Journal of Neonatal Nursing xxx (2017) 1e7

Contents lists available at ScienceDirect

Journal of Neonatal Nursing

journal homepage: www.elsevier.com/jneo

Original Article

Does Kangaroo care affect the weight of preterm/low birth-weight

infants in the neonatal setting of a hospital environment?
Colette Cunningham a, *, Zena Moore b, D. Patton b, T. O'Connor b, Linda E. Nugent b
a
Neonatal Intensive Care Unit, University Hospital Waterford, Dunmore Road, Waterford, Ireland
b
RCSI, Royal College of Surgeons in Ireland, School of Nursing and Midwifery, St. Stephens Green, Dublin 2, Ireland

a r t i c l e i n f o a b s t r a c t

Article history: The aim of this systematic review is to ascertain if kangaroo care (KC) affects the weight of preterm/LBW
Received 16 August 2017 infants in the neonatal setting of hospital environments. The following databases were searched:
Accepted 9 October 2017 PubMed, The Cochrane Library, The Cumulative Index to Nursing and Allied Health Literature, Web of
Available online xxx
Science, Embase and SCOPUS. Search terms include: kangaroo care, kangaroo mother care, kangaroo
ward care, skin to skin care, skin to skin contact, skin to skin mother care, weight, neonatal infant,
Keywords:
neonatal care and neonatal unit. 10 RCT's demonstrated that KC increases weight of preterm/LBW infants
Kangaroo care
in the neonatal setting of a hospital environment. 7 quantitative studies also reported an increase in
Weight
Neonatal/LBW infant
weight. Increased rates of breastfeeding were also consistently associated with regard to KC across the 17
Neonatal unit studies. KC effects weight gain of preterm/LBW infants in the neonatal setting of a hospital environment.
Hospital environment Exclusive breastfeeding rates were positively influenced through KC.
© 2017 Neonatal Nurses Association. Published by Elsevier Ltd. All rights reserved.

Introduction A Conde-Agudelo and Diaz-Rossello (2016) SR aimed to deter-

Kangaroo Care (KC) or Skin to Skin Care (SSC) is the method of
placing an infant between or on the mother's breasts dressed only
in a hat and nappy so that the frontal contact of mother and baby is
skin to skin (Bigelow et al., 2012). KC was first identified and
introduced in Bogota, Columbia in 1978, when incubator shortages
necessitated keeping preterm infants warm through the “natural
incubator” of skin to skin contact with a mother or carer (Leonard
and Mayers, 2008). Physical growth and development is an inte-
gral part of neonatal recovery (Samra et al., 2013; Sharma et al.,
2016). Weight gain in the neonatal setting is considered an indi-
cation of health and thriving and dictates the discharge home of
these preterm and/or low birth weight infants (Dodd, 2005). This
systematic review was undertaken to ascertain if kangaroo care
contributes to the weight gain of preterm/low birth weight infants
in the neonatal setting of a hospital environment. Weight gain of
preterm/low birth weight infants through the provision of kanga-
roo care in a hospital setting only is unclear in previously under-
taken systematic reviews.
* Corresponding author.
E-mail addresses: col.cunningham79@gmail.com (C. Cunningham), zmoore@
rcsi.ie (Z. Moore), declanpatton@rcsi.ie (D. Patton), tomoconnor@rcsi.ie
(T. O'Connor), lindanugent@rcsi.ie (L.E. Nugent).
mine whether KC with LBW infants reduces morbidity and mor-
tality. It was concluded that kangaroo care does promote weight
gain in LBW infants but the review followed infants that were not
all in a hospital environment or a neonatal setting. They compared
LBW infants including data from the hospital and at home. Addi-
tionally, they only included RCT's and disregarded the information
from other quantitative studies. Boundy et al. (2016) conducted a
systematic review to estimate the association between KMC and
neonatal outcomes. Infants of any birth weight or gestational age
were included and the literature was not limited to a hospital
environment. Moore et al. (2012) SR assessed the effects of early
SSC on breastfeeding, physiological adaptation and behaviour in
healthy newborns weighing greater than 2500 g. Johnston et al.
(2014) SR reviewed the effect of SSC on pain in neonates under-
going painful procedures. It did not detail any other effects on the
infants. Chan et al. (2015) SR examined barriers and enablers of KC
with regard to qualitative articles. Therefore, although the area of
KC/SSC is well researched in recent years, a gap still remains with
regards to how KC effects weight gain of preterm and/or LBW in-
fants in the neonatal setting of a hospital environment only. It is
therefore timely that a SR on KC with regard to weight gain in these
infants is undertaken.

https://doi.org/10.1016/j.jnn.2017.10.001
1355-1841/© 2017 Neonatal Nurses Association. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Cunningham, C., et al., Does Kangaroo care affect the weight of preterm/low birth-weight infants in the
neonatal setting of a hospital environment?, Journal of Neonatal Nursing (2017), https://doi.org/10.1016/j.jnn.2017.10.001
2 C. Cunningham et al. / Journal of Neonatal Nursing xxx (2017) 1e7
Methods
The PICO (population, intervention, comparison, and outcome)
acronym gives structure to the framework of a SR (Higgins and
Green, 2011). Population in this SR is preterm and/or LBW infants,
intervention is KC, comparison is no KC and outcome is weight gain.
Review question
Does KC promote weight gain in preterm and/or LBW infants in
the neonatal setting of a hospital environment?
