Women and Birth 30 (2017) e214–e220

Contents lists available at ScienceDirect

Women and Birth

journal homepage: www.elsevier.com/locate/wombi

Original Research - Quantitative

Maternal singing of lullabies during pregnancy and after birth: Effects

on mother–infant bonding and on newborns’ behaviour. Concurrent
Cohort Study
Giuseppina Persicoa,* , Laura Antolinib , Patrizia Vergania , Walter Costantinic ,
Maria Teresa Nardid, Lidia Bellottie
a
School of Medicine and Surgery, University of Milano Bicocca, Italy
b
Department of Health Sciences, Center of Biostatistics for Clinical Epidemiology, University of Milano Bicocca, Italy
c
Department of Clinical and Community Sciences, University of Milan, Italy
d
Musicologist and Music Therapist, Via Vittorio Emanuele II 24, 20842 Besana Brianza (MB), Italy
e
Maternal Neonatal Ward of San Gerardo Hospital, Monza, Italy

A R T I C L E I N F O A B S T R A C T

Article history: Background: Mother–infant bonding is of great importance for the development and the well-being of the
Received 30 June 2016 baby. The aim of this Concurrent Cohort Study was to investigate the effects of mothers singing lullabies
Received in revised form 12 December 2016 on bonding, newborns’ behaviour and maternal stress.
Accepted 18 January 2017
Methods: Eighty-three (singing cohort) and 85 (concurrent cohort) women were recruited at antenatal
classes at 24 weeks g.a. and followed up to 3 months after birth. The Prenatal Attachment Inventory (PAI) and
Keywords: the Mother-to-Infant Bonding Scale (MIBS) were used to assess maternal-foetal attachment and postnatal
Lullabies
bonding.
Singing mothers
Mother–infant bonding
Findings: No significant influence was found on Prenatal Attachment; by contrast, Postnatal Bonding was
Newborns’ behaviour significantly greater (i.e. lower MIBS) in the singing group 3 months after birth (mean 1.28 vs 1.96;
Midwifery p = 0.001). In the same singing group, the incidence of neonatal crying episodes in the first month was
significantly lower (18.5% vs 28.2; p < 0.0001) as were the infantile colic (64.7% vs 38.3%; p = 0.003) and
perceived maternal stress (29.6% vs 36.5%; p < 0.05). Infantile colic was reduced in the singing group, even
in the second month after birth (22.8% vs 36.5; p = 0.002). At the same time, a reduction was observed in the
neonatal nightly awakening (1.5% vs 4.7; p < 0.0001).
Conclusions: Mothers singing lullabies could improve maternal-infant bonding. It could also have positive
effects on neonatal behaviour and maternal stress.
© 2017 Australian College of Midwives. Published by Elsevier Ltd. All rights reserved.

communication between the mother and child is represented

Statement of significance by maternal singing, which is an expression of maternal love
Problem or issue and is generally recognised as being beneficial for both
mother and newborn.
Mother–infant bonding is of great importance for the
development and well-being of the newborn What this paper adds

What is already known Maternal singing during pregnancy and after birth could
both improve maternal infant interaction and contribute to
The sound experience during the prenatal period is so incisive preventing neonatal colic.
that after birth a newborn is able to discriminate between
maternal and other female voices and demonstrates a clear
preference for the former. An interesting form of acoustic
1. Introduction

* Corresponding author at: School of Medicine and Surgery, Via Cadore 48, 20900
According to the Theory of Attachment, mother–infant bonding
Monza (MB) Italy, Fax: +39 2 64488257.
E-mail address: giuseppina.persico@unimib.it (G. Persico).
is of utmost importance for the development and well-being of the

