Open access Original research

Pregnancy anaemia, child health and

development: a cohort study in
rural India
Esther Heesemann  ‍ ‍,1,2 Claudia Mähler,3 Malavika A Subramanyam,4
Sebastian Vollmer5

To cite: Heesemann E, ABSTRACT

Mähler C, Subramanyam MA, Objective  To assess how pregnancy anaemia affects STRENGTHS AND LIMITATIONS OF THIS STUDY
et al. Pregnancy the offspring’s early childhood development, child ⇒ The study used a unique cohort of pregnant women
anaemia, child health and and their 2-­year-­old children.
haemoglobin (Hb) levels child growth and diseases
development: a cohort study ⇒ We controlled for maternal haemoglobin (Hb) and
incidence 2 years after birth in a low-­income setting.
in rural India. BMJ Open
Furthermore, we investigate the mediating role of household food security after pregnancy, together
2021;11:e046802. doi:10.1136/
bmjopen-2020-046802 childhood Hb levels with disease incidences and skills. with other relevant confounding factors.
Design  Prospective cohort study. ⇒ The data contain only one measure of Hb during
► Prepublication history and pregnancy and the attrition between the waves is
Setting and participants  The study participants are
additional supplemental material high.
941-­999 mother–child dyads from rural Madhepura in
for this paper are available
Bihar, India. In 2015, the women were recruited during
online. To view these files,
please visit the journal online pregnancy from registers in mother–child centres of 140
villages for the first wave of data collection. At the time having a low level of haemoglobin (Hb) level
(http://dx.doi.org/10.1136/​
bmjopen-2020-046802). of the second wave in 2017, the children were 22–32 in the blood, document the precarious state
months old. of nutrition among many pregnant women
Received 10 November 2020 Primary and secondary outcome measures  The worldwide. In 2011, 32 million women were
Accepted 24 August 2021 recruited women were visited at home for a household estimated to be anaemic, the vast majority
survey and the measurement of the women’s and child’s living in South Asia.2 A focus on anaemia is
Hb level, child weight and height. Data on the incidence of imperative because Hb is a crucial ingredient
diarrhoea and respiratory diseases or fever were collected of red blood cells and thereby responsible for
from interviews with the mothers. To test motor, cognitive,
the transport of oxygen to the body tissues.
© Author(s) (or their language and socioemotional skills of the children,
Low Hb during pregnancy is a known risk
employer(s)) 2021. Re-­use we used an adapted version of the child development
assessment FREDI. factor for premature birth, low birth weight
permitted under CC BY-­NC. No
commercial re-­use. See rights Results  The average Hb during pregnancy was 10.2 g/ and in extreme cases leads to death.3–7 Yet,
and permissions. Published by dL and 69% of the women had pregnancy anaemia. At also high Hb levels, especially in the first
BMJ. the age of 22–32 months, a 1 g/dL increase in Hb during trimester, are associated with adverse birth
1
Department of Economics, pregnancy was associated with a 0.17 g/dL (95% CI: 0.11 outcomes, suggesting a U-­ shaped relation-
University of Mannheim, to 0.23) increase in Hb levels of the child. Children of ship.8 One of the most common causes of
Mannheim, Germany moderately or severely anaemic women during pregnancy
2
Center for Evaluation and
anaemia in low-­income and middle-­income
showed 0.57 g/dL (95% CI: −0.78 to −0.36) lower Hb than countries, such as India, is iron deficiency.9 10
Development, Mannheim,
children of non-­anaemic women. We find no association Anaemia can however also result from other
Germany
3 between the maternal Hb during pregnancy and early
Institute for Psychology, micronutrient deficiencies such as folic acid,
University of Hildesheim, skills, stunting, wasting, underweight or disease incidence.
vitamin B12 and vitamin A, as well as infec-
Hildesheim, Niedersachsen, While childhood anaemia does not correlate with childhood
Germany diseases, we find an association of a 1 g/dl increase in the tious diseases and genetic disorders.
4
Social Epidemiology, IIT child's Hb with 0.04 SDs higher test scores. In this paper, we investigate the conse-
Gandhinagar, Gandhinagar, Conclusions  While pregnancy anaemia is a risk factor quences of pregnancy anaemia on child Hb
Gujarat, India for anaemia during childhood, we do not find evidence for levels, early skills and other health indicators
5
Department of Development an increased risk of infectious diseases or early childhood in the first 1000 days. With a prevalence of
Economics, Center for Modern
development delays. 58%, anaemia of children below the age of 5
Indian Studies, University
of Göttingen, Goettingen, remains a significant global health challenge
Niedersachsen, Germany in South Asia and its causes are worth investi-
INTRODUCTION gating.2 The known adverse consequences of
Correspondence to
Sufficient intake of macronutrients and childhood anaemia on human development,
Sebastian Vollmer;
​svollmer@​uni-​goettingen.​de and micronutrients during pregnancy is a prereq- in particular cognitive skills, add importance
Esther Heesemann; uisite for healthy child development.1 Yet, to this matter.11–13 Most of the iron require-
​esther.​heesemann@​gmail.​com the high rates of pregnancy anaemia, that is, ment in the first year of life is met by the