Primary & secondary outcomes
This SR details how preterm and/or LBW infants' weight gain is
affected by KC in a neonatal care setting in a hospital environment
only. This does not include studies that involved follow up at home
weights, or outpatient follow up weights. The secondary outcome
identified was the effect KC had on exclusive breastfeeding rates for
the infants. This secondary outcome was chosen because it is re-
ported on in all RCT's and all but one of the other quantitative
studies (Kambarami et al., 1998).
Inclusion and exclusion criteria
Any English language studies of a quantitative design that
measured the effect of KC on the weight of a neonatal infant in the
neonatal setting of a hospital environment were included. Foreign
language studies were excluded due to resources. There was no
limit on the year of publication. The aim of this was to ascertain
evidence and data from earlier years that could contribute to the
overall clarity of the answer to the research question.
Search strategy
Databases searched included; PubMed, The Cochrane Library,
The Cumulative Index to Nursing and Allied Health Literature
(CINAHL), Web of Science, Embase and SCOPUS. The search terms
used for the strategy were; kangaroo care/kangaroo mother care/
kangaroo ward care/skin to skin care/skin to skin contact/skin to
skin mother care; weight/neonatal infant/neonatal care/neonatal
unit. MESH were used to search the terms in PubMed. Open Grey
was searched for full text conference papers. LENUS was searched
for publications by the Health Service Executive (HSE) in Ireland,
which is where the authors are based. The National Institute for
Health and Care Excellence (NICE) guidance website was searched
for guidelines and standards of relevance. The bibliographies and
reference lists of primary studies were also searched for relevant
studies.
Data extraction
Data was extracted from 17 articles that met the inclusion
criteria. 10 were RCT's and 7 were of other quantitative designs.
Data analysis
All studies included had continuous data that was analysed in
terms of mean differences. Each study had an intervention and a
control group. The 10 RCT's were put through a meta-analysis using
the RevMan 5.2 software. The remaining other quantitative studies
are narratively analysed.
Please cite this article in press as: Cunningham, C., et al., Does Kangaroo care affect the weight of preterm/low birth-weight infants in the
neonatal setting of a hospital environment?, Journal of Neonatal Nursing (2017), https://doi.org/10.1016/j.jnn.2017.10.001
Quality appraisal
All RCT's were appraised for quality using the risk of bias
assessment tool in RevMan 5.2. All other quantitative studies
included were appraised using the EBL Critical Appraisal Checklist
(Glynn, 2006). All studies had a clear and separate weight outcome
presented in their results section.
Results
The search results (See Fig. 1) identified 839 records which were
then screened by two independent reviewers through reading titles
and abstracts. This resulted in the exclusion of 824 records. 17
eligible quantitative records were included in the SR.
Overview of the included articles
Study design
One article used a purposive sampling design (Kambarami et al.,
1998). Three articles used a quasi-experimental design (El Moniem
and Morsy, 2011; Samra et al., 2013; Kashaninia and Dehghan,
2015). Two articles used a pre test/post test design (Ahn et al.,
2010; Lee and Sook, 2011). One article used a prospective cohort
design (Lamy-Filho et al., 2008).
Geographical location
Four studies took place in India (Ali et al., 2009; Kadam et al.,
2005; Ramanathan et al., 2001; Suman Rao et al., 2008), one in
Kenya (Mwendwa et al., 2012), one in Australia (Roberts et al.,
2000), one in the USA (Rojas et al., 2003), one in Malaysia (Boo
and Jamli, 2007), one across Ethiopia, Indonesia and Mexico
(Cattaneo et al., 1998), two in Iran (Kashaninia and Dehghan, 2015;
Mohammadzadeh et al., 2011), one in Zimbabwe (Kambarami et al.,
1998), two in Egypt (El Moniem and Morsy, 2011; Samra et al.,
2013), two in Korea (Ahn et al., 2010; Lee and Sook, 2011) and
one in Brazil (Lamy-Filho et al., 2008).
Study settings
Four studies took place in tertiary level NICU's, seven in un-
classified level NICU's, two in level two NICU's, one across two
neonatal nurseries in two different hospitals but in the same
country, one across three different neonatal departments in three
different hospitals and three different countries, one in a neonatal
unit, one across 16 neonatal units in 16 different hospitals but the
same country.
Participants
All infants across the studies were either LBW ± preterm. Birth
weights eligible for inclusion across the studies varied; 500 g-
1749 g (Lamy-Filho et al., 2008)/<1500 g (Ramanathan et al., 2001;
Rojas et al., 2003; Boo and Jamli, 2007)/<1600 g (Kambarami et al.,
1998)/1000 g-1750 g (Mwendwa et al., 2012)/1000 g-1990g
(Cattaneo et al., 1998)/1200 g-1800g (Ali et al., 2009)/<1800g
(Kadam et al., 2005)/<2000g (Mohammadzadeh et al., 2011; Suman
Rao et al., 2008)/<2500 g (Samra et al., 2013). Fifteen of the studies
stipulated that infants should be medically stable whereas two did
not (Roberts et al., 2000; Rojas et al., 2003). This included factors
such as not being ventilated/no NCPAP/no oxygen therapy/no
inotropic support/no chromosomal or congenital abnormalities/no
infants awaiting transfer out of the hospital/no grade three or four
IVH's/no HIE/no infants of critically ill mothers/no CNS impairment/
no sepsis/no UTI's. Two studies included infants on NCPAP ± oxygen
(Roberts et al., 2000; Rojas et al., 2003). One study included infants
on IV fluids and IV antibiotics (Roberts et al., 2000). One study only
included vaginally delivered infants (Ali et al., 2009). One study
 C. Cunningham et al. / Journal of Neonatal Nursing xxx (2017) 1e7 3

Fig. 1. Prisma flow diagram.