http://dx.doi.org/10.1016/j.wombi.2017.01.007
1871-5192/© 2017 Australian College of Midwives. Published by Elsevier Ltd. All rights reserved.
 G. Persico et al. / Women and Birth 30 (2017) e214–e220
baby.1 From conception to birth, all human beings establish a
relationship with their mother that allows them to interact with
her. The mother–infant relationship starts during pregnancy when
the mother perceives foetal movements, and from then on a special
dialogue develops between mother and baby. This bonding is
further strengthened at the moment of childbirth.2–5 After
pregnancy, the relationship between mother and baby has a
special channel of communication: the maternal voice. The
auditory foetal system already reaches full maturity between
the 24th and 28th weeks’ gestation. From this moment the foetus is
able to react to auditory stimuli. Kisilevsky et al.6 investigated
foetal responses to music played with a headset placed on the
maternal abdomen and found an increase in foetal motor activity
and in foetal heart rate in the foetus from 28 weeks. Among all the
acoustic signals perceived during pregnancy, the maternal voice is
the predominant one and the main source of sensory stimulation.7
From the 28th to the 36th week, the phenomenon of
“habituation” occurs: the foetus is able to preserve the sound
memory of the vibro-acoustic stimuli to which it is repeatedly
exposed.7,8 This is one of the most important higher brain
functions involving learning and ability to retrieve information.7,8
Some authors have suggested that the newborn recognises and
responds positively to the maternal voice as a result of prenatal
exposure, and is soothed when exposed to the same stimulus in the
early postnatal period.9,10 The sound experience during the
prenatal period is so incisive that after birth a newborn is able
to discriminate between maternal and other female voices and
demonstrates a clear preference for the former.7,10–13
Evidence of the beneficial effects of exposure to music during
pregnancy for both, mother and baby, has been quoted in the
literature.14–18 The adequate vibro-acoustic stimulation by expo-
sure to music can produce changes in foetal and neonatal
behaviour and has a direct influence on the emotions and
neurotransmitter system.14 During pregnancy, symptoms of stress,
anxiety and depressive mood are common and there is a clear
association between a high level of stress and a range of poorer
maternal and infant outcomes.16,17 It has been proven that
pregnant women benefit from music therapy in terms of reduction
of stress, anxiety and depression.18 An interesting form of acoustic
communication between the mother and child is represented by
maternal singing in pregnancy which is an important precursor for
the relationship. Even after birth, maternal singing to the infant is
an expression of maternal love and is generally recognised as being
beneficial for both mother and newborn.19–21 Lullabies are
differentiated from other types of songs due to their repetitiveness,
soothing quality, softness, simplicity and slow tempo.12 Infant-
directed play songs and lullabies are part of traditions and cultures
all around the world. However, in contemporary society, the
maternal singing tradition in Italian culture is changing due to
increased modernisation and the tendency of women to go back to
work soon after the birth of their baby.
1.1. Aim
This study aims to explore the potential of maternal singing of
lullabies during pregnancy and after birth in developing and
strengthening the relationship between the mother and baby. The
choice of lullabies, as opposed to other kinds of music, is a result of
their communicative value and their importance as one of the
symbols of mother–infant relationship in all the cultures around
the world.12,22
The main aim is to assess the effects of mothers singing lullabies
on both prenatal attachment and postnatal mother–infant bonding
by means of established and validated questionnaires. In addition,
the foetus and newborn behaviour are also investigated together
with the mother’s feelings and perceived stress.
e215
2. Methods
2.1. Participants
The study was approved by the Hospital Ethics Committee on
November 17, 2011 (approval Number 1497-2011). Recruitment
took place from December 2011 to August 2013, during the first
session of antenatal classes at the San Gerardo Hospital, a
metropolitan maternity hospital in Monza (Italy). The recruited
women attending the antenatal classes were aged 18 years or
more, had adequate knowledge of the Italian language, had
uncomplicated singleton pregnancies and gave their informed
