Background
Existing information on breastfeeding in low income countries such as Nepal has been largely derived from cross-sectional demographic health surveys. This study investigated exclusive breastfeeding rates, and compared the duration of exclusive breastfeeding between rural and urban mothers in central Nepal using an alternate cohort methodology.

Methods
A community-based prospective cohort study was conducted among 639 recently delivered mothers representative of the Kaski district of Nepal. Breastfeeding information was obtained at birth (n = 639), 4 weeks (n = 639), 12 weeks (n = 615; 96.2%) and 22 weeks (n = 515; 80.6%) through repeated interviews using validated questionnaires. Risk of cessation of exclusive breastfeeding was assessed by Cox regression analysis.