Breastfeeding Initiation and Duration: A 1990-2000 Literature Review


  • Cindy-Lee Dennis RN, PhD

    Corresponding author
    1. Cindy-Lee Dennis is an assistant professor, Faculty of Nursing, University of Toronto, Ontario, Canada.
    Search for more papers by this author

Address for correspondence: Cindy-Lee Dennis, RN, PhD, Faculty of Nursing, University of Toronto, 50 St. George St., Toronto, Ontario, Canada M5S 3H4. E-mail: E-mail:


Objective: To review the literature on breastfeeding initiation and duration and to delineate effective strategies for promoting positive breastfeeding behaviors.

Data Sources: Computerized searches on MEDLINE, CINAHL, and the Cochrane Library.

Study Selection: Articles from indexed journals relevant to the objective and published after 1990 (except for classic findings) were reviewed. Although a myriad of pertinent articles was located, referenced citations were limited to three per point. When article selection was required for a specific point, preferences were given to (a) randomized controlled trials; (b) meta-analyses; (c) studies with the largest, most representative samples; and (d) investigations conducted in North America.

Data Extraction: Data were extracted and organized under the following headings: benefits of breastfeeding, breastfeeding initiation and duration, personal characteristics, attitudinal and intrapersonal characteristics, hospital policies and intrapartum experience, sources of support, breastfeeding interventions, and review implications.

Data Synthesis: Although the health benefits of breastfeeding are well documented and initiation rates have increased over the past 20 years, most mothers wean before the recommended 6-months postpartum because of perceived difficulties with breastfeeding rather than due to maternal choice. Women least likely to breastfeed are those who are young, have a low income, belong to an ethnic minority, are unsupported, are employed full-time, decided to breastfeed during or late in pregnancy, have negative attitudes toward breastfeeding, and have low confidence in their ability to breastfeed. Support from the mother's partner or a nonprofessional greatly increases the likelihood of positive breastfeeding behaviors. Health care professionals can be a negative source of support if their lack of knowledge results in inaccurate or inconsistent advice. Furthermore, a number of hospital routines are potentially detrimental to breastfeeding. Although professional interventions that enhance the usual care mothers receive increase breastfeeding duration to 2 months, these supportive strategies have limited long-term effects. Peer support interventions also promote positive breastfeeding behaviors and should be considered.

Conclusions: A promising intervention is the complementation of professional services with peer support from a mother experienced in breastfeeding. This lay support appears to be an effective intervention with socially disadvantaged women.