ArticlePDF Available

Maternal Time Use and Nurturing: Analysis of the Association Between Breastfeeding Practice and Time Spent Interacting with Baby

Authors:

Abstract and Figures

Background: Breastfeeding supports child development through complex mechanisms that are not well understood. Numerous studies have compared how well breastfeeding and nonbreastfeeding mothers interact with their child, but few examine how much interaction occurs. Subjects and methods: Our study of weekly time use among 156 mothers of infants aged 3-9 months investigated whether lactating mothers spend more time providing emotional support or cognitive stimulation of their infants than nonbreastfeeding mothers, and whether the amount of such interactive time is associated with breastfeeding intensity. Mothers were recruited via mother's and baby groups, infant health clinics, and childcare services, and used an electronic device to record their 24-hour time use for 7 days. Sociodemographic and feeding status data were collected by questionnaire. Statistical analysis using linear mixed modeling and residual maximum likelihood analysis compared maternal time use for those giving "some breastfeeding" and those "not breastfeeding." Analysis was also conducted for more detailed feeding subgroups. Results: Breastfeeding and nonbreastfeeding mothers had broadly similar socioeconomic and demographic characteristics. Breastfeeding was found to be associated with more mother-child interaction time, a difference only partially explained by weekly maternal employment hours or other interactive care activities such as play or reading. Conclusion: This study presents data suggesting that lactating mothers spent significantly more hours weekly on milk feeding and on carrying, holding, or soothing their infant than nonlactating mothers; and on providing childcare. Understanding the mechanisms by which child mental health and development benefits from breastfeeding may have important implications for policies and intervention strategies, and could be usefully informed by suitably designed time use studies.
Content may be subject to copyright.
Breastfeeding, maternal time use and nurturing
1
Full title: Maternal time use and nurturing; Analysis of the association between breastfeeding
practice and time spent interacting with baby
Running title: Breastfeeding, maternal time use and nurturing
Authors names and academic degrees:
Julie P Smith, PhD, B Econs(hons)/BA
i
Robert Forrester, MSc, Dip Ed
ii
Word count for abstract and text: 4614
Number of Tables: 5 ; Number of Figures: 1
Affiliation name and address for correspondence; Associate Professor Julie P Smith, Menzies
Centre for Health Policy, School of Regulation and Global Governance (RegNet), College of
Asia and the Pacific, H.C. Coombs Extension Building #8, Fellows Road, The Australian
National University, Canberra, ACT, 0200.
Telephone number; and e-mail address: T:+ 61 416 099 630: E: julie.smith@anu.edu.au
Keywords: Breastfeeding, Maternal-infant interactions, Maternal time use, Child care, Oxytocin,
Weaning
Funding support: The Australian Research Council (Discovery Project DP0451117)
contributed funding for conduct of the TUSNM alongside an Australian Postdoctoral Fellowship.
i
Menzies Centre for Health Policy, The Australian National University, Canberra, Australia.
ii
Statistical Consulting Unit, The Australian National University, Canberra, Australia..
Page 1 of 37
Breastfeeding, maternal time use and nurturing
2
Maternal time use and nurturing; Analysis of the association between breastfeeding
practice and time spent interacting with baby
ABSTRACT
Breastfeeding supports child development through complex mechanisms which are not well
understood. Numerous studies have compared how well breastfeeding and non-breastfeeding
mothers interact with their child, but few examine how much interaction occurs. Our study of
weekly time use among 156 mothers of infants aged 3-9 months investigated whether lactating
mothers spend more time providing emotional support or cognitive stimulation of their infants
than non-breastfeeding mothers, and whether the amount of such interactive time is associated
with breastfeeding intensity. Mothers were recruited via mother’s and baby groups, infant health
clinics and childcare services, and used an electronic device to record their 24-hour time use for
7 days. Socio-demographic and feeding status data was collected by questionnaire. Statistical
analysis using linear mixed modelling and residual maximum likelihood analysis compared
maternal time use for those giving ‘some breastfeeding’ and those ‘not breastfeeding’. Analysis
was also conducted for more detailed feeding subgroups. Breastfeeding and non-breastfeeding
mothers had broadly similar socio-economic and demographic characteristics. Breastfeeding was
found to be associated with more mother-child interaction time, a difference only partially
explained by weekly maternal employment hours or other interactive care activities such as play
or reading. This study presents data suggesting that lactating mothers spent significantly more
hours weekly on milk feeding and on carrying, holding or soothing their infant than non-lactating
mothers; and on providing childcare. Understanding the mechanisms by which child mental
health and development benefits from breastfeeding may have important implications for
Page 2 of 37
Breastfeeding, maternal time use and nurturing
3
policies and intervention strategies, and could be usefully informed by suitably designed time use
studies.
Page 3 of 37
4
Introduction
Breastfeeding is recommended by health authorities for promoting child and maternal health and
wellbeing.
1-7
A growing body of high quality research shows improved child neurological and
cognitive development and mental health associated with breastfeeding in early life.
8-11
The mechanisms by which breastfeeding contributes to such outcomes are not well understood.
Some argue that the link with cognitive development gains simply reflects breastfeeding as a
proxy for more nurturing parenting and socioeconomic advantage.
12, 13
Nevertheless, recent
investigations do not support the ‘nurturing hypothesis’,
14, 15
and have urged the investigation of
other mechanisms for the observed effect of breastfeeding on child development outcomes.
An alternative mechanism commonly cited is that unique biochemical components of human
milk, such as fatty acids influence infant brain development and attachment (the ‘nutritional
hypothesis’).
14
However, the importance of epigenetic processes is also increasingly
recognized.
16
By altering maternal behaviors, lactation itself may influence the early social and
environmental experience (the ecology) of the child.
17-19
For example maternal affectionate
touch is known to be crucial for infant neurobiological, cognitive, and social-emotional
growth,
20, 21
and it has been suggested that the physical and/or emotional act of breastfeeding
may result in permanent physiologic changes which accelerate the infant’s neurocognitive
development.
22
Both animal and human studies show that lactation hormones such as oxytocin and prolactin
affect maternal behaviors and increase maternal nurturing, including increasing proximity
seeking.
6, 23-26
Mothers influenced by lactation hormones to seek proximity to their infant might
spend more time interacting with the child. While existing research has compared the quality of
interaction of breastfeeding and non-breastfeeding mothers, there has been no study comparing
Page 4 of 37
5
how much time mothers spend interacting with their infants. A novel way to explore the
potential bio-behavioral development pathway from breastfeeding to child development is
therefore by investigating whether the amount of maternal time spent in interactive child-caring
activities differs by infant feeding practice. The aim of this study was therefore to investigate
whether there are significant differences between breastfeeding and non-breastfeeding mothers
in the amount of time they spend in activities involving close physical interaction with their
infant such as emotional support or cognitive stimulation. Because we expected that mothers
spend less time on time-consuming interactive care activities such as feeding and carrying as
their infant matures,
27, 28
our objectives for the study included distinguishing how the effects of
infant feeding practice on maternal time use change as the infant gets older.
An important example is maternal employment, which is known to be associated with infant age
and infant feeding practice.
29, 30
As physical separation of mother and infant limits the number of
hours that mother and baby are together, our analysis explored how the interaction of infant age
and feeding practice was related to maternal time use activities, including maternal employment
hours.
Methodology
Participants
The Time Use Survey of New Mothers (TUSNM) was a nationwide Australian study conducted
through the Australian National University (ANU) between April 2005 and April 2006. All
participants gave written informed consent before enrolment (Protocol 2005/51 approved on 10
March 2005 by the ANU Human Research Ethics Committee under the National Statement on
Ethical Conduct in Research Involving Humans (1999)). The survey purpose was described to
participants as ‘measuring the time it takes to care for a baby’. Recruitment was through national
Page 5 of 37
6
playgroup and breastfeeding support organizations, maternal and child health professional
networks, infant health clinics, and childcare centers. Mothers with infants up to age nine
months were eligible to participate in the survey and could participate in tracking sessions at
time points when the target infant age was three, six and/or nine months.
Data collection and measures
Data on infant feeding method for the youngest child (the ‘target infant’) during seven days of
time use tracking was collected by questionnaire filled out by participant mothers. We excluded
participants who recorded time use for less than one 24-hour day. Feeding method was self-
categorized as (A) exclusive breastfeeding; (B) exclusive formula feeding; (C) mixed breastmilk
and formula milk – no solids; (D) breastmilk and solids; (E) formula milk and solids; (F) mixed
breastmilk and formula milk with solids. Self-report of feeding method was verified by cross
checking against individual time use data on feeding activities. Mothers using breast pumps
were instructed to record expressing breastmilk as ‘preparing feeds’; these few mothers were
categorized as breastfeeding mothers even though feeding this milk might be by someone other
than the mother.
Data collected from the mother by questionnaire also included the number of hours and minutes
another family member (usually her partner or husband) was caring for, and feeding the infant,
and how many hours the infant spent in paid childcare. Fathers were not invited to participate in
time use tracking mainly because of ethical concerns to limit response burdens on households.
Socio-demographic data were also collected via the written questionnaire.