excluded infants whose mothers were <18years old (Rojas et al.,
2003). As regards feeding; one study included infants who were
only tolerating full feeds (Lamy-Filho et al., 2008), one where in-
fants were tolerating 50% of feeds (Boo and Jamli, 2007). Rojas et al.
(2003) reported the infants as having had parenteral ± enteral
nutrition. Types of feeds varied amongst participants of studies
also; five had infants who BF ± EBM PO/NG or spoon (Roberts et al.,
2000; Ramanathan et al., 2001; Kadam et al., 2005; Suman Rao
et al., 2008; Ali et al., 2009)/four had infants who were combined
BF þ EBM fed and formula fed (Kambarami et al., 1998; Ahn et al.,
2010; Mwendwa et al., 2012; Samra et al., 2013)/two didn't state
the type of infants' feeds (Cattaneo et al., 1998; El Moniem and
Morsy, 2011)/two didn't clarify whether EBM or formula or both
(Boo and Jamli, 2007; Mohammadzadeh et al., 2011).
Sample size
The total sample size was 2487 participants. The mean was 146.
The smallest sample was 20 infants (Ahn et al., 2010) and the
largest was 985 (Lamy-Filho et al., 2008).
Interventions
N ¼ 17 studies compared an experimental or intervention group
against a control group. All experimental group infants matched
the inclusion and exclusion criteria of the control groups. All
experimental groups received KC or Kangaroo Mother Care (KMC).
All control groups received Conventional Mother Care (CMC) or
Traditional Holding Care (THC). N ¼ 17 measured weight of infants
in experimental and control groups, reported weight in grams and
whether up or down per day/week, or a length of hospitalisation
mean weight loss or gain.
Quality appraisal of included studies
Validity is deemed apparent if the overall validity score (the Yes/
Total) is >/ ¼ 75% or if the No þ Unclear/Total is </ ¼ 25%. The re-
sults of the validity scores were; 76% (Kambarami et al., 1998), 88%
(El Moniem and Morsy, 2011), 88% (Samra et al., 2013), 85% (Lee and
Sook, 2011), 81% (Lamy-Filho et al., 2008), 85% (Kashaninia and
Dehghan, 2015) which indicates conclusions can be generalisable
to the population. Ahn et al. (2010) scored 68% which indicates a
lower validity score.
SELECTION BIAS (randomisation)
Mohammadzadeh et al. (2011) doesn't state any sequence. Rojas
et al. (2003), Boo and Jamli (2007), Mohammadzadeh et al. (2011)
and Suman Rao et al. (2008) don't state whether randomisation
was undertaken after initial screening. Cattaneo et al. (1998),
Roberts et al. (2000), Ramanathan et al. (2001), Kadam et al.
(2005), Ali et al. (2009) and Mwendwa et al. (2012) all state that

Please cite this article in press as: Cunningham, C., et al., Does Kangaroo care affect the weight of preterm/low birth-weight infants in the
neonatal setting of a hospital environment?, Journal of Neonatal Nursing (2017), https://doi.org/10.1016/j.jnn.2017.10.001
4 C. Cunningham et al. / Journal of Neonatal Nursing xxx (2017) 1e7
randomisation took place after screening.
SELECTION BIAS (allocation)
Cattaneo et al. (1998), Kadam et al. (2005) and Suman Rao et al.
(2008) report concealed allocation through a sealed envelope
technique. Roberts et al. (2000), Ramanathan et al. (2001), Rojas
et al. (2003), Lamy-Filho et al. (2008), Ali et al. (2009), Ahn et al.
(2010), Mohammadzadeh et al. (2011) and Mwendwa et al. (2012)
don't state anything with reference to allocation concealment so
it is unclear if it took place or not. Boo and Jamli (2007) report a lack
of allocation concealment. Samra et al. (2013) and Kashaninia and
Dehghan (2015) report convenience sampling as a method of
allocation but not if it was concealed. Kambarami et al. (1998)
report allocation by a trained research midwife and El Moniem
and Morsy (2011) by a NICU consultant, but neither state if allo-
cation was concealed.
Performance bias
Boo and Jamli (2007) kept groups separate to try to avoid
contamination. El Moniem and Morsy (2011) had the experimental
group in one hospital and the control group in another hospital.
Kashaninia and Dehghan (2015) prevented the exchange of infor-
mation between groups by studying the experimental group first
and then afterwards conducting the study of the control group.
Cattaneo et al. (1998), Boo and Jamli (2007), Lamy-Filho et al.
(2008), Ali et al. (2009), El Moniem and Morsy (2011), Lee and Sook
(2011), Mwendwa et al. (2012) and Samra et al. (2013) state clearly
that the studies were not blinded. Suman Rao et al. (2008) report
how it was impossible to blind the investigators. Ramanathan et al.
(2001) says nurses assisted the mothers of both groups but didn't
clarify if the nurses were the investigators. Rojas et al. (2003) and
Kadam et al. (2005) report that parents knew what the studies
involved. Lee and Sook (2011) describe how mothers in the
experimental group were asked not to share information with
control group mothers until the study was over. Lamy-Filho et al.
(2008) state that the investigators supervised quality control with
regards to data collection carried out by field workers for the study,
but were not directly involved in the intervention. Kashaninia and
Dehghan (2015) report that participants were blinded but only to
each other. Kambarami et al. (1998), Roberts et al. (2000), Ahn et al.
