Filetype PDF | Posted on 09 Jan 2023 | 2 years ago
Authentication
digital resources
129x Filetype PDF File size 0.98 MB Source: bmjopen.bmj.com
Open access Original research
Associations between antenatal BMJ Open: first published as 10.1136/bmjopen-2020-046790 on 21 February 2022. Downloaded from
maternal diet and other health aspects
with infant temperament in a large
multiethnic cohort study: a path
analysis approach
1,2 3 1 3
Anja Schoeps , Teresa Gontijo de Castro, Elizabeth R Peterson, Clare Wall,
4,5 1 6
Stephanie D'Souza, Karen E Waldie, Susan Morton
To cite: Schoeps A, Gontijo ABSTRACT Strengths and limitations of this study
de Castro T, Peterson ER, Objectives The aim of this study was to assess the
et al. Associations between association of antenatal maternal dietary patterns (DPs)
antenatal maternal diet and and other health aspects with infant temperament in a ► This is the first study to look at the association be-
other health aspects with large multiethnic cohort, taking maternal personality and tween the whole of diet and infant temperament.
infant temperament in a large prenatal stress into account. ► A major strength is the use of path analysis, which
multiethnic cohort study: a path Design and methods Using data from 3968 children born enables the exploration of the complex inter-
analysis approach. BMJ Open relations between antenatal maternal diet, antenatal
2022;12:e046790. doi:10.1136/ in 2009/2010 and their mothers from the Growing Up in stress and maternal personality.
bmjopen-2020-046790 New Zealand cohort, infant temperament was assessed at ► This analysis uses data from roughly 4000 chil-
► Prepublication history and 9 months using the Infant Behavior Questionnaire- Revised dren and their mothers from a large multiethnic
additional supplemental material Very Short Form. Maternal antenatal diet and other health population- representative cohort.
for this paper are available aspects were assessed antenatally. Maternal DPs (n=4) were ► It is a limitation of this study that all information was
online. To view these files, derived using principal components analysis based on food mother- reported and could be not validated by ob-
please visit the journal online intake reported on a 44- item food frequency questionnaire. served information.
(http://dx.doi.org/10.1136/ Path analyses investigated factors associated with infant ► A further limitation, which comes as a result of the
bmjopen-2020-046790). temperament, namely maternal personality, prenatal maternal large number of variables in the analysis, is that
stress, DPs and other health aspects, including potential inter-
Received 09 November 2020 relations and mediating effects. more than one- third of the initial study population http://bmjopen.bmj.com/
Accepted 19 January 2022 Results Women who scored higher in the fusion DP of about 6500 children could not be considered in
(standardised beta (β)=0.05; 95% CI 0.02 to 0.09) and the analysis.
healthy DP (β=0.05; 95% CI 0.02 to 0.09), who exercised
more (β=0.04; 95% CI 0.01 to 0.07), and who drank
less alcohol (β=−0.05; 95% CI −0.08 to –0.02) were influenced over time by heredity, maturation,
more likely to have infants with an overall less difficult and experience’ (Rothbart and Derryberry,
temperament. Sex- specific differences were found 1
in the associations between maternal DP and infant p.37), and is typically conceptualised in
temperament. Maternal personality and prenatal stress the three broad dimensions: Negative Affec- on January 9, 2023 by guest. Protected by copyright.
were significantly associated with all dimensions of infant tivity—tearful, easily frustrated, difficult to
temperament. The strongest predictors for a more difficult soothe; Surgency—happy, active, sensation-
temperament were prenatal stress (β=0.12; 95% CI 0.08 seeking; and Effortful Control—attentive,
2
to 0.15) and the personality dimensions neuroticism concentrated, easy to please. Temperament
(β=0.10; 95% CI 0.07 to 0.14) and extraversion (β=−0.09; in infancy is of public health interest, because
© Author(s) (or their 95% CI −0.12 to –0.06). it is the foundation on which behaviour and
employer(s)) 2022. Re- use Conclusions Associations of antenatal maternal diet and personality develop, and thus has been shown
permitted under CC BY- NC. No health aspects with infant temperament were statistically to predict later problem behaviour, cognitive
commercial re- use. See rights significant but small. While they should not be overinterpreted 3–8
and permissions. Published by as being deterministic, the findings of this study support the outcomes and mental and physical health.