Participants were asked to track their time use for seven days, 24 hours a day, using
TimeCorder
time tracking devices. These were posted to the mothers along with the
aforementioned questionnaire at each tracking time point, that is, within two weeks of the
Page 6 of 37
7
relevant anniversary of the target infant’s birthdate. Participants could record at target infant age
three months, six months and/or nine months. Data on the frequency, duration, and time of day
of each activity was recorded through participants pressing one of 25 buttons on the device
corresponding to their current activity.
Measures
The factor of interest was infant feeding practice and this was measured using two different
categorisations. Maternal hours in ‘interactive feeding and emotional care’ were compared for
the 6 detailed feeding categories noted above (A to F). We also compared those giving ‘some’
breastfeeding (A, C, D, F) with those ‘not’ breastfeeding (B & E), that is, ‘lactating’ with ‘non-
lactating’ mothers.
TUSNM design and measurement of time use was based on the official Australian Bureau of
Statistics (ABS) Time Use Survey (TUS).
31
The TUS defines ‘Care of children’ as comprising
the sum of ‘Physical’ and ‘Emotional’ Care, plus ‘all other interactive and non-interactive
childcare’ activities.
iii
‘Physical care’ in TUS includes feeding, holding, bathing, dressing,
changing, and teeth cleaning, while ‘Emotional care’ is cuddling, hugging or soothing a child.
The TUSNM created subcategories to these ‘main activity’ categories in the TUS to more fully
describe infant care activities. The relevant activity categories are illustrated in Figure 1. For
example, consistent with TUS, time interacting with the infant to ‘breastfeed or give expressed
milk’ or ‘feed formula’ was categorized under ‘Physical Childcare’. ‘Physical care’ in TUSNM
measures bathing, dressing, changing, and teeth cleaning, but not feeding. Time taken to ‘carry,
soothe, hold’ the infant was an interaction labelled ‘Emotional care’. Other childcare activities
iii
This refers to household care and does not mean non-parental or out of home care.
Page 7 of 37
8
recorded by TUS include ‘teaching, helping, reprimanding children’, and ‘playing with, reading,
talking to the child’; corresponding interactive care activities in TUSNM included playing,
reading, talking to, or teaching the infant (‘play teach’), and teaching infants to eat solids (‘solid
feeding’).
Time use was measured for mothers of infants aged 3-9 months for all interactive care activities
expected to involve emotional support and/or cognitive stimulation. A dependent variable was
created and labelled ‘interactive feeding and emotional care’, measured as the sum of maternal
time spent interacting with the infant through breastfeeding, feeding formula, breastmilk or
solids (‘interactive feeding’), plus ‘emotional care’ (time recorded as ‘carrying, holding,
soothing, or hugging the infant’); it excluded ‘preparing feeds’.
To provide context for differences in time spent on these activities, we also report analyses of
data on maternal time spent in playing with or teaching the infant, other physical care,
childcaring activity in total, and hours of paid employment, as well as how much time others had
spent caring for, or feeding the infant during the tracking period.
Data analysis
T-tests, chi-square and analysis of variance techniques were used to compare socio-demographic
characteristics of the two infant feeding groups. Characteristics of participants who provided
time tracking data records at one, two or three time points (that is, at infant ages three, six,
and/or nine months) were also compared, using the same techniques.
We used a residual maximum likelihood (REML) analysis in GenStat
32
to perform a linear
mixed effects analysis of the relationship between infant feeding category, and the specified
maternal time use activities, for participants observed at three time points. Data for time use
variables were transformed by taking the square root prior to analysis, to stabilize the variance.
Page 8 of 37
9
A repeated measures approach to statistical analysis was appropriate because at least one but up
to three successive time use and sociodemographic data observations were provided by
participants when the target infant was aged 3, 6 and/or 9 months, and because the numbers of
observations in the cells of the two-way tables were highly variable. A REML mixed effect
model was preferred over traditional approaches such as repeated measures analysis of variance
(ANOVA). This technique addresses the repeated measures and unbalanced nature of the data,
exploiting all available observations for each participant at the three time points, rather than
omitting information from participants with missing data. With the focus of the study being the
effect of infant feeding practice on maternal time use, mixed effect modelling allowed us to
estimate the separate (main) effects of feeding category and infant age, whilst also exploring
how the interaction of these two (fixed) factors affected maternal time use.
REML analyses were conducted for the two factors of interest described earlier; ‘detailed
feeding category’ (six levels), and breastfeeding (including mixed) versus non-breastfeeding
(two levels). The fixed terms in the model were “infant age” * “factor of interest”, and the
interaction between these. The random terms were “target infant”/ “infant age” (which expands
to “target infant” and “infant age” within “target infant”).
The significance of fixed effects (p<0.05) was assessed on the transformed data using Wald
statistics and approximate F-statistics. This is appropriate for assessing the significance of both
the main effects and the interaction term in the model for fixed effects specified above.
Results
Of 185 mothers giving consent, 162 participated in postnatal time use tracking, and 156
successfully contributed time use and socio-demographic data records, generating 327 usable
data records of maternal time use activities and socio-demographic data. The number of data
Page 9 of 37
10
records/observations exceeds the number of participants because most provided time use
tracking data at more than time point.
Table 1 summarizes the dataset presenting information on the age of the target infants at the
maternal time use tracking time points, the feeding categories of the target infants at that time
point, and tracking categories (describing the number of participants who tracked at one, two or
three points). The dataset of 327 data records contains 86 observations for mothers whose infant
was three month old at time of tracking (26.3%), 134 for those with six month old infants
(41.0%), and 107 for those with nine month olds (32.7%), Of the 327 observations, 95 (29.1%)
were from mothers when they were exclusively breastfeeding, 185 (56.6%) from those when
breastfeeding with solids, and the rest (14.3%) were when the child was mixed breastfed and
formula fed (20) or formula fed, with or without solids (27). These observations are for 156
individual infants who may be in different feeding groups at different ages, for example, being
exclusively breastfed at 3 months, but contributing an observation in the ‘breastfeeding with
solids’ feeding group when time use is tracked at 6 or 9 months. Regarding tracking categories,
those (mothers) who did a single tracking contributed 31 observations, 134 observations were
from those who tracked twice, and 162 of the 327 observations (49.5%) were from a participant
who tracked three times.
The TUSNM sample population had similar characteristics to the Australian population of
mothers of infants on most key socio-demographic variables, though participants were more
likely to be first-time mothers, and more highly educated. TUSNM also contained a higher
prevalence of breastfeeding mothers than would be expected from population based studies of
breastfeeding in Australia.
33, 34
Page 10 of 37
11
Table 2 reports analyses comparing socio-demographic characteristics of participants by infant
feeding group during the week of time use tracking. Maternal age, number of children, and age
of second youngest child were not significantly different between the breastfeeding and non-
breastfeeding groups, although a difference in mean infant age approached statistical
significance (p=.10). There was no significant difference between the breastfeeding and non-
breastfeeding groups in the proportion with only one child. Differences between the feeding
groups in maternal employment, education levels, and family income were also not significant,
though there was a trend for maternal education differences (p=.06). Comparison of socio-
demographic characteristics for those doing one, two and three time use trackings also showed
no statistically significant differences in age of the target child, age of the 2
nd
youngest child,
number of children, mother’s age, mother’s education, or mother’s employment status between
mothers who provided data at one, two, or three time points. However, there was a trend for a
difference between groups in family income category (p=.05).
Table 3 reports predicted means (with back-transformed means in parentheses) for maternal time
spent in various activities, by detailed infant feeding group, and by infant age. On average,
mothers spent 38-49 hours a week on childcare.
iv
There were statistically significant differences
(p < .01) of several hours a week between feeding groups in maternal weekly hours spent on
emotional care. Those exclusively breastfeeding spent the most time (8.3 hours weekly) and
those who were formula feeding with solids spent least time (3.2 hours) in emotional care
activity. Differences between the six feeding groups in total interactive feeding and emotional
iv
Note that back-transformed means will be similar, but not the same as the means for the original data, due to the
transformation and the unbalanced nature of the data. Hence, for example, ‘interactive feeding and emotional care’
is the sum of time spent feeding the infant (i.e. ‘milk feeding’ + ‘solid feeding’) plus ‘emotional care’, but the
figures for each of these individual activities do not add up exactly to the total mean value for the combined
activities.
Page 11 of 37
12
care time (p<.01) mainly arise from differences in weekly hours spent on milk feeding, and on
emotional care, as time spent in feeding solids was not large and did not differ greatly between
feeding groups. For example, the mothers who were breastfeeding with solids spent 22.5 hours a
week on interactive feeding and emotional care, while those who were formula feeding with
solids spent 12.1 hours a week in these activities. Small cell sizes for formula fed infants (n=3,
n=5, n=17), some of whom were also breastfed, suggest the need for caution in interpreting
differences, as differences between any particular detailed feeding group categories may or may
not be statistically significant.
Overall, there were also no statistically significant differences between the feeding groups on
time spent on playing with the infant, or on maternal childcaring activity as a whole. Nor were
there any significant differences when maternal education was included as a covariate in the
analyses in Tables 3 and 4.
