Methods	A randomized, placebo‐controlled trial conducted in India
Participants	Inclusion criteria:
	(1)Neonates born at 27–33 weeks gestation in our hospital (2)aged <96 h of life (3)who were likely to either remain admitted in hospital or reside within 30 km of the hospital for the next 28 days (4)who were tolerating at least 15 ml/kg/day of milk feeds
	Exclusion criteria:
	(1)a gastro‐intestinal malformation (2)prior NEC or sepsis (3)any life‐threatening malformation that limited estimated life expectancy to less than a month
Interventions	Intervention group: Probiotic Lactobacillus acidophilus (662.5 million), Lactobacillus rhamnosus (362.5 million), Bifidobacterium longum (87.5 million), and Saccharomyces boulardii (137.5 million)
	Four groups
	A.High‐dose long course 1010 cells 12 hourly for 21 days B.High‐dose short course 1010 cells 12 hourly for days 1–14; followed by placebo from days 15–21 C.Low‐dose long course 109 cells 12 hourly for 21 days D. Control group: Placebo for 21 days
Outcomes	NEC, mortality and sepsis
Notes	We combined all the probiotics groups (group A + B + C) to avoid the double counting of the placebo.