Global Health Action

ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/zgha20

Effect of homemade peanut oil consumption

during pregnancy on low birth weight and preterm
birth outcomes: a cohort study in Southwestern
China

Yanxu Zhong, Huan Lu, Yuyan Jiang, Minyan Rong, Xiangming Zhang &
Tippawan Liabsuetrakul

To cite this article: Yanxu Zhong, Huan Lu, Yuyan Jiang, Minyan Rong, Xiangming Zhang &
Tippawan Liabsuetrakul (2024) Effect of homemade peanut oil consumption during pregnancy
on low birth weight and preterm birth outcomes: a cohort study in Southwestern China, Global
Health Action, 17:1, 2336312, DOI: 10.1080/16549716.2024.2336312

To link to this article: https://doi.org/10.1080/16549716.2024.2336312

© 2024 The Author(s). Published by Informa

UK Limited, trading as Taylor & Francis
Group.

Published online: 17 Apr 2024.

Submit your article to this journal

Article views: 48

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=zgha20
 GLOBAL HEALTH ACTION
2024, VOL. 17, 2336312
https://doi.org/10.1080/16549716.2024.2336312

RESEARCH ARTICLE

Effect of homemade peanut oil consumption during pregnancy on low birth

weight and preterm birth outcomes: a cohort study in Southwestern China
Yanxu Zhong a,b, Huan Luc, Yuyan Jianga, Minyan Rongd, Xiangming Zhange
and Tippawan Liabsuetrakul b
a
Food Safety Monitoring and Evaluation Department, Guangxi Zhuang Autonomous Region Centre for Disease Control and Prevention
(Guangxi CDC), Nanning, China; bDepartment of Epidemiology, Faculty of Medicine, Prince of Songkla University, Hat Yai, Songkhla,
Thailand; cInfectious Diseases Department, The Fourth People’s Hospital of Nanning, Nanning, China; dInfectious Disease Control and
Prevention Department, Guiping Center for Disease Control and Prevention (Guiping CDC), Guigang, China; eMaternity Department,
Guiping People’s Hospital, Guigang, China

ABSTRACT ARTICLE HISTORY

Background: Homemade peanut oil is widely consumed in rural areas of Southwestern China, Received 30 August 2023
which is easily contaminated by aflatoxins (AFs) and associated with adverse birth outcomes. Accepted 25 March 2024
Objective: To identify the effect of exposure to homemade peanut oil consumption on low
RESPONSIBLE EDITOR
birth weight (LBW), preterm birth (PB) and other associated factors. Julia Schröders
Methods: A prospective cohort study was conducted among pregnant women in Guangxi
province, Southwestern China. Information of all eligible women on homemade peanut oil KEYWORDS
consumption and potential factors associated with LBW and PB was collected, and all were Aflatoxins; homemade
followed up until delivery. The effect of homemade peanut oil exposure was analyzed using peanut oil; low birth weight;
multiple logistic regression models using the directed acyclic graph (DAG) approach. preterm birth; cohort study;
logistic regression
Results: Of 1611 pregnant women, 1316 (81.7%) had consumed homemade peanut oil, and
the rates of LBW and PB were 9.7% and 10.0%, respectively. Increased risks of LBW and PB in
women with homemade peanut oil consumption were found with aORs of 1.9 (95% CI 1.1–
3.2) and 1.8 (95% CI 1.1–3.0), respectively. Women with a history of PB or LBW were 3–5 times
more likely to have higher rates of LBW or PB compared with those without this type of
history. The odds of PB were approximately double in those taking medicine during preg­
nancy. Advanced maternal age, lack of physical exercise during pregnancy, passive smoking,
or pregnancy complications were also more likely to have a higher risk of LBW.
Conclusions: Homemade peanut oil consumption was a potential risk factor for both LBW
and PB, of which health authorities who are responsible for food safety of the country should
pay more attention to providing recommendation for oil consumption during pregnancy.

PAPER CONTEXT
● Main findings: Homemade peanut oil consumption was associated with increased risk of low
birth weight and preterm birth, in addition to advanced age, adverse obstetric histories, and
health risk behaviors during pregnancy in a county in Southwestern China.
● Added knowledge: This study identifies the direct and total effects of homemade peanut
oil consumption on low birth weight and preterm birth and explains the factors associated
with low birth weight and preterm birth in a county in Southwestern China.
● Global health impact for policy and action: Evidence of associated risk factors for low
birth weight and preterm birth should be informed to the community, and precautionary
policies for the protection of aflatoxin exposure during pregnancy are needed.

Background mills, which can be contaminated by AFs. Homemade

Aflatoxins (AFs) are commonly found in contaminated
staple foods such as maize, nuts, peanuts, corn, and oils,
leading to chronic exposure in humans [1] which are
shown to be associated with carcinogenicity and geno­
toxicity [2]. Maternal immunity and fetal growth can also
be impaired when pregnant women are exposed to AFs
[3–5]. Homemade peanut oil is made naturally by pea­
nut-squeezing machines in households or small-scale oil
peanut oil has a lower price and a special taste and has
been the main cooking oil in the southwestern rural area
of China [6,7], such as rural areas of Guangxi province;
however, the rate of homemade peanut oil use has not
been locally and nationally studied. A previous study in
Guangxi reported that aflatoxin B1 (AFB1), the most
toxic and frequent type of AF, was detected in 78.1% of
oil samples [8]. Health risks of cancer from dietary