Maternal time spent in infant care activities was clearly affected by infant age. Time spent with
infants in interactive feeding and emotional care activity was significantly less (p<.001) for
older than for younger infants (15.5 vs 22.5 hours weekly).
Table 4 compares lactating mothers with non-lactating mothers. It shows that lactating mothers
spent more hours on milk feeding (12.6 hours, p<0.001) and emotional care (7.3, p<0.01) than
non-lactating mothers (6.2 vs. 3.9). Lactating mothers also spent 8.5 hours more weekly in
childcare activity (p<0.01). Maternal time in other interactive care of infants (feeding solids,
playing or reading, and physical care) did not differ significantly. Average weekly employment
hours were higher for lactating mothers (2.7 vs. 1.2 hours, p<0.01).
The time that infants were cared for by someone other than the mother was not large, as shown
in Table 4. Differences in weekly time that other family members cared for the infant were not
Page 12 of 37
13
statistically significant, though infants of non-breastfeeding mothers spent more time in paid
childcare (average of 119.5 minutes weekly vs. 6.2, p<0.001). Notably, but not unexpectedly,
non-breastfed infants spent more time being fed by other family members and the difference was
statistically significant (52 minutes weekly vs. 12 minutes,).
Mothers of older infants spent less time on milk feeding, more time solid feeding, and less time
in emotional care than those with younger infants (p<0.01). Mothers of older infants also spent
less time in physical care, and in childcare overall, and more time in employment (.7 hours at 3
months compared to 4.2 hours at 9 months).
Testing using Wald statistics and approximate F-statistics did not reveal statistically significant
interactions of target infant age and feeding group in effects on maternal time use activities,
except for employment. Infant feeding category interacted significantly with age of infant for
maternal employment hours (p=.006). We therefore calculated predicted means for maternal
employment hours to explore how this related to age of target infant and feeding group (Table
5). Notably, employment hours were low for all the mothers of younger infants regardless of
feeding group, with less than one hour a week on average for those with infants aged three
months. At nine months, however, those breastfeeding with solids spent 1.3 hours weekly in
employment whereas those feeding formula and solids spent 5.9 hours. Mothers who were both
breastfeeding and formula feeding also had high hours in employment at this age (8.0 hours
weekly). Exclusively formula feeding mothers had the highest employment hours at nine
months, averaging around 48 hours weekly. Likewise, looking at the combined feeding groups,
at younger ages the differences between lactating and non-lactating mothers in hours of
employment were not substantial. However, at infant age 9 months, mothers who were not
Page 13 of 37
14
breastfeeding spent around 7.6 hours a week in employment while mothers who were still
breastfeeding spent 1.8 hours weekly in employment.
The above analysis shows that mothers in the breastfeeding groups spent several hours a week
more in close feeding-related interactions with their infant, than did mothers who were not
giving any breastmilk. This was not offset by time in other interactions with the infant such as
through playing, reading or physical care. Differences in overall time spent on childcare activity
between the feeding groups, of 8 hours a week, were also not accounted for by small differences
in employment hours, which averaged around 2 hours weekly.
Discussion
This study is the first to address the question of how lactation status relates to maternal time use,
and contributes new data on maternal time spent in interactive care of their infants. On average
mothers of infants aged 3-9 months spent the equivalent of a full time working week on
childcare activity, regardless of feeding method. The key finding from our primary analysis is
that time spent in interactive infant feeding or emotional care activities differed significantly by
infant feeding group. Specifically, lactating mothers spent 8.5 hours a week more time feeding,
carrying, holding, soothing or hugging the infant than non lactating mothers. This was not
compensated by additional time in other interactive care activities such as feeding solids, or
playing with the infant.
Our findings of differences of around 4.6 hours weekly in emotional care time by breastfeeding
mothers of infants aged 6 months are consistent with an Australian population-based study of
nearly 3000 infants’ time use which showed using multivariate analysis that infants who had
ceased breastfeeding by around five months of age spent 3-4 hours fewer hours each week being
cuddled or held by their mother than infants still breastfed at this age.
35
Page 14 of 37
15
Previous analyses have shown that exclusively breastfeeding mothers spend more time milk-
feeding
36
and cuddling, holding or soothing their infants
28
than mothers who are not exclusively
breastfeeding. Not addressed in existing research, however, is how the amount of time that
mothers spend touching their infants varies with the intensity of breastfeeding (i.e. the
relationship of breastfeeding “dose” to the amount of interactive activity) at key time points
during infancy. A second important finding is on the relationship between the intensity of
breastfeeding and the number of weekly hours of interactive feeding and emotional care time. In
this study, exclusively breastfed infants experienced the most hours of this interactive care time
with their mother, and exclusively formula fed infants on solids experienced the least, though
small cell sizes limit strong conclusions about comparisons for some mixed feeding categories.
This is consistent with how lactation hormones may be expected to alter maternal behavior
towards proximity seeking.
Thirdly, our consideration of maternal time use context showed that employment of mothers of
older babies is related to feeding method and may influence, or be influenced by, the amount of
time that mothers are in close interaction with their infant. Maternal labour force participation is
well established as influencing infant feeding practices.
30, 36, 37
The non-lactating mothers in this
Australian study spent slightly more time in employment than lactating mothers, but this small
difference did not fully account for the substantial difference in maternal interactive care time.
Our other important finding, that other family members helped less with the feeding of breastfed
infants, is not surprising, but reinforces that excessive time demands on the mother may result in
early weaning from breastfeeding if the mother has unsupportive family (or employment)
arrangements.
Page 15 of 37
16
The statistical analysis shows that the observed differences between feeding groups are very
unlikely to be explained by chance. Nor do they reflect socio-economic and demographic
differences (such as number of siblings, family type, maternal age or education, workforce
status, family income, State and rural-urban residence) between feeding groups.
The importance of this study is that no previous research has used maternal time use data to
explore potential mechanisms for differences in developmental outcomes for breastfed and non-
breastfed infants.
38
Existing time use studies of mother and infant interactions compare the
quality, rather than the amount, of mother-infant interactions; for example, the study by Lavelli
and colleagues compared twenty breastfed and twelve bottle-fed newborns and found different
mother-infant interactions during and just after feeding.
39
The dearth of research on maternal time use and infant feeding is partly because suitable time
use data on mother-infant dyads is rare.
40
Population-based stratified sample surveys of time use
do not identify activities of specific relevance to mothers with infants, such as feeding, while the
few existing studies of infants’ time use focus on activities such as playing, crying or sleeping
and lack data on general activities needed to give context.
41
Infant studies of this kind are also
often limited to very small, non-representative samples; some large time use datasets for infants
and older children exist but often do not allow comparison by infant feeding method,
42
and have
not been used to examine infant feeding.
43
The Australian Longitudinal Study of Australian
Children (LSAC) collects comprehensive data on both infant time use and breastfeeding status,
but not maternal time use.
35
A strength of this study is that it draws on a unique dataset which provides information on
sociodemographic characteristics and feeding status as well as comprensive data on maternal
time use for 24 hours a day over a full week, for a sizable sample of 156 mothers of infants. The
Page 16 of 37
17
use of time tracking devices rather than relying on written logs of activities or retrospective
reports is a major strength of this study. Previous studies of such mother-child interactions
lacked contexual data to explore, for example, how employment hours influence how much
time a mother can spend interacting with the infant. Our analysis was able to include the
interplay of such care time with the number of hours mothers spent in employment. Recent
research also illustrates the important role of fathers in supporting breastfeeding, including by
sharing housework and child care;
44, 45
using contextual data on time use of other caregivers
from the TUSNM, we could explore and compare associations between maternal time
interacting with the infant, and time contributed by other family caregivers, for breastfeeding
and non-breastfeeding mothers.
Our study has several limitations. It does not measure health or development outcomes arising
from maternal time use differences; such a study could provide important new insights. Nor was
it practicable to measure oxytocin or prolactin levels in TUSNM participants. The study’s
generalisability to different populations may be limited by self-selection bias, as participation
was voluntary. The small sample size limits the conclusions that can be drawn about the
statistical significance of comparisons between particular feeding subgroups. Time use of other
family members was by maternal report and may differ from time use measured through self
report or tracking using a time tracking device. Also, results from Australia where paid parental
leave (PPL) is 18 weeks and statutory job protection is for 12 months might differ in other
country settings having different maternal employment patterns and maternity leave access; the
average age of babies when their mothers return to employment in Australia is around 7
months.
46
While participants in the TUSNM had broadly homogenous socio-demographic
characteristics, there is potential confounding from missing variables, as mothers self-select into
Page 17 of 37
18
breastfeeding based on a complex combination of personal and social characteristics. These may
include parenting efficacy, mental health variables such as anxiety or depression, or the
availability of social and health services support. Reverse causation is also an alternative
explanation of our findings – the amount of time a mother spends interacting with her infant
may increase the likelihood of maintaining exclusive or partial breastfeeding.
In conclusion, to explore potential mechanisms behind differential child development and
mental health outcomes for breastfed and non-breastfed infants, we presented data showing that
lactating mothers spent significantly greater time interacting with the infant than non lactating
mothers, and such interaction increased with more exclusive breastfeeding. This is consistent
with lactation hormones influencing the amount of time mothers provide important interactive
emotional support or cognitive stimulation of their infant.