(2010) and Mohammadzadeh et al. (2011) fail to refer in any way to
performance bias measures.
Detection bias
Investigators involvement in the intervention is unclear in
Kambarami et al. (1998), Roberts et al. (2000), Ramanathan et al.
(2001), Kadam et al. (2005), Suman Rao et al. (2008), Ali et al.
(2009), Ahn et al. (2010), Mohammadzadeh et al. (2011), Samra
et al. (2013) and Kashaninia and Dehghan (2015). Cattaneo et al.
(1998), El Moniem and Morsy (2011), Lee and Sook (2011) and
Mwendwa et al. (2012) report investigators had direct involvement
in the intervention. Rojas et al. (2003) and Lamy-Filho et al. (2008)
state investigators did not have direct participation in the inter-
vention. Boo and Jamli (2007) report nurses who weighed the in-
fants were unaware of the objectives or the design of the study.
Attrition bias
Cattaneo et al. (1998), Kambarami et al. (1998),Rojas et al.
(2003), Kadam et al. (2005), Boo and Jamli (2007), Lamy-Filho
et al. (2008), Suman Rao et al. (2008), Lee and Sook (2011) and
Mwendwa et al. (2012) reported factors that may have influenced
results in favour or against the intervention. Roberts et al. (2000),
Ramanathan et al. (2001), Ali et al. (2009), Ahn et al. (2010),
Mohammadzadeh et al. (2011), El Moniem and Morsy (2011),
Samra et al. (2013) and Kashaninia and Dehghan (2015) did not
Please cite this article in press as: Cunningham, C., et al., Does Kangaroo care affect the weight of preterm/low birth-weight infants in the
neonatal setting of a hospital environment?, Journal of Neonatal Nursing (2017), https://doi.org/10.1016/j.jnn.2017.10.001
report any incomplete or loss of data which makes attrition bias
unclear.
Reporting bias
All 17 studies reported outcomes for the interventions being
studied.
Contamination bias
Contamination was deemed to have occurred if the infants in
the control groups were breastfed or held skin to skin by the
mother whenever they wanted to. This could inadvertently in-
crease the weight of control infants. This occurred in studies from
Roberts et al. (2000), Ramanathan et al. (2001), Rojas et al. (2003),
Kadam et al. (2005), Lamy-Filho et al. (2008), Suman Rao et al.
(2008), Ali et al. (2009), Lee and Sook (2011) and Mwendwa et al.
(2012). This factor is unclear in studies from Cattaneo et al.
(1998), Kambarami et al. (1998), El Moniem and Morsy (2011)
and Kashaninia and Dehghan (2015). Mohammadzadeh et al.
(2011) recorded the time control group infants had with their
mothers if it was requested. Samra et al. (2013) reports that the
intervention group could breastfeed ad lib in addition to feeding
requirements where control group infants were not offered any
extra breastfeeding. Rojas et al. (2003) describes how the experi-
mental group were not discouraged from offering traditional
holding in lieu of KC.
Primary outcome measures: weight
RCT results
N¼10 RCT's six had a statistically significant increase in weight
gain for infants of the KC (intervention) group: Ali et al. (2009)/
p < 0.001; Cattaneo et al. (1998)/p < 0.00003; Mohammadzadeh
et al. (2011)/p < 0.001; Mwendwa et al. (2012)/p < 0.001;
Ramanathan et al. (2001)/p < 0.05; Suman Rao et al. (2008)/
p < 0.0001. Three of the RCT's reported an increase in weight of
infants in the intervention group but of no statistical significance:
Boo and Jamli (2007)/p ¼ 0.6; Kadam et al. (2005)/p ¼ 0.47; Rojas
et al. (2003)/p ¼ 0.1). The remaining RCT of Roberts et al. (2000)
reported no difference between the intervention and control
groups as regards weight (both gained a mean of 23 g/day ± 7 g).
Meta-analysis of these RCT's (as shown in Fig. 2) shows a statistical
significance between the intervention and control groups
(p < 0.00001) which would indicate that KC promoted weight gain
in these neonatal infants. Kadam et al. (2005) had high standard
deviation numbers (211 g for KMC and 205 g for CMC) making it
difficult to plot through meta-analysis. The df value (degree of
freedom) is 9 which indicates the number of values in the final
calculation of a statistic that are free to vary. There is a Chi squared
value of 65.50 which detects whether distributions of categorical
variables differ from one another. The Z score is 15.81 which in-
dicates that there are 15.81 elements greater than the standard
deviation above the mean. The I square is 86% which indicates a
considerable degree of heterogeneity between the studies.
Other quantitative studies results
Four reported a statistically significant difference in the increase
in the weight of infants through the provision of KC (El Moniem and
Morsy (2011)/p0.000; Kambarami et al. (1998)/p < 0.001;
Kashaninia and Dehghan (2015)/p < 0.009; Samra et al. (2013)/
p < 0.001). Two reported an increase in weight gain of infants in the
KC group but of no statistical significance (Ahn et al. (2010)/
p ¼ 0.733; Lee and Sook (2011)/p ¼ 0.107). Lamy-Filho et al. (2008)
reported a statistical significance in favour of the control group
with regards to weight gain (p ¼ 0.012).
 C. Cunningham et al. / Journal of Neonatal Nursing xxx (2017) 1e7
Fig. 2. Kangaroo care versus Conventional care.
Secondary outcome measure: breastfeeding
RCT results
All RCT's in the systematic review reported on or discussed
breastfeeding as a secondary outcome of significance in KC groups.