consent to participate in the study.
The enrolled women were divided into a cohort invited to sing
lullabies during pregnancy (singing cohort) and a control cohort of
non-singing women (concurrent cohort). Antenatal classes were
organized in 14 weekly sessions, starting at 24 weeks of gestational
age (g.a.) with the participation of at least 30 women in each group.
The groups were progressively numbered from 1 to 6. Odd
numbers were assigned to the singing cohort group, and even
numbers were assigned to the concurrent cohort group to mimic
randomization. All the groups were conducted by the same
midwife. Women were monitored from 24 weeks g.a. to 3 months
after giving birth.
A required sample size of 156 (N = 78 for both the singing cohort
and the concurrent cohort), was calculated to detect a minimum
difference of 0.45 standard deviations between the expected
values of the mother–infant-bonding score (a continuous summa-
ry measure calculated from the MIBS questionnaire) with a 5%-
significance level and 80%-power. The sample size was increased to
about 180 participants to take into account a drop-out rate of 7.6%
due to premature birth revealed by the Italian official statistics,23
the presence of drop out during follow-up, and the standard size of
the antenatal classes (at least 30 women).
2.2. Measures and data collection
Demographic and obstetric data were collected at 24 weeks g.a.
and 48 h after birth. On this occasion, the neonatal data were also
collected. The Prenatal Attachment Inventory (PAI) question-
naire24,25 was used to assess the mother’s attachment to the foetus
at baseline (about 24 weeks g.a.) and at 36 weeks g.a.
This validated questionnaire24,25 consists of 21 items that
describe the mother’s feelings and thoughts towards the baby
during pregnancy. For each item, the woman is asked to rate her
feelings by setting an intensity rate on a 4-level scale ranging from
1 (weaker feelings) to 4 (stronger feelings). A final summary score
is calculated by summing up the 21 single scores. It ranges from a
minimum of 21 (very low prenatal attachment) to a maximum of
84 (very strong prenatal attachment). Mother to infant bonding
was assessed by the Mother-To-Infant Bonding Scale (MIBS)
questionnaire,26,27 administered 48 h after birth, and three months
later. The MIBS questionnaire26,27 is a structured multidimensional
and validated self-report scale. It consists of 8 items focused on key
adjectives which describe possible main emotions the mother
could feel for her newborn in the first weeks after birth.26,27 For
each item the woman is asked to rate her perception of the
emotion by setting an intensity rate on a 4-level scale ranging from
0 to 3. A final summary score is calculated by summing up the 8
single scores, with a total score ranging from a minimum of 0 (very
strong bonding) to a maximum of 24 (very weak bonding).
The singing frequency was collected for the singing cohort by
asking how many times per week the mother used to sing. The
frequency was then categorized into 1–3 times/week; 4–5 times/
week; 6–7 times/week; >7, times/week, as previously reported.21
The reaction of the foetus and the mother’s feelings while singing
e216 G. Persico et al. / Women and Birth 30 (2017) e214–e220
in pregnancy were investigated by asking the women in the singing
cohort questions about the perceived movement changes during
singing (yes/no, Foetal reaction data) and their feelings in terms of
positive feeling during singing (yes/no, Maternal reaction data).
In the concurrent cohort, although the topic of singing was not
included during the antenatal class, women where asked whether
or not they used to sing lullabies regularly.
The newborn behaviour was investigated in both cohorts19 by
asking questions on:
i) sleep-wake rhythm, in terms of: number of night awakenings
(1–3, 4–6, 7–9, 10 or more), ease in falling asleep (yes, no), need
of specific methods such as rocking or singing to fall asleep
(yes, no)
ii) crying episodes (yes, no)
iii) suffering from neonatal colic, defined as at least three hours of
crying for three days/week, over three consecutive weeks19,28
(yes, no)
Collected data on these above-mentioned aspects will be
referred to further on as “Neonatal behaviour data”.
The mother’s perceived stress was analysed in terms of:
i) easiness in falling asleep after the awakenings (yes, no)
ii) perceived stress related to the awakening (yes, no)
Collected data on these aspects will be referred to at a later
stage as “Maternal stress data”.
The data of the foetal reaction, the mother’s feelings, the