Future research - with study designs that combine time use data collection with neuroimaging
and physiological data, in larger, randomly selected populations, and with follow up of relevant
outcomes – would be of value to confirm whether lactation-driven maternal behaviour is an
important causal pathway through which breastfeeding contributes to child development and
health.
If our findings are generalisable, there are potentially major implications. Programs or
interventions to reduce child development disadvantage and promote child mental health might
achieve improved outcomes by including breastfeeding support for new mothers. Policies and
practices such as paid maternity leave or breastfeeding breaks that give mothers more time for
close positive interactions with their baby may be needed to protect appropriate development of
infants. Our study also suggests the importance for health professionals working with new
Page 18 of 37
19
mothers to advise of the need to ensure that weaning from exclusive or partial breastfeeding
does not prematurely reduce the mother’s time spent interacting positively with an infant.
Acknowledgements
The TimeCorder is Canadian technology of workplace productivity consultant, Pace
Productivity, Toronto, Canada, which supplied devices for use in time tracking.
The assistance
of Mark Ellwood, President of Pace Productivity, in initial design and conduct of the TUSNM is
gratefully acknowledged. The authors appreciate the important contribution of the new mothers
participating in the study. Research assistance and survey administration by Louise Bartlett is
also acknowledged.
Disclosure Statement
No competing financial interests exist.
Page 19 of 37
20
Page 20 of 37
21
References
1 Ip S, Mei Chung MPH, Gowri Raman, et al. Breastfeeding and maternal and infant
health outcomes in developed countries. AHRQ Publication No. 07-E007. Boston,
Massachusetts: Tufts-New England Medical Center Evidence-Based Practice Center,
2007.
2 Horta BL, Victora CG. Long-term effects of breastfeeding. Geneva: World Health
Organization, 2013.
3 National Health and Medical Research Council. Dietary Guidelines for Children and
Adolescents in Australia incorporating the Infant Feeding Guidelines for Health
Workers. Canberra: National Health and Medical Research Council, 2013.
4 American Academy of Pediatrics, Eidelman AI, Schanler RJ, et al. Breastfeeding and
the Use of Human Milk. Pediatrics 2012; 129:e827-e841.
5 World Health Organization/UNICEF (WHO/UNICEF). Global strategy for infant and
young child feeding. Geneva, Switzerland: World Health Organization (WHO)
UNICEF, 2003.
6 Strathearn L, Mamun AA, Najman JM, et al. Does Breastfeeding Protect Against
Substantiated Child Abuse and Neglect? A 15-Year Cohort Study. Pediatrics 2009;
123:483-493.
7 Chowdhury R, Sinha B, Sankar MJ, et al. Breastfeeding and maternal health
outcomes: a systematic review and meta-analysis. Acta Paediatr 2015; 104:96-113.
Page 21 of 37
22
8 Kramer MS, Aboud F, Mironova E, et al. Breastfeeding and child cognitive
development: new evidence from a large randomized trial. Archives of General
Psychiatry 2008; 65:578-84.
9 Horta BL, Loret de Mola C, Victora CG. Breastfeeding and intelligence: a systematic
review and meta-analysis. Acta Paediatr 2015; 104:14-9.
10 Luby JL, Belden AC, Whalen D, et al. Breastfeeding and Childhood IQ: The
Mediating Role of Gray Matter Volume. Journal of the American Academy of Child
& Adolescent Psychiatry 2016; 55:367-375.
11 Oddy WH, Kendall GE, Li J, et al. The Long-Term Effects of Breastfeeding on Child
and Adolescent Mental Health: A Pregnancy Cohort Study Followed for 14 Years.
The Journal of Pediatrics 2010; 156:568-574.
12 Der G, Batty GD, Deary IJ. Effect of breast feeding on intelligence in children:
prospective study, sibling pairs analysis, and meta-analysis. Bmj 2006.
13 Zhou SJ, Baghurst P, Gibson RA, et al. Home environment, not duration of breast-
feeding, predicts intelligence quotient of children at four years. Nutrition 2007;
23:236-41.
14 Huang J, Vaughn MG, Kremer KP. Breastfeeding and child development outcomes:
an investigation of the nurturing hypothesis. Maternal & child nutrition 2015.
15 Edwards RC, Thullen MJ, Henson LG, et al. The association of breastfeeding
initiation with sensitivity, cognitive stimulation, and efficacy among young mothers:
a propensity score matching approach. Breastfeed Med 2015; 10:13-9.
Page 22 of 37
23
16 McInerny TK. Breastfeeding, early brain development, and epigenetics--getting
children off to their best start. Breastfeed Med 2014; 9:333-4.
17 Uauy R, De Andraca I. Human milk and breast feeding for optimal mental
development. J Nutr 1995; 125:2278S-2280S.
18 Burgard P. Critical evaluation of the methodology employed in cognitive
development trials. Acta Paediatr Suppl 2003; 92:6-10.
19 Feldman R. Parent-infant synchrony: a biobehavioural model of mutual influences in
the formation of affliliative bonds. Monographs of the Society for Research in Child
Development 2012; 77:42-51.
20 Feldman R, Eidelman AI. Direct and indirect effects of breast milk on the
neurobehavioral and cognitive development of premature infants. Dev Psychobiol
2003; 43:109-19.
21 Feldman R. The handbook of touch: Neuroscience, behavioral, and health
perspectives. In: Hertenstein M, Weiss SJ, eds: Springer Publishing Company,
2011:373-408.
22 Kramer MS, Aboud F, Mironova E, et al. Breastfeeding and child cognitive
development: new evidence from a large randomized trial. Arch Gen Psychiatry
2008; 65:578-84.
23 Uvnas-Moberg K, Petersson M. [Oxytocin, a mediator of anti-stress, well-being,
social interaction, growth and healing]. Z Psychosom Med Psychother 2005; 51:57-
80.
Page 23 of 37
24
24 Tu MT, Lupien SJ, Walker C-D. Measuring stress responses in postpartum mothers:
Perspectives from studies in human and animal populations. Stress: The International
Journal on the Biology of Stress 2005; 8:19-34.
25 Carter C, Altemus M, Pchrousos G. Neuroendocrine and emotional changes in the
post-partum period. Progress in Brain Research 2001:133.
26 Kim P, Feldman R, Mayes LC, et al. Breastfeeding, brain activation to own infant
cry, and maternal sensitivity. Journal of Child Psychology and Psychiatry 2011;
52:907-915.
27 Smith JP, Forrester R. Who pays for the health benefits of exclusive breastfeeding?
An analysis of maternal time costs. Journal of Human Lactation 2013; 29:547-555.
28 Smith JP, Ellwood M. Feeding Patterns and Emotional Care in Breastfed Infants.
Social Indicators Research 2011; 101:227-231.
29 Mandal B, Roe BE, Fein SB. The differential effects of full-time and part-time work
status on breastfeeding. Health Policy 2010; 97:79-86.
30 Mandal B, Roe BE, Fein SB. Work and breastfeeding decisions are jointly
determined for higher socioeconomic status US mothers. Review of Economics of the
Household 2012; published online 28 July:1-21.
31 Australian Bureau of Statistics. Time Use Survey Australia: Users Guide. Canberra,
1997.
32 Payne RW, Murray DA, Harding SA, et al. GenStat for Windows (17th Edition)
Introduction. Hemel Hempstead: VSN International, 2014.
Page 24 of 37
25
33 Australian Institute of Family Studies (AIFS). Growing up in Australia: The
Longitudinal Study of Australian Children, Annual Report 2006-07, 2008.
34 Baxter J. Breastfeeding, employment and leave. Family Matters 2008; 80:17-26.
35 Baxter J, Smith JP. Breastfeeding and time use. Australian Institute of Family Studies
Research Paper No. 43. Melbourne, 2009
36 Smith JP, Forrester R. Who pays for the health benefits of exclusive breastfeeding?
An analysis of maternal time costs Journal of Human Lactation 2013; 29:547 - 555.
37 Baxter J, Cooklin AR, Smith J. Which mothers wean their babies prematurely from
full breastfeeding? An Australian cohort study. Acta Pædiatrica 2009; 98:1274-1277.
38 Golding J, Rogers IS, Emmett PM. Association between breast feeding, child
development and behaviour. Early Hum Dev 1997; 49 Suppl:S175-84.
39 Lavelli M, Poli M. Early mother-infant interaction during breast- and bottle-feeding.
Infant Behavior and Development 1998; 21:667-683.
40 Ben-Arieh A, Ofir A. Opinion, Dialogue, Review: Time for (More) Time-Use
Studies: Studying the Daily Activities of Children. Childhood 2002; 9:225.
41 St James-Roberts I. Infant Crying and Sleeping in London, Copenhagen and When
Parents Adopt a" Proximal" Form of Care. Pediatrics 2006; 117:e1146.
42 Ramey G, Ramey VA. The Rug Rat Race. Brookings Papers on Economic Activity
2010:129-176.
43 Drago R. The parenting of infants: a time-use study. Monthly Labor Review 2009;
132:33-43.
