.72, 95% CI 0.29, 1.78). Consumption of flesh foods (OR 0.58, 95% CI 0.38, 0.89), eggs (OR 0.60, 95% CI
0.38, 0.93), and having a meal frequency of three times or more per day (OR 0.68, 95% CI 0.48, 0.96) were also
associated with a lower likelihood of anemia.
Conclusions: Anemia was a mild public health problem among young adolescents in rural western China.
Nutritional and social determinants were identified as predictors, warranting interventions to reduce the risk of
anemia among this critical age group.
Keywords: Rural western China, Adolescent, Anemia, Associated factors, Puberty development, Dietary intake, Social
determinant

* Correspondence: [email protected]
1
Department of Epidemiology and Biostatistics, School of Public Health, Xi’an
Jiaotong University Health Science Center, No.76, Yanta West Road, Xi’an,
Shaanxi 710061, People’s Republic of China
8
Key Laboratory of Environment and Genes Related to Diseases, Xi’an
Jiaotong University, Xi’an 710061, Shaanxi, People’s Republic of China
Full list of author information is available at the end of the article

© The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License,
which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give
appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if
changes were made. The images or other third party material in this article are included in the article's Creative Commons
licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons
licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain
permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
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 in a credit line to the data.
Zhu et al. BMC Public Health (2021) 21:218 Page 2 of 14

Background adolescents may still have a high burden of anemia due

Adolescence is a critical period of growth, reproductive to the suboptimal intake of iron-rich foods, vitamins,
maturation, and developmental transitions which re- and other micronutrients [26].
quires increased nutritional intake and therefore makes In this paper, we analyzed data from adolescents in
adolescents more vulnerable to nutritional deficiencies rural western China to assess factors associated with
[1]. Nearly 90% of adolescents live in low- and middle- anemia among young adolescents aged 10–14 years.
income countries (LMICs) where undernutrition, includ- These data are intended to inform intervention strategies
ing anemia and micronutrient deficiencies, remains for this critical age group.
public health problems [2]. Anemia in adolescence may
cause a wide range of functional consequences across Methods
the life course, including reduced resistance to infection, Study design and participants
impaired physical performance and neurodevelopment, This study was a secondary analysis of data from a
and suboptimal schooling outcomes [3, 4]. follow-up study of adolescents born to women who par-
Globally, many prior studies have investigated the ticipated in a cluster-randomized controlled trial of pre-
prevalence of anemia and associated factors among ado- natal micronutrient supplementation. The details of the
lescent girls [5]; these studies suggest that, the preva- original trial and the adolescent follow-up study were
lence of anemia differs by country and region with described elsewhere [27, 28].
estimates ranging from 5.3% in high-income country set- Briefly, all pregnant women in villages from two coun-
tings to over 50.0% in some LMICs [6, 7]. Further, there ties in rural western China, where an anemia prevention
is limited data that shows on anemia prevalence among program for adolescents was not implemented, were
boys in LMICs. One study from India reported that the randomly assigned to take a daily capsule of folic acid,
prevalence of anemia among adolescent boys was as high folic acid plus iron, or multiple micronutrients between
as the prevalence among girls at 50.0 and 56.5%, respect- 2002 and 2006. A total of 4488 singleton live births were
ively [8]. In addition, studies have reported inconsistent eligible for long-term follow-up. We followed 1517 ado-
associations between pubertal development and adoles- lescents aged 10–14 years between June and December
cent anemia among boys and girls [9]. 2016. Due to the rapid urbanization, moving out of the
Multiple studies have reported that factors associ- study area was the primary reason for loss to follow-up.
ated with anemia differ for adolescents as compared Venous blood samples were collected from each par-
to adults [10, 11]. Additionally, most of growth gained ticipant by a nurse in a local hospital and were immedi-
during adolescence occurs at early adolescence (age ately tested for hemoglobin concentration using an
10–14 years) defined by the United Nations [12]. automated hematology analyzer (BC-3000Plus, mindray).
Studies have also reported that the prevalence of Anthropometric measurements were assessed, includ-
anemia may differ between early adolescents and ing height and weight. After removal of shoes and heavy
older adolescents (age 15–19 years) [7, 13, 14]. Given clothing, standing height was measured to the nearest
that iron deficiency may be a key factor for anemia 0.1 cm using a steel strip stadiometer (SZG − 210, Shang-
[15] and that early adolescence is a critical period of hai JWFU Medical Apparatus Corporation), and body
diet transition from childhood to adolescence [16, weight was measured to the nearest 0.1 kg (BC-420,
17], improving adolescent dietary intake may be a Tanita Corporation, Tokyo, Japan). Two measurements
reasonable approach to reduce anemia [18]. However, for height were performed and if a discrepancy occurred,
studies have reported inconsistent results on the asso- repeated measures were taken until consensus obtained.
ciations of eggs, dairy products, and heme-iron con- Body mass index (BMI) was calculated as body weight
taining food sources with anemia [13, 19–21]. divided by height squared (kg/m2). BMI-for- age and sex
China has experienced rapid development and Z-score (BAZ) and height-for-age and sex Z-score
urbanization in the past few decades that has corre- (HAZ) were used to classify the nutritional status as
sponded with major changes in dietary patterns [22]. Re- thinness (BAZ < -2 SD) or stunting (HAZ < -2 SD), and
cent data on Chinese adolescent anemia has been overweight (BAZ > 1 SD) or stature above average
obtained from the national survey [23–25], providing an (HAZ > 1 SD) according to the World Health
incomplete examination of the epidemiology of anemia Organization (WHO) growth standards [29].
in this age group, e.g., missing to examine influencing Puberty was assessed with the Tanner scale by rigor-
factors. The prevalence of anemia among Chinese ado- ously trained medical graduate students using standard
lescents aged 12 and 14 years was 9.6 and 8.4%, respect- procedures. The Tanner scale included five-level assess-
ively, and ranged from 2.0% in Beijing to 24.1% in ment of developmental stages of the scrotum and pubic
Hainan [25]. These results suggest that despite some im- hair in males, and breasts and pubic hair in females, re-
provement of wealth and diets in rural China, spectively [30, 31]. Specifically, among females, both
Zhu et al. BMC Public Health (2021) 21:218 Page 3 of 14

