# WHO Immediate KMC Study (iKMC) Study · id `maternal:ikmc-study` · Maternal & Newborn Atlas · release `15708f235d73cd42` (public) > General information from published sources, not medical advice, and not reviewed by clinicians. Check the cited source before relying on anything here. The atlas's own text and curation may be reused under Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0, https://creativecommons.org/licenses/by-nc/4.0/): credit the atlas, and do not use it commercially. See the atlas About page. Text quoted from a source stays under that source's own terms, given as the source's reuse\_basis. The atlas's reuse terms do not cover it. The downloads list each quotation with its source in third\_party\_quotations. Also called: iKMC; I-KMC Browse, life stage: Newborn ## In plain language A trial in five hospitals in Africa and India that tested starting skin-to-skin care right after birth for small babies, compared with starting it after they stabilized. ## Technical summary Multicentre, open-label randomized trial (ACTRN12618001880235; CTRI/2018/08/015369). 3211 infants and their 2944 mothers were randomized, against a registry target of 4200 infants. Birth weight 1.0 to under 1.8 kg. Primary outcomes: death to 28 days and to 72 hours. Enrolment stopped early on data and safety monitoring board advice. Primary sponsor: WHO. Funder: the Bill & Melinda Gates Foundation, per the abstract and both registries; PubMed's grant field separately lists the World Health Organization with grant ID "001". ## Key facts - Design: Multicentre, non-blinded (open-label), parallel-group randomized trial in five tertiary hospitals. Randomization was stratified by site and birth weight (1.0 to under 1.5 kg and 1.5 to under 1.8 kg) in variable blocks; twins were allocated to the same group. An independent team assessed outcomes. Mother-infant pairs were randomized: 3211 infants with 2944 mothers. Screening ran from 30 November 2017 to 20 January 2020. - Phase: Not applicable - Enrolled: 3211 infants (randomized count; Immediate "Kangaroo Mother Care" and Survival of Infants with Low Birth Weight, snapshot `maternal:pubmed-34038632-2026-10-09`, retrieved 2026-10-09); 2944 mothers (randomized count; Immediate "Kangaroo Mother Care" and Survival of Infants with Low Birth Weight, snapshot `maternal:pmc-pmc8108485-2026-10-09`, retrieved 2026-10-09) - Started: 2017-12-01 (Effect of immediate Kangaroo Mother Care on neonatal mortality for mothers and babies, snapshot `maternal:anzctr-actrn12618001880235-2026-10-09`, retrieved 2026-10-09) - Primary completion: Not reported - Completion: Not reported - Countries: Ghana; India; Malawi; Nigeria; Tanzania (Immediate "Kangaroo Mother Care" and Survival of Infants with Low Birth Weight, snapshot `maternal:pmc-pmc8108485-2026-10-09`, retrieved 2026-10-09) - Care setting: Five tertiary-level hospitals. The intervention arm used a Mother-NICU, a newborn intensive care unit with beds for mothers. (Immediate "Kangaroo Mother Care" and Survival of Infants with Low Birth Weight, snapshot `maternal:pmc-pmc8108485-2026-10-09`, retrieved 2026-10-09) - Registry status: "Stopped early" (Australian New Zealand Clinical Trials Registry, snapshot `maternal:anzctr-actrn12618001880235-2026-10-09`, retrieved 2026-10-09); "Completed" (Clinical Trials Registry - India, read through the WHO ICTRP mirror, snapshot `maternal:ictrp-ctri-2018-08-015369-2026-10-09`, retrieved 2026-10-09). The registries disagree; each status is given as its record states it. - Stopped early: Enrollment stopped early, date not reported. After the second interim analysis, the data and safety monitoring board recommended stopping because of reduced mortality with immediate KMC. The CTRI record, seen through the WHO ICTRP mirror, gives the status as completed. (Immediate "Kangaroo Mother Care" and Survival of Infants with Low Birth Weight, snapshot `maternal:pmc-pmc8108485-2026-10-09`, retrieved 2026-10-09) ## Claims, with the population each applies to ## Context: Liveborn infants weighing 1.0 to under 1.8 kg at birth in five tertiary hospitals in sub-Saharan Africa and India (iKMC trial) Recipient: Newborn; Population: Liveborn infants with a birth weight of 1.0 to under 1.8 kg (abstract: "between 1.0 and 1.799 kg"), of any gestational age, singleton or twin, by any mode of delivery, enrolled with their mothers within 