Page 25 of 37
26
44 Maycock B, Binns CW, Dhaliwal S, et al. Education and Support for Fathers
Improves Breastfeeding Rates: A Randomized Controlled Trial. Journal of Human
Lactation 2013; 29:484-490.
45 Rempel LA, Rempel JK. The breastfeeding team: the role of involved fathers in the
breastfeeding family. Journal of Human Lactation 2011; 27:115-121.
46 Australia. Department of Social Services. Paid Parental Leave scheme. Review
Report. Canberra, Australia: Australian Government, 2014.
Page 26 of 37
1
Table 1: Summary of data records/observations: detailed feeding group by age of target infant, and tracking category
Detailed feeding group Data records/observations by age of target child (months)
Data
records/observations
3
6
9
A Breastfed only 79
15
1
95 (29.1%)
B Formula only 4
0
1
5 (1.5%)
C Breastfed and formula fed 2
1
0
3 (0.9%)
D Breastfed and solids 1
101
83
185 (56.6%)
E Formula fed and solids 0
11
11
22 (6.7%)
F Breastfed and formula fed and solids 0
6
11
17 (5.2%)
All feeding groups 86 (26.3%)
134 (41.0%)
107 (32.7%)
327 (100%)
Tracking category
one tracking 9
15
7
31 (9.5%)
two trackings 23
65
46
134 (41.0%)
three trackings 54
54
54
162 (49.5%)
All tracking categories 86 (26.3%)
134 (41.0%)
107 (32.7%)
327 (100%)
Detailed feeding group Data records/observations by tracking
category Number of trackings (n)
Page 27 of 37
2
one
tracking
two
trackings
three
trackings
A Breastfed only 11
27
57
95
B Formula only 1
1
3
5
C Breastfed and formula fed 1
1
1
3
D Breastfed and solids 15
88
82
185
E Formula fed and solids 3
8
11
22
F Breastfed and formula fed and solids 0
9
8
17
All feeding groups
31 (9.5%)
134
(41.0%)
162 (49.5%)
327 (100%)
Page 28 of 37
1
Table 2: Socio-demographic characteristics of participants
Comparison by feeding group: t-test and chi-square analyses
No breastfeeding (feeding
groups 2 and 5), obs=27
Some breastfeeding (feeding
groups 1, 3 4, and 6), obs=299
p-value
Age of target child (months),
mean
6.9
6.1
0.10
Age of 2
nd
youngest child
(years), mean
3.7
3.7
0.98
Number of children (N),
mean
1.9
1.7
0.39
Target infant is only child
(%)
40.7
48.3
0.45
Mothers age (years), mean
31.7
32.8
0.17
Mothers education (N)i ii ≤ 12 years 4
17
0.06
> 12 years 23
282
Mothers employment status
(N)
ii
full time
or part time
1
0
8
5
0.
340
unemployed/not in
labour force
17
215
Family income (A$ weekly)
ii
0-899 13
113
0.53
900-1199 5
79
i
Excludes one record with missing data for these variables.
ii
1 cell has expected count less than 5, minimum expected is 1.74
Page 29 of 37
2
1200+/ no answer
8
102
Page 30 of 37
1
Table 3: Residual maximum likelihood analysis of maternal weekly hours spent in interactive care and other activities, by detailed feeding group and by
age of target infant, using linear mixed model. Predicted means with back-transformed means in parentheses.
a)
A. Maternal weekly hours spent in activity, by detailed feeding group (for all target infant ages)
Activity A. Breastfed
only (obs=95)
B. Formula
only (obs=5)
C. Breastfed
& formula
fed (obs=3)
D. Breastfed
& solids
(obs=185)
E. Formula
fed &solids
(obs=22)
F. Breastfed &
formula fed &
solids (obs=17)
Av. Standard
errors of
differences
p-
value
Physical care
2.369 (5.61)
2.216 (4.91)
2.415 (5.83)
2.373 (5.63)
2.220 (4.93)
2.255 (5.09)
0.252
NS
Milk feeding 3.812 (14.5)
3.233 (10.5)
3.173 (10.1)
3.422 (11.7)
2.218 (4.92)
3.219 (10.4)
0.394
<0.001
Solid feeding
.2039 (.0416)
.3041 (.0925)
.7021 (.493)
1.352 (1.83)
1.416 (2.01)
1.496 (2.24)
0.356
<0.001
Emotional Care 2.876 (8.27)
2.211 (4.89)
2.526 (6.38)
2.650 (7.02)
1.794 (3.22)
2.296 (5.27)
0.464
0.036
Interactive feeding plus
emotional care
4.996 (25.0)
4.242 (18.0)
4.176 (17.4)
4.740 (22.5)
3.484 (12.1)
4.374 (19.1)
0.410
<0.00
1
Play teach 2.480 (6.15)
2.652 (7.03)
2.610 (6.81)
2.620 (6.86)
2.582 (6.67)
2.289 (5.24)
0.470
NS
Childcare 7.034 (49.5)
6.228 (38.8)
6.437 (41.4)
6.899 (47.6)
6.172 (38.1)
6.581 (43.3)
0.448
NS
B. Maternal weekly hours spent in activity, by age of target infant (for detailed feeding group)
Age of target infant (months) Av. Standard
errors of
differences
p-value
Activity 3
(obs=86)
6
(obs=134)
9
(obs=107)
Page 31 of 37
2
Physical care
2.357 (5.56)
2.368 (5.61)
2.200 (4.84)
.1118
NS
Milk Feeding 3.568 (12.7) 3.196 (10.2) 2.774 (7.7) .1775
<0.001
Solid feeding
.7299 (.5328)
.9556 (.9132)
1.052 (1.128)
.1639
NS
Emotional Care 2.724 (7.42) 2.378 (5.65) 2.074 (4.30) .202
0.004
Interactive feeding and
emotional care
4.742 (22.49) 4.326 (18.7) 3.939 (15.5) .1807
<0.001
Play teach 2.581 (6.66) 2.456 (6.03) 2.583 (6.67) .2077
NS
Childcare 6.731 (45.3) 6.546 (42.9) 6.399 (40.9) .1967
NS
a) Note that back-transformed means will be similar, but not be the same as the means for the original data, due to the transformation and the unbalanced nature of the data.
Page 32 of 37
1
Table 4 Residual maximum likelihood analysis of maternal weekly hours spent in interactive care and other activities, by some or no breastfeeding and
by age of target infant,
using linear mixed model.
Predicted means with back-transformed means in parentheses.
a)
Maternal weekly hours spent in activity, by some or no breastfeeding
Breast feeding including mixed Av. Standard errors of
differences
p-value
Activity no breastfeeding
(obs=27)
Breastfeeding
(obs=300)
Physical care
2.365 (5.59)
2.233 (4.99)
.129
NS
Milk Feeding
2.493 (6.22)
3.553 (12.62)
.1977
<.001
Solid feeding
1.003 (1.006)
.9355 (.875)
.169
NS
Emotional Care
1.983 (3.93)
2.708 (7.33)
.253
.004
Interactive feeding and emotional care
3.732 (13.93)
4.805 (23.08)
.219
<.001
Play teach
2
.610 (6.81)
2.550 (6.50)
.242
NS
Childcare total
6.276 (39.39)
6.920 (47.89)
.238
.007
Employment
1.106 (1.223)
1.651 (2.726)
.418
.036
Context
Care by father (weekly mins.)
10.28 (105.7)
12.28
(150.80)
2.3
NS
Feeding by father (weekly mins.)
7.21 (51.98)
3.46
(11.97)
1.11
<.001
Page 33 of 37
2
Time in paid childcare (weekly mins.)
10.93 (119.5)
2.49
(6.20)
2.38
<.001
Maternal weekly hours spent in activity, by age of target infant (for some or no breastfeeding)
Age of target infant (months)
Activity 3
(obs=86)
6
(obs=134)
9
(obs=107)
Av. Standard errors of
differences
p-value
Physical care 2.352 (5.53)
2.360 (5.57)
2.184 (4.77)
.0624
0.004
Milk Feeding 3.677 (13.5)
2.930 (8.58)
2.462 (6.06)
.0981
<0.001
Solid feeding .086 (.0073)
1.301 (1.69)
1.517 (2.30)
.0992
<0.001
Emotional Care 2.838 (8.05)
2.273 (5.17)
1.924 (3.70)
.1069
<0.001
Interactive feeding and emotional care
4.862 (23.6)
4.189 (17.5)
3.755 (14.1)
.0977
<0.001
Play teach 2.561 (6.59)
2.530 (6.40)
2.648 (7.01)
.112
NS
Childcare total
6.863 (47.1)
6.555 (43.0)
6.376 (40.7)
.105
<0.001
Employment 0.828 (.686)
1.253 (1.57)
2.055 (4.22)
.368
0.003
a) Note that back-transformed means will be similar, but not be the same as the means for the original data, due to the transformation and the unbalanced nature of the data.