CONTACT Tippawan Liabsuetrakul tippawan.l@psu.ac.th Department of Epidemiology, Faculty of Medicine, Prince of Songkla University, Hat
Yai, Songkhla 90110, Thailand
© 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the
posting of the Accepted Manuscript in a repository by the author(s) or with their consent.
2 Y. ZHONG ET AL.
exposure AFs in China, not fetal growth effects, were only
studied [8,9], although a study conducted in Gambia
showed the association of AFs and fetal weights [10].
The World Health Organization defines low birth
weight (LBW) as a birth weight below 2500 g regard­
less of gestational age, and preterm birth (PB) as
neonate born before 37 completed weeks of gestation
[11]. The rates of LBW and PB were globally estimated
at 14.6% of live births in 2015 [12] and 10.6% of live
births in 2014 [13], respectively, of which the majority
occurred in low-income and middle-income countries.
LBW is an important factor of neonatal mortality and
morbidity and is closely associated with short and
long-term consequences [14]. Likewise, the PB was
the second most common cause of death in children
aged under 5 years in the world [15]. A systematic
review, involving the studies mostly from the United
States, identified considerable long-term economic
impact of low birth weight and preterm birth [16].
A previous study in Guiping county, Guangxi in
Southwestern China found higher prevalence of LBW
(10.4%) and PB (11.1%) [17] compared to the national
average rates of LBW (6.1%) and PB (5.7%) in China
[18]. Most residents living in Guiping county consume
homemade peanut oil. However, there have been no
studies to date examining whether LBW or PB are
related to the consumption of homemade peanut oil.
Based on the high consumption of homemade peanut
oil and higher prevalence of LBW and PB in Guiping
country, our study aimed to examine the effect of con­
sumption of homemade peanut oil during pregnancy
on LBW and PB adjusted for confounders and explore
other independent risk factors for LBW and PB.
Materials and methods
Study design and study setting
A prospective cohort study was conducted in Guiping
county, Southwestern China, during December 2021-
Figure 1. Study flow diagram of the cohort study.
May 2022. Pregnant women of at least 28 weeks of gesta­
tion living in Guiping county who come for antenatal
care and planned to give birth at the study hospital were
included. A gestational age of 28 weeks was used based on
the definition of live birth in accordance with the defini­
tion of stillbirth defined from the World Health
Organization [19]. In this county, a county hospital
where both services of antenatal and labor/childbirth
care are available with the highest records of delivery,
approximately 4000 each year, was chosen to be the study
hospital. Those with multiple pregnancies and those hav­
ing known medical diseases, such as hypertension, dia­
betes, autoimmune diseases, and neurological disorders
before pregnancy due to well-known risk of LBW and PB,
or verbal communication disorders, such as dumbness,
deafness, or mental retardation, were excluded.
Study sample size and participants
We used a formula for the difference in the two
proportions of LBW birth outcomes between home­
made peanut oil and non-homemade peanut oil con­
sumption to be at 5% (10.5% vs 5.5%). Because the
percentage of residents in Guiping who consumed
homemade peanut nut oil was 80%, we set the ratio
of exposure and non-exposure to homemade peanut
oil at 4:1. Given a type I error of 5%, power of 80%
and a ratio of 4, at least 291 non-homemade and 1164
homemade peanut oil participants were required.
Figure 1 shows the flow diagram for study recruit­
ment. A total of 1804 participants met the inclusion
criteria during the study period, of which 126 (7.0%)
were excluded due to having unknown medical dis­
eases before pregnancy and 67 women did not come
to deliver at the setting hospital as planned (3.7%).
Finally, the data of 1611 participants were included in
the analysis, of whom 1316 (81.7%) were in the
homemade peanut oil group and 295 (18.3%) were
in the non-homemade peanut oil group.

