We will conduct an individually randomised, controlled, superiority trial with two parallel groups; an intervention arm allocated to receive KMC and a control arm receiving 'standard' care. The primary aim is to examine the impact of KMC initiated before stabilisation on mortality within 7 days relative to standard care amongst neonates ≤2000g at four hospitals in Uganda. We hypothesise that neonates in the arm allocated to receive KMC before stabilisation will have a 25% overall reduction in mortality within 7 days compared to neonates allocated to receive standard care.

METHODS Study setting, design, and participants The OMWaNA trial was an individually randomised, controlled, superiority trial with two parallel arms allocated in a 1:1 ratio to receive KMC initiated before stabilisation (intervention) or standard care (control). The trial was conducted in the neonatal units (WHO level-2) of five government hospitals in Uganda (appendix p 5). Infrastructure and equipment improvements as well as training were required prior to trial initiation, as described previously.10 All hospitals were provided with essential equipment and supplies to support the provision of KMC and level-2 newborn care. Details regarding study context and methods have been published in the trial protocol9 and are briefly summarised here. Ugandan practice is that maternity and newborn care should be provided free of charge in government health facilities;11 however, families are expected to support caregivers to meet essential needs such as meals.

The trial is reported here in accordance with the CONSORT guidelines. All live-born neonates, aged 1 to <48 hours and weighing 700-2000g, who were admitted to participating hospitals and for whom the indication for KMC was "uncertain" according to WHO guidance concerning clinical stability, defined as receiving ≥1 therapy (oxygen, continuous positive airway pressure [CPAP] where available, intravenous [IV] fluids, therapeutic antibiotics, anti-seizure medication), were eligible for inclusion. Exclusion criteria included triplet or higher multi-fetal pregnancy (unless pregnancy resulted in demise of ≥1 fetus) and parent/caregiver being unable/unwilling to provide either consent, KMC, and/or attend follow-up visits. Neonates with life-threatening instability, severe jaundice requiring immediate management, active seizures, or major congenital malformations were also excluded. The study was approved by the Research Ethics Committees of Uganda Virus Research Institute (GC/127/19/06/717), Uganda National Council of Science and Technology (HS 2645), and London School of Hygiene & Tropical Medicine (LSHTM, 16972). The trial was overseen by a steering committee and an independent data and safety monitoring board (DSMB).

Screening and informed consent All admitted neonates weighing ≤2000g were screened for eligibility by study staff. Stable neonates (meeting WHO 2019 criteria for KMC eligibility) were excluded and remaining neonates were assessed for eligibility. Those who met criteria for 'life-threatening instability' or who had conditions precluding KMC (e.g., seizures, jaundice), were reassessed every 3 hours up to 48 hours, after which they were excluded. Written informed parental consent was obtained for all participants.

Randomisation, allocation concealment and masking A random allocation sequence was computer-generated using permuted blocks of varying sizes stratified by birthweight (<1000g, 1000-1499g, ≥1500g) and recruitment site. Allocation concealment was done by programming the allocation sequence into the screening database and revealing treatment group only when screening of eligible neonates was complete. Neonates from multiple births (twins or triplets) were allocated to the same arm according to first-born allocation.14 Masking of parents, caregivers, or healthcare workers was not possible due to the nature of the KMC intervention.

Procedures In the intervention arm, KMC was initiated soon after randomisation. Neonates were naked except for hat and diaper, placed prone and skin-to-skin on caregiver's chest, and secured using a KMC wrap. Adjustable beds were provided to facilitate continuous skin-to-skin care. KMC duration was charted by caregivers and verified by study staff. When not in KMC (e.g., during maternal bathing), incubator or radiant heater care was commenced. Study personnel, in addition to hospital staff, provided continuous KMC counselling throughout the hospitalisation. Control arm neonates were cared for in an incubator or radiant heater, as per hospital practice. Caregivers could have physical contact with their newborn but skin-to-skin contact was not initiated until stability criteria were met.9 Once stable, newborns were transferred to routine (intermittent) KMC. Neonates in both arms received standard clinical care according to hospital guidelines.

Inclusion criteria

• Liveborn at Jinja Hospital, Masaka Hospital, Entebbe Hospital, or Iganga Hospital
• Singleton or twin pregnancy
• Birthweight ≥700g and ≤2000g
• Chronological age 1-48 hours at time of screening
• Alive at time of recruitment
• Parent/caregiver able and willing to provide KMC
• Parent/caregiver willing to attend follow-up visit
• Indication for KMC "uncertain" according to WHO guideline concerning clinical stability: pragmatically defined as receiving ≥1 therapy: oxygen, CPAP, IV fluids, therapeutic antibiotics, phenobarbital

Exclusion criteria

• Outborn

• Result of triplet or higher order multifetal pregnancy

• Indication for KMC "certain" according to WHO guidelines: pragmatically defined as clinically well neonates receiving none of the above therapy-based criteria

• Severely life-threatening instability defined as SpO2 <88% in oxygen AND ≥1 of:

  • Respiratory rate <20 or >100 breaths/min
  • Apnoea requiring bag-mask ventilation
  • HR <100 or >200 bpm

• Severe jaundice requiring immediate management

• Active neonatal seizures

• Major congenital malformation

• Parent does not provide written informed consent to participate in trial

• Mother or neonate enrolled in another MRC/UVRI research project

LSHTM Data Compass

When main results paper is published - currently under review

email to LSHTM data compass and agreement by study data sharing committee

28 days of age

The site paediatrician was informed about SAEs within 24 h. SAEs were followed up by the site paediatrician until resolution, or until causality was determined as unrelated to the trial intervention. If an unexpected SAE occurred, potentially related to the trial intervention, a report was submitted to the Research Ethics Committees (REC) at UVRI and LSHTM within 48h, with a follow-up report provided. All SAEs were reported to Sponsor and DSMB as part of their respective reports.

Masking of parents, caregivers, or healthcare workers was not possible due to the nature of the KMC intervention. The independent statistician who conducted the analyses was masked to treatment allocation.