The team’s research explored what really happens after babies are admitted to neonatal units, especially  in low- and middle-income settings. Dr Narayanan worked with a multidisciplinary team to analyse one full year of admissions at Greater Accra Regional Hospital in southern Ghana and Tamale Teaching Hospital in northern Ghana. Together, these facilities serve millions of people and manage thousands of births each year, making them an important window into everyday neonatal care.

The study focused on “inborn” babies, meaning infants delivered within the hospitals themselves. In theory, neonatal units are designed for babies who are premature, or very small, or seriously ill, and who require specialised equipment or constant monitoring. In practice, however, many newborns are transferred there for much simpler reasons. As Dr Narayanan and her collaborators explain, midwives on postnatal wards often provide only essential newborn care, such as support with feeding and temperature control. If a baby needs something slightly beyond this – glucose monitoring, brief observation, or precautionary antibiotics – the default option is often admission to the neonatal unit.

By reviewing the medical records of nearly 3,000 newborns admitted in 2022, the research team was able to map not just who was admitted, but what care those babies actually received. The results were striking. Overall, 11.5% of babies admitted to the neonatal units across the two hospitals did not receive any interventions that truly required neonatal unit expertise. These babies did not need incubators, respiratory support, tube feeding, or infusions/transfusions. Some received only glucose checks, phototherapy for jaundice, or short courses of antibiotics. A small but notable group received no recorded interventions at all.

This pattern was especially pronounced at Tamale Teaching Hospital, where more than one in five admitted babies received no neonatal unit interventions. This was the case even though the hospital already has a Special Baby Ward – an intermediate care area where mothers can stay with their babies and where less intensive monitoring and treatment can be provided. According to Dr Narayanan, this finding highlights how system constraints – such as whether a mother is ready to be discharged from the postnatal ward – can push babies into higher-intensity settings than they truly need.

Why does this matter? One reason is pressure on neonatal units themselves. These units are often overcrowded and understaffed, with nurses caring for far more babies than is ideal. When beds and staff time are taken up by stable newborns, it may become harder to focus resources on the smallest and sickest babies, who face the greatest risk of death. The study found that mortality was, as expected, much higher among babies who required intensive interventions, particularly those weighing less than 1,800 grams at birth. Reducing unnecessary admissions could help staff devote more attention to these high-risk infants.

Another major concern is separation. Admission to a neonatal unit often means that mothers and babies are physically separated, sometimes for days. This can disrupt bonding and make breastfeeding far more difficult. Breastfeeding outcomes were a central focus of the study, reflecting Dr Narayanan’s longstanding interest in use of human milk for babies and maternal-infant dyadic care. The data revealed large differences between the two hospitals. At Tamale Teaching Hospital’s neonatal unit, more than 90% of babies were exclusively breastfed at discharge, compared with just over 66% at Greater Accra Regional Hospital.

One key reason appears to be feeding practices during separation. At Tamale, babies who were separated from their mothers were routinely given intravenous fluids rather than formula, specifically to protect breastfeeding. While this approach carries its own risks and should not be overused, it was associated with substantially higher breastfeeding rates, even among very small babies. The contrast was especially stark for infants weighing under 1,800 grams, where exclusive breastfeeding rates at discharge were nearly 30 percentage points higher in Tamale than in Accra.

The benefits of keeping mothers and babies together became even clearer when the researchers compared neonatal unit admissions with babies cared for in Tamale’s Special Baby Ward. In this ward, mothers stay in the room with their babies and provide most of the day-to-day care, supported by nurses and paediatricians. Almost every baby discharged from the Special Baby Ward was exclusively breastfed. Even stable babies in the neonatal unit who received no intensive interventions had lower breastfeeding rates than sick babies cared for alongside their mothers in the Special Baby Ward.

For Dr. Narayanan, these findings reinforce a simple but powerful idea: care environments shape outcomes. When systems are designed in ways that prioritise separation, even briefly, there can be lasting effects on feeding and family experience. Conversely, when care is organised to support the mother-baby pair, many babies thrive without ever needing high-tech intervention.

The study also raises important questions about how hospitals define “necessary” care. Some practices, such as routine glucose monitoring for all admissions, or automatic intravenous fluids to avoid formula use during separation, may inflate the appearance of medical need. Dr Narayanan and her colleagues are careful not to criticise individual clinicians, who are working within constrained systems and doing their best for newborns. Instead, they argue for clearer intermediate levels of care that sit between basic postnatal wards and neonatal units to deal with these issues without separation of mothers and babies.

Such intermediate care could take many forms. It might involve additional training for midwives in postnatal wards alongside greater involvement of paediatric staff, or designated areas close to the nurses’ station where stable at-risk babies can be monitored while staying with their mothers. The Special Baby Ward in the Pediatric unit at Tamale offers one example, but the researchers note that ideally this care should be physically located in or near postnatal wards, so mothers can participate in the care of their at-risk babies before being discharged themselves.

The insights revealed in this important study echo concerns raised in other low-resource settings around the world and also in some neonatal units in high-income countries. Neonatal units are vital, lifesaving spaces, but they are not always the right answer for every newborn who needs a little extra attention. Rethinking how newborn care is organised – with a stronger emphasis on intermediate, mother-baby dyad centred models – could improve both efficiency and equity. In the process, it could help ensure that neonatal units remain focused on what they do best: providing intensive care for the babies who truly need them most.