Heesemann E, et al. BMJ Open 2021;11:e046802. doi:10.1136/bmjopen-2020-046802 1

Open access

body iron a child is born with, either in the form of Hb be noted that our sample is not representative of the full
or in iron stores (ferritin). The amount of ferritin and population of pregnant women in that area, but only for
Hb at birth depends heavily on the iron transfer from those who registered in the centres. In 2015/2016, this
mother to child in-­utero, which occurs in the second and covered 76% of pregnant women.30
third trimesters of pregnancy.14 Studies have shown that All women listed in the registries in March/April 2015
anaemia during pregnancy correlates with low cord Hb were visited by teams of trained, local survey enumera-
levels and anaemia during infancy.11 15–19 To the best of tors, medical data collectors and child development
our knowledge, there are no studies investigating the testers, and invited to participate in the baseline survey
relationship of pregnancy anaemia and anaemia of chil- and medical tests. During the follow-­up in November/
dren older than 18 months. December 2017, we attempted to revisit the households
While several experimental and non-­ experimental of all formerly pregnant women. In addition to the survey
studies have examined the impact of pregnancy anaemia and medical measures for mothers and children, a devel-
or iron deficiency on early skills, the empirical findings opment test was administered to the child that resulted
are inconclusive.20 21 This might be a result of different from the pregnancy. The household survey, the medical
study designs, age variation of the study participants, measurements and the child development test were
the dimensions of skills measured or geographical diver- conducted within one week for each household. Online
sity.22–29 Furthermore, as pregnancy anaemia is a risk for supplemental figure S1 visualises the data collection time-
adverse birth outcomes, the existing experimental studies line, the age of the children and the gestational stage of
exclude moderately and severely anaemic women from the women during the data collections.
the trials. Hence, the external validity of their findings
for the general population of pregnant women is unclear. Outcome measures
Observational studies without the appropriate quasi-­ Anaemia is defined over the Hb level in the blood, which
experimental methods are unable to identify a causal in our case was obtained from a finger prick in the respon-
impact of anaemia during pregnancy on child outcomes dents’ homes, collected by trained local enumerators.
due to omitted, endogenous variables. For instance, Using point-­of-­care HemoCue 301 machines for capillary
ignoring time-­invariant environmental factors is likely to blood, the Hb level can be determined immediately and
upwardly bias the results of non-­experimental studies. communicated to the tested individual or its caregivers
We contribute to the existing literature by analysing the on the spot.31 (The bias of HemoCue 301 anaemia assess-
consequences of pregnancy anaemia on child outcomes ments compared with laboratory tests is with 0.25 g/
in three essential ways. First, we eliminate an important dL well below the WHO recommended threshold for
confounder in the analysis by controlling for the maternal point-­of-­care machines.) According to the WHO and the
Hb levels and the food diversity of the household after Indian Council of Medical Research definitions, a preg-
birth. While not being able to fully capture the unobserv- nant woman is anaemic if her Hb concentration falls
able differences between children exposed to pregnancy below 11.0 g/dL.32 33 Pregnancy anaemia is further distin-
anaemia and without, our set of covariates will omit the guished into mild anaemia (10.0–10.9 g/dL), moderate
bias emerging from differences in micronutrient diver- anaemia (7.0–9.9 g/dL) and severe anaemia (<7.0 g/dL).
sity in the postnatal period. Second, by following the The anaemia thresholds for children between 6 and 59
children more than 2 years after birth, we can observe if months are the same as for pregnant women.33
potential initial disadvantages persisted over time. Finally, During data collections, the field teams followed a
in addition to cognitive and non-­ cognitive functions, strict protocol on detection of anaemia. In case of mild or
and anaemia, we also assess the influence of pregnancy moderate anaemia, the women/caregivers were advised
anaemia on secondary health outcomes, namely child to go to the nearest primary healthcare centre soon to
growth and disease incidence. This analysis will help to seek treatment for anaemia. In case of severe anaemia,
get a deeper understanding of the adverse consequences the household was alerted that immediate attention was
of anaemia during pregnancy. needed. In the follow-­up survey, we also offered to cover
the treatment costs and transport to a health facility for
all severely anaemic children.
METHODS Child development was measured with a variation of the
Data and procedures FREDI 0–3, a German development test similar in struc-
Our data set is a panel of two waves, consisting of ture to the Bayles Scales of Infant and Toddler Develop-
household surveys, anthropometric and blood sample ment.34–36 Due to the different home environments of the
collections and child development tests conducted in children, certain items of the original FREDI 0–3 were
Madhepura in the North-­Eastern state Bihar. Bihar is one adjusted to the Bihari context. The development test
of the poorest states of India and Madhepura belongs consists of a parent questionnaire and a child assessment,
to its socioeconomically most deprived districts. Our and covers four areas: fine and gross motor development,
study sample was taken from pregnancy registers in local receptive and expressive language development, cogni-
mother–child-­centres (Anganwadi centres) in 140 villages tion and socioemotional development. Two age-­specific
in six subdistricts (blocks) of Madhepura. It should hence tests were administered, each covering skills over an

2 Heesemann E, et al. BMJ Open 2021;11:e046802. doi:10.1136/bmjopen-2020-046802

 Open access

age range of 5 months. Each test consisted of around 40 anaemia. Dummy variables were created for no, mild,
items. All raw scores have been standard normalised and and moderate-­ to-­
severe anaemia, and simultaneously
are hence presented as z-­scores. In addition to the four added to the regression equation. No anaemia served as
individual test scores, we calculate the total FREDI z-­score the reference category. To test a possible U-­shaped rela-
over all 40 test items. A brief validation of the FREDI tionship between pregnancy Hb and the primary child
with regard to physical growth and maternal education is outcomes, we included a quadratic term to the linear
presented in online supplemental figures S2 and 3). regression model.
Stunting (ie, being too short for their age), wasting (ie, All estimations controlled for the maternal Hb levels
being too light for their height) and underweight (ie, and the household food diversity scores at the time of the
being too light for their age) are used as secondary health follow-­up. This was done to avoid an overestimation of
outcomes. Children’s height and weight were measured the relationship between pregnancy and child outcomes
during the field visits by the medical testing team. We age-­ due to an overall poor food environment or chronic
standardised the raw height and weight values following diseases of the mothers. Overall poor household nutri-
the WHO Growth Standards.37 With a respective z-­score tion during childhood is likely to correlate with a poor
of 2 SD below the median of the WHO references popu- nutrition of women during pregnancy. As both factors are
lation, a child is defined as being either stunted, wasted likely to adversely affect child health and development,
or underweight. Any value above 6 SD or below 6 SD was ignoring the nutritional environment at the time of the
coded as measurement error and dropped from the anal- outcome measure might lead to an overestimation of the
yses. (This was the case for 21 weight-­for-­height z-­scores, correlation of pregnancy anaemia and child well-­being.
7 weight-­for-­age z-­scores and 11 height-­for-­age z-­scores.) We further add age, sex and current breast-­feeding status
The information on diarrhoea and respiratory disease of the child, as well as development test facilitator (FREDI
or fever incidences in the 2 weeks before the survey were fixed effects) or HemoCue machine/medical tester fixed
collected from maternal reports during the household effects as control variables to obtain more precise esti-
survey and coded as binary variables. mates. Additional covariates from the baseline data relate
to the socioeconomic status of the household (caste cate-
Patient and public involvement gory, wealth quintile and maternal literacy) and preg-
The aims and the survey design were shared at a meeting nancy characteristics which might be correlated with
of state- and district-­ level government officials who both the Hb levels during pregnancy and child outcomes
provided services in Madhepura through the Women (maternal age, pregnancy history (first birth dummy),
and Child Development ministry, village-­level leaders of trimester of gestation at the time of the Hb measurement
women’s groups prior to the baseline. At this meeting, and take-­up of antenatal care (ANC) services). (To test
there was a detailed discussion of the types of questions the robustness of our findings, we also conducted an
that needed to be asked during the data collection. Several analysis in which we replaced gestational trimester during
of these suggestions were incorporated in the baseline pregnancy Hb measurement with the gestational months,
questionnaire. Residents of Madhepura were involved to and a subsample analysis for each of the three gestational
the extent that they participated in the pretesting of the trimesters during the pregnancy Hb measurements.)
baseline questionnaire and the FREDI tool. Patients had Finally, we added subdistrict (block) fixed effects and
not been involved in the interpretation of results, writing clustered the SEs on village level to take spatial correla-
or editing of the final document. tion of the outcome variables into account. For the sensi-
tivity analysis, we included birth spacing, macronutrient
Statistical analysis deficiency and postnatal depression to the estimation and
In our main analysis, we estimated the association between replaced block with panchayat fixed effects. (We did not
pregnancy anaemia and the child Hb level and child include those variables in the main specification due to
development approximately 2 years after birth using an missing information, which would have further reduced
ordinary least-­square regression model. Our secondary the sample size. A panchayat is a subdivision of a block and
outcomes of interest, being stunted, wasted or under- comprises several villages.)
weight, incidence of diarrhoea and respiratory diseases In light of strong son preferences in the study region,38 39
or fever in the 2 weeks before the survey, were analysed we investigated heterogeneous effects for boys and girls.
with a logistical regression model. Furthermore, we tested for heterogeneous effects by ANC
We considered both continuous and discrete Hb levels take-­up, as a proxy for health preferences, caste catego-
to allow for linear and non-­linear relationships between ries, maternal literacy levels and gestational trimester at
pregnancy anaemia and the child outcomes. For the the time of the baseline survey.
linear relationship, our explanatory variable of interest As child anaemia could be a result as well as a mediator
was the Hb levels of the women during pregnancy, for pregnancy anaemia, we investigated the association of
measured at the time of the baseline data collection. For childhood Hb levels with early childhood development
the non-­linear relationship, we used the expressions of and infectious diseases in separate analyses. Using the
anaemia status as predictors: no pregnancy anaemia, mild same set of covariates as described above, we controlled
pregnancy anaemia and moderate-­ to-­
severe pregnancy for household level and child specific characteristics that