Statistical significance reflecting an increased uptake in breast-
feeding because of KC were reported in five out of the ten RCT's; Ali
et al. (2009)/p < 0.01; Boo and Jamli (2007)/p ¼ 0.04; Cattaneo et al.
(1998)/p ¼ 0.0003; Ramanathan et al. (2001)/p < 0.05; Rojas et al.
(2003)/p ¼ 0.06. However, the p value shown for Ramanathan
et al. (2001) was not a figure for breastfeeding within hospital. It
is indicative of exclusive breastfeeding at six weeks follow up.
Suman Rao et al. (2008) reported that at the end of their study 98%
vs. 76% in favour of KC group were exclusively breastfeeding.
Mwendwa et al. (2012) reported similar breastfeeding rates in their
study (93.6% in KMC vs. 92.8% in CMC). However, they found that
the KMC group breastfed earlier with 40.3% of infants breastfeeding
within the first week of life compared to 14.9% in the first week of
life of the CMC group. Roberts et al. (2000) reported equal rates of
breastfeeding (11 CMC vs. 10 KMC). Kadam et al. (2005) reported an
earlier initiation of breastfeeding in the KMC group (4.7 days)
compared with the CMC group (5.6 days), but the difference is not
of statistical significance. Mohammadzadeh et al. (2011) did not
report any breastfeeding outcomes but state that infants at enrol-
ment were breastfeeding or orogastric tube fed EBM in the two
groups.
Other quantitative studies
Lamy-Filho et al. (2008) report exclusive breastfeeding at
discharge statistically significant in favour of the KC group
(p ¼ 0.022). El Moniem and Morsy (2011), Kashaninia and Dehghan
(2015), Lee and Sook (2011) and Samra et al. (2013) all discuss and
narrate how KC is attributable to an increase in breastfeeding rates
because of increased opportunities to breastfeed. Ahn et al. (2010)
report no differences between control and intervention groups
with regards to breastfeeding. Kambarami et al. (1998) do not
discuss breastfeeding uptake in their results.
Discussion
It is apparent from this SR that KC is a positive factor in the
increase of weight of preterm/LBW infants in a neonatal hospital
environment. However, the reason weight gain occurs through KC
can be considered physiologically subjective, but is yet supported
through recent literature. Chiu and Anderson (2009) describe how
Please cite this article in press as: Cunningham, C., et al., Does Kangaroo care affect the weight of preterm/low birth-weight infants in the
neonatal setting of a hospital environment?, Journal of Neonatal Nursing (2017), https://doi.org/10.1016/j.jnn.2017.10.001
5
quiet sleep states promote better physiological outcomes in infants,
including weight gain. This state could be applied to the KMC
method. Tourneux et al. (2009) report how infants expend energy
through thermoregulation, metabolism and growth. Samra et al.
(2013) describe how through KC the infant saves energy on meta-
bolism and thermoregulation therefore focusing the majority of
energy on growth. Field et al. (2011) and Keshavarsand et al. (2012)
studied therapeutic touch as a method of weight gain promotion in
neonates. KC could also be incorporated in these results as thera-
peutic touch through SSC would be implied.
It is clear from this SR that deficiencies and constraints on
healthcare systems, in particular of lower socio-economic class can
affect how KC was accommodated. Disadvantages in surroundings
through lack of privacy or lack of comfortable chairs or beds
reduced the amount of time KC was carried out by the mother. Mills
(2014) details how in the past 10 years a focus of drugs and medical
supplies is emphasised for low income and socio-economic coun-
tries and that the broader health care needs of institutions are
being overlooked. This is comparable to this instance, where the
basic needs of mother and baby could be easily facilitated. Advan-
tages for the health care institute could then be recognisable
through earlier discharge and better outcomes for the infant
(Namnabati et al., 2016). This factor is confounded in the study by
Ahn et al. (2010) where the provision of comfortable beds, chairs,
private rooms and even front opening gowns for the mothers had a
positive and statistically significant effect on the infants’ weight
gain. Overcrowding and sharing of incubators or cots was described
in Kambarami et al. (1998) and Mwendwa et al. (2012) in their CMC
groups. The results of both studies showed statistical significance in
favour of the KC group who were nursed in quiet areas and pro-
vided with comfortable surroundings. The inconsistent and
disruptive nature of the CMC group nursing practices would have
influenced slower weight gaining opportunities for these infants.
The longer KC was provided the more positive were the results.
For example, Boo and Jamli (2007), Lee and Sook (2011) and Roberts
et al. (2000) had short duration KC hours per day in their studies
(1hr/day; 30mins/day; 2hrs/day). All three of these studies found
no statistically significant differences between the results of control
and intervention groups. Moreover, the mothers in the study by
Cattaneo et al. (1998) provided 20 h per day of KC with their infants
and the study found a statistical significance of p < 0.00003 in their
results. It can be deduced that this finding causes an effect on
weight gain that hasn't been previously reported on.
The degree of stress endured also played a role in affecting how
the infant gained weight. A study by Motil and Duryea (2017) de-
tails how physiological stress, including respiratory distress,
6 C. Cunningham et al. / Journal of Neonatal Nursing xxx (2017) 1e7
requires the infant to expend a higher than normal amount of
calories compared to an infant without this stress. In this system-
atic review both studies by Boo and Jamli (2007) and Roberts et al.