pubic hair and breasts showing the characteristics of variables were used to describe characteristics of the
Stage 1 indicated pre-puberty; both pubic hair and study population. Bivariate and multivariable logistic re-
breasts showing the characteristics of Stage 2 indicated gression models were performed to examine factors as-
mild stages of puberty; either pubic hair or breast that sociated with anemia including socioeconomic status,
showed the characteristics of Stage 3 or beyond indi- randomized regimens during pregnancy, adolescent age,
cated above mild stage (Due to scarce data, Stages 3, 4 sex, nutrition status, disease history, stages of puberty
and 5 were merged in the present study). For boys, the and dietary intake. All potential factors were considered
frequencies of Tanner stage 1 to 5 were 182 (21.0%), 420 in the multivariable analyses. We also examined influen-
(48.5%), 182 (21.0%), 78 (9.0%) and 4 (0.5%), respectively. cing factors of continuous hemoglobin concentrations
As for girls, the corresponding frequencies were 78 using generalized linear models. Finally, we performed
(12.2%), 226(35.2%), 286 (44.6%), 52 (8.1%) and 0 (0.0%), stratified analysis by sex with other variables included in
respectively. Similar cut-off criteria were applied to the the model. P values for interactions by sex were esti-
developmental stages of the scrotum and pubic hair in mated from likelihood ratio tests comparing models in-
males. cluding and excluding interaction terms between sex
A structured questionnaire was used to collect infor- and factors.
mation on socioeconomic status and adolescent disease Data were analyzed using STATA 15.0 (StataCorp,
history by a trained public health graduate student. Ado- College Station, TX, USA). A two-sided P value < 0.05
lescents’ primary caregivers provided all information ex- was considered statistically significant.
cept for dietary intake which was directly collected from
the adolescents. Adolescent dietary intake was collected Results
using a food group-based frequency questionnaire. As an A total of 1517 adolescents age 10–14 years were inter-
example, adolescents were asked “How often do you viewed, had hemoglobin data, and were included in the
consume flesh foods such as meat, poultry, and fish, i.e., analyses. Table 1 presents the socioeconomic and nutri-
almost never, almost 1/month, 1-3/month, 1/week, 2-4/ tional characteristics of the study participants. The mean
week, 5-6/week, and ≥1/day?”. Consumption of food age of adolescents, their mothers and fathers were
groups including beans, dairy products and egg, and 11.8 ± 0.9, 37.3 ± 4.4 and 39.5 ± 4.1 years, respectively. Of
meal frequency in 24 h were similarly collected. We then adolescents, 873 (57.6%) were female. Majority of
transformed these frequencies into the estimated num- mothers and fathers had secondary education or higher.
ber of times an adolescent consumed each food group More than half of mothers were farmers, but only 34.4%
per day and classified intake into tertiles within the sam- of fathers were farmers.
ple. Socioeconomic status data included parental age, The prevalence of thinness and stunting was 5.5 and
education, occupation, and household wealth. Adoles- 2.1%, respectively, while the prevalence of overweight
cent health history was defined as whether the adoles- and stature above average was 14.6 and 19.8%, respect-
cent had seen a doctor in the prior 2 weeks. Household ively. Puberty onset had occurred among 82.8% of par-
wealth index was constructed by principal component ticipants and 42.8% were in the mild stage of pubertal
analysis of 17 household assets and the ownership of development. Regarding dietary intake, the frequencies
goats, cattle, horses, and poultry, which were then cate- of high protein-based foods consumptions including
gorized into tertiles that indicated low-, middle- and flesh foods, beans, dairy products, and eggs were fairly
high-income households [32]. low and ranged from 0.23 to 0.38 times per day. Nearly
half (44.2%) of the adolescents reported to have a meal
Statistical analysis frequency of two times per day.
Anemia was defined and adjusted for age as recom-
mended by WHO [33], and was defined by hemoglobin Factors associated with adolescent anemia
concentrations < 115 g/L among adolescents 10–11 years The average hemoglobin concentration was 133.0 ± 14.9
old and < 120 g/L among adolescents 12–14 years old. In g/L. In total, age-adjusted anemia was found among 178
terms of severity, mild anemia was defined as a (11.73%) adolescents. Of these anemic adolescents, 93
hemoglobin concentration between 110 and 114 g/L (52.2%) were mild, 82 (46.1%) were moderate, while only
among adolescents aged 10–11 or between 110 and 119 3 (1.7%) were severely anemic.
g/L among adolescents aged 12–14. Moderate and se- Unadjusted and multivariable analyses of factors asso-
vere anemia was defined by a hemoglobin concentration ciated with adolescent anemia are presented in Table 2.
of between 80 and 109 g/L and lower than 80 g/L, Adolescent girls were 1.73 (95% confidence interval [CI]
respectively. 1.21, 2.48) times more likely to have anemia as com-
Counts/percentages for categorical variables and pared to boys. We also found that the odds (odds ratio
means±standard deviations (SDs) for continuous [OR] 0.50, 95% CI 0.29, 0.87) of anemia were lower in
Zhu et al. BMC Public Health (2021) 21:218 Page 4 of 14

Table 1 Background characteristics and anemia status among adolescents in rural Western China, 2016a
Parental characteristics Total/n(%) Adolescent characteristics Total/n(%)
N 1517 (100.0) N 1517 (100.0)
Maternal age/yearsb Adolescent age/years (Mean ± SD) 11.8 (0.9)
Mean (SD) 37.3 (4.4) 10 73 (4.8)
Q1: ≤35 642 (42.3) 11 538 (35.5)
Q2: 36–39 408 (26.9) 12 547 (36.1)
Q3: ≥40 467 (30.8) 13–14 (only two children were 14) 359 (23.7)
Maternal education Sex
< 3 years 87 (5.8) Male 873 (57.6)
Primary 433 (28.6) Female 644 (42.5)
Secondary 757 (50.1) Hemoglobin concentrations (g/L, Mean ± SD) 133.0 (14.9)
High school+ 235 (15.5) BMI for age z score
Maternal occupation Thinness (<−2SD) 82 (5.5)
Farmer 923 (60.9) Normal weight 1202 (80.0)
Others 593 (39.1) Overweight (>1SD) 219 (14.6)
Paternal age/yearsb Height for age z score
Mean (SD) 39.5 (4.1) Stunting (<−2SD) 32 (2.1)
Q1: ≤37 571 (37.8) -2 to 1 SD 1184 (78.1)
Q2: 38–41 477 (31.6) Stature above average (>1SD) 300 (19.8)
Q3: ≥42 464 (30.7) Whether having an illness in last 2 weeks
Paternal education Yes 590 (39.0)
< 3 years 19 (1.3) No 923 (61.0)
Primary 218 (14.4) Puberty developmentc
Secondary 894 (59.1) Pre-puberty 260 (17.2)
High school+ 383 (25.3) Mild 646 (42.8)
Paternal occupation Above mild 602 (39.9)
Farmer 507 (34.4) Times of consuming flesh foods per day/Mean (SD)d 0.23 (0.28)
Others 968 (65.6) Q1 (< 0.03) 374 (24.7)
Household wealth Q2 (0.03–0.14) 523 (34.5)
Low 525 (34.6) Q3 (> 0.14) 620 (40.9)
Medium 446 (29.4) Times of consuming beans per day/Mean (SD)d 0.29 (0.33)
High 546 (36.0) Q1 (< 0.03) 512 (34.0)
Randomized regimens Q2 (0.03–0.43) 723 (48.0)
Folic acid 529 (34.9) Q3 (> 0.43) 270 (17.9)
Iron/folic acid 502 (33.1) Times of consuming dairy products per day/Mean(SD)d 0.33 (0.38)
Multiple micronutrients 486 (32.0) Q1 (< 0.03) 583 (38.7)
Q2 (0.03–0.43) 54 (30.5)
Q3 (> 0.43) 36 (20.3)
Times of consuming egg per day/Mean(SD)d 0.38 (0.37)
Q1 (< 0.07) 526 (34.9)
Q2 (0.07–0.43) 564 (37.4)
Q3 (> 0.43) 419 (27.8)
Meal frequency in 24 h
Two times 663 (44.2)
e
Three times and four times 836 (55.8)
a
Data are missing for maternal education (n = 5), maternal occupation (n = 1), paternal education (n = 3), paternal occupation (n = 42), BMI for age z score (n = 14), height for
age z score (n = 1), high protein-based food (n = 4), puberty development (n = 9), beans (n = 2), dairy products (n = 2), egg (n = 8), and meal frequency (n = 18)
b
Parents’ age was categorized by its tertiles
c
Puberty development was defined by the Tanner stages
d
The frequency of consuming foods was converted into continuous variables, namely times per day, which were then categorized by its tertiles. Flesh foods included meat,
poultry and fish
e
Only 14 (0.9%) adolescents had a meal frequency of four times per day
Zhu et al. BMC Public Health (2021) 21:218 Page 5 of 14

