Acta Pdiatrica ISSN 08035253

REGULAR ARTICLE

Long-term effects of earlier initiated continuous Kangaroo Mother Care

(KMC) for low-birth-weight (LBW) infants in Madagascar
Shuko Nagai1,2, Naohiro Yonemoto3, Norotiana Rabesandratana2, Diavolana Andrianarimanana2,
Takeo Nakayama (nakayama.t@at2.ecs.kyoto-u.ac.jp)1, Rintaro Mori4
1.Department of Health Informatics, Kyoto University, School of Public Health, Kyoto, Japan
2.Centre Hospitalier Universitaire Mahajanga, Madagascar
3.National Center of Neurology and Psychiatry, Japan
4.Department of Global Health Policy, The University of Tokyo, Kodaira, Tokyo, Japan

Keywords
Kangaroo mother care, Long-term followup, Low
birth weight infants, Randomized controlled trial,
Resource-limited country
Correspondence
Takeo Nakayama, M.D., Ph.D., Department of Health
Informatics, Kyoto University, School of Public
Health, Yoshidakonoe Sakyo, Kyoto 606-8501,
Japan.
Tel: +81-75-753-4488 |
Fax: +81-75-753-4497 |
Email: nakayama.t@at2.ecs.kyoto-u.ac.jp
Received
6 February 2011; revised 23 May 2011;
accepted 30 May 2011.
DOI:10.1111/j.1651-2227.2011.02372.x

ABSTRACT
Aim: To examine the long-term effects of earlier initiated continuous Kangaroo
Mother Care (KMC) for relatively stable low-birth-weight (LBW) infants in a resource-limited
country.
Methods: A randomized controlled trial with long-term follow-up was performed in
LBW infants in Madagascar. Earlier continuous KMC (intervention group) was initiated as
soon as possible within 24 h postbirth, and later continuous KMC (control group: conventional care) was initiated after complete stabilization. Outcome measures were mortality or
readmission, nutritional indicators at 612 months postbirth and feeding condition at
6 months postbirth (ClinicalTrials.gov, NCT00531492).
Results: A total of 72 infants were followed for mortality or readmission at 612
months postbirth. There was no difference between the two groups (7 36 vs. 7 36, Risk
ratio (RR), 1.00; 95% CIs, 0.392.56; p = 1.00). The proportion of exclusive breast feeding (EBF) at 6 months postbirth was significantly higher with earlier KMC than later KMC
(12 29 vs. 4 26; RR 2.69; 95% CIs, 1.007.31; p = 0.04). There were no differences in
nutritional indicators between the two groups at 612 months postbirth.
Conclusion: Earlier initiated continuous KMC results in a significantly higher proportion of EBF at 6 months postbirth. Further larger-scale long-term evaluations of earlier
initiated continuous KMC for LBW infants are needed.

INTRODUCTION
Kangaroo Mother Care (KMC) is an important postnatal
intervention for low-birth-weight (LBW) infants in
resource-limited settings. KMC, which is defined by continuous skin-to-skin contact (SSC) between the infant and
mother to prevent hypothermia, promote breastfeeding and
strengthen motherinfant bonding, has been performed on
LBW infants for more than 30 years as a complement to
incubator care (1,2). KMC is listed as one of the five most
efficient interventions in postnatal care in resource-limited
settings (3). Despite its widespread use, KMC had not been
reported to have a conclusive effect on mortality, and evidence was limited to recommend its routine use for LBW
infants (4). A recent updated review, which discussed 12
additional trials and more data from individual studies,
reported the use of KMC in stabilized LBW infants as an
alternative to conventional neonatal care mainly in
resource-limited settings (5). In some of these trials, however, KMC was commenced several days postbirth.
Because 2545% of neonatal deaths occur within 24 h
postbirth (6), interventions must be initiated as soon as
possible.

For healthy term and late preterm infants, a meta-analysis

of more than 30 randomized controlled trials (RCTs) has
shown that early SSC results in better breastfeeding and a
better motherinfant relationship (7). In this review, early
SSC was defined as SSC initiated any time between 1 and
24 h postbirth. However, continuous care was not tested as
a complement to an incubator in these trials, and most were
not performed in resource-limited settings.