2 hours of birth. About 20% of infants in this weight band born in the study hospitals were not enrolled because mother or newborn was too sick.; Care setting: Tertiary hospitals. Intervention infants: a Mother-NICU (a neonatal intensive care unit with beds for mothers), then a KMC ward. Control infants: the NICU without the mother, then a KMC ward after stabilization.; Geography: Ghana (Kumasi), India (New Delhi), Malawi (Blantyre), Nigeria (Ile-Ife) and Tanzania (Dar es Salaam); cities per ANZCTR. The ICTRP mirror of the CTRI record does not name the fifth country; it lists it only as other.; Life stage: Neonatal period; Gestational age: Any, not restricted; Postnatal age: 0 days to 0 days; Follow-up: From enrolment to 28 days of age, with a home visit on day 29. The second primary outcome window was enrolment to 72 hours. ### The iKMC trial stopped enrolling babies early because its data and safety monitoring board found fewer deaths with immediate skin-to-skin care. - Category: Description - Evidence stage: Evidence stage not applicable - Certainty: Certainty not assessed - Applies to: Recipient: Newborn; Population: Liveborn infants with a birth weight of 1.0 to under 1.8 kg (abstract: "between 1.0 and 1.799 kg"), of any gestational age, singleton or twin, by any mode of delivery, enrolled with their mothers within 2 hours of birth. About 20% of infants in this weight band born in the study hospitals were not enrolled because mother or newborn was too sick.; Care setting: Tertiary hospitals. Intervention infants: a Mother-NICU (a neonatal intensive care unit with beds for mothers), then a KMC ward. Control infants: the NICU without the mother, then a KMC ward after stabilization.; Geography: Ghana (Kumasi), India (New Delhi), Malawi (Blantyre), Nigeria (Ile-Ife) and Tanzania (Dar es Salaam); cities per ANZCTR. The ICTRP mirror of the CTRI record does not name the fifth country; it lists it only as other.; Life stage: Neonatal period; Gestational age: Any, not restricted; Postnatal age: 0 days to 0 days; Follow-up: From enrolment to 28 days of age, with a home visit on day 29. The second primary outcome window was enrolment to 72 hours. - Limitations: The registries disagree on the trial's status: ANZCTR records it as stopped early, while the WHO ICTRP mirror of the CTRI record gives its status as completed. The primary CTRI record could not be fetched. This is recorded, not reconciled. - Last checked: Not recorded - Source problem: a cited source is unknown, not verified. Check the sources before relying on this claim. - Sources: 1. Supports: Immediate "Kangaroo Mother Care" and Survival of Infants with Low Birth Weight (Publication, edition Version of record not fetched (HTTP 403). The abstract comes from PubMed. Full-text facts come from the PMC author accepted manuscript PMC8108485.) Locator: Abstract, RESULTS, last sentence: "The trial was stopped early on the recommendation of the data and safety monitoring board owing to the finding of reduced mortality". Retrieved 2026-10-09T15:56:27Z. Source state: Current. Snapshot hash: 5dde5eaa4d684f398f7f3cfab05816eb73207b8adbaac5c4163e7e480567ecf6. How it applies: Abstract of the primary report, for the same infants and time window as this context. 2. Supports: Effect of immediate Kangaroo Mother Care on neonatal mortality for mothers and babies (Registry, edition Record as last updated 2021-04-07. Submitted 2018-09-10 and registered 2018-11-19, marked as retrospectively registered. Universal trial number U1111-1219-1230.) Locator: Recruitment, Reason for early stopping. Retrieved 2026-10-09T15:57:53Z. Source state: Current. Snapshot hash: 3db33e1ff7a7695daedf52ac75f4a3dcdd1a7a29cd4bf650354891f0b8cceb11. How it applies: ANZCTR record as last updated 2021-04-07. 3. Supports: Immediate "Kangaroo Mother Care" and Survival of Infants with Low Birth Weight (Publication, edition Version of record not fetched (HTTP 403). The abstract comes from PubMed. Full-text facts come from the PMC author accepted manuscript PMC8108485.) Locator: Methods, Statistical analysis, para 1: "After the second interim analysis, the DSMB recommended stopping enrolment in the trial because of a clear benefit in neonatal survival.". Retrieved 2026-10-09T15:57:07Z. Source state: Current. Snapshot hash: 0945a417a4ffacfded839e4e8df7cf03c99c581e34fbca8ddd6daa7229b79bac. How it applies: Author accepted manuscript (CC BY 4.0); not checked against the version of record. 