BMJ. link between maternal modifiable health- related behaviours Current research is making advances in
For numbered affiliations see and infant temperament outcomes. explaining differences in temperament,
end of article. behaviour and mental health not only by
Correspondence to examining genetic differences but also
Dr Anja Schoeps; INTRODUCTION through a multitude of early life exposures.
anja. schoeps@ uni- heidelberg. Temperament is defined as ‘constitutional Based on the Developmental Origins of
de differences in reactivity and regulation… Health and Disease approach, which puts
Schoeps A, et al. BMJ Open 2022;12:e046790. doi:10.1136/bmjopen-2020-046790 1
Open access
BMJ Open: first published as 10.1136/bmjopen-2020-046790 on 21 February 2022. Downloaded from
the first 1000 days between conception and the second The aim of our study was to assess the association
birthday into focus, research has increasingly focused between antenatal maternal dietary patterns and other
on common in utero exposures, such as stress, health health aspects with infant temperament in a large multi-
9–11
behaviour and variations in maternal diet. There is ethnic cohort using path analyses, which enable the
now growing evidence that the offspring’s psychological, exploration of the complex inter-relations between ante -
behavioural and cognitive development is associated with natal maternal diet and other health aspects, antenatal
such variations in the in utero environment, especially stress and maternal personality.
11–16
with regards to maternal stress. Apart from its poten-
tial direct effect on the fetus’s brain, stress experience
has also been shown to coincide with and to alter the
mother’s behaviour with regards to dietary intake, phys- METHODS
17 18 Study population
ical activity or medication and drug intake, which
may in turn have direct effects on the fetus’s develop- This study was conducted within the contemporary New
11 12 19 20 Zealand birth cohort study Growing Up in New Zealand
ment.
A number of studies have associated particular diets or (GUiNZ), which enrolled 6822 pregnant women and their
6853 children who survived to age 6 weeks.29
specific nutrient intakes or deficiencies with changes in Written
21–23 informed consent was obtained from all mothers in this
temperament or behaviour in children, however, to
date, no published study has looked at the associations study. The cohort of children is broadly generalisable
between the whole of diet and temperament or behaviour to New Zealand birth statistics with respect to ethnicity,
24 29
in infancy. During the last few years, a small body of maternal age and socioeconomic status. This study used
studies has evolved which investigated potential effects information from three data collection waves, completed
of certain aspects of maternal diet in conjunction with antenatally and when the children were approximately
11 15 9 months and 2 years old. From the antenatal question-
prenatal stress. In their literature review, Monk et al
state, ‘These prenatal experiences exert their influence naire, we used information on maternal stress, sociode-
in the context of one another and yet, almost uniformly, mographics and health aspects, including diet (assessed
are studied independently’. Since 2013, a small number with a food frequency questionnaire (FFQ)). Infant
of studies have shown that a bidirectional effect between temperament was assessed at the 9-month inter view and
stress and diet exists, and that there are possible inter- maternal personality was assessed at the 2-year inter view.
action effects of prenatal stress and diet/nutrition on Baby’s sex and gestational age were collected perinatally
infant and child temperament and behaviour develop- and linked to GUiNZ data set from perinatal and mater-
11 17 25–27 nity records.30
ment. However, only two of the above studies
looked at infant temperament, one at the association with Infants were included in the analysis: if information
26 on infant temperament was available from the 9- month
antenatal chocolate intake and the other at the associa-
tion with antenatal omega 3 to omega 6 fatty acid intake interview; if mothers had provided information on stress, http://bmjopen.bmj.com/
27 sociodemographics, diet and other health aspects ante-
ratio, and none has investigated the influence of the
whole of maternal diet during pregnancy on the offspring natally; if mothers had provided information on maternal
temperament. personality at the 2-year inter view; and if perinatal infor-
In summary, while there is a body of studies on the asso- mation could be linked from the official records (n=6129
ciations between certain domains of prenatal stress, diet of the 6476 children who took part in the 9- month inter-
and other health aspects with infant and child tempera- view) (online supplemental figure 1). Children were
ment, the literature in this field is very patchy, excessively excluded from analysis if factors were present, which
heterogeneous and conspicuously devoid of null results. would likely affect the validity of the infant temperament on January 9, 2023 by guest. Protected by copyright.