Key notes

2011 The Author(s)/Acta Pdiatrica 2011 Foundation Acta Pdiatrica 2011 100, pp. e241e247

To examine the long-term effects of earlier initiated

continuous Kangaroo Mother Care (KMC) for relatively
stable low-birth-weight (LBW) infants, a randomized,
controlled trial was performed in a resource-limited
country. A total of 72 infants were followed, and earlier
initiated continuous KMC resulted in a significantly
higher proportion of EBF at 6 months postbirth. Further
adequately designed and larger-scale long-term evaluations of earlier initiated continuous KMC for LBW infants
are needed.

e241

Long-term effect of earlier initiated continuous KMC for LBW

Nagai et al.

We conducted a RCT of earlier initiated (within 24 h

postbirth) continuous KMC for relatively stable LBW
infants in Madagascar. We previously reported short-term
outcomes (28 days postbirth) and observed higher, but not
statistically significant, neonatal mortality and significantly
lower weight loss from birth to 24 and 48 h postbirth (8).
However, information on short-term outcomes is insufficient to assess the overall benefits and risks of common
postnatal interventions. The aim of this study was to examine the long-term effects (612 months postbirth) of earlier
continuous KMC for relatively stable LBW infants on mortality or readmission, nutritional indicators at 612 months
postbirth and feeding conditions at 6 months postbirth in a
resource-limited setting.

METHODS
Study design
Infants born at the University Hospital of Mahajanga,
Madagascar were assessed for eligibility to participate in
the trial. Eligibility criteria were (i) birth weight under
2500 g, (ii) less than 24 h postbirth, (iii) no serious malformation, (iv) relatively stable clinical condition (oxygen saturation, 95% or more; heart rate, >100 beats per minute;
respiratory rate, <60 times per minute; capillary refilling
time, <3 sec), (v) mother and other family members were
willing to practise KMC and (vi) mother and or family
willing to practise KMC were healthy. Exclusion criteria
were (i) prolonged apnoea (more than 20 sec) and (ii)
intravenous infusion.
If an infant satisfied all eligibility criteria, we obtained
written informed consent from the mother and or father at
the time of enrolment in the study. After obtaining a signature for consent, randomized allocation was carried out
using the minimization method with Minim software. The
prescribed sample size at the beginning of the trial or the
number of infants that may be registered in the trial in
1 year was 100 infants. The sample size was based on an
effect size of 5 in risk ratio (RR) and estimated mortality of
25% in the control group. Details on exclusion criteria, randomization procedures and short-term outcomes have been
reported previously (8).
A total of 73 infants were enroled from August 2007 to
August 2008 and randomly assigned to two groups. The earlier initiated continuous KMC group was instructed to begin
KMC as soon as possible within 24 h postbirth. KMC was
defined as direct and continuous SSC (without any underwear except for a diaper, a warm hat and socks for the
infant) for as long as possible. The later continuous KMC
group initially followed conventional care. Hospital staff
used an incubator or radiant warmer first, and they later
covered the infants with cotton cloth and laid them beside
their mothers in the same bed, but without SSC. KMC was
initiated when the infant and mothers families were completely settled and ready, which was approximately 24 h
postbirth or later. After initiating KMC, all participants were
encouraged by hospital staff to continue KMC for as long as
possible during hospitalization and after discharge. Other