4. Opposes: Immediate skin to skin contact of the baby with mother or the surrogate (Registry, edition WHO ICTRP mirror, last refreshed on 24 November 2021. Registered 2018-08-17 (ICTRP marks it as not prospectively registered). The primary CTRI record could not be fetched: the http request timed out and the https request redirected to an error page.) Locator: Main record fields, recruitment status. Retrieved 2026-10-09T15:58:10Z. Source state: Unknown, not verified. Snapshot hash: ca9ae31bf890b23b30bf87167bc3cb5876db63618b15f9ad0c3658270c549e4c. How it applies: WHO ICTRP mirror of CTRI, refreshed 2021-11-24; the primary register could not be fetched. ## Study findings Findings are grouped by the claim they support. Each group keeps its own population and follow-up, and each finding keeps its own denominator and time window. Not reported is not zero. ### In the iKMC trial, babies weighing 1.0 to under 1.8 kg at birth who started continuous skin-to-skin care with their mother within hours of birth were less likely to die in their first 28 days than babies who started it only after they were stable. 191 of 1596 babies died in the immediate group and 249 of 1587 in the control group (adjusted risk ratio 0.75, 95% CI 0.64 to 0.89). Claim `maternal:claim-ikmc-28d-mortality`. Applies to: Recipient: Newborn; Population: Liveborn infants with a birth weight of 1.0 to under 1.8 kg (abstract: "between 1.0 and 1.799 kg"), of any gestational age, singleton or twin, by any mode of delivery, enrolled with their mothers within 2 hours of birth. About 20% of infants in this weight band born in the study hospitals were not enrolled because mother or newborn was too sick.; Care setting: Tertiary hospitals. Intervention infants: a Mother-NICU (a neonatal intensive care unit with beds for mothers), then a KMC ward. Control infants: the NICU without the mother, then a KMC ward after stabilization.; Geography: Ghana (Kumasi), India (New Delhi), Malawi (Blantyre), Nigeria (Ile-Ife) and Tanzania (Dar es Salaam); cities per ANZCTR. The ICTRP mirror of the CTRI record does not name the fifth country; it lists it only as other.; Life stage: Neonatal period; Gestational age: Any, not restricted; Postnatal age: 0 days to 0 days; Follow-up: From enrolment to 28 days of age, with a home visit on day 29. The second primary outcome window was enrolment to 72 hours. Limitations: One trial in five tertiary hospitals in low- and middle-income countries, with an open-label design. It stopped early for benefit before reaching its target of 4200 infants. About 20% of infants in the eligible weight band were not enrolled because mother or baby was too sick. The authors say the effect of starting early cannot be separated from the mother's continuous presence (author manuscript, Discussion para 5), and that applicability to low-mortality, high-resource settings is untested (author manuscript, Discussion para 3). The per-arm denominators (1596 and 1587) are smaller than the numbers randomized (1609 and 1602). They come from the author manuscript's Table 3 only, its text does not state their basis, and they were not checked against the version of record. - Infant death, enrolment to 28 days (Primary): Infants who died (count) 191 infants; reported as "191/1596 (12.0%)"; arm Immediate KMC (intervention), no comparator arm; interval Not applicable; denominator 1596 (As reported in Table 3 of the author accepted manuscript only, not checked against the version of record. The fetched text does not state the basis. 1609 infants were assigned to this arm.); time window Enrolment to 28 days of age - Infant death, enrolment to 28 days (Primary): Infants who died (count) 249 infants; reported as "249/1587 (15.7%)"; arm Conventional care, then KMC after stabilization (control), no comparator arm; interval Not applicable; denominator 1587 (As reported in Table 3 of the author accepted manuscript only, not checked against the version of record. The fetched text does not state the basis. 