adolescents who were above mild stages of puberty de- puberty was statistically associated with a reduced risk of
velopment as compared to those in pre-puberty. Com- anemia for male adolescents but not for females.
pared with adolescents whose mothers had < 3 years of The prevalence of anemia in our study population
formal education, those whose mothers completed edu- (11.7%) would be classified as a mild public health prob-
cation beyond high school were 0.35 (95% CI 0.13, 0.93) lem according to the WHO [15]. The prevalence is also
times less likely to be anemic. Adolescents from high- lower than the prevalence of 25.5% from a cross-
income households relative to those from low-income sectional survey among middle-school students con-
households were also less likely to be anemic, with an ducted in the study area conducted in 2006 [34]. Similar
adjusted OR of 0.55 (95% CI 0.34, 0.88). decrements in anemia prevalence have also been ob-
In terms of dietary intake, adolescents who were in the served in other rural areas in western China [35, 36].
highest tertile of daily egg consumption had reduced Our findings further support that the prevalence of ado-
odds (OR 0.60, 95% CI 0.38, 0.93) of having anemia as lescent anemia in China has improved to some extent.
compared to those in the lowest tertile. We also ob- Nevertheless, the prevalence of adolescent anemia in our
served that being in the middle tertile of flesh foods in- study population was still higher than that in more de-
take relative to the lowest was associated with a 0.58 veloped areas of China such as 2.9% among adolescents
(95% CI 0.38, 0.89) times lower odds of anemia. Simi- aged 12–14 in Yiwu, a city in eastern China [37], which
larly, the results for dairy products approached signifi- warrants the development of intervention strategies for
cance with an adjusted OR 0.68 (95% CI 0.46, 1.01). In adolescent anemia in rural western China.
addition, having a meal frequency of three times or more We found positive associations between being fe-
compared with two times per day was also associated male and adolescent anemia which was in line with
with reduced odds (OR 0.68, 95% CI 0.48, 0.96) of hav- previous studies conducted in Indonesia and Turkey
ing anemia. [9, 18]. This finding may be explained by the occur-
We examined the associations between factors and rence of menarche and regular blood loss [8]. Further
hemoglobin concentrations in adolescents (Table 3). differences in diet may also be a contributor. One
Similar influencing factors of anemia were identified. Be- study from India reported that adolescent girls tended
sides, mothers with the non-farmer occupation, not ex- to consume fewer protein- and vitamin-rich foods as
periencing diseases in the prior 2 weeks and stature compared with boys [38].
above average were also significantly associated with In addition, we found that males with higher stages of
higher hemoglobin concentrations in adolescents. puberty development had reduced odds of anemia while
the statistical association was not achieved among fe-
males, which was similar to the results of a study from
Stratified analysis by sex
Indonesia [9]. Some studies have reported that
In addition, we found that adolescent sex modified the
hemoglobin concentrations in adolescent boys may in-
associations of stages of puberty and maternal education
crease with the onset of puberty and may be attributed
with anemia with P values for interaction of 0.04 and
to testosterone and other androgen effects [39, 40].
0.01, respectively (Table 4). The statistical association
However, our study cannot provide evidence for the
between higher stages of puberty and reduced odds of
causality between puberty development and anemia due
anemia was achieved only among males (OR 0.35, 95%
to the cross-sectional design. Based on a cohort study in
CI 0.15, 0.83) but not among females (OR 0.72, 95% CI
Pakistan, Campisi and colleagues reported that anemia
0.29, 1.78). Similar results for continuous hemoglobin
and stunting in childhood may delay the onset of adoles-
concentrations were presented in Supplementary Table 1
cent puberty [41]. As a result, further research is needed
(see Additional file 1).
on the relationship of puberty and anemia among ado-
lescents in LMICs.
Discussion Some studies have reported that the risk of having
We found that the overall prevalence of anemia in our anemia increases with age among adolescents that may
study population of adolescents aged 10–14 years in be explained by puberty development [13], however
rural western China was 11.7%, with a prevalence of 9.7 these studies did not adjust for stages of puberty. We
and 14.4% among adolescent males and females, respect- noted significant associations between continuous ado-
ively. Multivariable analysis identified that lower mater- lescent age and anemia even after adjusting for stages of
nal education, lower household wealth, female sex, pre- puberty. Further, the models produced a condition index
puberty development, lower consumption of flesh foods, of 8.16 which indicated that the multicollinearity be-
eggs and dairy products, and lower meal frequency per tween age and stages of puberty was not a concern (data
day were associated with an increased likelihood of not shown) [42]. We noted, in the same study popula-
anemia. Moreover, we found that a higher stage of tion, that adolescent age was positively associated with
Zhu et al. BMC Public Health (2021) 21:218 Page 6 of 14

Table 2 Factors associated with adolescent anemia in rural western China, 2016 (N = 1517)
Anemia/ Unadjusted Adjusteda P
n(%) values
OR 95% CI p values OR 95% CI p values
for
trenda
Maternal age/yearsb 0.50
Q1: ≤35 84 (13.1) 1.00 1.00
Q2: 36–39 45 (11.0) 0.82 0.56, 1.21 0.32 0.85 0.53, 1.34 0.47
Q3: ≥40 49 (10.5) 0.78 0.54, 1.13 0.19 0.89 0.46, 1.69 0.71
Maternal education 0.07
< 3 years 17 (19.5) 1.00 1.00
Primary 61 (14.1) 0.68 0.37, 1.22 0.20 0.60 0.31, 1.15 0.13
Secondary 89 (11.8) 0.55 0.31, 0.97 0.04 0.56 0.28, 1.12 0.10
High school+ 11 (4.7) 0.20 0.09, 0.45 < 0.001 0.35 0.13, 0.93 0.04
Maternal occupation –
Farmer 130 (14.1) 1.00 1.00
Others 48 (8.1) 0.54 0.38, 0.76 < 0.001 0.81 0.53, 1.22 0.31
b
Paternal age/year 0.42
Q1: ≤37 71 (12.4) 1.00 1.00
Q2: 38–41 58 (12.2) 0.97 0.67, 1.41 0.89 0.97 0.62, 1.51 0.89
Q3: ≥42 49 (10.6) 0.83 0.56, 1.22 0.35 0.71 0.36, 1.43 0.34
Paternal education 0.88
< 3 years 4 (21.1) 1.00 1.00
Primary 32 (14.7) 0.65 0.20, 2.07 0.46 0.57 0.16, 1.97 0.37
Secondary 114 (12.8) 0.55 0.18, 1.68 0.29 0.63 0.18, 2.13 0.45
High school+ 28 (7.3) 0.30 0.09, 0.95 0.04 0.60 0.16, 2.20 0.44
Paternal occupation
Farmer 72 (14.2) 1.00 1.00
Others 98 (10.1) 0.68 0.49, 0.94 0.02 0.87 0.59, 1.27 0.46
Household wealth 0.78 0.63, 0.98 0.03
Low 75 (14.3) 1.00 1.00
Medium 70 (15.7) 1.12 0.78, 1.59 0.54 1.17 0.80, 1.72 0.38
High 33 (6.0) 0.39 0.25, 0.59 < 0.001 0.55 0.34, 0.88 0.01
Randomized regimens
Folic acid 55 (10.4) 1.00 1.00
Iron/folic acid 62 (12.4) 1.21 0.83, 1.79 0.32 1.10 0.72, 1.66 0.68
Multiple micronutrients 61 (12.6) 1.24 0.84, 1.82 0.28 1.27 0.85, 1.91 0.25
Adolescent age 1.28 1.02, 1.62 0.04
10 6 (8.2) 1.00 1.00
11 55 (10.2) 1.27 0.53, 3.07 0.59 1.01 0.40, 2.54 0.98
12 71 (13.0) 1.67 0.70, 3.98 0.25 1.48 0.58, 3.77 0.41
13–14 46 (12.8) 1.64 0.67, 4.00 0.28 1.64 0.61, 4.42 0.33
Sex
Male 85 (9.7) 1.00 1.00
Female 93 (14.4) 1.56 1.14, 2.14 0.01 1.73 1.21, 2.48 0.003
Height for age z score 0.23
Stunting (<−2SD) 7 (21.9) 1.89 0.80, 4.44 0.68 1.20 0.46, 3.13 0.71
Zhu et al. BMC Public Health (2021) 21:218 Page 7 of 14

