Objectives To investigate whether breast feeding is effective for pain relief during venepuncture in term neonates and compare any effect with that of oral glucose combined with a pacifier. Design Randomised controlled trial. Participants 180 term newborn infants undergoing venepuncture; 45 in each group. Interventions During venepuncture infants were either breast fed (group 1), held in their mother's arms without breast feeding (group 2), given 1 ml of sterile water as placebo (group 3), or given 1 ml of 30% glucose followed by pacifier (group 4). Video recordings of the procedure were assessed by two observers blinded to the purpose of the study. Main outcome measures Pain related behaviours evaluated with two acute pain rating scales: the Douleur Aiguë Nouveau-né scale (range 0 to 10) and the premature infant pain profile scale (range 0 to 18). Results Median pain scores (interquartile range) for breast feeding, held in mother's arms, placebo, and 30% glucose plus pacifier groups were 1 (0-3), 10 (8.5-10), 10 (7.5-10), and 3 (0-5) with the Douleur Aiguë Nouveau-né scale and 4.5 (2.25-8), 13 (10.5-15), 12 (9-13), and 4 (1-6) with the premature infant pain profile scale. Analysis of variance showed significantly different median pain scores (P < 0.0001) among the groups. There were significant reductions in both scores for the breast feeding and glucose plus pacifier groups compared with the other two groups (P < 0.0001, two tailed Mann-Whitney U tests between groups). The difference in Douleur Aiguë Nouveau-né scores between breast feeding and glucose plus pacifier groups was not significant (P=0.16). Conclusions Breast feeding effectively reduces response to pain during minor invasive procedure in term neonates.