neonatal behaviour and maternal stress were coded dichotomous-
ly. The data collection is summarized in Table 1.
Data were collected by two different midwives from the one
teaching the courses. Face-to-face interview modality was used in
the prenatal period and at the 48 h postnatal follow-up. Telephone
interviews were used for the data collection at 1, 2 and 3 months
after birth. All data were collected on each woman individually.
2.3. Intervention
At the first session of antenatal classes, women in the singing
cohort received the text of nine lullabies. The recommended
lullabies were given by the musicologist who included traditional,
loving and playful songs. Women learned and practised all nine
lullabies without accompaniment, together with the midwife,
before each antenatal session. After four weeks, women were
invited to choose one or two lullabies as a “leitmotiv” for their own
babies and to continue singing at home, in their own time, paying
attention to their emotions and their babies’ behaviour. The topic
of singing lullabies was not covered in standard antenatal classes
attended by women in the concurrent cohort. Women in
concurrent cohort were, however, left free to singing whatever
type of lullabies they wanted.
Table 1
Type of data (listed in rows) by time of collection (set in columns).
Pregnancy After birth
Baseline (24 weeks g.a.) 36 weeks g.a. 48 h
Demographic
Obstetric Obstetric and neonatal
Singing characteristicsa
Foetus’ reactiona
Mother’s reactiona
PAI PAI MIBS
PAI = Prenatal Attachment Inventory; MIBS = Mother Infant Bonding Scale.
a
Data collected only for the singing cohort.
2.4. Data analysis
Collected data were first checked for coherence and consisten-
cy. Categorical and continuous variables were described by the
proportion of subjects falling into each category and by mean and
standard deviation, respectively. Comparisons across the singing
cohort and the concurrent cohort were performed by the Chi-
square test for categorical variables and by the Wilcoxon test for
continuous variables. All tests were two-sided (5%-significance
level). Data checking and analysis were conducted with SAS
Package.29 Data were analysed according to the “Intention To Treat”
principle.
3. Results
A total of 205 women were considered for analysis. Of these, 196
were eligible and thus recruited (97 allocated to the singing cohort
and 99 allocated to the concurrent cohort). Eighty-three out of 97
completed the follow-up schedule for the singing cohort and 85
out 99 for the concurrent cohort. The sample identification is
represented in Fig. 1.
Table 2 summarizes the demographic data and the baseline PAI
score. The majority of women were Italian (92.8% vs 96.5%), with an
average age of 33 years for both the cohort and a relatively high
level of education (97.6% vs 93%). Most of them were married or
cohabitating (97.6% vs 98.8%).
Table 3 shows the obstetric and neonatal characteristics. The
majority of women were primiparous (84.3% vs 85.9%) as expected,
since priority in the antenatal classes enrolment was given to
primiparous women.
All baseline characteristics were comparable in the two study
cohorts.
In Table 4 the data on singing characteristics (collected at 36
weeks g.a. and 1 month after birth) are summarized for the singing
group. It is interesting to note that 55.4% of the women sang
lullabies during pregnancy 1–3 times/week, 22.9% sang lullabies
4–5 times/week and 21.7% sang 6 or more times/week. With
respect to the moment in the day when the women sang (data not
shown in Table 4), a percentage of 41.0% of participants used to sing
exclusively in the evening, 10.8% in the morning and 14.5% in the
afternoon. The remaining women were not restricted to a specific
time. As regards the foetal reaction, 52.4% of women reported a
change in the perceived foetal movements. A remarkable
percentage of women (90.6%) reported positive emotions after
singing, such as: serenity, relaxation or a feeling of being on the
same wavelength as their baby. At one month after birth, mothers
continued singing lullabies, especially when the baby was fidgety
(42%), cried (3.7%), went to sleep (32%) or during play time and
cuddling (15%) (data not shown in Table 4). The percentage of
babies who reacted positively and the percentage of mothers who
perceived positive emotions increased with respect to the prenatal
period.
1 month 2 months 3 months
Singing characteristicsa Singing characteristicsa
Newborn’s behaviour Newborn’s behaviour Newborn’s behaviour
Maternal stress Maternal stress Maternal stress
MIBS
 G. Persico et al. / Women and Birth 30 (2017) e214–e220 e217