Heesemann E, et al. BMJ Open 2021;11:e046802. doi:10.1136/bmjopen-2020-046802 3

Open access

might affect both child Hb levels and the outcome vari- were on average in a higher gestational trimester at the
ables of interest. time of measurement. This is unsurprising considering
Finally, given the existing evidence of maternal anaemia the elevated iron needs as pregnancy progresses. Yet
for adverse pregnancy outcomes, we conducted a survival as gestational trimester also correlates with child age,
analysis that assessed whether the Hb level of pregnant which might affect the development test outcomes and
women correlates with child loss. Such a correlation anaemia status, we controled for gestational trimester in
would downward bias the estimates of our main analysis, all estimations, to avoid biases. Importantly, half of the
as the most deprieved children would systematically be women reported not having received any ANC during
missing in the group with higher exposure. We tested this pregnancy. This figure sets our sample strongly apart
hypothesis by estimating a probit model for non-­survival from most existing studies, many of which recruited their
on the pregnancy anaemia and Hb levels. participants during ANC visits22 28 41 or delivered ANC
All estimations were weighted with inverse probability visits as part of the study.27 The uptake of ANC services
weight, as proposed by Fitzgerald, Gottschalk and Moffitt is the highest in the group of moderate or severe preg-
(1998) adjusting the sample for selective attrition between nancy anaemia and lowest for women of mild pregnancy
the waves.40 The statistical analyses were conducted with anaemia.
the statistical software Stata V.16 (StataCorp LP). The sample size during follow-­ up data collection
reduced considerably from the baseline (n=1918) due to
the unavailability of the women or children at the time
RESULTS follow-­up visits, inaccurate location information, refusal
Sample description to provide a blood sample or to participate in the child
The final sample consisted of 941–1000 mother–child-­ testing and child death (figure 1). We account for this loss
dyads, depending on the outcome variable. We calcu- in sample size by using inverse probability weights in all
lated an unadjusted minimum detectable effect of 0.22 g/ estimations, assigning higher weights to those households
dL for child Hb and 0.12 SD for skill outcomes, statisti- that had a higher probability of dropping out. (Relevant
cally significant at the 10% level, comparing children of weighting variables were used despite missing informa-
mothers with mild or moderate pregnancy anaemia to tion, which reduces the estimation samples by two more
children of mothers without pregnancy anaemia. At the observations.).
time of the endline data collection in 2017, the age of the
children lied between 22 and 32 months. Association of pregnancy anaemia with childhood anaemia
Table 1 presents the explanatory variable, covariates and early skills
and the outcome variables of interest of the estimation We found a strong association of Hb during pregnancy,
sample separately for three anaemia categories: no preg- mild and moderate or severe pregnancy anaemia with the
nancy anaemia (columns 1–3), mild pregnancy anaemia child’s Hb levels (table 2). An increase of 1 g/dL in Hb
(columns 4–6) and moderate-­ to-­
severe pregnancy during pregnancy was associated with 0.17 g/dL higher
anaemia (columns 7–9). Hb levels in the offspring. Children born to women with
The distribution shown in the table is suggestive of a mild anaemia had 0.20 g/dL lower Hb level than their
gradient in child Hb levels and growth indicators across non-­anaemic peers did. The coefficient was more than
pregnancy anaemia levels. We do not observe a clear twice the size for children born to mothers with moderate
trend for the skill outcomes or disease incidences. Inter- or severe pregnancy anaemia. Current HB of the mothers
estingly, the distribution of Hb levels of women after preg- are consistently positively correlated with the child Hb
nancy across the anaemia groups mirrors the Hb levels levels, at a statistical significance level of 1%. Testing for
during pregnancy suggesting that suboptimal micronu- a U-­shape relationship showed a positive, but decreasing
trient intake during pregnancy continued after delivery. correlation of pregnancy Hb and child Hb, statistically
This is however not the case for the household food diver- significant at the 5% level (Panel C, table 2).
sity scores, which is the highest for the mild pregnancy We did not find a statistically significant relationship
anaemia group. of Hb or any type of pregnancy anaemia with the child
The majority of women in our sample (69%) were at development in general, or any specific dimension of
least mildly or moderately anaemic during pregnancy, development. The coefficients were small in magnitude,
with an average Hb level of 10.2 g/dL. Of all anaemic indicating indeed zero-­ effects, rather than an impre-
women, 48% showed signs of moderate anaemia, and cise estimation. For cognitive development, we found a
only 4% severe anaemia. Less than 2% of women had Hb small, negative and weakly statistically significant associ-
level above 13 g/dL, which can be considered as high. ation with Hb level during pregnancy and moderate or
Overall, the literacy level of the study population was severe pregnancy anaemia. We found weak evidence for a
low with a quarter of the women being able to read and U-­shaped relationship between pregnancy Hb and cogni-
write, similar in all three presented anaemia groups. Such tive skills, but not for motor, language or socioemotional
low levels of literacy are comparable with district-­wide skills.
statistics of a nationally representative survey from 2015 The heterogeneity analyses showed that there were only
to 2016.30 Women with more severe pregnancy anaemia small differences in the regressions of pregnancy Hb on

