Background and aims Breastmilk contains factors important for immunity. However, the benefit of prolonged full breastfeeding for 4 compared to 6 months as infection prevention in an affluent society is unclear.

Methods In a prospective birth cohort study recruiting from 1999–2008, questionnaires were completed at 6 and 18 months of age regarding infant diet and the occurrence and hospitalisation for infections. The study sample included 67 856 children (50.9% girls).

Results After adjustments for maternal age, parity, sectio delivery, smoking, gender, birthweight, gestational age and daycare, hospitalisation for infections was similar for infants fully breastfed for 6 (14%, reference category) vs 4 (41%) months (adjusted OR 0.88, 95% CI 0.77–1.00) but increased for never breastfed (aOR 1.41, 0.99–2.00) or partially breastfed <4 months (aOR 1.25, 1.06–1.47). Gastroenteritis was associated with no breastfeeding (aOR 1.77, 1.42–2.21), partial breastfeeding < 4 months (aOR 1.90, 1.72–2.09) or 4–6 months (aOR 1.33, 1.22–1.45). From 6 to 18 months the risk of hospitalisation for infection was higher for those who stopped any breastfeeding < 4 months (aOR 1.40, 1.25–1.57), 4–6 (aOR 1.29, 1.13–1.49) and for 6–8 months (aOR 1.11, 1.00–1.24), whereas similar risk was observed for breastfeeding 9–11 compared to breastfeeding ≥12 months.

Conclusions Full breastfeeding for 6 months was associated with a reduction in hospitalisation for infections compared with absent or partial breastfeeding from <4 months, but not compared to full breastfeeding for 4 months. Reduced risk of hospitalisation was found with any breastfeeding up to 9 months’ age but not beyond.