Page 34 of 37
1
Table 5: Residual maximum likelihood analysis of maternal weekly hours spent in employment using a linear mixed model. Predicted means with back-
transformed means in parentheses.
a)
Interaction between age of target infant and detailed feeding group (P=.006, Av. SED=1.172)
Age of target infant (months)
Feeding group
3
(obs=86)
6
(obs=134)
9
(obs=107)
A. Breastfed only (obs=95) 0.888 (0.789)
1.445 (2.088)
2.150 (4.62)
B. Formula only (obs=5) 0.675 (0.456)
-
6.915 (47.8)
C. Breastfed & formula fed (obs=3) 0.04 (0)
4.063 (16.51)
-
D. Breastfed & solids (obs=185) 0.504 (0.254)
1.043 (1.09)
1.127 (1.27)
E. Formula fed &solids (obs=22) -
1.285 (1.65)
2.432 (5.91)
F. Breastfed & formula fed & solids (obs=17)
-
1.097 (1.203)
2.821 (7.96)
Interaction between age of target infant and some or no breastfeeding (P=0.079, Av. SED=.5677)
Some breastfeeding (obs=300) .855 (.731)
1.117 (1.248)
1.346 (1.812)
No breastfeeding (obs=27) .800 (.640)
1.390 (1.932)
2.765 (7.645)
a) Note that back-transformed means will be similar, but not be the same as the means for the original data, due to the transformation and the unbalanced nature of the data.
Page 35 of 37
Figure 1: Relevant TUS and TUSNM time use activity categories, and summary measures
Page 36 of 37
259x154mm (300 x 300 DPI)
Page 37 of 37
... Lactation hormones also influence maternal nurturing behavior and desire for proximity to the infant, in animals as well as humans. For example, time use research on new mothers in Australia suggests differences in the amount of time they spent holding and soothing breastfed compared to non-breastfed infants and this was related to the degree of breastfeeding [28,29]. ...
... Time is increasingly recognised to be a resource needed for good health [32], including for nurturing care and breastfeeding of infants [28,33]. Who has access to (spare) time resources has been shown in a recent study to be significantly shaped by both socioeconomic class and gender [32]. ...
... Lactating mothers spent 8.5 h more weekly in childcaring activity (p = 0.007). As shown elsewhere [28], average weekly employment hours were higher for lactating mothers (2.7 vs. 1.2 h, p < 0.01), as were childcare hours (47.89 vs 39.39, p < .01). There were no significant differences in free time between lactating and non lactating mothers. ...
Article
Full-text available
Background Infant sleep is of great interest to new parents. There is ongoing debate about whether infants fed with breastmilk substitutes sleep longer than those exclusively or partially breastfed, but what does this mean for the mother? What expectations are realistic for mothers desiring to exclusively breastfeed as recommended by health authorities? There are both biological and social influences on infant and maternal sleep. More accurate information on average maternal sleep hours for diverse feeding practices may help guide realistic expectations and better outcomes for mothers, infants and families. Methods Using a unique time use dataset purposefully designed to study the time use of new mothers, this study investigated whether the weekly duration of maternal sleep, sleep disturbance, unpaid housework, and free time activities differed by detailed feeding method. The study collected 24/7 time use data from 156 mothers of infants aged 3, 6 and/or 9 months between April 2005 and April 2006, recruited via mother’s groups, infant health clinics, and childcare services throughout Australia. Sociodemographic and feeding status data were collected by questionnaire. Statistical analysis used linear mixed modelling and residual maximum likelihood analysis to compare effects of different infant feeding practices on maternal time use. Results There were no significant differences in time spent asleep between lactating and non lactating mothers, though lactating mothers had more time awake at night. Lactating mothers spent more time (8.5 h weekly) in childcaring activity ( p = 0.007), and in employment (2.7 vs. 1.2 h, p < 0.01), but there were no significant differences in free time. Those not breastfeeding spent more time in unpaid domestic work. Exclusive breastfeeding was associated with reduced maternal sleep hours (average 7.08 h daily). Again, free time did not differ significantly between feeding groups. Exclusively breastfeeding mothers experienced reduced sleep hours, but maintained comparable leisure time to other mothers by allocating their time differently. Domestic work hours differed, interacting in complex ways with infant age and feeding practice. Conclusions Optimal breastfeeding may require realistic maternal sleep expectations and equitable sharing of paid and unpaid work burdens with other household members in the months after the birth of an infant.
... Furthermore, elucidation of the SROI from mothers' perspectives would be of great value in further clarifying the social impact of implementing Ten Steps. Mothers' time and effort to breastfeed were not included in the calculation, with only limited data on this important investment available in the current literature [86][87][88]. Our examination of the SROI of implementation of the Ten Steps in one public hospital in Surabaya, Indonesia provides the foundation for future research in other hospitals and community settings with more comprehensive stakeholder involvement, such as other healthcare staff, family members, workplace, and community settings. ...
Article
Full-text available
Background: Despite the known importance of breastfeeding for women's and children's health, global exclusive prevalence among infants under 6 months old is estimated at only 41%. In 2018, Indonesia had a lower exclusive breastfeeding rate of 37% at 6 months postpartum; ranging from 20% to 56%, showing unequal breastfeeding support throughout the country. The World Health Organization (WHO) launched the Ten Steps to Successful Breastfeeding (Ten Steps) in 1989, later embedded in UNICEF's Baby-Friendly Hospital Initiative (BFHI) program in 1991. The BFHI aims to encourage maternity facilities worldwide to ensure adequate education and support for breastfeeding mothers by adhering to the Ten Steps and complying with the International Code of Marketing of Breastmilk Substitutes. An Indonesian survey in 2011 found that less than one in 10 government hospitals implemented the Ten Steps. It has been common for Indonesian health services to collaborate with infant formula companies. While no Indonesian hospitals are currently BFHI-accredited, the WHO/UNICEF Ten Steps (updated in 2018) have been adopted in Indonesia's national regulation of maternity facilities since 2012. Internationally, implementation of the Ten Steps individually and as a package has been associated with benefits to breastfeeding rates and maternal and infant health. However, to date, few studies have examined the impact of implementing the Ten Steps in economic terms. This study aims to measure the economic benefit of Ten Steps implementation in an Indonesian hospital. Methods: The study was conducted in January 2020 in Airlangga University Hospital, Surabaya, Indonesia, which has implemented the Ten Steps since it was established in 2012. To understand and generate evidence on the social value of the Ten Steps, we conducted a "Social Return on Investment (SROI)" study of implementing the Ten Steps in this maternity facility. To estimate the costs relating to the Ten Steps we interviewed the financial and nursing managers, a senior pediatrician, and senior midwife due to their detailed understanding of the implementation of the Ten Steps in the hospital. The interview was guided by a questionnaire which we developed based on the 2018 WHO/UNICEF Ten Steps to Successful Breastfeeding. The analysis was supported with peer-reviewed literature on the benefits of Ten Steps breastfeeding outcomes. Results: The total per annum value of investment (cost) required to implement Ten Steps in Airlangga University Hospital was US$ 972,303. The estimate yearly benefit was US$ 22,642,661. The social return on the investment in implementing Ten Steps in this facility was calculated to be US$ 49 (sensitivity analysis: US$ 18-65). Thus, for every US$ 1 invested in Ten Steps implementation by Airlangga Hospital could be expected to generate approximately US$ 49 of benefit. Conclusions: Investment in the Ten Steps implementation in this Surabaya maternity facility produced a social value 49 times greater than the cost of investment. This provides novel evidence of breastfeeding as a public health tool, demonstrating the value of the investment, in terms of social impact for mothers, babies, families, communities, and countries. Breastfeeding has the potential to help address inequity throughout the lifetime by providing the equal best start to all infants regardless of their background. Indonesia's initial moves towards implementing the WHO/UNICEF Ten Steps can be strengthened by integrating all elements into the national regulation and health care system.
... Because respondents report every activity they engage in on a single day, when the activity happened, how long it lasted, where it occurred, and who else was with them, the possible questions that can be answered with these data extend far beyond a simple accounting of time in major activities. They are used to study numerous issues, classically topics such as gender inequality [4], parental investments in children [5], behavior during economic recessions versus times of a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 economic growth [6], when and where people work [7], the use of time outside of paid work [8], and time spent with others [9]. ...
Article
Full-text available
Time-use data can often be perceived as inaccessible by non-specialists due to their unique format. This article introduces the ATUS-X diary visualization tool that aims to address the accessibility issue and expand the user base of time-use data by providing users with opportunity to quickly visualize their own subsamples of the American Time Use Survey Data Extractor (ATUS-X). Complementing the ATUS-X, the online tool provides an easy point-and-click interface, making data exploration readily accessible in a visual form. The tool can benefit a wider academic audience, policy-makers, non-academic researchers, and journalists by removing accessibility barriers to time use diaries.
... There is also no literature on the benefits to healthcare professionals in implementing and/or achieving the BFHI accreditation. Moreover, mothers' time and effort to breastfeed were not included in the calculation, with only limited data on this important investment available in the current literature [98][99][100]. The impact of breastfeeding support programs on mothers is well documented [101][102][103]; however, elucidation of the SROI from mothers' perspectives would be of great value in further clarifying the social impact of implementing the BFHI. ...