4 Heesemann E, et al. BMJ Open 2021;11:e046802. doi:10.1136/bmjopen-2020-046802

 Open access

Table 1  Summary statistics across exposure categories in the Hb sample

Moderate or severe preg.
No preg. anaemia Mild preg. anaemia anaemia
Mean SD N Mean SD N Mean SD N
Outcome variables in 2017
 Hb (child) 10.97 1.35 292 10.71 1.23 305 10.26 1.39 344
 Motor skills z-­score −0.02 1.04 284 −0.04 1.00 298 0.07 0.93 335
 Language skills z-­score 0.01 0.96 291 −0.01 1.00 304 0.00 1.02 344
 Cognition skills z-­score −0.01 0.99 290 −0.04 1.02 304 0.07 0.97 340
 Socioemo. skills z-­score 0.01 0.98 291 −0.02 1.00 304 −0.02 1.02 342
 Height-­for-­age z-­score −2.36 1.39 287 −2.40 1.27 300 −2.54 1.34 338
 Weight-­for-­age z-­score −1.91 1.10 286 −1.94 1.05 294 −2.07 1.08 339
 Weight-­for-­height z-­score −0.98 1.47 288 −0.97 1.71 300 −0.98 1.60 338
 Respiratory disease or fever 0.20 0.40 291 0.21 0.41 305 0.20 0.40 344
 Diarrhoea incidence 0.24 0.43 287 0.25 0.43 304 0.26 0.44 344
Other child characteristics in 2017
 Age of child (months) 27.27 2.43 292 27.26 2.41 305 27.54 2.16 344
 Currently breastfed 0.44 0.50 292 0.46 0.50 305 0.44 0.50 344
 Sex (male=1) 0.53 0.50 292 0.50 0.50 305 0.53 0.50 344
Pregnancy characteristics at baseline
 Hb (preg.) 11.76 0.72 292 10.44 0.29 305 8.68 1.07 344
 Gestational trimester during Hb 1.89 0.76 292 2.01 0.69 305 2.16 0.69 344
(preg.) measurement
 First pregnancy 0.22 0.42 292 0.22 0.41 305 0.19 0.40 344
 Any ANC visits 0.53 0.50 292 0.48 0.50 305 0.55 0.50 344
 Supplementary iron intake during 0.61 0.49 292 0.60 0.49 305 0.57 0.50 344
pregnancy
 Other micronutrient intake during 0.43 0.50 259 0.40 0.49 265 0.44 0.50 306
pregnancy
Mother characteristics
 Hb (mother) in 2017 12.09 1.32 292 11.86 1.35 305 11.14 1.62 344
 Mother can read at baseline 0.25 0.43 292 0.27 0.44 305 0.24 0.43 344
 Age of mother (years) at baseline 24.65 3.66 292 24.90 3.89 305 24.67 3.93 344
Household characteristics
 Food diversity index in 2017 7.01 1.50 292 7.30 1.46 305 7.10 1.52 344
 Scheduled case or tribe at baseline 0.29 0.46 284 0.29 0.46 295 0.33 0.47 331
 Improved sanitation facility at 0.14 0.34 292 0.12 0.33 305 0.12 0.32 344
baseline
 Asset index quintile at baseline 3.09 1.44 292 3.01 1.42 305 2.85 1.45 344
ANC, antenatal care; Hb, haemoglobin.

child Hb and skills by gestational trimester, caste category, Our findings were robust to the inclusion of additional,
maternal literacy, sex and ANC take-­up (online supple- potentially confounding, covariates, namely birth spacing,
mental table S1). In the case of language skills, we found body-­mass-­
index and postnatal depression, panchayat
a small, weakly statistically significant, negative interac- fixed effects and the exclusion of all covariates and fixed
tion effect with ANC visits and a statistically significant, effects (online supplemental table S2 and 3). Robustness
but positive interaction term for households belonging checks including gestational month instead of gesta-
to a scheduled caste or tribe. Hence, for scheduled castes tional trimester, and the subgroup analysis by gestational
and tribes, we find evidence for a positive correlation of trimester also confirmed our main results (online supple-
pregnancy anaemia and early language skills. mental table S4 and 5). We found no evidence for survival

Heesemann E, et al. BMJ Open 2021;11:e046802. doi:10.1136/bmjopen-2020-046802 5

Open access

Figure 1  Attrition between 2015 (wave 1) and 2017 (wave 2)

in childhood haemoglobin and skills sample.
Figure 2  Weighted regression results on the association
between maternal haemoglobin (Hb) levels during pregnancy
bias, that is, that the surviving children were exposed to and growth and disease incidence indicators (ORs).
higher maternal Hb levels in-­utero (online supplemental
table S6).