Variables
The main outcomes of this study were LBW and PB,
based on the WHO definitions. Low birth weight was
defined as a newborn weight below 2500 g (yes, no).
Preterm birth was defined as a baby born alive before
37 complete weeks of gestation but after 28 weeks of
gestation (yes, no). The main exposure of interest was
consumption of either homemade or non-homemade
peanut oil. Since there is no standard threshold for
recommended consumption of homemade peanut oil
available, in cases where a woman consumed both types
of oil (n = 30, 1.9%), we used the consumption at least
60% of the time to be grouped into homemade peanut oil.
The demographic characteristics included age group,
ethnicity, residency, education level, current occupation,
and family income. Age groups were calculated from age
(years) and categorized into <20, 20–34, ≥35 [18].
Ethnicity was categorized into Han, Zhuang, and others.
Residency was categorized as urban and rural. Education
level was categorized as middle school and below, high
school, college and above. Current occupation was cate­
gorized as government staff, farmer or worker, self-
employed, and others. Family income was converted
from Chinese Yuan (CNY) into United States Dollars
(USD) and categorized as ≤450 USD/month, 450–750
USD/month, 750–1500 USD/month, >1500 USD/
month according to the average income of Guangxi
residents in 2020 [20].
The behavior and obstetric information included phy­
sical exercise during pregnancy, drinking history, smok­
ing history, passive smoking, history of PB, history of
LBW, history of abortion, history of hepatitis B virus
(HBV), parity, body mass index (BMI) before pregnancy
(pre-pregnancy BMI), pregnancy complications, folic
supplementation during pregnancy, calcium supplemen­
tation during pregnancy, iron supplementation during
pregnancy, vitamin D supplementation during preg­
nancy, and medicine intake during pregnancy. All the
variables were categorized into yes and no, except passive
smoking, parity, and pre-pregnancy BMI. Passive smok­
ing was categorized into no, workplace or home, or work­
place and home, and parity was categorized into 1, 2 and
≥3. Pre-pregnancy BMIs were calculated by pre-
pregnancy weight (kg) divided by height in meters
squared and categorized into underweight (<18.5 kg/
m2), normal (18.5–22.9 kg/m2), and overweight or
obese (≥23 kg/m2) [21].
The birth outcome variables included birth gestation
age (weeks), mode of delivery, newborn sex, and weight
of newborn (g). The birth gestation age was determined
according to the mother’s last menstrual cycle, with ultra­
sound used to estimate the gestational age if the men­
strual date was uncertain. Modes of delivery were
categorized into vaginal delivery, cesarean section or
assisted vaginal delivery. Newborn sex was categorized
as male and female. The weight of the newborn was the
GLOBAL HEALTH ACTION 3
weight when the newborn was delivered. All of these
variables were collected from the Guangxi maternal and
child healthcare system.
Data collection
All eligible pregnant women were invited to participate in
the study, and after signing the information sheet and
consent form, they were interviewed using a structured
questionnaire and a pregnant woman’s health and
antenatal care (ANC) questionnaire. We followed the
study of pregnant women until they delivered and col­
lected the newborn-related factors and examination data
from the Guangxi maternal and child healthcare system.
The questionnaires were checked by investigators by
days, and all unconfirmed or missing data were recol­
lected the next day or double-checked through Guangxi
maternal and child healthcare system.
Statistical analysis
All analyses were conducted using R software version
4.1.2 (The R Foundation for Statistical Computing 2020,
Vienna, Austria). The characteristics of the participants
for continuous data were descriptively analyzed using
means ± standard deviations or median, with interquar­
tile range depending on the normal distribution of data or
percentages for the categorical data. A directed acyclic
graph (DAG) diagram was used to represent the causal
relationships between homemade peanut oil consump­
tion and LBW and PB outcomes based on direct and total
effects [22] regarding the associated factors obtained
from a previous systematic review [23] and additional
exploratory factors from a univariate analysis of our
study. The direct effect estimates of the causal pathways
considered only confounders, not mediators, while the
total effect estimates considered both confounders and
mediators [24]. The factors used to construct DAGs were
the main exposures and outcomes as well as the factors
with a p value lower than 0.05 in univariate analysis. The
effects of homemade peanut oil consumption on LBW
and PB in DAGs were analyzed using multiple logistic
regression and presented in terms of crude odds ratios
(cOR) and adjusted odds ratios (aOR) with 95% confi­
dence intervals (CIs) in both direct and total effects. Apart
from the DAG models, other independent risk factors
were also analyzed using a multiple logistic regression
model considering factors with a p value less than 0.2 in
univariate analysis included in the first model with
a backward stepwise approach using the lowest Akaike
Information Criterion (AIC) value. A two-sided p < 0.05
was considered statistically significant.
Results
Their demographic characteristics, divided by con­
sumption or not of using peanut oil, are shown in
4 Y. ZHONG ET AL.

Table 1. Demographic characteristics of study women who currently consumed and did not consume homemade peanut oil.
Variable Total (N = 1611) Homemade peanut oil (N = 1316) Non-homemade peanut oil (N = 295) p
Age group 0.070
<20 64 (4.0) 56 (4.3) 8 (2.7)
20–34 1229 (76.3) 989 (75.2) 240 (81.4)
35 or above 318 (19.7) 271 (20.6) 47 (15.9)
Ethnicity 0.939
Han 1458 (90.5) 1190 (90.4) 268 (90.9)
Zhuang 42 (2.6) 92 (7.0) 19 (6.4)
Others 111 (6.9) 34 (2.6) 8 (2.7)
Residency 0.002
Urban 149 (9.2) 107 (8.1) 42 (14.2)
Rural 1462 (90.8) 1209 (91.9) 253 (85.8)
Education 0.003
Middle school and below 844 (52.4) 713 (54.2) 131 (44.4)
High school 590 (36.6) 471 (35.8) 119 (40.3)
College and above 177 (11.0) 132 (10.0) 45 (15.3)
Occupation <0.001
Government staff 230 (14.3) 175 (13.3) 55 (18.6)
Famer or worker 711 (44.1) 627 (47.6) 84 (28.5)
Self-employed 246 (15.3) 197 (15.0) 49 (16.6)
Others 424 (26.3) 317 (24.1) 107 (36.3)
Family income (USD/month) <0.001
<450 196 (12.2) 173 (13.2) 23 (7.8)
450–750 383 (23.8) 333 (25.3) 50 (16.9)
751–1500 803 (49.8) 624 (47.4) 179 (60.7)
>1500 229 (14.2) 186 (14.1) 43 (14.6)

