To determine the association between sucking in infants born preterm and developmental outcomes at 5 years. Thirty-four infants were included (mean gestational age 30wks 4d, mean birthweight 1407g). The Neonatal Oral-Motor Assessment Scale was used longitudinally from 37 to 50weeks postmenstrual age. At 5years, we assessed motor skills, intelligence, language, verbal memory, and behavioural problems. Linear regression analyses were performed to test whether aspects of sucking behaviour predicted these developmental outcomes. Where linear regression was not appropriate, Spearman's correlation coefficients were calculated between sucking and developmental outcomes. Sucking was associated with total motor skills (B [unstandardized correlation coefficient for normally distributed data]=22.66, 95% confidence interval [CI] 6.61 to 38.71), balance (Spearman's ρ=0.64, p<0.001), total intelligence (B=-1.16, 95% CI -1.89 to -0.44, B=10.48, 95% CI 0.39 to 20.71, B=-2.22, 95% CI -3.42 to -1.02), verbal intelligence (B=-0.95; 95% CI -1.83 to -0.07, B=-2.02; 95% CI -3.55 to -0.49), performance intelligence (B=-1.34, 95% CI -2.13 to -0.54, B=12.36, 95% CI 1.13 to 23.60, B=-2.37, 95% CI -3.75 to -0.96), and language (B=-1.78, 95% CI -3.36 to -0.19). All associations were in the same direction: the better the sucking, the higher the test scores. Verbal memory and behavioural problems were not associated with sucking. Abnormal sucking between 42 weeks and 50 weeks postmenstrual age may reflect abnormal neurological functioning in children born preterm.