Association of pregnancy anaemia and child health indicators

We found no indication that suboptimal Hb levels during
pregnancy were associated with stunting, wasting or

Table 2  Weighted regression results on the association between maternal haemoglobin (Hb) levels and anaemia during
pregnancy and early childhood development and childhood Hb levels
Hb (child) Motor skills Language skills Cognition skills Socioemo. skills
Panel A          
 Hb (preg) 0.17*** (0.11, 0.23) −0.01 (−0.05, 0.03) −0.01 (−0.05, 0.04) −0.03* (−0.07, 0.00) −0.02 (−0.05, 0.02)
 Hb (mother) 0.13*** (0.07, 0.20) 0.01 (−0.03, 0.05) 0.04 (−0.01, 0.08) 0.06*** (0.02, 0.10) 0.05** (0.01, 0.10)
 R2 0.174 0.246 0.218 0.303 0.321
Panel B          
 Mild preg. anaemia −0.20* (−0.41, 0.00) 0.03 (−0.13, 0.19) 0.01 (−0.13, 0.15) 0.06 (−0.09, 0.22) 0.05 (−0.08, 0.17)
 Moderate/severe −0.57*** (−0.78 to –0.36) 0.06 (−0.10, 0.22) 0.03 (−0.12, 0.19) 0.12* (−0.02, 0.26) 0.01 (−0.12, 0.15)
preg. anaemia
 Hb (mother) 0.15*** (0.08, 0.22) 0.01 (−0.03, 0.05) 0.04* (−0.01, 0.08) 0.06*** (0.02, 0.10) 0.05** (0.00, 0.09)
2
 R 0.172 0.246 0.218 0.303 0.321
Panel C          
 Hb (preg.) 0.64*** (0.17, 1.12) 0.16 (−0.12, 0.45) 0.03 (−0.26, 0.32) 0.21 (−0.05, 0.47) −0.01 (−0.28, 0.25)
2
 Hb (preg.) −0.02** (−0.05 to –0.00) −0.01 (−0.02, 0.01) −0.00 (−0.02, 0.01) −0.01* (−0.03, 0.00) −0.00 (−0.01, 0.01)
 Hb (mother) 0.13*** (0.06, 0.19) 0.01 (−0.03, 0.05) 0.04 (−0.01, 0.08) 0.06*** (0.02, 0.10) 0.05** (0.01, 0.10)
2
 R 0.178 0.247 0.218 0.305 0.321
 Controls Yes Yes Yes Yes Yes
 Tester fixed effects Yes No No No No
 FREDI fixed effects No Yes Yes Yes Yes
 N 939 972 996 990 994

Outcome variables in columns (2)–(5) are standardised test scores and the coefficients are shown in SD. Panel A uses pregnancy Hb level of
the mother as main explanatory variable. In Panel B, the two explanatory variables of interest are mild pregnancy anaemia and moderate/severe
pregnancy anaemia, while the omitted category is no pregnancy anaemia. In Panel C, the level of pregnancy Hb in quadratic from is included. 95%
CIs using SEs clustered on village level are in parentheses. Included control variables: caste category, wealth quintile, food diversity in 2017, breast
feeding status, maternal age and literacy, first pregnancy (dummy), gestational trimester during Hb (preg.) measurement, ANC visit (dummy), child’s
sex and age, and block dummies. Additional control variable in columns (2)–(5): test version. Tester fixed effects are anthropometric test conductor
fixed effects. FREDI fixed effects are child development test conductor fixed effects. Haemoglobin is measured in g/L. Inverse probability weight
accounting for attrition applied.
Conventional significance level: *p<0.1, **p<0.05, ***p<0.01.

6 Heesemann E, et al. BMJ Open 2021;11:e046802. doi:10.1136/bmjopen-2020-046802

 Open access

Table 3  Association of child’s haemoglobin (Hb) with their early skills and infectious disease incidence
Respiratory disease or
Cum. development z-­score fever Diarrhoea
Marginal effects ORs ORs
Hb (child) 0.04*** (0.01 to 0.08) 1.02 (0.89 to 1.16) 0.99 (0.88 to 1.11)
Hb (mother) 0.02 (−0.01 to 0.05) 1.00 (0.89 to 1.12) 0.98 (0.88 to 1.08)
(Adjusted) R2 0.334 0.061 0.047
Controls Yes Yes Yes
Tester fixed effects Yes Yes Yes
FREDI fixed effects Yes No No
N 915 938 933
The outcome variable in column (1) is the standardised total test score and the coefficients are shown in SD. Outcomes in columns (2) and
(3) are binary variables and the coefficient are shown as ORs. 95% CIs using SEs clustered on village level are in parentheses. Included
control variables: caste category, wealth quintile, food diversity in 2017, breast feeding, maternal age and literacy, first pregnancy (dummy),
gestational trimester during Hb (preg.) measurement, ANC visit (dummy), child’s sex and age, and block dummies. Additional control variable
in column (1): test version. Tester fixed effects are anthropometric test conductor fixed effects. FREDI fixed effects are child development test
conductor fixed effects. Haemoglobin is measured in g/L. Inverse probability weight accounting for attrition applied.
Conventional significance level: *p<0.1, **p<0.05, ***p<0.01.

underweight of children, or increased the incidence of anaemia more than 2 years after birth is unclear as the
respiratory diseases or fever, or diarrhoea (figure 2). iron transferred from mother to child in-­utero is typically
consumed by the child’s needs within the first year of life,
Association of childhood anaemia with child skills and health before complementary feeding starts.14 Instead, our find-
The lack of association between pregnancy anaemia, early ings might be explained by a lower Hb trajectory since
skills and infectious diseases, despite the strong correla- birth, initiated by the low Hb environment in-­utero as
tion with early Hb levels, might be an indication that has been found in another study for Benin.15 It is also
childhood anaemia is not a risk factor for early childhood possible that the high rate of prolonged breastfeeding
development or disease incidence in our study population. (44% of the sample still being breastfed at the age of 2)
The cross-­sectional analysis confirmed this hypothesis for hinders children with low iron stores at birth to catch
diarrheal and respiratory diseases or fever (table 3). Yet, up through an iron-­rich diet. The negative association
for the cumulative development scores, we found a small,
between continued breastfeeding, anaemia and iron defi-
positive and statistically significant coefficient.
ciency of young children found in other studies supports
this hypothesis.10 44
DISCUSSION Overall, our findings on childhood anaemia are in
Interpretation line with a recent study from the USA, that showed an
Our cohort study from rural Bihar, India, shows a strong elevated risk ratio of infant anaemia from anaemia
association between maternal Hb levels during preg- during pregnancy, in particular for moderate and severe
nancy and the Hb levels of the offspring between 22 and anaemic conditions.18 Similarly, an analysis with data from
32 months after birth. The association is strongly statis- China showed reduced Hb levels of infants for maternal
tically significant and robust to the inclusion of several, anaemia during the 24–28 gestational week. The authors
potentially confounding, variables, such as current nutri- find however no association for pregnancy anemia in
tional status of the mother or child age. The relationship the first gestational trimester.19 The suggestive evidence
does not differ by child sex, caste category, gestational for an inverse U-­shaped relationship with childhood Hb
trimester, ANC take-­up or maternal literacy. Moderate is in line with findings on preterm birth and small-­for-­
or severe pregnancy anaemia is associated with lower Hb gestational age as presented in a review of 19 studies
levels of children than mild pregnancy anaemia. With across the world.8
increased pregnancy Hb level, the association with child- We did not find any correlation between low Hb level
hood Hb becomes weaker, yet in our sample, it does not during pregnancy and other child health outcomes, such
reach the tipping point to a full reversal. This might be as disease incidence or growth indicators. The disadvan-
due to the small sample size on the higher end of the Hb tage of children born to anaemic mothers, compared
spectrum. with their peers, hence seems to be limited to the low
Our analysis extends the current literature that Hb levels. Our study does also not show any associa-
connects pregnancy anaemia with direct birth outcomes tion between childhood Hb and infectious diseases, even
in India, such as prematurity birth weight and size.42 43 though iron deficiency, a major cause of anaemia, is
The biological link between pregnancy anaemia and child known to weaken the immune functions.45 In this regard,