e242

family members occasionally assisted the mother in performing continuous KMC. In our previous study, participants were followed until 28 days postbirth (8). For this
long-term follow-up study, we observed these participants
until 612 months postbirth (i.e. from September 2008 to
March 2009).
The protocol for this follow-up study was approved by
the Institutional Ethics Committee of the Ministry of Health
in Madagascar and the Research Ethics Board of Kyoto
University in Japan. After approving, the registration of
ClinicalTrials.gov was revised (number NCT00531492).
Outcomes
The primary outcome was mortality or readmission at 6
12 months postbirth (i.e. from 1 week before 6 months
postbirth to 1 week after 12 months postbirth). Readmission was surrogated for morbidity during the trial period.
Data were collected from hospital charts, maternal and
child health handbooks and family interviews at 6
12 months postbirth. Secondary outcomes were (i) feeding
status at 6 months postbirth and (ii) nutritional indicators
at 612 months postbirth. For feeding status, we ensured
exclusive breastfeeding (EBF), in which the infant received
only breast milk and nothing else. Medicine, oral rehydration solution, vitamins and minerals, as recommended by
health providers, were allowed during EBF (9). Nutritional
indicators were assessed by (i) stunting (measured by
height-for-age), (ii) wasting (measured by weight-for-height)
and (iii) underweight (measured by weight-for-age). We followed the World Health Organization (WHO) child growth
standards (10) and the growth standards of Madagascar
(11). WHO Anthro 2005 software (WHO, Geneva, Switzerland, http://www.who.int/childgrowth/software/en/) was
used for calculating the Z-score of each indicator. The cutoff point for malnutrition was a Z-score < )2, and severe
malnutrition corresponded to a Z-score < )3 (12). The
growth standard of Madagascar is used for rapid assessment
of acute malnutrition screening, which is based on the
National Center for Health Statistics growth reference (13).
This standard only observes wasting. The cutoff point for
malnutrition was <75%, and severe malnutrition corresponded to <60% (11).
Data collection
Data were collected daily by a research coordinator through
interviews with mother family and from medical records
during hospitalization. To observe health conditions at 6
12 months postbirth, another research coordinator visited
all participants at home. If the family was absent at the
appointed time, the research coordinator visited repeatedly
until a meeting with the infant and mother family was
achieved, and the participants information (health status,
measurements of weight and height) was collected. If the
family relocated outside the follow-up area, the research
coordinator telephoned the family and or visited a relatives
house and obtained the participants information (health
status). The research coordinator was blinded to participant
allocation.

2011 The Author(s)/Acta Pdiatrica 2011 Foundation Acta Pdiatrica 2011 100, pp. e241e247

Nagai et al.

Long-term effect of earlier initiated continuous KMC for LBW

Statistical analysis
Participant baseline characteristics and outcomes are
reported through means and standard deviation (SD), frequency and percentage for comparison between the two
groups. Fishers exact test was used for binary comparisons,
and analysis of variance was used to compare continuous
data. RR and 95% confidence intervals (95% CIs) were used
to compare incidences in primary and secondary outcomes
between the two groups. Adjusted analyses were conducted
with birth weights, Apgar score, gender for EBF at 6 months
postbirth and nutritional indicators at 612 months postbirth in secondary outcomes using a logistic regression
model. Nonparametric analysis by Wilcoxon rank-sum test
was conducted for distribution of nutritional indicators
(Z-score of WHO child growth standard, percentage of the
growth standards of Madagascar). The two-sided level of
significance was set at p < 0.05. JMP software version 6.0

(SAS Institute, Cary, NC, USA: http://www.jmp.com/) was

used for statistical analyses.

RESULTS
Study participants
The number of infants who were screened, randomly
assigned to earlier or later continuous KMC and assessed at
612 months postbirth are shown in Figure 1. Follow-up
assessments began in September 2008 and ended in March
2009. One infant was lost to follow-up because the family
relocated and could not be contacted by telephone, and
neighbours had no information on their whereabouts. Adequate data for analysis of the primary outcome were available for 72 (98.6%) infants. Thirty-one infants (42.5%)
relocated from where they had lived at birth at least once
during the 612 months postbirth. Of these, 17 were

1126 newborn babies were born

at University hospital of Mahajanga
1005 babies were excluded
892 > 2500 g at birth (not LBW)
71 LBW but could not get stable condition <24 h
42 LBW but mother and family were not motivated
/available for KMC

121 babies were eligible

48 babies did not participate
21 Living outside study follow-up area
2 Refused to sign for Informed consent
6 Started KMC before Informed consent
6 Called study staff after 24h of birth
4 Randomization was not available(Network problem)
9 Study staff was not available within 24h

73 babies underwent randomization

37 were assigned to
earlier KMC group

36 were assigned to
later KMC group

1 loss to follow-up

36 infants had adequate data

for analysis of the primary
outcome

36 infants had adequate data

for analysis of the primary
outcome

2 dead
5 moved to outside
study follow-up area

2 dead
8 moved to outside
study follow-up area

29 had adequate data for

analysis of the secondary
outcomes

26 had adequate data for

analysis of the secondary
outcomes

Figure 1 Enrolment, randomization and follow-up of the study participants.

2011 The Author(s)/Acta Pdiatrica 2011 Foundation Acta Pdiatrica 2011 100, pp. e241e247

e243

Long-term effect of earlier initiated continuous KMC for LBW

Nagai et al.