205
Eligible Women

9 Excluded

- 5 pre-term delivery
- 1 foetal pathology
- 3 not compliance

196
Recruited

97 99
Singing Cohort Concurrent Cohort

14 Drop out 14 Drop out

83 85
Concluded Study Concluded Study

Fig. 1. Sample definition.

Table 2
Demographic characteristics and baseline PAI total score, by study cohort.

Type of parameter Singing cohort (n = 83) Concurrent cohort (n = 85)

Demographic
Mother's age (years) Mean SD Mean SD
32.6 4.8 32.7 3.8
Nationality (Italian) n % n %
77 92.8 82 96.5
Education n % n %
Primary school 2 2.4 6 7.1
Secondary school 49 59.0 34 40.0
University degree 32 38.6 45 52.9
Married/cohabitating n % n %
81 97.6 83 98.8
PAI total score at 24 weeks g.a. Mean SD Mean SD
60.6 6.9 59.3 8.3

PAI = Prenatal Attachment Inventory.

A modest decline in the tendency to continue singing was
observed at two months after birth (88.0%). It is worth noting that
at two months the percentage of women who used to sing during
play time and cuddles rose to 25%. A response was observed in
68.6% of the babies.
A comparison of the PAI and MIBS scores between the two
groups is presented in Table 5. The PAI total score at 36 weeks g.a.
and MIBS total score 48 h after birth were not significantly different
across the cohorts. The MIBS total score, however, was on average
slightly lower for the singing cohort (1.35 versus 1.68 of the
concurrent cohort).
By contrast, at 3 months after birth, the MIBS total score was
significantly lower in the singing cohort than in the concurrent
cohort, with an average of 1.28 versus 1.96 in the concurrent cohort
e218 G. Persico et al. / Women and Birth 30 (2017) e214–e220

Table 3
Obstetric and Neonatal characteristics, by study cohort.

Type of parameter (at birth) Singing cohort (n = 83) Concurrent cohort (n = 85)
Obstetric
Primiparous n % n %
70 84.3 73 85.9
Gestational age at birth (weeks) Mean SD Mean SD
39.5 1.28 39.6 1.36
Delivery n % n %
Vaginal 62 74.7 69 81.2
C-section 15 18.1 13 15.3
Vacuum assisted 6 7.2 3 3.5
Newborn
Weight (gr) Mean SD Mean SD
3276.3 406.1 3262.9 451.1
Gender (male) n % n %
40 48.2 49 57.7

In the singing cohort, the percentage of newborns defined by

Table 4
Singing characteristics during pregnancy and one month after birth (singing cohort, their mothers as “crying often” at 1 month of age was smaller in the
83 women). singing cohort (18.5% vs 28.2% respectively; p < 0.0001). Finally,
the proportion of babies suffering from neonatal colic was
During pregnancy
Singing Frequency (times/week) n % significantly higher in the concurrent cohort, both in the first
1–3 46 55,4 month (64.7% vs 38.3% for the singing cohort; p = 0.003) and in the
4–5 19 22,9 second month (36.5% vs 22.8% for the singing cohort; p = 0.002).
6–7 8 9,6 At the last follow up, three months after birth, the percentage of
7 10 12,1
neonatal colic was 8,6% in the singing cohort and 17,9% in the
Foetus reaction 43 52,4
Mother’s positive feeling 75 90.6 concurrent cohort (p = 0.059).
1 month after birth n %
Singing still performed 81 97,6 4. Discussion
Baby reaction 48 63,2
Mother’s positive feeling 76 93.8
This is a quasi-randomized study to compare a cohort of women
who were encouraged to sing lullabies to their babies, during
(p = 0.001). Perceived maternal stress at one month was reduced in pregnancy and after birth, with a control group left free to sing
the singing cohort in terms of ease in going back to sleep after an whenever they so desired. Existing evidence has already demon-
awakening (29.6% for the singing cohort vs 36.5% for the strated a beneficial effect of maternal singing after birth.12,13,21 The
concurrent cohort; p < 0.05). novelty of our study lies in the provision of data from pregnancy to
In relation to the newborn’s sleep-wake rhythm, it can be seen postpartum with a comparison between mothers invited to sing
that in both study groups a small percentage of newborns used to and a concurrent group. Such aspects have not been addressed up
wake up four times or more overnight. This rate tended to be lower to now, not even in recent literature.4,25,30 We also investigated,
in the singing cohort at one month postpartum (6.4% vs 10.6%; comprehensively, the effect of maternal lullaby singing on both
p = 0.240). At two months after birth the corresponding percen- mothers and newborns. The two cohorts here analysed were
tages were 1.5% for the singing cohort vs 4.7% for the concurrent homogeneous in terms of the social, demographic and obstetric
cohort (p < 0.001). aspects and all data concerning the mode of birth are consistent

Table 5
Response characteristics by study cohort.

Singing cohort Concurrent cohort p-Value

(n = 83) (n = 85)

PAI total score at 36 weeks' gestation mean SD mean SD

65.6 9.3 65.4 6.7 0.692
MIBS total score mean SD mean SD
48 h after birth 1.35 1.36 1.68 1.55 0.152
3 months after birth 1.28 1.29 1.96 1.48 0.001
Maternal Stress at 1 month after birth n % n %
Stress perceived due to night awakening 24 29.6 31 36.5 <0.05
Easiness in falling asleep 67 82.7 75 88.2 0.598
Neonatal behaviour
1 month after birth n % n %
4 or more nightly awakening 5 6.4 9 10.6 0.24
Crying often 15 18.5 24 28.2 <0.0001
Suffering due to Colic 31 38.3 55 64.7 0.003
2 months after birth n % n %
4 or more nightly awakening 1 1.5 4 4.7 <0.0001
Suffering due to Colic 19 22.8 31 36.5 0.002
3 months after birth n % n %
Suffering due to Colic 7 8.6 15 17.9 0.059

PAI = Prenatal Attachment Inventory; MIBS = Mother Infant Bonding Scale.