Table 2 Factors associated with adolescent anemia in rural western China, 2016 (N = 1517) (Continued)
Anemia/ Unadjusted Adjusteda P
n(%) values
OR 95% CI p values OR 95% CI p values
for
trenda
-2 to 1 SD 153 (12.9) 1.00 1.00
Above average (>1SD) 18 (6.0) 0.43 0.26, 0.71 0.43 0.71 0.40, 1.24 0.22
Whether having an illness in the last 2 weeks
Yes 81 (13.7) 1.00 1.00
No 95 (10.3) 0.72 0.53, 0.99 0.04 0.76 0.54, 1.06 0.10
Puberty developmentc 0.70 0.53, 0.93 0.01
Pre-puberty 40 (15.4) 1.00 1.00
Mild 71 (11.0) 0.68 0.45, 1.03 0.07 0.64 0.40, 1.01 0.06
Above mild 66 (11.0) 0.68 0.44, 1.03 0.07 0.50 0.29, 0.87 0.01
Times of consuming flesh foods per day (Mean/SD)d 0.11
Q1 (Lowest) 62 (16.6) 1.00 1.00
Q2 54 (10.3) 0.58 0.39, 0.86 0.01 0.58 0.38, 0.89 0.01
Q3 (Highest) 62 (10.0) 0.56 0.38, 0.82 0.003 0.72 0.48, 1.08 0.11
Times of consuming beans per day (Mean/SD)d 0.48
Q1 (Lowest) 62 (12.1) 1.00 1.00
Q2 94 (13.0) 1.08 0.77, 1.53 0.64 1.20 0.83, 1.73 0.34
Q3 (Highest) 22 (8.2) 0.64 0.39, 1.07 0.09 0.75 0.44, 1.31 0.31
Times of consuming dairy products per day (Mean/SD)d 0.79 0.64, 0.99 0.04
Q1 (Lowest) 87 (14.9) 1.00 1.00
Q2 54 (10.2) 0.64 0.45, 0.93 0.02 0.68 0.46, 1.01 0.06
Q3 (Highest) 36 (9.2) 0.58 0.38, 0.88 0.01 0.70 0.45, 1.09 0.11
Times of consuming egg per day (Mean/SD)d 0.76 0.62, 0.95 0.01
Q1 (Lowest) 76 (14.5) 1.00 1.00
Q2 63 (11.2) 0.74 0.52, 1.06 0.11 0.79 0.54, 1.15 0.22
Q3 (Highest) 39 (9.3) 0.61 0.40, 0.92 0.02 0.60 0.38, 0.93 0.02
Meal frequency in 24 h –
Two times 100 (15.1) 1.00 1.00
Three times and four times 76 (9.1) 0.56 0.41, 0.77 < 0.001 0.68 0.48, 0.96 0.03
a
The adjusted model included all the variables in the table except for dietary variables. And then, each of the dietary variables were put in the adjusted model
above one at a time. The p values for trend were calculated by treating the factors as ordinal variables in the adjusted models above
b
Parents’ age was categorized by its tertiles
c
Puberty development was defined by the Tanner stages
d
The frequency of consuming foods was converted into continuous variables, namely times per day, which were then categorized by its tertiles. Flesh foods
included meat, poultry and fish

the prevalence of stunting. Taken together, these find- dairy products was associated with decreased likelihood
ings suggest that data on adolescent health outcomes of anemia. A meta-analysis of randomized controlled tri-
e.g., anemia, should be reported by sex and age and that als in China also reported that dietary interventions such
when possible stages of puberty should be factored in as consuming eggs may significantly reduce the risk of
the interpretation of findings. iron deficiency among children with iron-deficiency
Although early adolescence is a critical period in the anemia [45]. Further, a study conducted in a refugee
dietary transition from mid-childhood through adoles- camp in Ethiopia found that adolescent girls who con-
cence to adulthood [43, 44], dietary data on anemia in sumed greater heme-iron containing food sources were
this age group are limited. In the adjusted models, we less likely to be anemic [13]. Foods such as eggs, dairy
found that higher consumption of eggs, flesh foods and products, and flesh foods, are important sources of
Zhu et al. BMC Public Health (2021) 21:218 Page 8 of 14

Table 3 Factors associated with hemoglobin concentrations (g/L) in adolescents from rural Western China, 2016 (N = 1517)
Mean (SD) Unadjusted Adjusteda
Mean differences 95% CI p values Mean differences 95% CI p values
Maternal age/yearsb
Q1: ≤35 131.9 (15.1) Ref. Ref.
Q2: 36–39 133.6 (14.1) 1.65 −0.20, 3.49 0.08 0.54 −1.48, 2.56 0.60
Q3: ≥40 134.0 (15.3) 2.06 0.29, 3.83 0.02 1.63 −1.26, 4.52 0.27
Maternal education
< 3 years 129.3 (15.9) Ref. Ref.
Primary 131.9 (15.5) 2.58 −0.84, 6.00 0.14 3.28 −0.24, 6.79 0.07
Secondary 133.2 (15.1) 3.86 0.56, 7.15 0.02 4.34 0.68, 8.01 0.02
High school+ 135.8 (12.1) 6.49 2.83, 10.14 0.001 4.83 0.44, 9.22 0.03
Maternal occupation
Farmer 135.0 (14.1) Ref. Ref.
Others 131.8 (15.3) 3.21 1.68, 4.74 < 0.001 2.44 0.62, 4.27 0.01
Paternal age/yearb
Q1: ≤37 132.0 (13.9) Ref. Ref.
Q2: 38–41 133.6 (15.7) 1.60 −0.21, 3.41 0.08 0.84 −1.16, 2.83 0.41
Q3: ≥42 133.6 (15.3) 1.68 −0.14, 3.51 0.07 0.57 −2.46, 3.59 0.71
Paternal education
< 3 years 126.9 (14.9) Ref. Ref.
Primary 132.3 (16.8) 5.41 −1.56, 12.39 0.13 6.05 −0.94, 13.03 0.09
Secondary 132.6 (15.3) 5.70 −1.06, 12.46 0.10 4.98 −1.92, 11.88 0.16
High school+ 134.6 (12.6) 7.73 0.88, 14.59 0.03 4.25 −2.89, 11.39 0.24
Paternal occupation
Farmer 133.3 (13.50 Ref. Ref.
Others 132.9 (17.3) 0.39 −1.22, 1.99 0.64 −1.50 −3.29, 0.29 0.10
Household wealth
Low 131.6 (15.7) Ref. Ref.
Medium 131.9 (16.0) 0.37 −1.50, 2.24 0.70 0.46 −1.42, 2.35 0.63
High 135.3 (12.8) 3.71 1.93, 5.48 < 0.001 2.52 0.53, 4.50 0.01
Randomized regimens
Folic acid 133.4 (15.7) Ref. Ref.
Iron/folic acid 132.4 (13.7) −0.96 −2.78, 0.86 0.30 −0.78 −2.59, 1.02 0.39
Multiple micronutrients 133.2 (15.3) −0.22 −2.05, 1.62 0.82 −0.28 −2.10, 1.53 0.76
Adolescent age
10 131.2 (14.5) Ref. Ref.
11 131.0 (15.7) −0.24 −3.86, 3.81 0.90 1.08 −2.59, 4.76 0.56
12 134.0 (13.9) 2.75 −0.87, 6.37 0.14 3.05 −0.74, 6.84 0.12
13–14 135.0 (14.7) 3.78 0.06, 7.51 0.05 3.01 −1.05, 7.08 0.15
Sex
Male 134.9 (15.2) Ref. Ref.
Female 130.5 (14.1) −4.44 −5.94, −2.94 < 0.001 −5.04 −6.64, −3.45 < 0.001
Height for age z score
Stunting (<−2SD) 127.7 (13.5) −4.31 −9.48, 0.86 0.10 −2.21 −7.51, 3.08 0.41
-2 to 1 SD 132.0 (14.5) Ref. Ref.
Zhu et al. BMC Public Health (2021) 21:218 Page 9 of 14