Table 1 Characteristics of study participants for secondary outcomes

Earlier KMC
group (N = 29)
Birth weight (g)
Mean (SD)
2082.2 (250.3)
Birth weight*, n (%)
2000
11 (37.9)
20012500
18 (62.1)
Gestational age at birth (weeks)
Mean (SD)
36.8 (2.07)
Gestational age at birth, n (%)
3233
2 (6.9)
3436
13 (44.8)
37
14 (48.3)
Classification based on Lubchencos charts, n (%)
Preterm AGA
12 (41.4)
Preterm SGA
3 (10.3)
Term AGA
2 (6.9)
Term SGA
12 (41.4)
Gender, n (%)
Male
12 (41.4)
Delivery type*, n (%)
Normal delivery
21 (72.4)
Caesarean
8 (27.6)
Apgar score at 5 min*, n (%)
06
1 (3.4)
710
28 (96.6)
Elapsed time (h) from birth to randomization *, n (%)
012
12 (41.4)
1224
17 (58.6)

Later KMC
group (N = 26)

2074.3 (297.9)
12 (46.2)
14 (53.8)
36.0 (2.07)
3 (11.5)
11 (42.3)
12 (46.2)
13 (50.0)
1 (3.9)
3 (11.5)
9 (34.6)
19 (73.1)
20 (76.9)
6 (23.1)
1 (3.9)
25 (96.1)
11 (42.3)
15 (57.7)

*Characteristics were used in the randomization factors.

followed to their new house by contacting the family by telephone and or by asking the neighbourhood. Information
for 13 infants could only be obtained by telephone as the
family had relocated outside the follow-up area; only the
primary outcome was analyzed for these infants.

Characteristics of the 72 infants were similar in the two

groups at birth and at the time of randomization (8). The
secondary analysis characteristics of 55 infants who could
be followed were also similar (Table 1). For these 55 infants,
the mean number of hours postbirth to KMC initiation was
20.58 h (SD 15.85) in the earlier initiated KMC group and
31.46 h (SD 9.53) in the later KMC group.
Primary outcome
Results for primary and secondary outcomes are shown in
Table 2. There was no difference in mortality or readmission
between the two groups (RR, 1.00; 95% CIs, 0.392.56;
p = 1.00). Of the four infants who died, three died during
the first 28 days postbirth (earlier KMC, 2; later KMC, 1).
The primary causes of mortality in the earlier initiated KMC
group were neonatal infection (died on day 13) and
asphyxia (died during first 48 h postbirth). The cause of
mortality in one of the infants in the later KMC group was
maternal-foetal infection (died on day 17) (8). Another
infant in the later KMC group died 4 months postbirth to
malaria. This infant visited the hospital outpatient department three times in a week before death but could not be
saved. Primary causes of readmission were acute respiratory
tract infection (seven infants), acute enterocolitis (two
infants), high fever (one infant) and not sucking well (one
infant). There was no substantial difference between the
two groups.
Secondary outcomes
The earlier continuous KMC group showed a significantly
higher incidence of EBF at 6 months postbirth (12 29 vs.
4 26; RR, 2.69; 95% CI, 1.007.31; adjusted p = 0.04),
(Table 2). The difference would still be significant even if we
assumed that all those lost to follow-up were not EBF at
6 months postbirth (12 37 vs. 4 36; RR, 2.92; 95%CI, 1.048.21; adjusted p = 0.028).

Table 2 Primary and secondary outcomes

Primary outcome
Mortality or readmission to 612 months postbirth
Mortality
Readmission
Secondary outcomes
Exclusive breastfeeding at 6 months postbirth
Nutritional indicators at 612 months postbirth
Stunting* (Zscore < )2)
Severe stunting* (Zscore < )3)
Wasting* (Zscore < )2)
Severe wasting* (Zscore < )3)
Underweight* (Zscore < )2)
Severe underweight* (Zscore < )3)

Earlier KMC
group Incidence (%)

Later KMC group

Incidence (%)

N = 36
7 (19.4)
2 (5.6)
5 (13.9)
N = 29
12 (41.4)

N = 36
7 (19.4)
2 (5.6)
5 (13.9)
N = 26
4 (15.4)

12 (41.4)
3 (10.3)
0 (0.0)
0 (0.0)
6 (20.6)
1 (3.4)

13 (50.0)
4 (15.4)
4 (15.4)
0 (0.0)
11 (42.3)
4 (15.4)