 G. Persico et al. / Women and Birth 30 (2017) e214–e220
with those presented in the annual report of the hospital where the
study was conducted.
The study revealed a great interest by women in the proposal to
use the singing of lullabies. They tended to choose a single lullaby
as a leitmotif in the relationship with the baby. It is also interesting
to note that in the concurrent cohort, 8% of women spontaneously
sang lullabies every day, which suggests that women have a
natural tendency to sing to their newborns. Most pregnant women
chose the evening to sing lullabies because this time of day
promotes greater calmness and intimacy. The majority of mothers
(90%) reported positive feelings such as serenity, relaxation and
harmony with the baby, according to other authors.21,30 Moreover,
while singing, 53% of women perceived a change in foetal
behaviour at 36 g.a., as if there was a response to the sound
dialogue. These findings are in line with current literature.21,25,30,31
Regarding maternal attachment during pregnancy, we have
observed an increased level in both groups, which, however, was
not of statistical interest. We also observed a consistency between
the scores collected at recruitment and the findings presented in
the literature.2 Similarly, our findings were consistent with the
literature about mother-newborn bonding, suggesting a progres-
sive increase in the level of bonding from birth onwards.32
However, the present study has showed a statistically significant
difference between the two cohorts in the bonding three months
after birth, with a level of bonding stronger in the singing cohort.
This reduction in the MIBS total score suggests that maternal
singing of lullabies can improve the mother–infant bonding. After
birth, in the singing group, the majority of mothers continued to
sing lullabies to their newborns as a means of nurturing and most
of them (93%) reported that the singing of lullabies, which began in
pregnancy, enriched the relationship with their babies. The
particular expressiveness and intimacy that characterizes the
singing of lullabies, might just have contributed even more to
bringing mother and baby closer with a greater bonding in the
third month postpartum. This study also suggests that there could
be a positive effect of maternal singing lullabies on both the
newborn’s behaviour and the maternal stress. With respect to the
newborns’ behaviour, we found a frequency of crying episodes
significantly lesser in the singing cohort during the first month
postpartum (p < 0.0001). Moreover, for the mothers of this cohort
it was easier to calm their babies. Our findings of a reduction in
maternal stress related to singing lullabies is echoed in the extant
literature19,21,30 Maternal stress in the first month was significantly
reduced in the singing cohort (p < 0.05). As regards stress, some
authors found that levels of stress hormones, such as cortisol, are
reduced in response to music and this factor leads to a reduction in
anxiety and improved immune system functioning.33,34 A signifi-
cant reduction of neonatal colic in the newborns of the singing
cohort was present showing that singing could have a positive
effect in the baby’s relaxation.
With respect to the neonatal colic, some researchers found that
maternal stress is one of the possible causes,35 and that singing
could be a useful strategy or tool.36
Our findings suggest that maternal singing during pregnancy
and after birth could both improve maternal infant interaction and
contribute to preventing neonatal colic. The investigation of the
mechanisms of singing on the neuroendocrine function will be the
focus of future research.
4.1. Limits
Due to the study design, it was not possible to blind the two
cohorts and the researcher to the group allocation. Knowledge of
the group allocation could then have affected the behaviour of the
participants and their responses to subjective outcome measures.
In addition, for ethical reasons, the women of concurrent cohort
e219
could choose to sing lullabies as they so desired, and a percentage
of them (8%) sang lullabies during pregnancy and after birth.
Furthermore, data on the psychological state of the mothers and
mode of infant feeding were not collected since they were not the
focus of the study. The collection of data regarding mode of
feeding, however, could have been useful to help understand
whether the impact of singing on bonding is mediated by an
impact on the mode of infant feeding. In fact, it is acknowledged
that infant feeding mode has an effect on neuroendocrine factors
and those are related to bonding. Singing, however, could itself
have an independent impact on the neuroendocrine function. An
analysis, which includes both infant feeding modes and singing/
non singing allocation, could help distinguish the role of the
neuroendocrine function from the role of the two above-
mentioned factors.
5. Conclusions
To sum up, prenatal attachment and Maternal infant bonding in
the first few months of the life of a newborn are crucial areas of
research in midwifery given the importance of the mother and
baby’s well-being. This work goes beyond similar research
previously conducted using a large cohort of participants. The
novelty of our study lies in the provision of data from pregnancy to
3 months after birth with a comparison between mothers invited
to sing and a concurrent group, thus covering a longer time period
than what was previously investigated. Mothers in the singing
group unanimously emphasized the strong emotions and feelings
they experienced while singing, both in pregnancy and after birth.
The MIBS total score suggests that maternal singing of lullabies can
improve the mother–infant bonding. In the first few months after
giving birth, the mothers continued to sing lullabies to their
newborns and a remarkable percentage (93%) reported that this
activity enriched the relationship with their babies. This study
suggests that there was a positive effect of maternal singing both
on maternal stress, which was significantly reduced in the first few
months, and on the newborn behaviour. There were significantly
fewer crying episodes in the singing cohort during the first month
postpartum and mothers found it easier to calm their babies.
Moreover, the study suggests that singing lullabies has a positive
effect on the newborn’s behaviour, with a lower prevalence of
crying and colic. Since maternal singing was proved to have a
positive effect on both the reduction of perceived maternal stress
and on the frequency of neonatal colic, this research corroborates
the hypothesis that maternal stress has a direct impact on the
occurrence of neonatal colic. A comprehensive analysis could also
include a study of the psychological state of mothers which could
be investigated further in the future.
In conclusion, in our opinion these results offer midwives an
additional, easy to use, highly feasible tool that can be proposed to
pregnant women in antenatal classes in order to strengthen both
antenatal and after birth mother–infant bonding.
Conflict of interest
All the authors declare that there is neither conflict of interest
nor outside support for the research.
Disclosures
No financial assistance was provided for this project.
Acknowledgements
We are especially grateful to the women who so generously
gave their time to complete this study.
e220 G. Persico et al. / Women and Birth 30 (2017) e214–e220