(2000) included infants who were NCPAP dependant ± requiring
oxygen. Both studies found no statistical difference between groups
with regards to weight gain. All other studies in this review except
Lamy-Filho et al. (2008) had an increase in weight gain reported in
their studies and had no infants on NCAP or oxygen. This point
highlights the need for studies with regards to KC and ill preterm/
LBW infants. Cultures and traditions also became apparent in their
ability to sway the result of studies with regard to effective KC
provision in the NICU environment. For example, Boo and Jamli
(2007) conducted their study in Malaysia where KC was a rela-
tively new concept of care. Eight infants in the study received KC
only 50% of the time of the length of the study. It became apparent
that in Malaysia it is traditionally acceptable for the mother to stay
at home on bed rest for six weeks postpartum. This cultural influ-
ence contributed to the lack of attendance of these mothers to
provide KC for their infants.
If the infant was on full enteral feeds or not is another point of
discussion. Hay (2013) reports how full enteral feeding of preterm
infants allows faster regaining of birth weight in general. The
studies where infants were not on full feeds did not gain weight of a
statistical significance compared to the infants who were on full
milk feeds in this SR. Roberts et al. (2000) included infants who
were on IV fluids in their study, and Lee and Sook (2011) did not
exclude infants on IV fluids in their study. Both of these studies
reported no statistical difference with regards to weight gain in
their results between the two groups. Therefore differing inclusion
criteria influences differing results in the literature which needs to
be cautiously interpreted in its applicability to practice.
The exclusive breastfeeding rates of KC groups was of signifi-
cance in this SR. Exclusive breastfeeding is when an infant receives
only breast milk without the addition of other food or formula
(WHO, 2017). Breast milk is considered the best source of nutrition
for an infant and would be the optimal method of feeding for a
preterm/LBW infant (Motee and Jeewon, 2014). The Lamy-Filho
et al. (2008) study showed that exclusive breastfeeding occurred
2.34 times more frequently in the KC group than the control group.
Samra et al. (2013) found a statistical significance of p < 0.001 in
favour of KC promoting weight gain and within this group there
was 17.4 times more opportunities to breastfeed than the CMC
group. Outside of this SR a study by Heidarzadeh et al. (2013)
concurs with this finding. They found that KC had a statistically
significant effect on exclusive breastfeeding of preterm infants in a
NICU in Iran (p ¼ 0.00).
A recommendation for practice from the results of this SR would
be the facilitation by health institution managers of nurse educa-
tion in their role as KC facilitators. The implementation of KC as a
primary point of care for preterm/LBW infants and their mothers is
key to this education becoming a success. This is because nurses
have a key role in detecting barriers to KC and facilitating it effi-
ciently to reduce disruptions and maximise its potential benefits for
infant and mother. An international KC educational initiative re-
ported on by Bergh et al. (2012) advises change agents, educational
models, awareness, teamwork and managerial support as key to its
success. There are teaching models readily available within the
literature to assist in this education (Stikes and Barbier, 2013;
Namnabati et al., 2016; Cooper et al., 2014).
The results of this SR recommend that NICU's internationally
incorporate KC as a primary standard of care in clinical practice.
However, this recommendation is made with caution as there were
some methodological issue that may impact on included studies
validity. Differing enrolment demographics giving a variation in
figures effects clarity of results when considering application to
Please cite this article in press as: Cunningham, C., et al., Does Kangaroo care affect the weight of preterm/low birth-weight infants in the
neonatal setting of a hospital environment?, Journal of Neonatal Nursing (2017), https://doi.org/10.1016/j.jnn.2017.10.001
practice. Cattaneo et al. (1998) conducted their study in three
neonatal departments in three different countries (Addis Ababa,
Indonesia and Mexico) where there were differences in birth
weights of infants, gestational age and weight at enrolment. There
were also differing social, economic and health factors. Many dif-
ferences make a true comparison problematic. Nijhawan et al.
(2013) state that informed consent is a legal and ethical require-
ment for any research involving human participants. Yet, in the
study by Boo and Jamli (2007) informed consent was not obtained
from the CMC groups in order to prevent contamination bias. This
confounding ethical issue questions acceptance of the application
of this study's results to practice. Kambarami et al. (1998) reports
their method of allocation as problematic in that the researcher
could predict what group the next eligible infant was going to and
resisted allocating vulnerable infants to the experimental group.
Viera and Bangdiwala (2007) discuss how selection bias can be
removed through allocation concealment and how important this
is to the validity of a study's results. Lamy-Filho et al. (2008) was the
only study in this systematic review to find an increase in weight of
the CMC group (p ¼ 0.012). The sample size of 985 infants across 16
Brazilian NICU's is convincing in its transferability or results.
However, a major confounding factor in this study was potential
contamination. The authors report that health professionals in the
NICU's across Brazil had recently taken part in training in the
practice of KC that was promoted by the Brazilian Ministry of
Health. They report that almost all of the control units had KC
practices that could possibly cancel out any differences between
the two groups. Inclusion and exclusion criteria in the studies also
necessitated that the infants be well and stable. This is not a true
total representation of a NICU environment. Very well preterm
neonates would only make up a proportion of the population in the
NICU. Finally, the I squared score of the meta-analysis of this sys-
tematic review is 86% which indicates a considerable degree of
heterogeneity between the studies. This could be considered as a
limitation of how generalisable the results of this systematic review
are to practice (Groenwold et al., 2010). However, the authors argue
that the presentation of the results of the individual studies allows
transparency. Clarity can then be achieved in how the results may
have been affected by intrinsic confounding factors that occurred
within individual studies. Thus, the recommendation of KC as a
primary standard of care in NICU's globally is still justified as is the
recommendation of staff education to implement it effectively.