Table 1. Those living in rural areas, education of
middle school or below, working as farmer or worker,
and family income less than 450 US$/month were
significantly higher in the homemade peanut oil
group than in the non-homemade peanut oil group.
Behavioral factors and obstetric information of the
women between homemade peanut oil and non-
homemade peanut oil groups are presented in
Table 2. Most behavioral factors and obstetric infor­
mation between the two groups were not significantly
different, except for parity, pre-pregnancy BMI, and
folic supplementation, for which primiparous
women, women with pre-pregnancy BMI > 23, and
those who had received folic supplementation were
more likely to be in the homemade peanut oil group.
The birth outcomes of study women who con­
sumed or did not consume homemade peanut oil
are presented in Table 3. Of 1611 newborns, the
average gestational age at birth was 38.2 ± 1.8 weeks
in the homemade peanut oil group and 38.5 ± 1.6
weeks in the non-homemade peanut oil group.
Mode of delivery, newborn sex and average birth
weight were not significantly different between the
two groups. A significantly higher rate of LBW was
found in the homemade peanut oil group, at 10.6% vs
5.8% in the non-homemade peanut oil group (p =
0.015). The rate of PB was 10.7% in the homemade
peanut oil group compared with 6.8% in the non-
homemade peanut oil group, but the difference was
not significant.
All factors associated with homemade peanut oil as
the main exposure of interest and the LBW and PB
outcomes are shown in the DAG diagrams in
Figure 2(a) and (b), respectively. The direct and total
effects of homemade peanut oil consumption on LBW
and PB are shown in Table 4. After adjusting for
confounders under the DAG model, the odds of hav­
ing LBW (aORtotal 1.8, 95% CI 1.1–3.1 and aORdirect
1.9, 95% CI 1.1–3.2) and PB (aORtotal 1.7, 95% CI 1.0–
2.8 and aORdirect 1.8, 95% CI 1.1–3.0) were higher in
women who consumed homemade peanut oil in both
the direct and total effect models, respectively.
Other independent risk factors for LBW and PB
analyzed using multiple logistic regression models are
shown in Figures 3 and 4. Apart from consumption
of homemade peanut oil, age over 35 years, having no
physical exercise during pregnancy, exposure to pas­
sive smoking in both workplace and home, having
a history of PB or LBW, having pregnancy complica­
tions, and female newborn were significantly asso­
ciated with LBW (Figure 3). The odds of having PB
were higher in women aged over 35 years, urban
residency, having a history of PB or LBW, and med­
icine intake during pregnancy in addition to consum­
ing homemade peanut oil (Figure 4).
Discussion
The rates of LBW and PB in pregnant women who
consumed homemade peanut oil during pregnancy
were almost double compared to those who consumed
non-homemade peanut oil in our study in a county in
Southwestern China. Advanced maternal age and his­
tory of LBW or PB were associated with higher rates of
both LBW and PB. No physical exercise during preg­
nancy, passive smoking, pregnancy complications, and
female newborn were associated with LBW, while
urban residency and medicine intake during preg­
nancy were additionally associated with risk for PB.
To our knowledge, there have been no other studies
to date examining the association between risk of LBW
or PB and consumption of homemade peanut oil as in
 GLOBAL HEALTH ACTION 5

Table 2. Behavioral factors and obstetric information of study women who consumed homemade peanut oil and not.
Variable Total (N = 1611) Homemade peanut oil (N = 1316) Non-homemade peanut oil (N = 295) p
Behavioral factors
Physical exercise during pregnancy 0.372
Yes 1373 (85.2) 1127 (85.6) 246 (83.4)
No 238 (14.8) 189 (14.4) 49 (16.6)
Drinking 1.000
Yes 15 (0.9) 12 (0.9) 3 (1.0)
No 1596 (99.1) 1304 (99.1) 292 (99.0)
Smoking 0.127
Yes 13 (0.8) 8 (0.6) 5 (1.7)
No 1598 (99.2) 1308 (99.4) 290 (98.3)
Passive smoking 0.861
No 645 (40.0) 527 (40.0) 118 (40.0)
Work place or home 710 (44.1) 577 (43.9) 133 (45.1)
Work place and home 256 (15.9) 212 (16.1) 44 (14.9)
Obstetric information
History of PB 0.432
Yes 82 (5.1) 56 (4.3) 9 (3.1)
No 1529 (94.9) 1260 (95.7) 286 (96.9)
History of LBW 0.721
Yes 46 (2.9) 39 (3.0) 7 (2.4)
No 1565 (97.1) 1277 (97.0) 288 (97.6)
History of abortion 0.399
Yes 264 (16.4) 221 (16.8) 43 (14.6)
No 1347 (83.6) 1095 (83.2) 252 (85.4)
History of HBV 0.056
Yes 82 (5.1) 74 (5.6) 8 (2.7)
No 1529 (94.9) 1242 (94.4) 287 (97.3)
Parity <0.001
1 695 (43.2) 158 (53.6) 537 (40.8)
2 558 (34.6) 96 (32.5) 462 (35.1)
≥3 358 (22.2) 41 (13.9) 317 (24.1)
Pre-pregnancy BMI (kg/m2) 0.017
Underweight (<18.5) 317 (19.7) 256 (19.5) 61 (20.7)
Normal (18.5–22.9) 980 (60.8) 786 (59.7) 194 (65.8)
Overweight or obese (≥23) 314 (19.5) 274 (20.8) 40 (13.5)
Pregnancy complications 0.636
No 1350 (83.8) 1106 (84.0) 244 (82.7)
Yes 261 (16.2) 210 (16.0) 51 (17.3)
Folic supplementation 0.013
Yes 1471 (91.3) 1213 (92.2) 258 (87.5)
No 140 (8.7) 103 (7.8) 37 (12.5)
Calcium supplementation 0.491
Yes 1261 (78.2) 1035 (78.6) 226 (76.6)
No 350 (21.8) 281 (21.4) 69 (23.4)
Iron supplementation 0.110
Yes 609 (37.8) 510 (38.8) 99 (33.6)
No 1002 (62.2) 806 (61.2) 196 (66.4)
Vitamin D supplementation 0.562
No 1463 (90.8) 1192 (90.6) 271 (91.9)
Yes 148 (9.2) 124 (9.4) 24 (8.1)
Medicine intake 0.360
No 1410 (87.5) 1157 (87.9) 253 (85.8)
Yes 201 (12.5) 159 (12.1) 42 (14.2)
Abbreviations: PB, preterm birth; LBW, low birth weight; BMI, body mass index; HBV, hepatitis B virus.