Table 3 Factors associated with hemoglobin concentrations (g/L) in adolescents from rural Western China, 2016 (N = 1517)
(Continued)
Mean (SD) Unadjusted Adjusteda
Mean differences 95% CI p values Mean differences 95% CI p values
Above average (>1SD) 137.6 (15.7) 5.55 3.69, 7.42 < 0.001 2.96 0.92, 5.00 0.004
Whether having an illness in the last 2 weeks
Yes 131.7 (13.8) Ref. Ref.
No 133.9 (15.4) 2.14 0.61, 3.67 0.01 1.58 0.06, 3.10 0.04
c
Puberty development
Pre-puberty 128.7 (14.2) Ref. Ref.
Mild 132.7 (14.8) 3.95 1.83, 6.08 < 0.001 3.26 1.07, 5.44 0.004
Above mild 135.3 (14.9) 6.53 4.39, 8.68 < 0.001 5.83 3.27, 8.39 < 0.001
d
Times of consuming flesh foods per day (Mean/SD)
Q1 (Lowest) 130.9 (15.7) Ref. Ref.
Q2 133.7 (13.5) 2.79 0.82, 4.76 0.01 2.28 0.31, 4.24 0.02
Q3 (Highest) 133.7 (15.4) 2.73 0.83, 4.64 0.01 1.31 −0.64, 3.26 0.19
Times of consuming beans per day (Mean/SD)d
Q1 (Lowest) 132.7 (15.3) Ref. Ref.
Q2 132.6 (14.9) −0.04 −1.73, 1.64 0.96 −0.87 −2.56, 0.82 0.31
Q3 (Highest) 134.3 (14.2) 1.64 0.56, 3.84 0.14 0.27 −1.93, 2.47 0.81
d
Times of consuming dairy products per day (Mean/SD)
Q1 (Lowest) 132.2 (15.6) Ref. Ref.
Q2 133.3 (14.2) 1.13 −0.62, 2.87 0.21 0.81 −0.94, 2.56 0.37
Q3 (Highest) 133.7 (14.7) 1.50 −0.41, 3.41 0.12 0.85 −1.06, 2.77 0.38
d
Times of consuming egg per day (Mean/SD)
Q1 (Lowest) 131.3 (14.2) Ref. Ref.
Q2 134.2 (14.7) 2.86 1.14, 4.58 0.001 2.64 0.94, 4.34 0.002
Q3 (Highest) 133.4 (14.5) 2.11 0.25, 3.97 0.03 2.45 0.62, 4.29 0.01
Meal frequency in 24 h
Two times 131.1 (15.7) Ref. Ref.
Three times and four times 134.5 (14.1) 3.39 1.88, 4.90 < 0.001 1.69 0.13, 3.25 0.03
a
The adjusted model included all the variables in the table except for dietary variables. And then, each of the dietary variables were put in the adjusted model
above one at a time
b
Parents’ age was categorized by its tertiles
c
Puberty development was defined by the Tanner stages
d
The frequency of consuming foods was converted into continuous variables namely times per day, which were then categorized by its tertiles. Flesh foods
included meat, poultry and fish

protein, vitamin B12, bioavailable iron and other micro- was associated with reduced risk of anemia. We
nutrients which may influence the risk of anemia [46]. hypothesize that adolescents who eat three or more
In a recent study, it was estimated that less than half of meals per day have a higher likelihood of meeting their
adolescent girls consumed dairy products, flesh foods, or nutritional requirements. It is notable that in our study
eggs on a daily basis (41, 46, and 19%, respectively) in population majority of participants only had two meals
LMICs [47]. WHO recommends weekly iron and folic per day and ate the first meal at noon and skipped the
acid supplementation for menstruating adolescent girls breakfast. One study from China reported that skipping
in settings with 20% or higher levels of anemia preva- breakfast was associated with higher risk of stunting,
lence [32], however whether this public health program wasting and malnutrition among children aged 6–17
should be extended to adolescent boys remains unclear. years [49].
In agreement with a study in southern Ethiopia [48], we In addition to biomedical influences, we also found
found that consuming three or more of meals per day that adolescents from higher-income households and
Zhu et al. BMC Public Health (2021) 21:218 Page 10 of 14

Table 4 Factors associated with adolescent anemia stratified by adolescent sex in rural western China, 2016
Factors Male Female P values
for
No. (%) of Adjusted 95% CI No. (%) of Adjusted 95% CI
interaction
anemia ORa anemia ORa
between
sex and
factorsb
Maternal age/yearsc 0.01
Q1: ≤35 37 (11.4) 1.00 47 (14.9) 1.00
Q2: 36–39 15 (6.3) 0.54 0.24, 30 (17.9) 1.17 0.64,
1.19 2.14
Q3: ≥40 33 (10.8) 1.01 0.39, 16 (10.0) 0.54 0.20,
2.64 1.48
Maternal education 0.01
< 3 years 5 (8.9) 1.96 0.37, 12 (38.7) 5.84 1.51,
10.44 22.54
Primary 39 (14.5) 3.14 0.87, 22 (13.4) 0.90 0.30,
11.31 2.70
Secondary 37 (8.9) 1.89 0.56, 52 (15.3) 1.44 0.56,
6.38 3.68
High school+ 4 (3.1) 1.00 7 (6.7) 1.00
Maternal occupation 0.61
Farmer 25 (7.2) 1.00 23 (9.4) 1.00
Others 60 (11.4) 1.03 0.56, 70 (17.6) 0.58 0.33,
1.90 1.06
Paternal age/yearsc 0.35
Q1: ≤37 30 (10.5) 1.00 41 (14.3) 1.00
Q2: 38–41 23 (8.4) 0.83 0.40, 35 (17.2) 1.02 0.57,
1.74 1.84
Q3: ≥42 32 (10.3) 0.73 0.26, 17 (11.2) 0.72 0.25 2.04
2.07
Paternal education 0.48
< 3 years 2 (16.7) 1.00 2 (28.6) 1.00
Primary 14 (10.1) 0.32 0.05, 18 (22.5) 0.79 0.12,
2.01 5.38
Secondary 55 (10.6) 0.52 0.09, 59 (15.7) 0.56 0.08,
3.18 3.68
High school+ 14 (6.9) 0.68 0.10, 14 (7.8) 0.43 0.06,
4.65 3.17
Paternal occupation 0.09
Farmer 42 (7.6) 1.00 56 (13.6) 1.00
Others 40 (13.5) 0.58 0.33, 32 (15.2) 1.22 0.71,
1.02 2.12
Household wealth 0.40
Low 39 (12.6) 1.00 36 (16.7) 1.00
Medium 30 (11.7) 0.92 0.53, 40 (21.2) 1.62 0.93,
1.61 2.85
High 16 (5.2) 0.56 0.27, 17 (7.1) 0.61 0.30,
1.13 1.23
Randomized regimens 0.09
Folic acid 30 (9.6) 1.00 25 (11.6) 1.00
Iron/folic acid 32 (11.4) 1.16 0.65, 30 (13.6) 1.10 0.59,
2.07 2.08
Zhu et al. BMC Public Health (2021) 21:218 Page 11 of 14