Heesemann E, et al. BMJ Open 2021;11:e046802. doi:10.1136/bmjopen-2020-046802 7

Open access

our findings contradict a study on Bedouin children in The loss of follow-­up of around half of the study children
Israel, presenting a strong linkage between anaemia, diar- is reason for concern about the external validity of our
rhoea and respiratory illnesses of children.46 findings. Despite weighting the study sample according
Finally, we find no evidence that pregnancy anaemia is to their inverse probably of attrition, we are only able to
associated with lower early childhood development, even correct for observable differences in the study population
though children’s Hb levels do correlate with early skills in and the lost observations. If the unobserved characteris-
our sample. Our results thereby go against the outcomes tics which caused the loss in follow-­up also correlated with
of several observational studies, most prominently, Chang the explanatory and outcome variables, the results of our
et al’s analysis for rural China with children of a similar analysis would be biased.
age. While they also did not find an association of preg- Any systematic error in Hb measurement, depending
nancy anaemia and psychomotor development, children on the Hb level, could be a bigger concern. This could
of anaemic mothers showed lower scores in language and have been induced by the differential recommenda-
cognitive development when 18 and 24 months old.27 tions our field team gave based on the blood test results.
Differences in the socioeconomic characteristics of two However, our data do not support this hypothesis. We
study populations might explain this contrast. Not only did not observe a particular behaviour change in more
did the mothers in Chang et al’s study have higher levels severely anaemic women: even after delivery only 57% of
of education, but the prevalence of childhood undernu- the women with moderate or severe pregnancy anaemia
trition among them was about a fifth of that in our study reported consuming iron supplements during pregnancy,
population. The poor nutritional status of children in compared with 60%–61% in the mildly and non-­anaemic
Madhepura is likely not only a result of the insufficient group. During pregnancy, the reported supplementation
intake of macronutrients and micronutrients but also of rates across the groups were aswell similar (16% of the
frequent gastrointestinal infections. The low coverage of moderately or severely anaemic women versus 19%–20%
improved toilet facilities and high prevalence of unsafe of the mildly or non-­ anaemic women). This suggests
disposal of children’s stool possibly led to greater expo- that the advice given by the field teams upon anaemia
sure to faecal bacteria, thus facilitating the spread of diar- detected did not alter the women’s behaviour.
rheal diseases and parasites. The constant exposure to Two additional points should be kept in mind when
faecal bacteria could also cause environmental enterop- interpreting our findings. First, we collected Hb levels but
athy which hampers the absorption of nutrients and not ferritin levels, which leaves room for speculation on
worsens malnutrition.47–49 the origin of anaemia in the study sample. Although iron
Furthermore, it is important to note that the human deficiency is commonly believed to be the major reason
brain development is largely driven by experience.50 A for anaemia, to the best of our knowledge, no study has
lack of adequate learning opportunities and stimulation documented the actual share of iron deficiency anaemia
in the early years can have long-­lasting consequences for among the anaemic pregnant women in rural Bihar.
the functioning of the brain.51 52 The stimulation environ- Second, it should be noted that the share of severely
ment created by the caregivers for their children is limited anaemic pregnant women in our study is very small. The
in our study area. Only about half of the caregivers told association of moderate or severe anaemia with child
stories, sung songs or read a book to the children during development is hence mainly attributable to the group of
the 3 days prior to the survey. About a quarter of the women who were moderately anaemic during pregnancy.
mothers reported that no household member had played
with the child in that time.
Taken together, all of these lead us to argue that the CONCLUSION
additional, possible adverse impact of pregnancy anaemia We find strong, yet not causal, evidence that preg-
on child development in our study setting was not large nancy anaemia is a risk factor for childhood anaemia
enough to be detectable in our estimations. but not for any deficiency in the development of early
skills, on average. The relationship between pregnancy
Limitations anaemia and childhood Hb grows stronger with lower
Our analysis is based on the assumption that the single Hb levels of Hb during pregnancy. Yet, diarrhoea or respi-
measure taken at baseline is informative of the Hb status ratory disease or fever incidence or child growth is not
during the full course of pregnancy. Most Hb measure- affected by either childhood Hb or pregnancy Hb. This
ments of our participants were taken in the second study gives important insights into the consequences
trimester, allowing sufficient time for any improvement of pregnancy anaemia for populations underserved
in the Hb levels during the remaining months. Such by ANC services and with high rates of malnutrition.
improvements after the baseline data collection would Nevertheless, using a singular Hb measurement during
hence weaken the relationship of Hb that we measured pregnancy and facing high rate of attrition between the
with child outcomes. Given that half of the mothers in waves might affect the external validity of our results.
our sample reported not having consumed any iron Nevertheless, the strong association between preg-
supplements during pregnancy or received any ANC, we nancy anaemia and childhood anaemia we identified
believe that the extent of such attenuation bias is limited. in this study should be further investigated to observe