1602 infants were assigned to this arm.); time window Enrolment to 28 days of age - Infant death, enrolment to 28 days (Primary): Adjusted risk ratio, immediate vs control (log-binomial regression, adjusted for clustering from multiple births and for site, delivery mode, multiple pregnancy, age at randomization, infant sex and weight, maternal schooling and age, household toilet and family income, per the Table 3 footnote) 0.75 ratio; reported as "aRR 0.75 (0.64 to 0.89), P=0.001"; arm Immediate KMC (intervention) compared with Conventional care, then KMC after stabilization (control); interval 0.64 to 0.89 (95% confidence interval); denominator Not applicable (basis: Comparison of the two per-arm rows above); time window Enrolment to 28 days of age ### In the first 72 hours, 74 of 1606 babies died with immediate skin-to-skin care and 92 of 1599 with the control approach (adjusted risk ratio 0.77, 95% CI 0.58 to 1.04). The interval includes no difference, so the trial did not show a clear effect in this window. Claim `maternal:claim-ikmc-72h-mortality`. Applies to: Recipient: Newborn; Population: Liveborn infants with a birth weight of 1.0 to under 1.8 kg (abstract: "between 1.0 and 1.799 kg"), of any gestational age, singleton or twin, by any mode of delivery, enrolled with their mothers within 2 hours of birth. About 20% of infants in this weight band born in the study hospitals were not enrolled because mother or newborn was too sick.; Care setting: Tertiary hospitals. Intervention infants: a Mother-NICU (a neonatal intensive care unit with beds for mothers), then a KMC ward. Control infants: the NICU without the mother, then a KMC ward after stabilization.; Geography: Ghana (Kumasi), India (New Delhi), Malawi (Blantyre), Nigeria (Ile-Ife) and Tanzania (Dar es Salaam); cities per ANZCTR. The ICTRP mirror of the CTRI record does not name the fifth country; it lists it only as other.; Life stage: Neonatal period; Gestational age: Any, not restricted; Postnatal age: 0 days to 0 days; Follow-up: From enrolment to 28 days of age, with a home visit on day 29. The second primary outcome window was enrolment to 72 hours. Limitations: One trial in five tertiary hospitals in low- and middle-income countries, with an open-label design. It stopped early for benefit before reaching its target of 4200 infants. About 20% of infants in the eligible weight band were not enrolled because mother or baby was too sick. The authors say the effect of starting early cannot be separated from the mother's continuous presence (author manuscript, Discussion para 5), and that applicability to low-mortality, high-resource settings is untested (author manuscript, Discussion para 3). Fewer deaths in this window make the estimate less precise. The per-arm denominators (1606 and 1599) come from the author manuscript's Table 3 only, its text does not state their basis, and they were not checked against the version of record. - Infant death, enrolment to 72 hours (Primary): Infants who died (count) 74 infants; reported as "74/1606 (4.6%)"; arm Immediate KMC (intervention), no comparator arm; interval Not applicable; denominator 1606 (As reported in Table 3 of the author accepted manuscript only, not checked against the version of record. The fetched text does not state the basis. 1609 infants were assigned to this arm.); time window Enrolment to 72 hours of age - Infant death, enrolment to 72 hours (Primary): Infants who died (count) 92 infants; reported as "92/1599 (5.8%)"; arm Conventional care, then KMC after stabilization (control), no comparator arm; interval Not applicable; denominator 1599 (As reported in Table 3 of the author accepted manuscript only, not checked against the version of record. The fetched text does not state the basis. 1602 infants were assigned to this arm.); time window Enrolment to 72 hours of age - Infant death, enrolment to 72 hours (Primary): Adjusted risk ratio, immediate vs control (log-binomial regression, adjusted for clustering from multiple births and for site, delivery mode, multiple pregnancy, age at randomization, infant sex and weight, maternal schooling and age, household toilet and family income, per the Table 3 footnote) 0.77 ratio; reported as "aRR 0.77 (0.58 to 1.04), P=0.09"; arm Immediate KMC (intervention) compared with Conventional care, then KMC after stabilization (control); interval 0.58 to 1.04 (95% confidence interval); denominator Not applicable (basis: Comparison of the two per-arm rows above); time window Enrolment to 72 hours of age ### Fewer babies in the immediate skin-to-skin group had a recorded low temperature (under 36 °C at the armpit) during their hospital