Moreover, most of the studies have three major shortcom- scale: inadequate English language skills of the mother
ings in common: (1) failure to draw a complete picture (n=161); infant’s behaviour during the preceding week
of antenatal maternal diet and other health aspects: a less than somewhat typical (scores 1–2) (n=703); born
large proportion of publications studied single micro- preterm (n=393); or older than 52 weeks old at the inter-
nutrients or macronutrients in isolation, thus neglecting view (n=107), as this is outside the target range of the
31
the context of the whole of diet or other health aspects, infant temperament measure. Children whose mothers
such as exercise or smoking and alcohol consumption; had not been able to respond to the antenatal question-
(2) poor confounder adjustment: while the majority of naire prior to giving birth were excluded to minimise
studies adjusted for sociodemographic variables, such as recall bias for the FFQ (n=712). Finally, n=179 children
maternal age and education, maternal personality was had to be excluded from analysis because of missing
not considered as potential confounder, even though values in at least one of the explanatory or adjustment
a substantial part of temperament is thought to be variables other than body mass index (BMI) and gesta-
28 tional weight gain. Thus, the final number of children
genetic; and (3) limited statistical analyses techniques
to explore inter- related factors such as maternal person- assessed was 3968, corresponding to 61.3% of the chil-
ality, maternal stress, maternal diet, other health aspects dren who took part in the 9-month inter view (online
and their effect on temperament. supplemental figure 1).
2 Schoeps A, et al. BMJ Open 2022;12:e046790. doi:10.1136/bmjopen-2020-046790
Open access
BMJ Open: first published as 10.1136/bmjopen-2020-046790 on 21 February 2022. Downloaded from
Study variables In alignment with these recommendations, we calculated
Antenatal maternal diet a weighted exercise score, towards which vigorous physical
Antenatal maternal diet was assessed by a semiquantita- activity counted with a factor of 1, and moderate physical
tive 44- item FFQ during pregnancy, for which mothers activity with a factor of 0.4.
Pre-
recalled their food intake over the 4 weeks preceding the pregnancy BMI was calculated based on self-
reported pre-pregnancy weight and height and it was clas
antenatal interview. The food list of the FFQ was devel- -
37
offs. Monthly gestational
oped based on the New Zealand Ministry of Health guide- sified according to the WHO cut-
32 weight gain (kg) was calculated as the difference between
lines for healthy pregnant women, which recommends
the adequate intake of four core food groups: fruits and the current weight at the time of the antenatal interview
vegetables, breads and cereals, milk and milk products and the reported pre- pregnancy weight. To take account
and lean meat, meat alternatives and eggs. The frequency of different gestational ages at the time of the interview,
of the intake of individual foods was described by catego- this difference was divided by the elapsed months of preg-
ries increasing from not at all or less than once a month nancy up to the time of the interview.
to a number of times per month, week or day. Show cards Nausea experience was reported during the first
33 34 trimester and after the first trimester on a 4-point scale,
were used to illustrate a standard portion size.
A previous study identified four distinct dietary patterns ranging from 0 – no nausea, over 1 – mild nausea and
(DPs) during pregnancy using principal components 2 – moderate nausea, to 3 – severe nausea, respectively.
34 In this study, nausea was defined as the maximum severity
analysis and performing a varimax orthogonal rotation.
High loadings for the food items were defined as being of nausea experienced at any time during pregnancy.
larger than 0.30 or smaller than −0.30. The DP junk was Finally, we also captured self- reported alcohol consump-
characterised by high positive loadings for confectionary, tion and smoking during pregnancy. We created three
snacks, takeaways, hot chips, soft/energy drinks, cakes/ categories: women who did not drink any alcohol before
biscuits, processed meats, battered fried fish/seafood pregnancy; women who stopped drinking alcohol at the
and ice cream. The healthy DP had high positive loadings beginning of pregnancy; and women who drank alcohol
for vegetables, brown whole meal bread, cheese, non- during pregnancy. Self-reported smoking behaviour was
citrus fruits, yoghurt, dried fruits, high fibre cereal and categorised likewise.