Risk ratio
(95% CIs)

p-value adjusted
p-value**

1.00 (0.392.56)
1.00 (0.156.72)
1.00 (0.323.16)

1.00
1.00
1.00

2.69 (1.007.31)

0.04 0.04

0.83 (0.461.48)
0.67 (0.172.73)

0.49 (0.211.14)
0.22 (0.031.88)

0.59 0.78
0.70 0.79

0.14 0.15
0.18 ***

*Stunting: height-for-age, Wasting: weight-for-height, Underweight: weight-for-age

**p-value adjusted with birth weight, Apgar score and gender by logistic regression model.
***Multivariate analysis could not perform in the outcome, because this data was sparse for the analysis.

e244

2011 The Author(s)/Acta Pdiatrica 2011 Foundation Acta Pdiatrica 2011 100, pp. e241e247

Nagai et al.

Long-term effect of earlier initiated continuous KMC for LBW

14

12
p = 0.49*

12

p = 0.004*

10

10

2
0
(n) 4
3
2
(Z-score)

0
(n) 4
3
2
(Z-score)

A WHO child growth standard

B WHO child growth standard

(Stunting: height-for-age)

(Wasting: weight-for-height)

12

25
p = 0.04*

10

p = 0.008*

20

15

6
10

2
(n) 4
3
(Z-score)

0
2

60

70

75

80

85

100

(%)

D Growth standard in Madagascar

C WHO child growth standard

(Underweight: weight-for-age)

//

(n)

(Wasting: weight-for-height)

Right-side bar : Earlier continuous KMC group

Left-side bar : Later continuous KMC group
Solid line : cut-off point of malnutrition
Dot line : cut-off point of severe malnutrition
* Wilcoxon rank-sum test

Figure 2 Distribution of nutritional indicators. (A) WHO Child Growth standard (Stunting: height-for-age). (B) WHO Child Growth standard (Wasting: weight-for-height).
(C) WHO Child Growth standard (Underweight: weight-for-age). (D) Growth standard in Madagascar (Wasting: weight-for-height).

The frequency of wasting, stunting and being underweight

at 612 months postbirth was lower for the earlier initiated
continuous KMC group but did not differ significantly
between the two groups (Table 2). With respect to the
WHO child growth standard, the distribution of Z scores
for height-for-age was similar in the two groups (Wilcoxon
rank-sum test; p = 0.49). However, the distribution of Z
scores for weight-for-height and weight-for-age was significantly higher in the earlier initiated continuous KMC group
than the later continuous KMC group (Wilcoxon rank-sum
test; wasting, p = 0.004; underweight, p = 0.04) (Fig. 2).
With respect to the growth standard of Madagascar, the percent distribution for weight-for-height was significantly
higher in the earlier initiated continuous KMC group (Wilcoxon rank-sum test; p = 0.008) than the later continuous
KMC group (Fig. 2).

DISCUSSION
We performed this randomized controlled trial with followup to examine the long-term effects of earlier initiated continuous KMC on relatively stable LBW infants in a
resource-limited setting. Although our results indicate that

earlier initiated continuous KMC has little if any effect on

mortality or readmission at 612 months postbirth, we
could not get a conclusive answer given the smaller than
planned sample size and insufficient statistical power.
There are no published reports of earlier continuous
KMC for LBW infants in resource-limited countries.
Charpak et al. (14,15) showed that continuous KMC, but
not earlier KMC, was associated with reduced mortality
and morbidity (severe infectious episodes) at 12 months
postbirth, but this difference was not significant (mortality:
RR, 0.57; 95% CIs, 0.271.17; severe infectious episodes:
RR, 0.86; 95% CIs, 0.711.03). Sloan et al. (16) also
reported that continuous KMC, but not earlier initiated
KMC, had no effect on mortality but reduced morbidity
(severe illness) at 6 months postbirth, compared with
earlier initiated KMC (mortality: RR, 0.98; 95% CIs, 0.46
2.12; severe illness: RR, 0.30; 95% CIs, 0.140.67). These
two long-term studies did not involve earlier initiation of
intervention (i.e. within 24 h postbirth). The mean or median (range) age for commencing continuous KMC was 34
(160) days postbirth in the Colombian study (14,15), and
13 (070) days postbirth in the Ecuadorian study (16).
Because the targeted infants were much smaller and

