Department of Reproductive Health and Research
World Health Organization
Geneva

This chapter reviews the evidence on KMC, from both developing and developed countries,
with regard to the following outcomes: mortality and morbidity; breastfeeding and growth;
thermal protection and metabolism, and other effects. The experience with KMC has been
reviewed by several authors,12, 13, 16, 21, 22 and in a systematic review.23 We also present evidence
on the acceptability of the intervention for mothers and health-care staff.
While reviewing the evidence, regardless of the outcome, it became clear that it was
important to highlight two essential variables: time of initiation of KMC, and daily and overall
duration of skin-to-skin contact.
Time of initiation of KMC in the studies under consideration varied from just after birth
to several days after birth. Late initiation means that the preterm/LBW infants have already
overcome the period of maximum risk for their health.
Length of daily and overall duration of skin-to-skin contact also varied from minutes (e.g. 30
minutes per day on average) to virtually 24 hours per day; from a few days to several weeks.
The longer the care, the stronger the possible direct and causal association between KMC and
the outcome. Furthermore, when KMC was carried out over a long period of time, care was
predominantly provided by the mother rather than the nursing staff or the conventional
incubator.
Some other variables that might have affected the outcome of KMC are:
 the position in which the baby was kept;
 the changes in the type and mode of feeding;
 the timing of discharge from the institution and the transition to home care;
 condition at discharge;
 the intensity of support and follow-up offered to mothers and families after discharge from the
institution.
Many other factors (e.g. social conditions, environment and health care, especially services
offered for KMC) may be associated with the positive effects observed in KMC studies. It is
very important to separate the effects of these factors from those deriving from KMC. Below,
in reviewing the evidence, we try to address those additional factors.
No published study on KMC was found in the context of high HIV prevalence among mothers.

Evidence

KANGAROO MOTHER CARE
Mortality and morbidity
Clinical trials
Three published randomized controlled trials (RCT) comparing KMC with conventional
care were conducted in low-income countries.24-26 The results showed no difference in survival
between the two groups. Almost all deaths in the three studies occurred before eligibility,
i.e. before LBW infants were stabilised and enrolled for research. Infants weighing less than
2000g were enrolled after an average period of 3 -14 days on conventional care, in urban thirdlevel
hospitals. The KMC infants stayed in hospital until they fulfilled the usual criteria for
discharge, as the control infants did, in two of the studies,24, 26 while in the third study they were
discharged earlier and subjected to a strict ambulatory follow-up.25 The follow-up periods lasted
one,26 six24 and twelve months,25 respectively.
The RCT carried out in Ecuador by Sloan and collaborators showed a lower rate of severe
illness among KMC infants (5%) than in the control group (18%).24 The sample size required
for that study was 350 subjects per group for a total of 700 infants, but only 603 babies were
recruited. Recruitment, in fact, was interrupted when the difference in the rate of severe illness
became apparent. The other controlled studies conducted in low-income countries revealed no
significant difference in severe morbidity, but found fewer hospital infections and readmissions
in the KMC group. Kambarami and collaborators from Zimbabwe also reported reduced
hospital infections.27 High-income countries report no difference in morbidity. However, it is
notable that no additional risk of infection seems to be associated with skin-to-skin contact.24-27
Observational studies showed that KMC could help reduce mortality and morbidity in
preterm/LBW infants. Rey and Martinez,9 in their early account, reported an increase in
hospital survival from 30% to 70% in infants between 1000g and 1500g. However, the
interpretation of their results is difficult because numerators, denominators and follow-up in
the KMC group were different from those in the historical control group.28 Bergman and
Jürisoo, in another study with an historical control group conducted in a remote mission
hospital without incubator care in Zimbabwe,14 reported an increase in hospital survival from
10% to 50% in infants weighing less than 1500g, and from 70% to 90% in those weighing
between 1500 to 1999g. Similar results are reported from a secondary hospital in nearby
Mozambique.15 The difference in survival, however, may be due to some uncontrolled variables.
The studies in Zimbabwe and Mozambique, conducted in hospitals with very limited resources,
applied KMC very early on, well before LBW and preterm infants were stabilized. In the early
study by Rey and Martinez, KMC was applied later, after stabilization. In both cases the skinto-
skin contact was maintained virtually 24 hours a day.
Charpak and collaborators, in a two-cohort study carried out in Bogotá, Colombia,29 found
a crude death rate higher in the KMC group (relative risk = 1.9; 95%CI: 0,6 to 5.8), but their
results reverted in favour of KMC (relative risk = 0.5, 95%CI: 0.2 to 1.2) after adjustment for
birth weight and gestational age. The differences, however, were not statistically significant.
The two cohorts recruited in two third-level hospitals, showed many social and economic
differences. KMC was also applied after stabilization and 24 hours a day. In a controlled but
not randomized trial carried out in a tertiary-care hospital in Zimbabwe, there was a slight
difference in survival in favour of the KMC infants, but this might have been due to differences
in feeding.

Conclusion
On balance the evidence shows that although KMC does not necessarily improve survival, it
does not reduce it. After stabilization, there is no difference in survival between KMC and good
conventional care. The hypothesis that KMC might improve survival when applied before
stabilization needs to be further explored with well-designed studies. If such an effect on
survival exists, it will be more evident and easier to demonstrate in the poorest settings, where
mortality is very high.
As for morbidity, while there is no strong evidence of a beneficial effect of KMC, there is
no evidence of it being harmful. In addition to the little evidence already published,14, 15 some
preliminary results on a small number of newborn infants with mild respiratory distress seem to
confirm that very early skin-to-skin contact might have a beneficial effect.30 A word of warning
about discharge: KMC infants discharged during the cold season may be more susceptible to
severe illness, especially lower respiratory tract infections, than those discharged during the
warm season.31 A closer follow-up is needed in such cases.
It should be noted that all the studies so far have taken place in well-equipped hospitals,
yet arguably the most significant impact of KMC will be felt in settings with limited resources.
There is an urgent need for further research in these settings. In the meantime, it seems that
where poor conventional care is available, KMC offers a safe substitute, with little risk of raised
morbidity or mortality.