We are thankful to Nuala Tansey, Language Consultant at the 17. Monk C, Leight KL, Fang Y. The relationship between women’s attachement
University of Milano Bicocca, for kindly helping with the language style and perinatal mood disturbance: implications for screening and
treatment. Arch Womens Ment Health 2008;11:117–29.
editing. 18. Chang M, Chen C, Huang K. Effects of music therapy on psychological health of
The main author is deeply grateful to her mother for the lullaby woman during pregnancy. J Clin Nurs 2008;17:2580–7.
that she used to sing to her when she was a child and which 19. Bellieni C, Ceccarelli D, Rossi F, Buonocore G, Maffei M, Perrone S, et al. Is
prenatal bonding enhanced by prenatal education course? Minerva Ginecol
remains one of her most precious memories. 2007;59:125–9.
20. Odent M. L’amour scientifié: les mécanismes de l’amour. Saint-Julien-en-
References Genevois: Jouvence ed; 1999.
21. Cevasco A. The effects of mothers’ singing on full term and preterm infants and
maternal emotional responses. J Music Ther 2008;45(3):273.
1. Bowlby J. Attachment and loss, Vol. 1: attachment. New York: Basic Books; 1969.
22. Trainor LJ. Lullabies and playsongs: why we sing to children. Zero Three
2. Klaus MH, Kennel JH. Maternal-infant bonding. Saint Louis: The C.V. Mosby
2002;23(1):31–4.
Company; 1976.
23. Basili F, Cocchi M, Di Rosa A, Tamburini C. Certificato di assistenza al parto
3. Siddiqui A, Hagglof B. Does maternal prenatal attachment predict postnatal
(CEDAP). Analisi dell’evento nascita. Anno 2010. Roma: Direzione Generale del
mother- infant interaction? Early Hum Dev 2000;59:13–25.
Sistema Informativo e Statistico Sanitario; 2011.
4. Jurgens MA, Levy-Rueff M, Goffinet F, Golse B, Beauquier-Macotta B. Étude des
24. Muller ME. Development of a prenatal attachment inventory. West J Nurs Res
propriétés psychométriques d’une échelle d’attachement prenatal Version
1993;15:199–215.
française de la Prenatal Attachement Inventory (PAI, Müller, 1993). L’Encéphale
25. Della Vedova AM, Dabrassi F, Imbasciati A. Assessing prenatal attachment in an
2010;36:219–25.
Italian women sample. J Reprod Infant Psychol 2008;26(2):86–98.
5. Marazziti D, Roncaglia I, Piccinni A, Dell’Osso L. Aspetti neurobiologici
26. Taylor A, Atkins R, Kumar R, Adams D, Glover V. A new mother-to-infant
dell’attaccamento. G Ital Psicopatol 2008;14:58–71.
bonding scale: links with early maternal mood. Arch Womens Ment Health
6. Kisilevsky BS, Hains SM, Jacquet AY, Granier-Deferre C, Lecanuet JP. Maturation
2005;8:45–51.
of fetal responses to music. Dev Sci 2004;7:550–9.
27. Bienfait M, Maury M, Haquet A, Faillie JL, Franc N, Combes C, et al. Pertinence of