Vegemite/Marmite. The traditional DP had high positive
loadings for whole/standard milk, white bread, marga- Prenatal stress
rine, jam/honey/marmalade, peanut butter/Nutella and Prenatal perceived stress was assessed during the ante-
38
a high negative loading for skim/trim milk. The fusion natal interview by the 10- item Perceived Stress Scale, for
DP had high positive loadings for noodles/rice/pasta, which women have to report the frequency of different
seafood, chicken, green leafy vegetables, eggs, red meat, stress- related experiences and feelings during the
processed fish and citrus fruits (online supplemental preceding 4 weeks on a 5- point scale (1=Never, 2=Almost
34 never, 3=Sometimes, 4=Fairly often, 5=Very often). The http://bmjopen.bmj.com/
table 1). Together, these four DPs explained 23.4% of
the total dietary variance in this population and individu- perceived stress score was retrieved by summing up over
ally the junk, healthy, traditional and fusion DPs explained the 10 items.
7.7%, 5.6%, 5.4% and 4.8% of the dietary variance,
respectively. For all following analyses, the women were Maternal personality
ranked in quintiles of the score in each of the four respec- Maternal personality was assessed by the big five inven-
34 39 40
tive DPs. tory (BFI) at the 2- year interview. The 44 items for
the assessment of personality are a popular measure of
Other relevant health aspects personality due to the established factor structure, high on January 9, 2023 by guest. Protected by copyright.
Folic acid supplementation during pregnancy was reliability, good convergent validity with other longer
described in three categories: folic acid supplementation Big Five measures of personality and acceptable self-
39–41
that started before pregnancy; folic acid supplementation peer agreement. While few studies have looked at
that started in first trimester; and folic acid supplementa- the stability of the BFI 44 across adulthood, high stability
tion that started after the first trimester or never. has been reported with the BFI 44 over a 3-year period
42
Women reported on their moderate and vigorous across middle adulthood. In addition, research using
weekly physical activity before and during pregnancy by the Big Five personality traits more generally, has repeat-
providing the number of days with exercise and the dura- edly shown that the rank order stability of the Big Five
tion of an average exercise episode over the last 7 days. personality increases with age peaking at around the age
42 43
Using an abbreviated version of the International Phys- of 60 and does not vary much by assessment method.
35
ical Activity Questionnaire, we decided to focus on pre- The BFI 44 used in this study uses a 5-point Likert scale
pregnancy exercise because pre-pregnancy exercise levels which asks how strongly a person agrees or disagrees with
are not affected by pregnancy health problems. The Amer- statements about themselves. Each item is associated with
ican College of Sports Medicine and the American Heart one of the five personality factors: extraversion—sociable,
Association recommend 150 min of moderate or 60 min outgoing, energetic; agreeableness—forgiving, compliant,
36
of vigorous physical activity for adults aged 18–65 years. sympathetic; conscientiousness—organised, efficient,
Schoeps A, et al. BMJ Open 2022;12:e046790. doi:10.1136/bmjopen-2020-046790 3
Open access
BMJ Open: first published as 10.1136/bmjopen-2020-046790 on 21 February 2022. Downloaded from
thorough; neuroticism—irritable, shy, moody; and open- one at the household level. Area deprivation was based
ness—imaginative, curious, excitable.39 Scores for the five on results from the 2006 National Census around nine
personality factors are derived by averaging over all items socioeconomic indicators, which enabled assignment of
that load on the respective factors. a 10-point deprivation index to each geographical area,
where 1 was assigned to the least deprived areas and 10
Infant temperament 46
to the most deprived areas. Household deprivation was
The 37- item Infant Behavior Questionnaire-Revised V ery based on six questions from the 9- month interview, asking
Short Form (IBQ- R VSF) was used to assess infant temper- mothers if their financial situation during the preceding
ament at the 9- month interview.31 In a previous analysis it 12 months had affected their access to adequate food,
was shown that a five-factor structure was a better statis- clothing and/or heating. A mean score was calculated
tical fit for the Growing Up in New Zealand Cohort than over the six variables, with higher values indicating more
44
the original three- factor structure. The fit of the five- severe household deprivation.