Table 4 Factors associated with adolescent anemia stratified by adolescent sex in rural western China, 2016 (Continued)
Factors Male Female P values
for
No. (%) of Adjusted 95% CI No. (%) of Adjusted 95% CI
interaction
anemia ORa anemia ORa
between
sex and
factorsb
Multiple micronutrients 23 (8.2) 0.95 0.52, 38 (18.4) 1.92 1.06,
1.73 3.47
Adolescent age 0.39
10 4 (11.1) 1.00 2 (5.4) 1.00
11 31 (10.0) 0.62 0.19, 24 (10.5) 1.61 0.35,
2.02 7.54
12 29 (9.4) 0.75 0.23, 42 (17.6) 2.53 0.53,
2.51 12.08
13–14 21 (9.6) 0.97 0.27, 25 (17.9) 2.67 0.52,
3.52 13.79
Height for age z score 0.56
Stunting (<−2SD) 3 (18.8) 1.00 0.20, 4 (25.0) 2.14 0.58,
5.06 7.94
-2 to 1 SD 73 (11.1) 1.00 80 (15.2) 1.00
Above average (>1SD) 9 (4.5) 0.57 0.25, 9 (9.1) 0.98 0.43,
1.30 2.23
Whether having an illness in the last 2 weeks 0.27
Yes 41 (12.2) 1.00 40 (15.8) 1.00
No 42 (7.9) 0.63 0.39, 53 (13.6) 0.92 0.56,
1.02 1.50
Puberty developmentd 0.04
Pre-puberty 29 (15.9) 1.00 11 (14.1) 1.00
Mild 40 (9.5) 0.59 0.33, 31 (13.7) 0.92 0.39,
1.06 2.13
Above mild 15 (5.7) 0.35 0.15, 51 (15.1) 0.72 0.29,
0.83 1.78
Times of consuming flesh foods per day (Mean/ 0.57
SD)e
Q1 (Lowest) 31 (14.8) 1.00 31 (18.8) 1.00
Q2 22 (7.9) 0.51 0.27, 32 (13.2) 0.68 0.37,
0.97 1.24
Q3 (Highest) 32 (8.3) 0.58 0.33, 30 (12.7) 0.96 0.52,
1.05 1.77
Times of consuming beans per day (Mean/SD)e 0.36
Q1 (Lowest) 33 (11.5) 1.00 29 (12.9) 1.00
Q2 42 (10.2) 0.94 0.56, 52 (16.7) 1.55 0.90,
1.59 2.68
Q3 (Highest) 10 (6.0) 0.54 0.24, 12 (11.8) 0.94 0.42,
1.21 2.10
Times of consuming dairy products per day 0.63
(Mean/SD)e
Q1 (Lowest) 44 (13.0) 1.00 43 (17.6) 1.00
Q2 20 (6.9) 0.56 0.31, 34 (14.1) 0.80 0.46,
1.02 1.39
Q3 (Highest) 20 (8.4) 0.70 0.38, 16 (10.5) 0.67 0.34,
1.28 1.32
Times of consuming egg per day (Mean/SD)e 0.22
Zhu et al. BMC Public Health (2021) 21:218 Page 12 of 14

Table 4 Factors associated with adolescent anemia stratified by adolescent sex in rural western China, 2016 (Continued)
Factors Male Female P values
for
No. (%) of Adjusted 95% CI No. (%) of Adjusted 95% CI
interaction
anemia ORa anemia ORa
between
sex and
factorsb
Q1 (Lowest) 34 (11.6) 1.00 42 (18.1) 1.00
Q2 26 (8.3) 0.70 0.39, 37 (14.9) 0.87 0.52,
1.26 1.51
Q3 (Highest) 25 (9.5) 0.75 0.41, 14 (8.9) 0.46 0.23,
1.37 0.92
Meal frequency in 24 h 0.80
Two times 50 (13.1) 1.52 0.91, 50 (17.7) 1.53 0.91,
2.54 2.55
Three times and four times 33 (6.9) 1.00 43 (12.1) 1.00
a
The adjusted model included all the variables in the table except for dietary variables. And then, each of the dietary variables were put in the adjusted model
above one at a time
b
The p values for interaction between sex and factors were calculated using likelihood-ratio test between including interaction terms and not including in
the models
c
Parents’ age was categorized by its tertiles
d
Puberty development was defined by the Tanner stages
e
The frequency of consuming foods was converted into continuous variables namely times per day, which were then categorized by its tertiles. Flesh foods
included meat, poultry and fish

those whose mothers had higher education levels were on high protein-, vitamin-, and mineral-based foods, but
less likely to be anemic, which is in accord with other some studies reported that regular consumptions of
studies [6, 11]. Individuals from higher socioeconomic fruits and green leafy vegetables were also associated
status may have access and consume more iron- and with reduced odds of having anemia in adolescent girls
vitamin C-rich foods [6]. Tur and colleagues also re- [52], and this is an area to pursue in further research. In
ported that maternal education was positively associated addition, other factors associated with anemia were not
with the quality of dietary intake of mineral and vitamin accounted in the present study such as parasite infec-
among adolescents [50]. In the stratified analysis by sex, tion, thalassemia, and maternal nutritional status. Fi-
the association of maternal education with adolescent nally, owing to the cross-sectional study design,
anemia was only found in females but not in males. We prospective studies are needed to verify the relationships
hypothesize that adolescent girls are more likely to fol- between factors and anemia among adolescents. Our re-
low maternal advice on healthy behaviors as compared sults, however, provide evidence that can help develop
to boys [51]; however, this hypothesis warrants further intervention strategies and target at high-risk adolescent
study. Therefore, programmes that only emphasize bio- populations.
medical factors might not be sufficient to prevent ado-
lescent anemia. Conclusions
This study has limitations that should be noted. First, We found that anemia was a mild public health problem
the study included adolescents who were born to women among young adolescents aged 10–14 years in rural
who participated in an antenatal micronutrient supple- western China. Integrated interventions that address bio-
mentation trial, and this population may not be a truly medical determinants and targeting high-risk popula-
representative sample of our target population of adoles- tions of adolescents may be essential to reduce the risk
cents in rural western China. Our prior data had shown of anemia and improve health among this critical age
that the background characteristics between participants group.
followed and those lost to follow-up at adolescence were
balanced [28]. Besides, the cluster-randomized trial in-
cluded all pregnant women in villages, representing the Supplementary Information
The online version contains supplementary material available at https://doi.
community to some extent. Second, due to no available org/10.1186/s12889-021-10268-z.
data on older adolescents in the present study, we were
not able to compare the potential differences in the Additional file 1: Supplementary Table 1. Factors associated with
prevalence and risk factors for anemia between early adolescent hemoglobin concentrations (g/L) stratified by adolescent sex
in rural western China, 2016.
adolescence and older adolescence. Third, we focused
Zhu et al. BMC Public Health (2021) 21:218 Page 13 of 14