Conclusion
The authors recommend further research of the effects of KC on
the weight gain of ill preterm/LBW neonatal infants. It is also
apparent that further research is warranted in how many hours per
day of KC are advisable in order to promote weight gain in these
infants. Finally, the study of traditional and cultural practices in
differing NICU's and how this affects preterm/LBW KC practices is
an evident gap in the literature where a SR would be warranted.
References
Ahn, H.Y., Lee, J., Shin, H.J., 2010. Kangaroo care on premature infant growth and
maternal attachment and postpartum depression in South Korea. J. Trop.
Pediatr. 56 (5), 342e344.
Ali, S.M., Sharma, J., Sharma, R., Alam, S., 2009. Kangaroo mother care as compared
to conventional care for low birth weight babies. Dicle Med. J. 36 (3), 155e160.
Bergh, A.M., Charpak, N., Ezeonodo, A., Udani, R.H., Van Rooyen, E., 2012. Education
and training in the implementation of kangaroo mother care. South Afr. J. Child
Health 6 (2), 38e48.
Bigelow, A., Power, M., MacLellan-Peters, J., Alex, M., McDonald, C., 2012. Effect of
mother/infant skin-to-skin contact on postpartum depressive symptoms and
maternal physiological stress. J. Obs. Gynecol. Neonatal Nurs. 41 (3), 369e382.
Boo, N.Y., Jamli, F.M., 2007. Short duration of skin to skin contact: effects on growth
and breastfeeding. J. Paediatr. Child Health 43, 831e836.
 C. Cunningham et al. / Journal of Neonatal Nursing xxx (2017) 1e7
Boundy, E.O., Dastjerdi, R., Spiegelman, D., Fawzi, W.W., Missemer, S.A.,
Lieberman, E., Kajeepeta, S., Wall, S., Chan, G.J., 2016. Kangaroo mother care and
neonatal outcomes: a meta-analysis. Pediatrics 137 (1), 2e16.
Cattaneo, A., Davanzo, R., Worku, B., Surjono, A., Echeverria, M., Bedri, A., Haksari, E.,
Osorno, L., Gudetta, B., Setyowireni, D., Quintero, S., Tamburlini, G., 1998. Kan-
garoo mother care for low birth weight infants: a randomised controlled trial in
different settings. Acta Paediatr. 87, 976e985.
Chan, G.J., Labar, A.S., Wall, S., Atun, R., 2015. Kangaroo mother care: a systematic
review of barriers and enablers. Bull. World Health Organ. (Online) Available at:
https://www.who.int./bulletin/volumes/94/2/15-157818/en/ (Accessed 23
March 2017).
Chiu, S.H., Anderson, G.C., 2009. Effect of early skin to skin contact on mother-
preterm infant interaction through 18 months: randomised controlled trial.
Int. J. Nurs. Stud. 46 (9), 1168e1180.
Conde-Agudelo, A., Diaz-Rossello, J.L., 2016. Kangaroo mother care to reduce
morbidity and mortality in low birth weight infants (Online) Available at:
Cochrane Database Syst. Rev. https://doi.org/10.1002/14651858 . (Accessed 29
December 2016).
Cooper, L., Morill, A., Russell, R.B., Gooding, J.S., Miller, L., Berns, S.D., 2014. Close to
Me, enhancing kangaroo care practices for NICU staff and parents. Adv.
Neonatal Care 14 (6), 410e423.
Dodd, V., 2005. Implications of kangaroo care for growth and development in
preterm infants. J. Obstetric Gynecol. Neonatal Nurs. 34 (2), 218e232.
El Moniem, I.I.A., Morsy, M.A., 2011. The effectiveness of kangaroo technique on
preterm infant's weight gain. J. Am. Sci. 7 (1), 687e702.
Field, T., Diego, M., Hernandez-Reif, M., 2011. Potential mechanisms for weight gain
in massaged preterm infants. Infant Behav. Dev. 34, 383e389.
Glynn, L., 2006. Evidence Based Library Critical Appraisal Checklist (Online) Avail-
able at: http://bit.ly/IN8qOa. (Accessed 4 January 2017).
Groenwold, R.H.H., Rovers, M.M., Lubsen, J., van der Heijden, G.J.M.G., 2010. Sub-
group effects despite homogenous heterogeneity test results. Br. Med. Counc.
Med. Res. Meth. 10 (43).
Heidarzadeh, M., Hossein, M.B., Eishadmanesh, M., Tabari, M.G., Khazaee, S., 2013.
The effect of kangaroo mother care on breastfeeding at the time of NICU
discharge. Iran. Red Crescent Med. J. 15 (4), 302e306.
Higgins, J.P.T., Green, S., 2011. Cochrane Handbook for Systematic Reviews of In-
terventions: Version 5.1.0 the Cochrane Collaboration (Online) Available at:
https://www.cochrane-handbook.org. (Accessed 1 November 2016).
Johnston, C., Campbell-Yeo, M., Fernandes, A., Inglis, D., Streiner, D., Zee, R., 2014.
Skin to Skin Care for Procedural Pain in Neonates (Online) Available at: https://
www.ncbi.nlm.nih.gov/pubmed/24459000. (Accessed 12 February 2017).
Kadam, S., Binoy, S., Kanbur, W., Mondkar, J.A., Fernandez, A., 2005. Feasibility of
kangaroo mother care in Mumbai. Indian J. Pediatr. 72, 35e38.
Kambarami, R.A., Chidede, O., Kowo, D.T., 1998. Kangaroo care versus incubator care
in the management of well preterm infants-a pilot study. Ann. Trop. Paediatr. 18
(2), 81e86.