2011 The Author(s)/Acta Pdiatrica 2011 Foundation Acta Pdiatrica 2011 100, pp. e241e247

e245

Long-term effect of earlier initiated continuous KMC for LBW

Nagai et al.

preterm compared with ours, it took more time to achieve

a relatively stable condition. In this regard, it might be said
that earlier or later continuous KMC are different types of
interventions.
In the current study, earlier initiated continuous KMC on
relatively stable LBW infants resulted in a significantly
higher proportion of EBF at 6 months postbirth. We find it
interesting that only a half-day difference in the timing of
KMC initiation can make such a substantial difference in
EBF. The Cochrane SSC review (7) showed significant and
positive effects of early (initiated any time between 1 and
24 h postbirth) SSC on the proportion of breastfeeding at
14 months postbirth (10 trials, 552 participants; odds ratio
(OR), 1.82; 95% CIs, 1.083.07) and exclusive breastfeeding
up to 46 months postbirth (1 trial, 92 participants; OR,
5.67; 95% CIs, 2.27-14.16). These results support our findings. However, most of the trials in this review targeted fullterm infants, except for one trial that targeted late pre-term
(3436 weeks gestation) infants (17); none were continuous
interventions. In contrast, no significant differences were
seen for exclusive breastfeeding at 612 months follow-up
in the Cochrane KMC review (5) (3 trials, 810 participants;
RR,1,29; 95%CIs, 0.95-1.76). While two of the trials (14,16)
were continuous interventions, they were not earlier initiated (735 participants; RR,1,10; 95%CIs, 0.661.86).
What accounts for the EBF effect associated with earlier
initiated continuous KMC for LBW infants? Our analysis
suggests that the effects of earlier initiated continuous KMC
for relatively stable LBW infants are very similar to those
seen with early SSC for healthy term infants in developed
countries. As per the Cochrane SSC review (7), earlier SSC
stimulates the early initiation of breastfeeding.
To the best of our knowledge, the present study is the first
to show the effects of earlier initiated continuous intervention targeting LBW infants, particularly in a resource-limited setting. Generally, while the breastfeeding proportion
in most developing countries exceeds 90%, the major difficulty is maintaining EBF for 6 months (18). Our results support the opinion of Ruiz et al. (19) that many preterm LBW
infants can grow properly on EBF after continuous KMC.
In the present study, earlier initiated continuous KMC
reduced the incidences of wasting and underweight, but the
reductions were not significant. However, the distribution
of Z scores for wasting and underweight was significantly
higher in the earlier initiated continuous KMC group than
the later continuous KMC group. Vesel et al. reported that
the duration of EBF was not associated with malnutrition in
the first 6 months of life (12). The association between earlier initiated continuous KMC and nutritional indicators
was still inconclusive.
There is growing evidence of the benefits of early initiation of breastfeeding, particularly within the first hour
after birth (20,21). According to these reports, early initiation of breastfeeding contributes to an overall reduction
in neonatal mortality in resource-limited countries. Reduction of neonatal mortality is a global health priority. In
fact, the Millennium Development Goal (MDG) 4 calls
for a two-thirds reduction in mortality of children under

e246

5 years of age (22,23). The potential effects of earlier initiated continuous KMC are expected to be greater in
resource-limited settings. However, earlier initiated continuous KMC is not always applicable to all infants. For
instance, in our study, 48.3% (113 234) of LBW infants
were excluded because they were in unstable condition or
the familys conditions (mother and other family members
were willing to practise KMC and mother and or other
family willing to practise KMC were healthy) were not
available within 24 h postbirth (Fig. 1) (8). Additional
studies will be needed to examine the potential role of
earlier initiated postbirth interventions.
One potential limitation of the study was the earlier and
continuous interventions. Prior to this study, continuous
KMC was conventionally initiated approximately 4872 h
postbirth at the University Hospital of Mahajanga, Madagascar. Indeed, in the trial, only an 11 h difference exists
between initiating earlier and later continuous KMC, and
there was overlap between the intervention and control
group with respect to time of KMC onset. Because KMC
was officially introduced to Madagascar in 2000 (24), we
could not include a no KMC group as a control. In addition, the sample size was not large enough to show the
effects on outcomes, and the proportion of events was lower
to compare with the estimation before the trial. Another
reason for the reduced number of participants in this longterm follow-up study was that 31 infants (31 73, 42.5%)
had relocated between birth and 612 months postbirth.
Some families moved 23 times during this period, and they
occasionally moved to another town to look for employment. The frequency of address changes was much higher
than we anticipated.
The overall effects of earlier initiated SSC and KMC
targeting preterm LBW infants are still inconclusive even
in developed countries (25,26). In our study, we followed
only outcomes such as mortality, readmission, EBF and
nutritional indicators, but earlier initiated SSC and KMC
might have effects on other outcomes (e.g. mother-preterm infant interaction, reduction of maltreatment, infants
cognitive and or motor-development). Accordingly, evaluation of other outcomes will likely be informative. The
effects of the timing and or duration of intervention were
also insufficient. Recently, a prospective longitudinal study
from a developed country showed that the time spent in
KMC (duration of KMC per day) in the hospital is associated with breastfeeding duration in very preterm infants
(27). Furthermore, much longer follow-up outcomes
would be informative, in particular, with respect to EBF
(28).