Abbreviations 4. Pivina L, Semenova Y, Doşa MD, Dauletyarova M, Bjørklund G. Iron

BAZ: Body mass index for-age z score; CI: Confidence interval; HAZ: Height deficiency, cognitive functions, and neurobehavioral disorders in children. J
for-age z score; Hb: Hemoglobin; LMICs: Low- and middle-income countries; Mol Neurosci. 2019;68(1):1–10.
OR: Odds ratio; SD: Standard deviation; WHO: World Health Organization 5. Bhutta ZA, Lassi ZS, Bergeron G, Koletzko B, Salam R, Diaz A, McLean M,
Black RE, De-Regil LM, Christian P, et al. Delivering an action agenda for
Acknowledgements nutrition interventions addressing adolescent girls and young women:
We thank all the adolescents and their families. We are also grateful for the priorities for implementation and research. Ann N Y Acad Sci. 2017;1393(1):
support of the local hospital, government, and education bureau. 61–71.
6. Kim JY, Shin S, Han K, Lee KC, Kim JH, Choi YS, Kim DH, Nam GE, Yeo HD,
Authors’ contributions Lee HG, et al. Relationship between socioeconomic status and anemia
ZZ, YC, WY, SC, MJD, and LZ: planned and designed the study; ZZ, YC, QQ, prevalence in adolescent girls based on the fourth and fifth Korea National
SL and ME: conducted the study; ZZ, CS, and WF: analyzed data and Health and nutrition examination surveys. Eur J Clin Nutr. 2014;68(2):253–8.
interpreted results; ZZ: wrote the paper; LZ: had primary responsibility for 7. Mistry SK, Jhohura FT, Khanam F, Akter F, Khan S, Yunus FM, Hossain MB,
final content; and all authors: reviewed, revised, and approved the final Afsana K, Haque MR, Rahman M. An outline of anemia among adolescent
manuscript. girls in Bangladesh: findings from a cross-sectional study. BMC Hematol.
2017;17:13.
8. Sabale R, Kowli S, Chowdary P. Prevalence of anemia and its determinants
Funding
in urban school-going children of Mumbai. Int J Med Public Health. 2013;
The field work was supported by the National Natural Science Foundation of
3(4):325–9.
China (grant number 81872633, Lingxia Zeng) and the National Key Research
9. Soekarjo DD, de Pee S, Bloem MW, Tjiong R, Yip R, Schreurs WH. Muhilal.
and Development Program of China (grant number 2017YFC0907200 and
Socio-economic status and puberty are the main factors determining
2017YFC0907201, NA). This study was also supported by the China
anaemia in adolescent girls and boys in East Java, Indonesia. Eur J Clin Nutr.
Scholarship Council (grant number 201806280188, Zhonghai Zhu). The
2001;55(11):932–9.
funders had no role in the design, conduct, analysis, interpretation of the
10. Sekhar DL, Murray-Kolb LE, Kunselman AR, Weisman CS, Paul IM. Differences
results, writing, or decision to submit the manuscript for publication.
in risk factors for anemia between adolescent and adult women. J Women's
Health. 2016;25(5):505–13.
Availability of data and materials 11. Mchiza ZJ, Parker W, Sewpaul R, Job N, Chola L, Mutyambizi C, Sithole M,
All data generated or analyzed during this study are included in this Stokes A, Labadarios D. Understanding the determinants of hemoglobin
published article and its supplementary information files. Data sharing is and iron status: adolescent-adult women comparisons in SANHANES-1. Ann
available from the corresponding author on reasonable request. N Y Acad Sci. 2018;1416(1):31–47.
12. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition
Ethics approval and consent to participate for child overweight and obesity worldwide: international survey. BMJ. 2000;
This study was conducted according to the principles laid down in the 320(7244):1240–3.
Declaration of Helsinki, and approved by the Ethics Committee of Xi’an 13. Engidaw MT, Wassie MM, Teferra AS. Anemia and associated factors among
Jiaotong University. Written and verbal informed consent were obtained adolescent girls living in aw-Barre refugee camp, Somali regional state,
from each parent or legal guardian of participants and adolescent, Southeast Ethiopia. PLoS One. 2018;13(10):e205381.
respectively. 14. Teji K, Dessie Y, Assebe T, Abdo M. Anaemia and nutritional status of
adolescent girls in Babile District, Eastern Ethiopia. Pan Afr Med J. 2016;
Consent for publication 24:62.
Not applicable. 15. WHO. Iron deficiency anaemia: Assessment, prevention, and control-A guide
for programme managers. Geneva: WHO Press; 2001.
Competing interests 16. Haerens L, Vereecken C, Maes L, De Bourdeaudhuij I. Relationship of
Dr. Suying Chang is a nutrition specialist at the UNICEF China Office. The physical activity and dietary habits with body mass index in the transition
other authors declare that they have no competing interests. from childhood to adolescence: a 4-year longitudinal study. Public Health
Nutr. 2010;13(10A):1722–8.
Author details 17. Biazzi Leal D, Altenburg De Assis MA, Hinnig PDF, Schmitt J, Soares Lobo A,
1
Department of Epidemiology and Biostatistics, School of Public Health, Xi’an Bellisle F, Di Pietro P, Vieira F, de Moura Araujo PH, de Andrade D. Changes
Jiaotong University Health Science Center, No.76, Yanta West Road, Xi’an, in dietary patterns from childhood to adolescence and associated body
Shaanxi 710061, People’s Republic of China. 2Department of Global Health adiposity status. Nutrients. 2017;9(10):1098.
and Population, Harvard T.H. Chan School of Public Health, Boston, MA, USA. 18. Isik BY, Karabulut A, Gurses D, Ethem CI. Prevalence and risk factors of anemia
3
Department of Nutrition and Food Safety Research, School of Public Health, among adolescents in Denizli, Turkey. Iran J Pediatr. 2012;22(1):77–81.
Xi’an Jiaotong University Health Science Center, Xi’an, Shaanxi 710061, 19. Gonete KA, Tariku A, Wami SD, Derso T. Prevalence and associated factors
People’s Republic of China. 4National Institute of Public Health Research of anemia among adolescent girls attending high schools in Dembia
(INRSP), BP. 695, Nouakchott, Mauritania. 5Department of Obstetrics and District, Northwest Ethiopia, 2017. Arch Public Health. 2018;76:79.
Gynecology, The First Affiliated Hospital of Xi’an Jiaotong University, Xi’an, 20. Tesfaye M, Yemane T, Adisu W, Asres Y, Gedefaw L. Anemia and iron
Shaanxi, People’s Republic of China. 6United Nations Children’s Fund, China deficiency among school adolescents: burden, severity, and determinant
Office, Beijing 100600, People’s Republic of China. 7The Sydney School of factors in Southwest Ethiopia. Adolesc Health Med Ther. 2015;6:189–96.
Public Health, Faculty of Medicine, The University of Sydney, Sydney, NSW, 21. Ahankari AS, Myles PR, Fogarty AW, Dixit JV, Tata LJ. Prevalence of iron-
Australia. 8Key Laboratory of Environment and Genes Related to Diseases, deficiency anaemia and risk factors in 1010 adolescent girls from rural
Xi’an Jiaotong University, Xi’an 710061, Shaanxi, People’s Republic of China. Maharashtra, India: a cross-sectional survey. Public Health. 2017;142:159–66.
22. Chang JL, Wang Y. Chinese nutrition and health surveillance-the
Received: 10 May 2020 Accepted: 19 January 2021 comprehensive report between 2010–2013. Beijing: Peking University
Medical Press; 2016.
23. Li M, Hu Y, Mao D, Wang R, Chen J, Li W, Yang X, Piao J, Yang L. Prevalence
References of Anemia among Chinese rural residents. Nutrients. 2017;9(3):192.
1. Fuhrmann D, Knoll LJ, Blakemore S. Adolescence as a sensitive period of 24. Wu J, Hu Y, Li M, Chen J, Mao D, Li W, Wang R, Yang Y, Piao J, Yang L, et al.
brain development. Trends Cogn Sci. 2015;19(10):558–66. Prevalence of anemia in Chinese children and adolescents and its
2. Christian P, Smith ER. Adolescent undernutrition: global burden, physiology, associated factors. Int J Env Res Pubic Health. 2019;16(8):1416.
and nutritional risks. Ann Nutr Metab. 2018;72(4):316–28. 25. Luo D, Yan X, Hu P, Zhang J, Lei Y, Song Y, Ma J. Subnational disparity of
3. Beard JL. Iron biology in immune function, muscle metabolism and anemia among Chinese Han students aged 7-14 years in 2014. Chin J Sch
neuronal functioning. J Nutr. 2001;131(2S-2):568S–80S. Health. 2019;40(6):878–84.
Zhu et al. BMC Public Health (2021) 21:218 Page 14 of 14