Kashaninia, Z., Dehghan, M., 2015. The effect of kangaroo care on weight gain of
premature neonates in hospitalised in neonatal intensive care units. Biosci.
Biotechnol. Res. Asia 12 (2), 1405e1410.
Keshavarsand, M., Kiani, A., Nasani, L., Hoseini, F., 2012. Effect of touch therapy by
mothers on weight gaining of preterm newborns. Koomesh 13 (2), 240e246.
Lamy-Filho, F., da Silva, A.A.M., Lamy, Z.C., Gomes, M.A.S.M., Moreira, M.E.L.,
Canguru, G.A.M., Neonatais, R.B.P., 2008. Evaluation of the neonatal outcomes of
the kangaroo mother method in Brazil. J. Pediatr. 84 (5), 428e435.
Please cite this article in press as: Cunningham, C., et al., Does Kangaroo care affect the weight of preterm/low birth-weight infants in the
neonatal setting of a hospital environment?, Journal of Neonatal Nursing (2017), https://doi.org/10.1016/j.jnn.2017.10.001
7
Lee, J., Sook, B.K., 2011. The effects of kangaroo care on maternal self esteem and
premature infants' physiological stability. Korean J. Women’s Health 17 (5),
454e462.
Leonard, A., Mayers, P., 2008. Parents lived experience of providing kangaroo care to
their preterm infants. Health S. Afr. Gesondheid 13 (4), 16e27.
Mills, A., 2014. Healthcare systems in low and middle income countries. N. Engl. J.
Med. 370, 552e557.
Mohammadzadeh, A., Farhat, A., Jafarzadeh, M., Hasanzadeh, L., Esmaeli, H., 2011.
Advantages of kangaroo care in less than 2000grams low birth weight neo-
nates. Med. J. Islamic Repub. Iran 25 (1), 11e15.
Moore, E.R., Anderson, G.C., Bergman, N., Dowsell, T., 2012. Early Skin to Skin
Contact for Mothers and Their Healthy Newborn Infants (Online) Available at:
https://www.ncbi.nlm.nih.gov/pubmed/22592691. (Accessed 23 March 2017).
Motee, A., Jeewon, R., 2014. Importance of Exclusive Breastfeeding and comple-
mentary feeding among infants. Curr. Res. Nutr. Food Sci 2 (2).
Motil, K.J., Duryea, T.K., 2017. Patient Education: Poor Weight Gain in Infants and
Children (Beyond the Basics). Up To Date (Online) Available at: https://www.
uptodate.com/contents/poor-weight-gain-in-infants-and-children-beyond-the-
basics. (Accessed 5 March 2017).
Mwendwa, A.C., Musoke, R.N., Wamalwa, D.C., 2012. Impact of partial kangaroo
mother care on growth rates and duration of hospital stay of low birth weight
infants at the Kenyatta national hospital. Nairobi. East Afr. Med. J. 89 (2), 53e58.
Namnabati, M., Talakoub, S., Mohammadizadeh, M., Mousaviasl, F., 2016. The
implementation of kangaroo mother care and nurses perspective of barriers in
Iranian neonatal intensive care units. Iran. J. Nurs. Midwifery Res. 21 (1), 84e88.
Nijhawan, L.P., Janodia, M.D., Muddukrishna, B.S., Bhat, K.M., Bairy, K.L., Udupa, N.,
Musmade, P.B., 2013. Informed consent: issues and challenges. J. Adv. Pharm.
Technol. Res. 4 (3), 134e140.
Ramanathan, K., Paul, V.K., Deorari, A.K., Taneja, U., George, G., 2001. Kangaroo
mother care in very low birth weight infants. Indian J. Pediatr. 68, 1019e1023.
Roberts, K.L., Paynter, C., McEwan, B., 2000. A comparison of kangaroo mother care
and conventional cuddling care. Neonatal Netw. 19 (4), 31e35.
Rojas, M.A., Kapplan, M., Quevedo, M., Sherwonit, E., Foster, L.B., Ehrenkranz, R.A.,
Mayes, L., 2003. Develop. Behav. Pediatr. 24 (3), 163e168.
Samra, N.M., El Taweel, A., Cadwell, K., 2013. Effect of intermittent kangaroo mother
care on weight gain of low birth weight neonates with delayed weight gain.
J. Perinat. Educ. 22 (4), 194e200.
Sharma, D., Murki, S., Pratapo, T., 2016. The effect of kangaroo ward care in com-
parison with “intermediate intensive care” on the growth velocity in preterm
infant with birth weight <1100g: randomised controlled trial. Eur. J. Pediatr. 175
(10), 1317e1324.
Stikes, R., Barbier, D., 2013. Applying the plan-do-study-act model to increase the
use of kangaroo care. J. Nurs. Manag. 21, 70e78.
Suman Rao, P.N., Rekha, U., Ruchi, N., 2008. Kangaroo mother care for low birth
weight infants: a randomised controlled trial. Indian Pediatr. 45, 17e23.
Tourneaux, P., Libert, J.P., Ghyselen, L., Leke, A., Delanaud, S., Degrugilliers, L.,
Bach, V., 2009. Heat exchanges and thermoregulation in the neonate. Arch.
Pediatr. 16 (7), 1057e1062.
Viera, A.J., Bangdiwala, S.T., 2007. Eliminating bias in randomised controlled trials:
importance of allocation concealment and masking. Fam. Med. 39 (2), 132e137.
World Health Organisation, 2017. Exclusive Breastfeeding (Online) Available at:
www.who.int/nutrition/topics/exclusive_breastfeeding/en/. (Accessed 8 March
2017).