Table 3. Birth outcomes of study women who currently consumed and did not consume homemade peanut oil.
Variable Total (N = 1611) Homemade peanut oil (N = 1316) Non-homemade peanut oil (N = 295) p
Birth gestation age (weeks) 38.3 ± 1.8 38.2 ± 1.8 38.5 ± 1.6 0.005
Mode of delivery 0.065
Vaginal delivery 1127 (70.0) 907 (68.9) 220 (74.6)
Cesarean section or 484 (30.0) 409 (31.1) 75 (25.4)
assisted vaginal delivery
Newborn sex 0.377
Male 832 (51.6) 687 (52.2) 145 (49.2)
Female 779 (48.4) 629 (47.8) 150 (50.8)
Birth weight (gm) 3071.8 ± 508.2 3068.8 ± 518.1 3085.1 ± 462.2 0.593
Low birth weight 0.015
Yes 157 (9.7) 140 (10.6) 17 (5.8)
No 1454 (90.3) 1176 (89.4) 278 (94.2)
Preterm birth 0.054
Yes 161 (10.0) 141 (10.7) 20 (6.8)
No 1450 (90.0) 1175 (89.3) 275 (93.2)
6 Y. ZHONG ET AL.

Figure 2. The DAGs showing the related confounders of homemade peanut oil and LBW (a) and PB (b).
DSGs showing different relationship pathways of homemade peanut oil consumption with low birth weight (LBW) or preterm birth (PB) are
adjusted by other variables (created in a browser-based environment, DAGitty: https://www.dagitty.net). In the DAGs, each circle represents
a node of one variable (blue circle with vertical line = main outcome; green circle with triangle = main exposure; blank blue circle = mediating
circle or ancestor of outcome; blank orange circle = confounding or biasing variable). The green line with arrows indicates the only
unidirectional causal pathway that comes from main exposure to main outcome; orange and pink lines with arrows indicate unidirectional
causal pathways. DAG = directed acyclic graph.

Table 4. Direct and total effects of exposure to homemade peanut oil consumption on LBW and PB.
Main outcome Variable Adjusted variables Effect cOR aOR (95% CI)
LBW Consumption of homemade peanut oil Age group, cooking oil consumption duration, Direct 1.9 (1.2 - 3.3) 1.9 (1.1 - 3.2)
education, occupation
Age group, education, occupation, residency, Total 1.8 (1.1 - 3.1)
family income
PB Consumption of homemade peanut oil Age group, cooking oil consumption duration, Direct 1.6 (1.0 - 2.7) 1.8 (1.1 - 3.0)
education, occupation, residency
Age group, education, occupation, residency, Total 1.7 (1.0 - 2.8)
family income
Abbreviations: LBW, low birth weight; PB, preterm birth; cOR, crude odds ratio; aOR, adjusted odds ratio; CI, confidence interval.
 GLOBAL HEALTH ACTION 7

Figure 3. Factors associated with low birth weight using a multiple logistic regression model.

Figure 4. Factors associated with preterm using a multiple logistic regression model.

our study. A previous study reported lower birth weight
and gestational age at birth in Guangxi, where con­
sumption of homemade peanut oil is prevalent, than
in other areas having low homemade peanut oil con­
sumption, which implied that homemade peanut oil
consumption during pregnancy may impact the birth
weight and gestation age at birth [25]. We found
a similar effect of homemade peanut oil consumption
on LBW and PB regardless of confounding factors or
mediating adjustments that were age group, education,
occupation, residency, and family income, indicating
that homemade peanut oil consumption was almost
certainly related to the occurrence of LBW and PB.
Scientific support of this finding can be explained
8 Y. ZHONG ET AL.
through the contamination of AFs in homemade pea­
nut oil, as two previous studies conducted in Guangxi
reported high AFs in homemade peanut oil [8,9].
Guangxi is one of the provinces for homemade peanut
oil contamination by AFB1 in China [26], with the
finding double the AFB1 contamination in peanut sam­
ples than peanut oil samples. Another previous study
found that fetuses were at a higher risk of AFs than
adults when exposed to AFs [27].
Apart from homemade peanut oil consumption,
it was not surprising that maternal age, complica­
tions during pregnancy, medication intake, new­
born sex, parity, and residency were associated
with LBW and PB in our study. Increasing LBW
and PB in advanced maternal age was consistent
with previous studies for LBW [28–30] and for PB
[31–34]. However, our finding was different from
a study from China in which no significant asso­
ciation between maternal age and LBW was found
[17]. It was not surprising that women with
a history of PB or LBW had higher odds of having
LBW and PB, which was consistent with previous
studies [30,32,35,36]. This may be explained by
chronic exposure to unknown factors for LBW
and PB.
We found that LBW was higher in pregnant women
who had further complications or were taking medica­
tion during pregnancy, as was also found in previous
studies [28,35,37]. Likewise, female newborns had sig­
nificantly higher odds of having LBW than males [17],
and there were lower odds of having LBW in women
with higher parity [38]. The odds of having PB in
women living in an urban area were higher than in
women in a rural area, which was opposite to the
results of a recent systematic review study [35]. This
could be due to other unknown factors which are
different between rural and urban areas.
Those who reported no physical exercise during
pregnancy had higher odds of having an LBW infant,
which was consistent with a previous study in China
[39] but different from the conclusion of a systematic
review [40], which might be because the characteris­
tics of women and exercise measurement were not
the same. The association between passive smoking
and LBW in our study was consistent with a previous
study [39].
Our study highlights the relationship between
homemade peanut oil consumption and LBW and
PB using a prospective cohort study with a relatively
large sample. Even though there have been many
studies examining AF exposure, including diet and
food exposure [41–46], we were unable to find any
publications examining the effect of homemade pea­
nut oil consumption on pregnant women and their
newborn LBW and PB outcomes. In addition, our
study not only used DAGs to adjust for confounders
and mediators but also explored independent, related
factors, which can provide more information for
governments to control the risk of AFs exposure.
There were some limitations in the study. First,
the levels of oil consumption during pregnancy
were obtained through a questionnaire, which
may have had recall bias. However, this bias
would be minimal since the duration of time fol­
lowing the consumption ranged from 7 to 10
months and the lifestyles of the study group tend
to be static, particularly concerning things such as
the use of cooking oils. Second, the histories of
behavioral factors relied on women’s self-reports.
Third, we did not collect the cause of PB, therefore
we could not identify whether homemade peanut
oil consumption can be significantly related to
spontaneous or iatrogenic cause of PB. Fourth,
the sample size was calculated for the outcome
based on exposure to homemade peanut oil, not
other independent factors such as socioeconomic
or pre-pregnancy BMI. Fifth, our study did not
assess the dietary intake for calculating calorie
and nutrient assessments during pregnancy.
Finally, the levels of AFs in homemade peanut oil
samples and homemade peanut oil consumption
were not presented in this cohort study; however,
their findings from a subset of the study were
measured and submitted elsewhere.
Conclusions
Homemade peanut oil consumption was associated
with increased risk of LBW and PB, in addition to
advanced age, adverse obstetric histories, and health
risk behaviors during pregnancy in a county in
Southwestern China. Governments should pay more
attention to the consumption of homemade peanut
oil during pregnancy. Close monitoring and appro­
priate counseling are required for high-risk pregnant
women with LBW and PB. The hypothesis that AFs
in homemade peanut oil are associated with LBW
and PB should be tested.
Acknowledgments
This study was supported by the staffs of Guangxi CDC,
Guiping CDC and Guiping People's Hospital. We acknowl­
edge their great work in data collection and data manage­
ment. We thank Mr Dave Patterson of the International
Affairs Office, Faculty of Medicine in Prince of Songkla
University for his English editing.
Author contributions
Conceptualization: Yanxu Zhong, Tippawan Liabsuetrakul,
Huan Lu
Data curation: Yuyan Jiang, Huan Lu, Xingming Zhang,
Minyan Rong
Data analysis: Yanxu Zhong, Tippawan Liabsuetrakul

