Starting full milk feeds from birth in moderately preterm infants is associated with modestly lower healthcare costs compared with gradual feeding supported by intravenous (IV) fluids, although it does not reduce length of hospital stay, according to a UK multicentre economic evaluation.

Early nutrition is widely recognised as a key determinant of outcomes in preterm infants, yet feeding practices vary considerably across neonatal units.

To address uncertainty around the economic impact of early feeding strategies, researchers conducted a cost-consequence analysis alongside the Fluids Exclusively Enteral from Day 1 (FEED1) randomised controlled trial.

FEED1 showed that starting exclusively enteral feeding from birth does not increase the risk of necrotising enterocolitis in clinically stable preterm infants, nor does it impact length of hospital stay.

The within-trial economic evaluation was conducted across 46 NHS neonatal units in the UK and compared early full milk feeding from birth with gradual feeding supported by IV fluids in infants born between 30+0 and 32+6 weeks’ gestation.

A cost-consequence approach was used to prevent double-counting length of neonatal hospital stay, which was both a major cost driver and the trial’s primary clinical outcome, alongside economic outcomes.

Infants were randomised within three hours of birth to receive either full milk feeds from day one or standard care involving IV fluids with gradual introduction of milk feeds. Resource use and care costs were measured from birth to six weeks’ corrected age.

A total of 2,088 preterm infants were enrolled in the FEED1 trial. All were clinically stable at enrolment, were less than three hours old and had no congenital abnormalities affecting gastrointestinal function.

Early full milk feeding may reduce neonatal care costs

No statistically significant difference in mean hospital stay was observed between groups. The adjusted mean difference was −0.05 days (95% CI −0.64 to 0.54), equivalent to around 1.2 hours.

Mean total care costs were £670 lower in the full milk feeding group than in the gradual feeding group, although the difference did not reach statistical significance (95% CI −£1,562 to £223; P=0.141). Sensitivity analyses produced consistent findings across alternative analytical approaches.

Resource use patterns differed between groups. Infants receiving full milk feeds spent significantly fewer days in neonatal intensive care (mean difference −1.6 days; 95% CI −2.4 to −0.8; P<0.001) but more time in neonatal special care (mean difference 1.8 days; 95% CI 0.5 to 3.0; P=0.005). This reflected a redistribution of care rather than a reduction in overall hospitalisation.

Subgroup analyses suggested lower care costs among infants born at 30 weeks’ gestation (mean difference −£4,026; 95% CI −£6,420 to −£1,632; P=0.001) and those below the 10th birth weight centile (mean difference −£4,086; 95% CI −£7,034 to −£1,138; P=0.007). However, formal interaction testing found no evidence that treatment effects differed by gestational age or birth weight.

Starting full milk feeds from birth did not decrease hospital stays for moderately preterm infants but might offer modest economic benefits in certain subgroups and shift care from intensive- to lower-dependency wards, the authors stated, although the analysis had several limitations.

Missing data were more common in the gradual feeding group at six weeks, although the authors said multiple imputation methods were used to address this. Furthermore, follow-up was limited to six weeks’ corrected age and did not capture longer-term outcomes, quality of life or wider societal costs.

The authors concluded that further research is needed to assess longer-term clinical and economic outcomes and determine whether early feeding strategies provide sustained benefits for health services and families.

Reference
Naghdi S et al. Cost-consequence analysis of early full milk feeding versus gradual feeding with intravenous support in preterm infants: results from the FEED1 trial. Arch Dis Child Fetal Neonatal Ed 2025;0:F1–F8.