stay: 90 of 1609 versus 133 of 1602 (adjusted risk ratio 0.65, 95% CI 0.51 to 0.83). Claim `maternal:claim-ikmc-hypothermia`. Applies to: Recipient: Newborn; Population: Liveborn infants with a birth weight of 1.0 to under 1.8 kg (abstract: "between 1.0 and 1.799 kg"), of any gestational age, singleton or twin, by any mode of delivery, enrolled with their mothers within 2 hours of birth. About 20% of infants in this weight band born in the study hospitals were not enrolled because mother or newborn was too sick.; Care setting: Tertiary hospitals. Intervention infants: a Mother-NICU (a neonatal intensive care unit with beds for mothers), then a KMC ward. Control infants: the NICU without the mother, then a KMC ward after stabilization.; Geography: Ghana (Kumasi), India (New Delhi), Malawi (Blantyre), Nigeria (Ile-Ife) and Tanzania (Dar es Salaam); cities per ANZCTR. The ICTRP mirror of the CTRI record does not name the fifth country; it lists it only as other.; Life stage: Neonatal period; Gestational age: Any, not restricted; Postnatal age: 0 days to 0 days; Follow-up: From enrolment to 28 days of age, with a home visit on day 29. The second primary outcome window was enrolment to 72 hours. Limitations: Secondary outcome. Its confidence interval is not adjusted for multiplicity, and the authors say such intervals should not be used to infer definitive effects. Temperatures were measured in an open-label trial. The trial's hypothermia definition differs from other studies. One trial in five tertiary hospitals in low- and middle-income countries, with an open-label design. It stopped early for benefit before reaching its target of 4200 infants. About 20% of infants in the eligible weight band were not enrolled because mother or baby was too sick. The authors say the effect of starting early cannot be separated from the mother's continuous presence (author manuscript, Discussion para 5), and that applicability to low-mortality, high-resource settings is untested (author manuscript, Discussion para 3). - Hypothermia (iKMC definition) (Secondary): Infants with at least one hypothermia episode (count) 90 infants; reported as "90/1609 (5.6%)"; arm Immediate KMC (intervention), no comparator arm; interval Not applicable; denominator 1609 (As reported in Table 3 of the author accepted manuscript. It equals the number assigned to this arm.); time window From 2 hours after randomization to hospital discharge - Hypothermia (iKMC definition) (Secondary): Infants with at least one hypothermia episode (count) 133 infants; reported as "133/1602 (8.3%)"; arm Conventional care, then KMC after stabilization (control), no comparator arm; interval Not applicable; denominator 1602 (As reported in Table 3 of the author accepted manuscript. It equals the number assigned to this arm.); time window From 2 hours after randomization to hospital discharge - Hypothermia (iKMC definition) (Secondary): Adjusted risk ratio, immediate vs control (log-binomial regression, adjusted for clustering from multiple births and for site, delivery mode, multiple pregnancy, age at randomization, infant sex and weight, maternal schooling and age, household toilet and family income, per the Table 3 footnote) 0.65 ratio; reported as "aRR 0.65 (0.51 to 0.83)"; arm Immediate KMC (intervention) compared with Conventional care, then KMC after stabilization (control); interval 0.51 to 0.83 (95% confidence interval, not adjusted for multiplicity (Table 3 footnote)); denominator Not applicable (basis: Comparison of the two per-arm rows above); time window From 2 hours after randomization to hospital discharge ## Connected pages - Tests immediate KMC: Immediate kangaroo mother care (`maternal:immediate-kmc`) - Enrolled infants weighing 1.0 to under 1.8 kg at birth: Low birth weight (`maternal:low-birth-weight`) - Primary sponsor (ANZCTR, ICTRP): World Health Organization (`maternal:who`) - Funder (abstract, ANZCTR, ICTRP): Bill & Melinda Gates Foundation (`maternal:gates-foundation`) - Primary results, NEJM 2021: iKMC primary results, NEJM 2021 (`maternal:pub-ikmc-nejm-2021`) - Related to: Immediate KMC in high-resource neonatal intensive care units (`maternal:idea-immediate-kmc-high-resource-nicu`) ## Provenance Content hash: eafb76f42be76b70dd8346ea4c1cccf73f688b14957c482edefc716a62ab5ea5 Last substantive edit: 2026-10-09 Evidence last checked: Not recorded