Methodology: Yanxu Zhong, Tippawan Liabsuetrakul
Supervision: Tippawan Liabsuetrakul, Yuyan Jiang
Visualization: Yanxu Zhong
Writing – original draft: Yanxu Zhong, Tippawan
Liabsuetrakul
Writing – review & editing: Tippawan Liabsuetrakul
Disclosure statement
No potential conflict of interest was reported by the
author(s).
Ethics and consent
The study was reviewed and approved by Human Research
Ethics Committee, Faculty of Medicine, Prince of Songkla
University (REC.64-404-18-1) in full compliance with
International Guideline for human research subject such
as Deceleration of Helsinki, Belmont Report, CIOMS
Guideline and the International Conference on
Harmonization in Good Clinical Practice (ICH-GCP). All
eligible pregnant women were invited to participate in the
study, and the informed consent was obtained from all
subjects before data collection.
Funding information
This study was supported by the Guangxi Medical and
Health Appropriate Technology Development and
Promotion Application Project [Grant number S2022047]
and a scholarship from the TUYF Charitable Trust:
Research Capacity through Education and Networking on
Epidemiology in Asia, the Department of Epidemiology,
Faculty of Medicine, Prince of Songkla University [Grant
number 5/2022].
Paper context
Homemade peanut oil is commonly contaminated by afla­
toxin in Southwestern China. However, there were few
studies on adverse pregnancy outcomes in pregnant
women who consumed homemade peanut oil during preg­
nancy. Our study found that homemade peanut oil con­
sumption was one of the significant risk factors of low birth
weight and preterm birth. The findings of associated risk
factors should be informed to the community, and precau­
tionary policies for the protection of aflatoxin exposure
during pregnancy are needed.
ORCID
Yanxu Zhong http://orcid.org/0000-0002-5428-4198
Tippawan Liabsuetrakul http://orcid.org/0000-0001-
7687-5629
References
[1] Williams JH, Phillips TD, Jolly PE, Stiles JK, Jolly CM,
Aggarwal D. Human aflatoxicosis in developing coun­
tries: a review of toxicology, exposure, potential health
consequences, and interventions. Am J Clin Nutr.
2004;80:1106–1122. doi: 10.1093/ajcn/80.5.1106
GLOBAL HEALTH ACTION 9
[2] Benford D, Leblanc JC, Setzer RW. Application of the
margin of exposure (MoE) approach to substances in
food that are genotoxic and carcinogenic: example:
aflatoxin B1 (AFB1). Food Chem Toxicol. 2010;48:
S34–41. doi: 10.1016/j.fct.2009.10.037
[3] Khlangwiset P, Shephard GS, Wu F. Aflatoxins and
growth impairment: a review. Crit Rev Toxicol.
2011;41:740–755. doi: 10.3109/10408444.2011.575766
[4] Jiang Y, Jolly PE, Ellis WO, Wang JS, Phillips TD,
Williams JH. Aflatoxin B1 albumin adduct levels and
cellular immune status in Ghanaians. Int Immunol.
2005;17:807–814. doi: 10.1093/intimm/dxh262
[5] Smith LE, Prendergast AJ, Turner PC, Humphrey JH,
Stoltzfus RJ. Aflatoxin exposure during pregnancy, mater­
nal anemia, and adverse birth outcomes. Am J Trop Med
Hyg. 2017;96:770–776. doi: 10.4269/ajtmh.16-0730
[6] Li FQ, Yoshizawa T, Kawamura O, Luo XY, Li YW.
Aflatoxins and fumonisins in corn from the
high-incidence area for human hepatocellular carcinoma
in Guangxi, China. J Agric Food Chem.
2001;49:4122–4126. doi: 10.1021/jf010143k
[7] Woo CSJ, El-Nezami H. Mycotoxins in Asia: is China
in danger? Quality assurance and safety of crops &
foods. Qual Assur Saf Crops Food. 2015;7:3–25. doi:
10.3920/QAS2014.x005
[8] Cheng HY, Zhong YX, Chen J, Meng HY, Liao YH,
Chen H, et al. Dietary exposure assessment of aflatoxin
B1 in edible vegetable oil consumed by Guangxi residents.
Applied Prev Med. 2017;23:451–454. in Chinese
[9] Shi MM, Liang J, Zhao P, Zhong YX, Jiang YY. Dietary
exposure assessment of aflatoxins of residents in Guangxi.
Chin J Food Hyg. 2020;32: 432–436. in Chinese.
[10] Turner PC, Collinson AC, Cheung YB, Gong YY, Hall AJ,
Prentice AM, et al. Aflatoxin exposure in utero causes
growth faltering in Gambian infants. Int J Epidemiol.