26. Luo D, Xu R, Ma J, Yan X, Hu P, Song Y, Jan C, Raat H, Patton GC. The 49. Li L, Xu P, Yang T, Gan Q, Cao W, Pan H, Xu J, Hu X, Zhang Q. Relationship
associations of economic growth and anaemia for school-aged children in between breakfast and nutrition status study of children aged 6-17 in China
China. Matern Child Nutr. 2020;16(2):e12936. from 2010 to 2012. Wei Sheng Yan Jiu. 2018;47(3):373–7.
27. Zeng L, Dibley MJ, Cheng Y, Dang S, Chang S, Kong L, Yan H. Impact of 50. Tur JA, Puig MS, Benito E, Pons A. Associations between sociodemographic
micronutrient supplementation during pregnancy on birth weight, duration and lifestyle factors and dietary quality among adolescents in Palma de
of gestation, and perinatal mortality in rural western China: double blind Mallorca. Nutrition. 2004;20(6):502–8.
cluster randomised controlled trial. BMJ. 2008;337:a2001. 51. Greene A, Michelle GD. Age and gender differences in adolescents’
28. Zhu Z, Cheng Y, Zeng L, Elhoumed M, He G, Li W, Zhang M, Li W, Li D, preferences for parental advice: Mum's the word. J Adolesc Res. 1990;34(4):
Tsegaye S, et al. Association of antenatal micronutrient supplementation 351–80.
with adolescent intellectual development in rural western China. JAMA 52. Ghatpande NS, Apte PP, Naik SS, Kulkarni PP. Fruit and vegetable
Pediatr. 2018;172(9):832. consumption and their association with the indicators of iron and
29. Butte NF, Garza C, de Onis M. Evaluation of the feasibility of international inflammation status among adolescent girls. J Am Coll Nutr.
growth standards for school-aged children and adolescents. J Nutr. 2007; 2019;38(3):218–26.
137(1):153–7.
30. Marshall WA, Tanner JM. Variations in pattern of pubertal changes in girls.
Arch Dis Child. 1969;44(235):291–303.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in
31. Marshall WA, Tanner JM. Variations in the pattern of pubertal changes in
published maps and institutional affiliations.
boys. Arch Dis Child. 1970;45(239):13–23.
32. Filmer D, Pritchett LH. Estimating wealth effects without expenditure data--
or tears: an application to educational enrollments in states of India.
Demography. 2001;38(1):115–32.
33. World Health Organization. Haemoglobin concentrations for the diagnosis
of anaemia and assessment of severity. Geneva: WHO Press; 2011.
34. Wang X, Wang Z, Xiao S, Zhang J, Lei Y. Analysis of anaemic causes among
students of two middle schools in Guanzhong rural regions of Shaanxi
province. J Xi’an Jiaotong Univ (Med Sci). 2008;29(5):530–4.
35. Luo R, Kleiman-Weiner M, Rozelle S, Zhang L, Liu C, Sharbono B, Shi Y, Yue
A, Martorell R, Lee M. Anemia in rural China's elementary schools:
prevalence and correlates in Shaanxi Province's poor counties. Ecol Food
Nutr. 2010;49(5):357–72.
36. Luo R, Zhang L, Liu C, Zhao Q, Shi Y, Miller G, Yue E, Sharbono B, Medina A,
Rozelle S, et al. Anaemia among students of rural China's elementary
schools: prevalence and correlates in Ningxia and Qinghai's poor counties. J
Health Popul Nutr. 2011;29(5):471–85.
37. Sun C, Zhang R, Zhu Z, Dong X, Huang K, Huang L, Chen L, Li Y, Zong S.
Nutritional status among children and adolescents in Yiwu. Prev Med. 2018;
30(6):609–11.
38. Aurino E. Do boys eat better than girls in India? Longitudinal evidence on
dietary diversity and food consumption disparities among children and
adolescents. Econ Hum Biol. 2017;25:99–111.
39. Thomsen K, Riis B, Krabbe S, Christiansen C. Testosterone regulates the
haemoglobin concentration in male puberty. Acta Paediatr Scand. 1986;
75(5):793–6.
40. Hero M, Wickman S, Hanhijarvi R, Siimes MA, Dunkel L. Pubertal
upregulation of erythropoiesis in boys is determined primarily by androgen.
J Pediatr. 2005;146(2):245–52.
41. Campisi SC, Humayun KN, Rizvi A, Lou W, Söder O, Bhutta ZA. Later puberty
onset among chronically undernourished adolescents living in a Karachi
slum, Pakistan. Acta Paediatr. 2019;109(5):1019–25.
42. Belsley DA. Conditioning diagnostics, collinearity and weak data in
regression. New York: Wiley; 1991.
43. Winpenny EM, van Sluijs EMF, White M, Klepp K, Wold B, Lien N. Changes in
diet through adolescence and early adulthood: longitudinal trajectories and
association with key life transitions. Int J Behav Nutr Phy. 2018;15(1):86.
44. Winpenny EM, Penney TL, Corder K, White M, van Sluijs EMF. Change in diet
in the period from adolescence to early adulthood: a systematic scoping
review of longitudinal studies. Int J Behav Nutr Phy. 2017;14(1):60.
45. Sun J, Zhang L, Cui J, Li S, Lu H, Zhang Y, Li H, Sun J, Baloch Z. Effect of
dietary intervention treatment on children with iron deficiency anemia in
China: a meta-analysis. Lipids Health Dis. 2018;17(1):108.
46. USAID. Conceptual frameworks for anemia. 2013. https://www.spring-
nutrition.org/sites/default/files/events/multisectoral_anemia_meeting_
diagrams.pdf. Accessed 8 July 2019.
47. Keats EC, Aviva IR, Reena J, Christina OH, Shailja S, Zulfiqar AB. Diet and
eating practices among adolescent girls in low- and middle-income
countries: A systemic review. Arlington: Strengthening Partnerships, Results,
and Innovations in Nutrition Globally (SPRING) project; 2018.
48. Shaka MF, Wondimagegne YA. Anemia, a moderate public health concern
among adolescents in South Ethiopia. PLoS One. 2018;13(7):e191467.