Initiation and continuation of breastfeeding: theory of planned behaviour

Abstract
Aims. This paper reports a study investigating new mothers' subjective norms in relation to both breast- and bottle-feeding. The influence of norms on women's infant feeding decisions and the relative influence of social referents at varying degrees of social distance were assessed. Background. Increasing breastfeeding initiation and continuation rates is a key challenge for health educators. The perceived influence of other people's views (subjective norms), including the views of women's partners and health care professionals, is an important predictor of infant feeding behaviour. Methods. Semi-structured questionnaires were administered to 203 new mothers in central and northern Scotland and followed-up by postal questionnaire at 6 weeks. Infant feeding intentions, feeding behaviour at birth and follow-up, behavioural beliefs and subjective norms for both breastfeeding and bottle-feeding were assessed. The data were collected in 1998-1999. Results. Subjective norms were important determinants of initiation and continuation of breastfeeding for breast- and bottlefeeders. Breastfeeders rated close social referents as more in favour of bottle-feeding and more against breastfeeding at follow-up, whereas bottlefeeders' ratings did not change. Partner's and nurses'/midwives' views were an important influence at baseline and follow-up. Breastfeeding 'continuers' perceived their partners as more pro-breastfeeding at 6 weeks. Discontinuers perceived more overall social pressure to bottle-feed. However, sampling limitations may have led to over-representation of the views of breastfeeders at baseline and follow-up. Conclusions. Nurses and midwives have a crucial role in communicating positive views on breastfeeding to new mothers at different time points. Future interventions to promote breastfeeding could adopt a broad social approach, encouraging positive norms for existing and potential mothers and fathers, families and people in general.