CONCLUSION
Earlier initiated continuous KMC for relatively stable
LBW infants in a resource-limited country results in a significantly higher proportion of EBF at 6 months postbirth.
Further adequately designed, larger-scale long-term evaluations of earlier continuous KMC for LBW infants are
needed.

2011 The Author(s)/Acta Pdiatrica 2011 Foundation Acta Pdiatrica 2011 100, pp. e241e247

Nagai et al.

ACKNOWLEDGEMENTS
We thank research coordinators: A. Ramarijaona and M.E.
Raza-Fanomezanjanahary, head nurse;H.N. Ralibenja and
all the staff members in the neonatal unit at the University
Hospital of Mahajanga for their assistance, the president of
the KMC association in Madagascar;Y. Ranaivoson for her
KMC technical advice, and, most importantly, all the participants of this study. In addition, we thank anonymous
reviewers and the editor of Acta Paediatrica to their useful
comments. This study was supported by grants from FASID:
Foundation for Advanced Studies on International Development and St. Lukes Life Science Institute, Japan.

CONFLICTS OF INTEREST AND FUNDING

We have no conflicts of interest and no specific funding to
declare.

References
1. Charpak N, Ruiz JG, Zupan J, Cattaneo A, Figueroa Z, Tessier
R, et al. Kangaroo Mother Care: 25 years after. Acta Paediatr
2005; 94: 51422.
2. World Health Organisation. Kangaroo mother care A practical
guide. World Health Organisation; 2003.
3. Darmstadt GL, Bhutta ZA, Cousens S, Adam T, Walker N, de
Bernis L. Evidence-based, cost-effective interventions: how
many newborn babies can we save? Lancet 2005; 365: 97788.
4. Conde-Agudelo A, Diaz-Rossello JL, Belizan JM. Kangaroo
mother care to reduce morbidity and mortality in low birthweight infants. Cochrane Database Syst Rev 2003; (2):
CD002771.
5. Conde-Agudelo A, Belizan JM, Diaz-Rossello J Kangaroo
mother care to reduce morbidity and mortality in low birthweight infants. Cochrane Database Syst Rev
2011;(3):CD002771.
6. Lawn JE, Cousens S, Zupan J. 4 million neonatal deaths: when?
where? why? Lancet 2005; 365: 891900.
7. Moore ER, Anderson GC, Bergman N Early skin-to-skin contact for mothers and their healthy newborn infants. Cochrane
Database Syst Rev 2007 18:CD003519.
8. Nagai S, Andrianarimanana D, Rabesandratana N, Yonemoto
N, Nakayama T, Mori R. Earlier versus later continuous Kangaroo Mother Care (KMC) for stable low-birth-weight infants: a
randomized controlled trial. Acta Paediatr 2010; 99: 82735.
9. United Nations Childrens Fund (UNICEF). Tracking progress
on child and maternak nutrition. New York, NY: United
Nations Childrens Fund (UNICEF), 2009.
10. WHO Multicentre Growth Reference Study Group. WHO
Child Growth Standards: length height-for-age, weight-for-age,
weight-for-length, weight-for-height and body mass index-forage. Geneva: World Health Organization; 2006.
11. Ministe`re De La Sante, Du Planning Familial Et De La Protection Social Madagascar. Depistage et prise en charge de la malnutrition aigue. Antananarivo: Ministe`re De La Sante, Du
Planning Familial Et De La Protection Social Madagascar,
2007.
12. Vesel L, Bahl R, Martines J, Penny M, Bhandari N, Kirkwood
BR; WHO Immunization-linked Vitamin A Supplementation
Study Group. Use of new World Health Organization child
growth standards to assess how infant malnutrition relates to
breastfeeding and mortality. Bull World Health Organ
2010;88:3948.