7. Tomatis A. La nuit utérine. 2nd ed. Paris: Editions Stock; 1987.
the self-report mother-to-infant bonding scale in the neonatal unit of a
8. Soldera G. Premassaggio d’amore in gravidanza. Introduzione alla comunicazione
maternity ward. Early Hum Dev 2011;87(4):281–7.
psicotattile madre-padre-bambino con esercizi pratici. Ferrara, Italy: Editeam;
28. Kheir AE. Infantile colic: facts and fiction. Ital J Pediatr 2012;38:34.
2005.
29. SAS Institute Inc. SAS 9.1.3 Help and Documentation, Cary, NC: SAS Institute
9. James DK, Spencer CJ, Stepsis BW. Fetal learning: a prospective randomized
Inc. 2002–2004.
controlled study. Ultrasound Obstet Gynecol 2002;20:431–8.
30. Carolan M, Barry M, Gamble M, Turner K, Mascareñas O. The limerick lullaby
10. DeCasper AJ, Fifer WP. Of human bonding: newborns prefer their mother’s
project: an intervention to relieve prenatal stress. Midwifery 2012;28(2):173–
voices. Science 1980;208(4448):1174–6.
80.
11. Nardi MT. La relazione Sonora. Brescia, Italy: Ed. La Scuola; 2009.
31. Manfredi P, Ghilardi A, Imbasciati A. L’esposizione auditiva fetale: uno
12. Volta A. In principio era il suono. Quaderni ACP 2010;17(2):89–91.
strumento per l’indagine sulle origini dello sviluppo psichico. Imago
13. DeCasper AJ, Spence MJ. Prenatal maternal speech influences newborns’
1997;2:89–106.
perception of speech sounds. Infant Behav Dev 1986;9:133–50.
32. Figueiredo B, Costa R, Pacheco A, Pais A. Mother-to-infant and father-to-infant
14. Arya R, Chansoria M, Konanki R, Tiwari DK. Maternal music exposure during
initial emotional involvement. Early Child Dev Care 2007;177(5):521–32.
pregnancy influences neonatal behaviour: an open-label randomized con-
33. Kreutz G, Bongard S, Rohrmann S, Hodapp V, Grebe D. Effects of choir singing
trolled trial. Int J Pediatr 2012;2012(Feb (4)):1–614.
or listening on secretory immunoglobulin A, cortisol, and emotional state. J
15. Partanen E, Kujala T, Tervaniemi M, Huotilainen M. Prenatal music exposure
Behav Med 2004;27:623–63.
induces long-term neural effect. PLoS One 2013;8(10):e78946. doi:http://dx.
34. Avers L, Mathur A, Kamat D. Music therapy in paediatrics. Clin Pediatr
doi.org/10.1371/journal.pone.0078946.
2007;46:575–9.
16. Alder J, Fink N, Bitzer J, Hösli I, Holzgreve W. Depression and anxiety during
35. Savino F. Focus on infantile colic. Acta Paediatr 2007;96(9):1259–64.
pregnancy: a risk factor for obstetric, fetal and neonatal outcome? A critical
36. Gupta SK. Update on infantile colic and management options. Curr Opin
review of the literature. J Matern Fetal Neonatal Med 2007;20:189–209.
Investig Drugs 2007;8(11):921–6.