8 Heesemann E, et al. BMJ Open 2021;11:e046802. doi:10.1136/bmjopen-2020-046802

 Open access

if it will affect later life outcomes, commonly associ- 7 Rahman MM, Abe SK, Rahman MS, et al. Maternal anemia and risk
of adverse birth and health outcomes in low- and middle-­income
ated with iron deficiency and anaemia. countries: systematic review and meta-­analysis. Am J Clin Nutr
2016;103:495–504.
Twitter Malavika A Subramanyam @MalavikaSocEpi 8 Dewey KG, Oaks BM. U-­shaped curve for risk associated with
maternal hemoglobin, iron status, or iron supplementation. Am J Clin
Acknowledgements  The authors acknowledge the immensely helpful contribution Nutr 2017;106:1694S–702.
during the data collections and advice of Cara Ebert, Abhijeet Kumar and Mareike 9 Balarajan Y, Ramakrishnan U, Özaltin E, et al. Anaemia in low-­income
Schön. We further acknowledge support by the Open Access Publication Funds of and middle-­income countries. The Lancet 2011;378:2123–35.
the University of Göttingen. 10 Pasricha S-­R, Black J, Muthayya S, et al. Determinants of anemia
among young children in rural India. Pediatrics 2010;126:e140–9.
Contributors  EH, MS and SV designed research and the survey instruments, CM 11 Georgieff MK. Iron deficiency in pregnancy. Am J Obstet Gynecol
designed the child skill assessment. SV and MS coordinated the first wave data 2020;223:516–24.
collection, EH coordinated the second wave data collection, with support of CM 12 Lozoff B. Iron deficiency and child development. Food Nutr Bull
for the child skill assessment. EH analysed the data and wrote the first draft of the 2007;28:S560–71.
manuscript. All authors provided input to the manuscript and approved the final 13 Lukowski AF, Koss M, Burden MJ, et al. Iron deficiency in infancy
version. EH is the guarantor. and neurocognitive functioning at 19 years: evidence of long-­term
deficits in executive function and recognition memory. Nutr Neurosci
Funding  The study was approved by the ethics board of the IIT Gandhinagar 2010;13:54–70.
(IEC/2014-­15/2/MS /MS/006, IEC/2016-­17/2/MS /MS/025) and the University of 14 Chaparro CM. Setting the stage for child health and
Goettingen (no IRB number available). We obtained written informed consent for development: prevention of iron deficiency in early infancy. J Nutr
the interviews, growth measurements, blood sample collections and the child 2008;138:2529–33.
development testing during each wave. 15 Koura KG, Ouédraogo S, Cottrell G, et al. Maternal anaemia at
delivery and haemoglobin evolution in children during their first 18
Competing interests  None declared. months of life using latent class analysis. PLoS One 2012;7:e50136.
16 Abioye AI, McDonald EA, Park S, et al. Maternal anemia type during
Patient consent for publication  Not required. pregnancy is associated with anemia risk among offspring during
Ethics approval  This work was supported by the German Research Foundation infancy. Pediatr Res 2019;86:396–402.
(DFG) (Research Training Group 1723) and 3ie International Initiative for Impact 17 Kumar A, Rai AK, Basu S, et al. Cord blood and breast milk iron
Evaluation (CPW.01.GV.IE). The study was prepared in parts during the PhD status in maternal anemia. Pediatrics 2008;121:e673–7.
18 Leslie MS, Park J, Briggs LA, et al. Is anemia in low income pregnant
Fellowship at the United Nations University World Institute for Development
women related to their infants' having anemia? a cohort study of
Economic Research (UNUWIDER). pregnant Women-­Infant pairs in the United States. Matern Child
Provenance and peer review  Not commissioned; externally peer reviewed. Health J 2020;24:768–76.
19 Zhang Y, Jin L, Liu J-­M, et al. Maternal hemoglobin concentration
Data availability statement  Data access can be obtained through reasonable during gestation and risk of anemia in infancy: secondary analysis of
request from the corresponding authors. Data is not yet available to the public. a randomized controlled trial. J Pediatr 2016;175:106–10.
20 Janbek J, Sarki M, Specht IO, et al. A systematic literature review of
Supplemental material  This content has been supplied by the author(s). It has
the relation between iron status/anemia in pregnancy and offspring
not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been neurodevelopment. Eur J Clin Nutr 2019;73:1561–78.
peer-­reviewed. Any opinions or recommendations discussed are solely those 21 McCann S, Perapoch Amadó M, Moore SE. The role of iron in brain
of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and development: a systematic review. Nutrients 2020;12:2001.
responsibility arising from any reliance placed on the content. Where the content 22 Angulo-­Barroso RM, Li M, Santos DCC, et al. Iron supplementation
includes any translated material, BMJ does not warrant the accuracy and reliability in pregnancy or infancy and motor development: a randomized
of the translations (including but not limited to local regulations, clinical guidelines, controlled trial. Pediatrics 2016;137:e20153547.
terminology, drug names and drug dosages), and is not responsible for any error 23 Santos DCC, Angulo-­Barroso RM, Li M, et al. Timing, duration, and
severity of iron deficiency in early development and motor outcomes
and/or omissions arising from translation and adaptation or otherwise.
at 9 months. Eur J Clin Nutr 2018;72:332–41.
Open access  This is an open access article distributed in accordance with the 24 Tran TD, Tran T, Simpson JA, et al. Infant motor development in rural
Creative Commons Attribution Non Commercial (CC BY-­NC 4.0) license, which Vietnam and intrauterine exposures to anaemia, iron deficiency and
permits others to distribute, remix, adapt, build upon this work non-­commercially, common mental disorders: a prospective community-­based study.
BMC Pregnancy Childbirth 2014;14:8.
and license their derivative works on different terms, provided the original work is
25 Hanieh S, Ha TT, Simpson JA, et al. Effect of low-­dose versus
properly cited, appropriate credit is given, any changes made indicated, and the use higher-­dose antenatal iron supplementation on child health
is non-­commercial. See: http://creativecommons.org/licenses/by-nc/4.0/. outcomes at 36 months of age in Viet Nam: longitudinal follow-­
up of a cluster randomised controlled trial. BMJ Glob Health
ORCID iD 2017;2:e000368.
Esther Heesemann http://orcid.org/0000-0001-9453-035X 26 Tran TD, Biggs B-­A, Tran T, et al. Impact on infants' cognitive
development of antenatal exposure to iron deficiency disorder and
common mental disorders. PLoS One 2013;8:e74876.
27 Chang S, Zeng L, Brouwer ID, et al. Effect of iron deficiency anemia
REFERENCES in pregnancy on child mental development in rural China. Pediatrics
1 Morgane PJ, Austin-­LaFrance R, Bronzino J, et al. Prenatal 2013;131:e755–63.
malnutrition and development of the brain. Neurosci Biobehav Rev 28 Parsons AG, Zhou SJ, Spurrier NJ, et al. Effect of iron
1993;17:91–128. supplementation during pregnancy on the behaviour of children at
2 Stevens GA, Finucane MM, De-­Regil LM, et al. Global, regional, early school age: long-­term follow-­up of a randomised controlled
and national trends in haemoglobin concentration and prevalence trial. Br J Nutr 2008;99:1133–9.
of total and severe anaemia in children and pregnant and non-­ 29 Zhou SJ, Gibson RA, Crowther CA, et al. Effect of iron
pregnant women for 1995-­2011: a systematic analysis of population-­ supplementation during pregnancy on the intelligence quotient
representative data. Lancet Glob Health 2013;1:e16–25. and behavior of children at 4 Y of age: long-­term follow-­up of a
3 Allen LH. Biological mechanisms that might underlie iron's effects on randomized controlled trial. Am J Clin Nutr 2006;83:1112–7.
fetal growth and preterm birth. J Nutr 2001;131:581S–9. 30 IIPS. National family health survey 2015-­16 district fact sheet
4 Burke RM, Leon JS, Suchdev PS. Identification, prevention and Madhepura Bihar, 2017. Available: http://rchiips.org/NFHS/FCTS/BR/​
treatment of iron deficiency during the first 1000 days. Nutrients BR_FactSheet_213_Madhepura.pdf
2014;6:4093–114. 31 Yadav K, Kant S, Ramaswamy G, et al. Validation of point of care
5 De Pee S, Bloem MW, Sari M, et al. The high prevalence of low hemoglobin estimation among pregnant women using digital
hemoglobin concentration among Indonesian infants aged 3-­5 Hemoglobinometers (HemoCue 301 and HemoCue 201+) as
months is related to maternal anemia. J Nutr 2002;132:2215–21. compared with Auto-­Analyzer. Indian J Hematol Blood Transfus
6 Shao J, Lou J, Rao R, et al. Maternal serum ferritin concentration is 2020;36:342–8.
positively associated with newborn iron stores in women with low 32 Sharma JB, Shankar M. Anemia in pregnancy. JIMSA
ferritin status in late pregnancy. J Nutr 2012;142:2004–9. 2010;23:253–60.