2007;36:1119–1125. doi: 10.1093/ije/dym122
[11] World Health Organization. WHO recommendations
for care of the preterm or low-birth-weight infant.
Geneva: WHO; 2022.
[12] Blencowe H, Krasevec J, de Onis M, Black RE, An X,
Stevens GA, et al. National, regional, and worldwide
estimates of low birthweight in 2015, with trends from
2000: a systematic analysis. Lancet Glob Health. 2019;7:
e849–e860. doi: 10.1016/S2214-109X(18)30565-5
[13] Chawanpaiboon S, Vogel JP, Moller A-B, Lumbiganon P,
Petzold M, Hogan D, et al. Global, regional, and national
estimates of levels of preterm birth in 2014: a systematic
review and modelling analysis. Lancet Glob Health.
2019;7:e37–e46. doi: 10.1016/S2214-109X(18)30451-0
[14] McCormick MC. The contribution of low birth weight to
infant mortality and childhood morbidity. N Engl J Med.
1985;312:82–90. doi: 10.1056/NEJM198501103120204
[15] Liu L, Johnson HL, Cousens S, Perin J, Scott S, Lawn JE,
et al. Global, regional, and national causes of child
mortality: an updated systematic analysis for 2010 with
time trends since 2000. Lancet. 2012;379:2151–2161.
doi: 10.1016/S0140-6736(12)60560-1
[16] Petrou S, Sach T, Davidson L. The long-term costs of
preterm birth and low birth weight: results of
a systematic review. Child Care Health Dev. 2001
Mar;27:97–115. doi: 10.1046/j.1365-2214.2001.00203.x
[17] Zhang XM, Zhao P, Chen SC, Li KJ, Chen LP, Xin YJ,
et al. Analysis of related factors of adverse birth outcomes
of living newborns from 2016 to 2020 in Guiping county,
China. Chin J Dis Control Prev. 2022;26:1050–1056. in
Chinese
10 Y. ZHONG ET AL.
[18] Chen Y, Li G, Ruan Y, Zou L, Wang X, Zhang W. An
epidemiological survey on low birth weight infants in
China and analysis of outcomes of full-term low birth
weight infants. BMC Pregnancy Childbirth.
2013;13:242. doi: 10.1186/1471-2393-13-242
[19] Cousens S, Blencowe H, Stanton C, Chou D, Ahmed S,
Steinhardt L, et al. National, regional, and worldwide
estimates of stillbirth rates in 2009 with trends since
1995: a systematic analysis. Lancet. 2011;377:1319–1330.
doi: 10.1016/S0140-6736(10)62310-0
[20] Guangxi Zhuang Autonomous Region Statistics
Bureau. Guangxi statistical yearbook; 2021. [cited
2023 Oct 23]. Available from: http://tjj.gxzf.gov.cn//
tjsj/tjnj/material/tjnj20200415/2021/zk/indexch.htm
[21] World Health Organization. Regional office for the
Western Pacific. The Asia-Pacific perspective: redefin­
ing obesity and its treatment. Manila: WHO Western
Pacific Region; 2000.
[22] Textor J, van der Zander B, Gilthorpe MS, Liskiewicz M,
Ellison GT. Robust causal inference using directed acyclic
graphs: the R package “dagitty”. Int J Epidemiol.
2016;45:1887–1894. doi: 10.1093/ije/dyw341
[23] Kyei NNA, Boakye D, Gabrysch S. Maternal mycotoxin
exposure and adverse pregnancy outcomes: a systematic
review. Mycotoxin Res. 2020;36:243–255. doi: 10.1007/
s12550-019-00384-6
[24] Tennant PWG, Murray EJ, Arnold KF, Berrie L, Fox MP,
Gadd SC, et al. Use of directed acyclic graphs (DAGs) to
identify confounders in applied health research: review
and recommendations. Int J Epidemiol. 2021;50:620–632.
doi: 10.1093/ije/dyaa213
[25] Lei L, Liu S, Ye Y, Qiu X, Huang D, Pan D, et al.
Associations between serum aflatoxin-B1 and anemia
in pregnant women: evidence from Guangxi Zhuang
birth cohort in China. Toxins (Basel). 2021;13:806.
doi: 10.3390/toxins13110806
[26] Yang BL, Zhang XJ, Wang G, Zhang CX, Geng HR,
Li L. Investigation of aflatoxin B1 and cyclopiazonic
acid in bulk peanut oil in China. China Oils and Fats.
2020;45: 34–37. in Chinese.
[27] da Silva JVB, de Oliveira CAF, Ramalho LNZ. Effects
of prenatal exposure to aflatoxin B1: a review.
Molecules. 2021;26:7312. doi: 10.3390/
molecules26237312
[28] Çam HH, Harunoğulları M, Polat Y. A study of low
birth weight prevalence and risk factors among new­
borns in a public-hospital at Kilis, Turkey. Afr Health
Sci. 2020;20:709–714. doi: 10.4314/ahs.v20i2.22