Long-term effect of earlier initiated continuous KMC for LBW

13. Hamill PV, Drizd TA, Johnson CL, Reed RB, Roche AF, Moore
WM. Physical growth: National Center for Health Statistics
percentiles. Am J Clin Nutr 1979; 32: 60729.
14. Charpak N, Ruiz-Pelaez JG, Figueroa de CZ, Charpak Y. Kangaroo mother versus traditional care for newborn infants 2000
grams: a randomized, controlled trial. Pediatrics 1997; 100:
6828.
15. Charpak N, Ruiz-Pelaez JG, Figueroa de CZ, Charpak Y. A randomized, controlled trial of kangaroo mother care: results of
follow-up at 1 year of corrected age. Pediatrics 2001; 108:
10729.
16. Sloan NL, Camacho LW, Rojas EP, Stern C. Kangaroo mother
method: randomized controlled trial of an alternative method
of care for stabilised low-birthweight infants. Lancet 1994; 344:
7825.
17. Syfrett EB, Anderson GC. Very early kangaroo care beginning
at birth for healthy preterm infants and mothers who choose to
breastfeed: effect on outcome. A workshop on the Kangaroomother method for low birthweight infants. Trieste, Italy:
World Health Organisation, 1996 October.
18. Saloojee H. Early skin-to-skin contact for mothers and their healty newborn infants: RHL commentary (last reviced: 4 January
2008). The WHO Reproductive Health Library; Geneva: World
Health Organization, 2008. Available at: http://apps.who.int/
rhl/newborn/hscom2/en/index.html.
19. Ruiz JG, Charpak N, Figuero Z. Predictional need for supplementing breastfeeding in preterm infants under Kangaroo
Mother Care. Acta Paediatr 2002; 91: 11304.
20. Edmond KM, Zandoh C, Quigley MA, Amenga-Etego S,
Owusu-Agyei S, Kirkwood BR. Delayed breastfeeding initiation
increases risk of neonatal mortality. Pediatrics 2006; 117:
e3806.
21. Mullany LC, Katz J, Li YM, Khatry SK, LeClerq SC, Darmstadt
GL, et al. Breast-feeding patterns, time to initiation, and mortality risk among newborns in southern Nepal. J Nutr 2008;
138: 599603.
22. Rajaratnam JK, Marcus JR, Flaxman AD, Wang H, Levin-Rector A, Dwyer L, et al. Neonatal, postneonatal, childhood, and
under-5 mortality for 187 countries, 1970-2010: a systematic
analysis of progress towards Millennium Development Goal 4.
Lancet 2010; 375: 19882008.
23. World Health Organisation. The World Health Report 2005
make every mother and child count. Geneva: World Health
Organisation; 2005.
24. Ranaivoson Y, Rasoarimanana L, Rakotonirina S. Technique
Kangourou et Unite Trotro An-Koditra. Antananarivo:
Societe Malgache de Pediatrie; 2003.
25. Chiu SH, Anderson GC. Effect of early skin-to-skin contact
on mother-preterm infant interaction through 18 months:
randomized controlled trial. Int J Nurs Stud 2009; 46:
116880.
26. Nyqvist KH, Anderson GC, Bergman N, Cattaneo A, Charpak
N, Davanzo R, et al. Towards universal Kangaroo Mother Care:
recommendations and report from the First European conference and Seventh International Workshop on Kangaroo
Mother Care. Acta Paediatr 2010; 99: 8206.
27. Flacking R, Ewald U, Wallin L. Positive effect of Kangaroo
mother care on long-term breastfeeding in very preterm infants.
J Obstet Gynecol Neonatal Nurs 2011; 40: 1907.
28. Kramer MS, Matush L, Bogdanovich N, Aboud F, Mazer B, Fombonne E, et al. Health and development outcomes in 6.5-yold children breastfed exclusively for 3 or 6 mo. Am J Clin Nutr
2009; 90: 10704.

2011 The Author(s)/Acta Pdiatrica 2011 Foundation Acta Pdiatrica 2011 100, pp. e241e247

e247