Heesemann E, et al. BMJ Open 2021;11:e046802. doi:10.1136/bmjopen-2020-046802 9

 Open access

33 World Health Organization. Haemoglobin concentrations for the 43 Patel A, Prakash AA, Das PK, et al. Maternal anemia and
diagnosis of anaemia and assessment of severity. Geneva: World underweight as determinants of pregnancy outcomes: cohort study
Health Organization, 2011. http://www.who.int/vmnis/indicators/​ in eastern rural Maharashtra, India. BMJ Open 2018;8:10.1136/
haemoglobin.pdf bmjopen-­2018-­021623.
34 Albers CA, Grieve AJ. Test review: Bayley, N. (2006). Bayley scales 44 Maguire JL, Salehi L, Birken CS, et al. Association between
of infant and toddler Development– third edition. San Antonio, tx: total duration of breastfeeding and iron deficiency. Pediatrics
Harcourt assessment. J Psychoeduc Assess 2007;25:180–90. 2013;131:e1530–7.
35 Mähler C, Cartschau F, Rohleder K. Frühkindliche 45 Ekiz C, Agaoglu L, Karakas Z, et al. The effect of iron deficiency
Entwicklungsdiagnostik für Kinder von 0-­3 Jahren (FREDI 0-­3). anemia on the function of the immune system. Hematol J
Hogrefe: Goettingen, 2016. 2005;5:10.1038/​sj.​thj.​6200574:​579–83.
36 Hasselhorn M. FREDI 0–3. Frühe Bildung 2017;6:101–3. 46 Levy A, Fraser D, Rosen SD, et al. Anemia as a risk factor for
37 World Health Organization, UNICEF. WHO child growth standards infectious diseases in infants and toddlers: results from a prospective
and the identification of severe acute malnutrition in infants and study. Eur J Epidemiol 2005;20:277–84.
children: a joint statement by the World Health Organization and the 47 Kosek MN, Ahmed T, Bhutta Z, MAL-­ED Network Investigators.
United Nations Children’s Fund, 2009. Available: http://www.ncbi.​ Causal pathways from enteropathogens to environmental
nlm.nih.gov/books/NBK200775/ [Accessed 26 Sep 2020]. enteropathy: findings from the MAL-­ED birth cohort study.
38 Barcellos SH, Carvalho LS, Lleras-­Muney A. Child gender and EBioMedicine 2017;18:109–17.
parental investments in India: are boys and girls treated differently? 48 Ngure FM, Reid BM, Humphrey JH, et al. Water, sanitation,
Am Econ J Appl Econ 2014;6:157–89. and hygiene (wash), environmental enteropathy, nutrition, and
39 Ebert C, Vollmer S. Child-­specific son preference, birth order and early child development: making the links. Ann N Y Acad Sci
cognitive skills in early childhood. Courant Research Centre “Poverty, 2014;1308:118–28.
Equity and Growth.”, 2019. 49 Watanabe K, Petri WA. Environmental enteropathy: elusive but
40 Fitzgerald J, Gottschalk P, Moffitt R. An analysis of sample attrition in significant subclinical abnormalities in developing countries.
panel data: the Michigan panel study of income dynamics. Working EBioMedicine 2016;10:25–32.
paper. National Bureau of Economic Research, 1998. 50 Knudsen EI. Sensitive periods in the development of the brain and
41 Mireku MO, Davidson LL, Koura GK, et al. Prenatal hemoglobin behavior. J Cogn Neurosci 2004;16:1412–25.
levels and early cognitive and motor functions of One-­Year-­Old 51 Prado EL, Dewey KG. Nutrition and brain development in early life.
children. Pediatrics 2015;136:e76–83. Nutr Rev 2014;72:267–84.
42 Nair M, Choudhury MK, Choudhury SS, et al. Association between 52 Walker SP, Wachs TD, Grantham-­McGregor S, et al. Inequality
maternal anaemia and pregnancy outcomes: a cohort study in in early childhood: risk and protective factors for early child
Assam, India. BMJ Glob Health 2016;1:e000026. development. Lancet 2011;378:1325–38.

10 Heesemann E, et al. BMJ Open 2021;11:e046802. doi:10.1136/bmjopen-2020-046802