[29] Demelash H, Motbainor A, Nigatu D, Gashaw K,
Melese A. Risk factors for low birth weight in Bale
Zone Hospitals, South-East Ethiopia: a case-control
study. BMC Pregnancy Childbirth. 2015;15:264. doi:
10.1186/s12884-015-0677-y
[30] Badshah S, Mason L, McKelvie K, Payne R, Lisboa PJ.
Risk factors for low birthweight in the public-hospitals
at Peshawar, NWFP-Pakistan. BMC Public Health.
2008;8:197. doi: 10.1186/1471-2458-8-197
[31] Mao Y, Zhang C, Wang Y, Meng Y, Chen L,
Dennis CL, et al. Association between paternal age
and birth weight in preterm and full-term birth:
a retrospective study. Front Endocrinol.
2021;12:706369. doi: 10.3389/fendo.2021.706369
[32] Richterman A, Raymonville M, Hossain A, Millien C,
Joseph JP, Jerome G, et al. Food insecurity as a risk
factor for preterm birth: a prospective facility-based
cohort study in rural Haiti. BMJ Glob Health. 2020;5:
e002341. doi: 10.1136/bmjgh-2020-002341
[33] Rodríguez-Fernández A, Ruíz-De la Fuente M,
Sanhueza-Riquelme X, Parra-Flores J, Dolores
Marrodán M, Maury-Sintjago E. Association between
maternal factors, preterm birth, and low birth weight
of Chilean singletons. Children (Basel). 2022;9:967.
doi: 10.3390/children9070967
[34] Mao Y, Liu Z, Zhang C, Wang Y, Meng Y, Chen L,
et al. Association between paternal age and subtypes
of preterm birth: a retrospective study. Reprod
Biomed Online. 2022;44:1101–1109. doi: 10.1016/j.
rbmo.2021.12.012
[35] Muchie KF, Lakew AM, Teshome DF, Yenit MK,
Sisay MM, Mekonnen FA, et al. Epidemiology of pre­
term birth in Ethiopia: systematic review and
meta-analysis. BMC Pregnancy Childbirth.
2020;20:574. doi: 10.1186/s12884-020-03271-6
[36] Omar AI, Mohamed AD, Farah MG, Mahad IA,
Mohamed SA, Dimbil AH, et al. Maternal risk factors
associated with preterm births among pregnant
women in Mogadishu, Somalia. Children (Basel).
2022;9:1518. doi: 10.3390/children9101518
[37] Dreyfuss ML, Msamanga GI, Spiegelman D, Hunter DJ,
Urassa EJ, Hertzmark E, et al. Determinants of low
birth weight among HIV-infected pregnant women in
Tanzania. Am J Clin Nutr. 2001;74:814–826. doi: 10.
1093/ajcn/74.6.814
[38] Kaur S, Ng CM, Badon SE, Jalil RA, Maykanathan D,
Yim HS, et al. Risk factors for low birth weight among
rural and urban Malaysian women. BMC Public
Health. 2019;19:539. doi: 10.1186/s12889-019-6864-4
[39] Xi C, Luo M, Wang T, Wang Y, Wang S, Guo L, et al.
Association between maternal lifestyle factors and low
birth weight in preterm and term births: a
case-control study. Reprod Health. 2020;17:93. doi:
10.1186/s12978-020-00932-9
[40] Davenport MH, Meah VL, Ruchat SM, Davies GA,
Skow RJ, Barrowman N, et al. Impact of prenatal
exercise on neonatal and childhood outcomes:
a systematic review and meta-analysis. Br J Sports
Med. 2018;52:1386–1396. doi: 10.1136/bjsports-2018-
099836
[41] Qin M, Liang J, Yang D, Yang X, Cao P, Wang X,
et al. Spatial analysis of dietary exposure of aflatoxins
in peanuts and peanut oil in different areas of China.
Food Res Int. 2021;140:109899. doi: 10.1016/j.foodres.
2020.109899
[42] Waliyar F, Reddy SV, Subramaniam K, Reddy TY,
Devi KR, Craufurd PQ, et al. Importance of mycotox­
ins in food and feed in India. Asp Appl Biol.
2003;68:147–154.
[43] Sabran MR, Jamaluddin R, Mutalib A, Ahmad Z.
A mini review on aflatoxin exposure in Malaysia:
past, present and future. Front Microbiol. 2013;4:334.
[44] Coppa CFSC, Khaneghah AM, Alvito P, Assunção R,
Martins C, Eş I, et al. The occurrence of mycotoxins in
breast milk, fruit products and cereal-based infant
formula: a review. Trends Food Sci Technol.
2019;92:81–93. doi: 10.1016/j.tifs.2019.08.014
[45] Reddy KRN, Abbas HK, Abel CA, Shier WT,
Oliveira CAF, Raghavender CR. Mycotoxin contam­
ination of commercially important agricultural
commodities. Toxin Rev. 2009;28:154–168. doi: 10.
1080/15569540903092050
[46] Ali N. Aflatoxins in rice: worldwide occurrence and
public health perspectives. Toxicol Rep.
2019;6:1188–1197. doi: 10.1016/j.toxrep.2019.11.007