Article
Full-text available
Background Breastfeeding has positive impacts on the health, environment, and economic wealth of families and countries. The World Health Organization (WHO) launched the Baby Friendly Hospital Initiative (BFHI) in 1991 as a global program to incentivize maternity services to implement the Ten Steps to Successful Breastfeeding (Ten Steps). These were developed to ensure that maternity services remove barriers for mothers and families to successfully initiate breastfeeding and to continue breastfeeding through referral to community support after hospital discharge. While more than three in four births in Australia take place in public hospitals, in 2020 only 26% of Australian hospitals were BFHI-accredited. So what is the social return to investing in BFHI accreditation in Australia, and does it incentivize BFHI accreditation? This study aimed to examine the social value of maintaining the BFHI accreditation in one public maternity unit in Australia using the Social Return on Investment (SROI) framework. This novel method was developed in 2000 and measures social, environmental and economic outcomes of change using monetary values. Method The study was non-experimental and was conducted in the maternity unit of Calvary Public Hospital, Canberra, an Australian BFHI-accredited public hospital with around 1000 births annually. This facility provided an opportunity to illustrate costs for maintaining BFHI accreditation in a relatively affluent urban population. Stakeholders considered within scope of the study were the mother-baby dyad and the maternity facility. We interviewed the hospital’s Director of Maternity Services and the Clinical Midwifery Educator, guided by a structured questionnaire, which examined the cost (financial, time and other resources) and benefits of each of the Ten Steps. Analysis was informed by the Social Return on Investment (SROI) framework, which consists of mapping the stakeholders, identifying and valuing outcomes, establishing impact, calculating the ratio and conducting sensitivity analysis. This information was supplemented with micro costing studies from the literature that measure the benefits of the BFHI. Results The social return from the BFHI in this facility was calculated to be AU$ 1,375,050. The total investment required was AU$ 24,433 per year. Therefore, the SROI ratio was approximately AU$ 55:1 (sensitivity analysis: AU$ 16–112), which meant that every AU$1 invested in maintaining BFHI accreditation by this maternal and newborn care facility generated approximately AU$55 of benefit. Conclusions Scaled up nationally, the BFHI could provide important benefits to the Australian health system and national economy. In this public hospital, the BFHI produced social value greater than the cost of investment, providing new evidence of its effectiveness and economic gains as a public health intervention. Our findings using a novel tool to calculate the social rate of return, indicate that the BHFI accreditation is an investment in the health and wellbeing of families, communities and the Australian economy, as well as in health equity.
... There is also no literature on the bene ts to healthcare professionals in implementing the BFHI. Moreover, mothers' time and effort to breastfeed were not included in the calculation, with only limited data on this important investment available in the current literature (77)(78)(79). The impact of breastfeeding support programs on mothers is well documented (80)(81)(82); however, elucidation of the SROI from mothers' perspectives would be of great value in further clarifying the social impact of implementing the BFHI. ...
Preprint
Full-text available
Background Breastfeeding has positive impacts on the health, environment, and economic wealth of families and countries. Nevertheless, barriers to accessing high-quality breastfeeding support are evident in the low global exclusive breastfeeding rate of 41%. The World Health Organization (WHO) launched the Baby Friendly Hospital Initiative (BFHI) in 1991 as a global program to incentivize maternity services to implement the Ten Steps to Successful Breastfeeding (Ten Steps). These were developed to ensure that maternity services remove barriers for mothers and families to successfully initiate breastfeeding and to continue breastfeeding through referral to community support after hospital discharge. However, in 2020 only 26% of Australian hospitals were BFHI-accredited. This study aimed to examine the social return on investment (SROI) of implementing the BFHI in one public maternity unit in Australia.Method The study was non-experimental and conducted in the maternity unit of an Australian BFHI-accredited public hospital with around 1000 births annually. This facility illustrated costs for BFHI implementation in a relatively affluent urban population, and more than three in four births in Australia take place in public hospitals. Stakeholders considered within scope of the study were the mother-baby dyad and the maternity facility. We interviewed the hospital’s Director of Maternity Services and the Clinical Midwifery Educator, guided by a structured questionnaire, which examined the cost (financial, time and other resources) and benefits of each of the Ten Steps. Analysis was informed by the Social Return on Investment (SROI) framework, which uses monetary values to measure social, environmental and economic outcomes of change. This information was supplemented with micro costing studies from the literature that measure the benefits of the BFHI. ResultsThe social return from the BFHI initiative in this facility was calculated to be AU$ 1,398,140. The total investment required was AU$ 24,433 per year. Therefore, the SROI ratio was approximately AU$ 55:1 (sensitivity analysis: AU$ 16-112), which meant that every AU$1 invested in BFHI implementation by this maternal and newborn care facility generated approximately AU$55 of benefit. Conclusions Scaled up nationally, the BFHI could provide important benefits to the Australian health system and national economy. In this public hospital, the BFHI produced social value greater than the cost of investment, providing new evidence of its effectiveness and economic gains as a public health intervention. Our findings using a novel tool to calculate the social rate of return, indicate that implementation of the BHFI is an investment in the health and wellbeing of families, communities and the Australian economy, as well as in health equity.
... These findings are in line with prior studies reporting an increased bias towards expressions of happiness with increasing breastfeeding experience (Krol et al., 2015(Krol et al., , 2014. How exactly breastfeeding experience interacts with emotion processing is not certain, but a possible explanation is an increased closeness between mother and infants; breastfed infants on average spend more time interacting with their mother (Smith and Forrester, 2017) and show a higher attachment security (Gibbs et al., 2018). However, such reasoning would go against prior work suggesting that an enhanced fear response at seven months is indicative of better attachment quality (Peltola et al., 2015(Peltola et al., , 2020. ...
Article
Full-text available
Maternal odor is known to play an important role in mother-infant-interaction in many altricial species such as rodents. However, we only know very little about its role in early human development. The present study therefore investigated the impact of maternal odor on infant brain responses to emotional expression. We recorded the electroencephalographic (EEG) signal of seven-month-old infants watching happy and fearful faces. Infants in two control groups exposed to no specific odor (control 1) or the odor of a different infant's mother (control 2) showed the expected EEG fear response. Crucially, this response was markedly absent in the experimental group exposed to their mother's odor. Thus, infants respond differently to fear signals in the presence of maternal odor. Our data therefore suggest that maternal odor can be a strong modulator of social perception in human infants.
Chapter
This chapter reviews the role of responsive feeding in supporting breastfeeding, as well as the ways in which breastfeeding can reinforce responsive feeding. First, the demand and supply nature of lactation is outlined as a foundation for understanding the role of responsive feeding in supporting breastfeeding success. Next, connections between perceived insufficient milk supply and responsive feeding are elucidated. Finally, effects of breastfeeding on mothers’ responsiveness to infant cues via hormonal, neural, and behavioral mechanisms are discussed. Overall, this chapter highlights the importance of promoting caregivers’ engagement in responsive feeding during the perinatal period as a critical support for breastfeeding and the benefits of breastfeeding for infant feeding and weight outcomes.
Article
Our study was designed to assess the rates of exclusive breastfeeding (defined as direct breastfeeding) and the use of mother’s own milk (MOM) in preterm infants and sick term infants at discharge and to identify potential influencing factors such as gestational age, early colostrum, and privacy. The study was conducted at a German level III neonatal department. All preterm and sick term infants admitted to the neonatal intensive care unit, the pediatric intensive care unit, the intermediate care unit, and the low care ward were included in the study. Infants were recruited between March and October 2015 (phase 1) and April to July 2016 (phase 2). Due to an emergency evacuation, privacy was limited during the first phase. Breastfeeding and the use of MOM were assessed daily using a self-designed score. In total, 482 infants of 452 mothers were included. More than 90% initiated breastfeeding and one-third were exclusively breastfed at discharge. Extremely immature infants and late preterm infants were less likely to be exclusively breastfed at discharge. Privacy (p<0.001) and early colostrum (p=0.002) significantly increased exclusive breastfeeding. Conclusion Extremely immature and late preterm infants were least likely to be exclusively breastfed at discharge and need special support. Interventions such as privacy and early colostrum should be promoted to increase breastfeeding.
Article
Full-text available
This study was aimed at systematically reviewing evidence of the association between breastfeeding and performance in intelligence tests. Two independent searches were carried out using Medline, LILACS, SCIELO and Web of Science. Studies restricted to infants and those where estimates were not adjusted for stimulation or interaction at home were excluded. Fixed and random-effects models were used to pool the effect estimates, and a random-effects regression was used to assess potential sources of heterogeneity. We included 17 studies with 18 estimates of the relationship between breastfeeding and performance in intelligence tests. In a random-effects model, breastfed subjects achieved a higher IQ [mean difference: 3.44 points (95% confidence interval: 2.30; 4.58)]. We found no evidence of publication bias. Studies that controlled for maternal IQ showed a smaller benefit from breastfeeding [mean difference 2.62 points (95% confidence interval: 1.25; 3.98)]. In the meta-regression, none of the study characteristics explained the heterogeneity among the studies. Breastfeeding is related to improved performance in intelligence tests. A positive effect of breastfeeding on cognition was also observed in a randomized trial. This suggests that the association is causal. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved.
