Abstract Background Factors associated with duration of breastfeeding have been usually studied at specific times after birth. Little is known about how much time is added to breastfeeding by each associated factor. Methods A cohort of 969 mother-child dyads was followed-up for twelve months at the Marqués de Valdecilla University Hospital, Spain, in 2018. Data on mother characteristics, pregnancy, delivery and children characteristics were obtained from medical records. Length of breastfeeding was reported by the mothers and recorded in paediatric medical record at hospital discharge and 2, 4, 6, 9 and 12 months of life. Factors associated with duration of breastfeeding were analysed via multivariate Weibull regression parameterized as accelerated time of failure. Results are presented as time ratios. Results About four out of five children were breastfed at hospital discharge, although this proportion dropped to 65% in children born from smoker women, 70% in preterm children and 68% in neonates weighting less than 2500 g. Mother’s age was associated with longer breastfeeding, adding 2% more breastfeeding time per year (adjusted time ratio 1.02; 95% confidence interval 1.00, 1.04). Children born from mothers with university studies were breastfed 53% more time than those born from mothers with primary studies (adjusted time ratio 1.53; 95% confidence interval 1.21, 1.95); smoking in pregnancy decreased length of breastfeeding by 41% (adjusted time ratio 0.59; 95% confidence interval 0.46, 0.76). Other factors associated with longer breastfeeding were single pregnancy and newborn weight over 2500 g. Conclusions Analysing factors associated with duration of breastfeeding as time parameters allows us to quantify the amount of time gained or lost by each factor, which could make it easier to evaluate the relevance of programmes directed to promote facilitating breastfeeding factors.