44
factor model was found to be acceptable in this cohort
and the factors were found to be similarly precise across Statistical analysis
four ethnic groups (European, Maori, Pacific Peoples and We fitted path analysis models to assess the associations
45
Asians). In addition, within this cohort, the five factors between the dimensions of maternal personality, stress,
have also been related to mother- and partner- reported maternal diet/other health aspects and infant tempera-
infant closeness, parenting confidence and parenting ment. To determine the important paths, we first assessed
44
satisfaction in the expected direction. The five- factor pairwise associations between these dimensions. We used
model consists of Negative Affectivity (NEG), Positive Affect directed acyclic graphs (DAGs) to describe hypothesised
and Surgency (PAS) and a slightly altered factor Orienting associations between our exposures, outcome, and poten-
47
Capacity (OC), along with two new factors Affiliation/ tial confounding variables. A graphical presentation of
Regulation (AR) and Fear, which separated out from the our proposed DAGs can be found in online supplemental
broader factors of Orienting and Regulatory Capacity and figure 2. We made decisions about which confounders we
NEG, respectively (see online supplemental table 2) for a needed to adjust for in the respective analyses based on
brief description of these factors). the respective DAGs. All regression models and path anal-
In completing the IBQ-R VSF , mothers or caregivers yses were adjusted for the six sociodemographic variables.
were asked to indicate how frequently their child engaged
in specific behaviours during the past week on a scale Preparatory analysis
from 1 to 7 (1=Never, 2=Very rarely, 3=Less than half the time, To determine the important paths for the path analysis,
4=About half the time, 5=More than half the time, 6=Almost in the preparatory analyses we first assessed pairwise
always, 7=Always). The mothers could choose the option associations via multivariable linear regressions between
‘not applicable’ if they had not seen the child in the situ- the following dimensions: (A) maternal personality and
ation. Temperament scores were calculated by averaging prenatal perceived stress; (B) maternal personality/stress http://bmjopen.bmj.com/
over all items that were found to load on the respective and DPs/other health aspects; (C) maternal person-
temperament factors. Because infant behaviour is very ality/stress and infant temperament; and (D) DPs/other
changeable over any given week (eg, due to illness or health aspects and infant temperament. Associations
teething), and the reference period for the questionnaire between maternal personality and prenatal perceived
was the behaviour over the last week, an additional ques- stress were examined by regressing stress on the five
tion was asked: ‘In terms of your baby’s behaviour, how factors of maternal personality (online supplemental
typical has the last week been for your baby?’, in which figure 2, Panel A). Associations between maternal person-
mothers were asked to respond on a scale from 1 (not at ality/stress and DPs/other health aspects were assessed on January 9, 2023 by guest. Protected by copyright.
all typical) to 5 (very typical).44 by regressing each of the eleven maternal health aspects
(DPs and other health aspects) on the five factors of
Sociodemographic variables/adjustment set maternal personality and the one stress factor: a total of
We used six antenatal sociodemographic variables in our 11 distinct linear regression models, adjusting for the
analyses: parity; maternal education; maternal age; self- six major confounders from the adjustment set (online
prioritised ethnicity; deprivation at area level; and depri- supplemental figure 2, Panel B). Continuous outcomes
vation at household level. Parity is binary in these analyses are a requirement for linear regression. To be able to
and discriminates between first-born and subsequent chil- use diet and other health aspect variables as outcomes,
dren. Mothers’ education was assessed in five categories, these variables were transformed from categorical to
namely: no secondary school qualification; secondary ordinal scale, which implies a clear order of values:
school qualification; diploma or trade certificate; bach- DPs (in quintiles); folic acid supplementation; nausea;
elor’s degree; and higher degree. The maternal self- alcohol consumption; and smoking. To give an example,
prioritised ethnic group was categorised into: European; alcohol was used as a continuous variable, taking on the
Māori; Pacific Peoples; Asian; and Middle Eastern, Latin values 0–1–2 for women who: did not drink any alcohol
American, African and Other. We used two distinct vari- before pregnancy (0); who stopped drinking alcohol at
ables to account for deprivation, one at the area level and the beginning of pregnancy (1); and who drank alcohol
4 Schoeps A, et al. BMJ Open 2022;12:e046790. doi:10.1136/bmjopen-2020-046790
no reviews yet
Please Login to review.