Article
Full-text available
To evaluate the effect of breastfeeding on long (breast carcinoma, ovarian carcinoma, osteoporosis and type-2 diabetes mellitus) and short term (lactational amenorrhea, postpartum depression, postpartum weight change) maternal health outcomes. A systematic literature search was conducted in PUBMED, Cochrane Library and CABI databases. Outcome estimates of odds ratios or relative risks or standardized mean differences were pooled. In cases of heterogeneity, subgroup analysis and meta-regression were explored. Breastfeeding >12 months was associated with reduced risk of breast and ovarian carcinoma by 26% and 37% respectively. No conclusive evidence of an association between breastfeeding and bone mineral density was found. Breastfeeding was associated with 32% lower risk of type-2 diabetes. Exclusive and predominant breastfeeding were associated with longer duration of amenorrhea. Shorter duration of breastfeeding was associated with higher risk of postpartum depression. Evidence suggesting an association of breastfeeding with postpartum weight change was lacking. Review supports the hypothesis that breastfeeding is protective against breast and ovarian carcinoma and exclusive/predominant breastfeeding increase duration of lactation amenorrhoea. There is evidence that breastfeeding reduce the risk of type-2 diabetes. However, an association between breastfeeding and bone mineral density or maternal depression or postpartum weight change was not evident. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved.
Article
Full-text available
Background: The benefits of exclusive breastfeeding, including public health cost savings, are widely recognized, but breastfeeding requires maternal time investments. Objective: This study investigates the time taken to exclusively breastfeed at 6 months compared with not exclusively breastfeeding. Methods: Time use data were examined from an Australian survey of new mothers conducted during 2005-2006. Data from 139 mothers with infants age 6 months were analyzed using chi-square tests of independence to examine socioeconomic and demographic characteristics and 2-sided t tests to compare average weekly hours spent on milk feeding, feeding solids, preparing feeds, and the total of these. The comparison was of exclusively breastfeeding mothers with other mothers. We also compared exclusively breastfeeding with partially breastfeeding and formula feeding mothers using a 1-way between-groups analysis of variance (ANOVA). Results: The exclusively breastfeeding (vs other) mothers spent 7 hours extra weekly on milk feeding their infants but 2 hours less feeding solids. These differences were statistically significant. ANOVA revealed significant differences between exclusively breastfeeding mothers, breastfeeding mothers who had introduced solids, and mothers who fed any formula, in time spent feeding milk, and solids, and preparing feeds. Conclusion: Exclusive breastfeeding is time intensive, which is economically costly to women. This may contribute to premature weaning for women who are time-stressed, lack household help from family, or cannot afford paid help. Gaining public health benefits of exclusive breastfeeding requires strategies to share maternal lactation costs more widely, such as additional help with housework or caring for children, enhanced leave, and workplace lactation breaks and suitable child care.
Article
Full-text available
We provide the first study of time use patterns among,parents ofchildren under the age of one year. Using a sample of 2,081 time diary respondentsto the American Time Use Survey, diverse measures of childcare time uniformly find coupledfathers, coupled mothers, and single mothers spending more time on the care of infants relative to older children. A dditional time for infants among fathers is generated mainly by reductionsin housework and leisure time, while both coupled and single mothers scale back work along with other activities. The single mothers also sleep more when caring for infants. Regression analyses suggest that teenage girls and grandparents often help with childcare. Worktypically reduces childcare time, although shiftwork among,fathers has an opposite effect. Hispanics and particularly African Americans
Article
Full-text available
Concerns about children's time-use and daily activities are not new. Nevertheless it is evident that such studies are scarce and partial and that as of today the knowledge we have on the patterns of children's everyday activities and the way in which different patterns of activity impact on children's well-being is insufficient. This article makes a case for the importance of time-use studies of children: such studies that will provide a flexible resource for describing and analysing children's social lives and well-being. Based on an extensive and rigorous literature review, the article analyses various studies on children's time-use by their goals, target population and methods. An analysis that provides not only the basis for calling for more studies but suggests how children's time-use could be studied on a larger scale and on a regular basis as a part of society's effort to study and monitor its children's well-being.
Article
Full-text available
A number of studies have suggested breastfed infants have improved bonding and attachment or cognitive development outcomes. However, mechanisms by which these differences might develop are poorly understood. We used maternal time use data to examine whether exclusively breastfeeding mothers spend more time in close interactive behaviors with their infants than mothers who have commenced or completed weaning. Mothers (188) participating in a time use survey recorded infant feeding activities for 24 h over a 7 day period using an electronic device. Tracking was conducted at 3, 6, and 9 months postpartum. Data was collected for maternal activities including infant feeding and time spent in emotional care. The mothers of exclusively breastfed infants aged 3–6 months fed them frequently and total time spent in breastfeeding averaged around 17 h a week. Maternal time spent in emotional care was also substantial, and found to correlate positively with time spent breastfeeding. Exclusively breastfed infants received greatest amounts of emotional care from their mother, and exclusively formula fed infants the least. Mixed fed infants received more emotional care time than formula fed infants, but less than fully breastfed infants. These findings may help explain the differential cognitive developmental outcomes reported in the medical literature for breastfed and non breastfed infants.
Article
We study postpartum decisions about paid work and breastfeeding using a simultaneous equations model. For our sample of higher socioeconomic status mothers, we find a joint decision process for three sets of decisions modeled: work leave duration and duration of any, as well as of exclusive, breastfeeding, and daily work hours and daily breastfeedings at infant age 3 months. We find that returning to paid work 1 week earlier reduces any breastfeeding duration by about two-thirds of a week while extending breastfeeding by a week delays work participation by about one-third of a week. We find larger elasticities for decisions involving daily work hours and number of breastfeedings at 3 months than for work leave duration and any breastfeeding duration and the smallest elasticities for duration of exclusive breastfeeding and work leave. This marks the first study to find a joint decision-making process for postpartum work and breastfeeding decisions and suggests that, in addition to increased leave impacting breastfeeding behaviors, successful breastfeeding promotion policies can have nontrivial impacts on the US labor market.
Article
Synchrony, a concept coined by the first researchers on parenting in social animals (Rosenblatt, 1965; Schneirla, 1946; Wheeler, 1928), describes the dynamic process by which hormonal, physiological, and behavioral cues are exchanged between parent and young during social contact. Over time and daily experience, parent and child adjust to the specific cues of the at-tachment partner and this biobehavioral synchrony provides the foundation for the parent–infant bond (Fleming, O'Day, & Kraemer, 1999). Affiliative bonds—defined as selective and enduring attachments—are formed on the ba-sis of repeated exposure to the coordination between physiological states and interactive behavior within each partner, between partners, and between the physiology of one and the behavior of the other. Such social bonds, in turn, set the framework for the infant's emotional development and shape the life-long capacity to regulate stress, modulate arousal, and engage in coregulatory interactions, achievements that are central components of the child's social– emotional growth (Feldman, 2007a). Moreover, the experience of biobehav-ioral synchrony in the first months of life sets the biological and behavioral systems that enable the child to provide optimal parenting to the next gen-eration, thereby forming the cross-generation transmission of attachment patterns (Feldman, Gordon, & Zagoory-Sharon, 2010a). During the sensitive period of bond formation, infants' brains are sen-sitized to the mutual influences between physiological systems, behavioral indicators, and their interactions. Studies in mammals propose that this pro-cess of synchrony—the system's sensitivity to the coordination of physiology Research reported in this chapter was supported by the Israel Science Foundation (#01/941, #1318/08), the US-Israel Bi-National Science Foundation (2001-241, 2005-273), the March of Dimes Foundation (12-FY04-50), the NARSAD foundation (Independent Investigator Award), and the Irving B. Harris Foundation.
Article
After three decades of decline, the amount of time spent by parents on childcare in the U.S. began to rise dramatically in the mid-1990s. Moreover, the rise in childcare time was particularly pronounced among college-educated parents. Why would highly educated parents increase the amount of time they allocate to childcare at the same time that their own market returns have skyrocketed? After finding no empirical support for standard explanations, such as selection or income effects, we offer a new explanation. We argue that increased competition for college admissions may be an important source of these trends. The number of college-bound students has surged in recent years, coincident with the rise in time spent on childcare. The resulting “cohort crowding” has led parents to compete more aggressively for college slots by spending increasing amounts of time on college preparation. Our theoretical model shows that, since college-educated parents have a comparative advantage in college preparation, rivalry leads them to increase preparation time by a greater amount than less-educated parents. We provide empirical support for our explanation with a comparison of trends between the U.S. and Canada, and a comparison across racial groups in the U.S.
Article
Thirty-two primiparous mothers with their neonates, 20 breast-fed and 12 bottle-fed, were videotaped during feeding longitudinally (at 3, 10 days, 1 and 3 months) in order to investigate whether patterns of mother-infant interaction observed during breast- and bottle-feeding differ also for behavior not directly related to sucking. Infant state, sucking behavior, mother behavior and mother-infant interaction were analyzed by both sequential and non sequential analysis. Results suggest that the method of feedings affects mother-infant interaction observed during and just after feeding also for some behavior not directly related to sucking. The percentage of mutual touch, tactile stimulation and mother's gaze to infant was significantly more elevated during breast-feeding. Independently of the feeding mode, significant sequential dependencies between infant state of alertness and auditory stimulation, and auditory stimulation and mutual gaze were found.