Effect of an extended midwifery postnatal support programme on the duration of breast feeding: A randomised controlled trial
Introduction
Breast feeding is recognised as the optimal method of neonatal nutrition. Exclusive breast feeding is currently promoted for the first six months after birth, with breast milk as the major nutrient source for at least the first 12 months of life (World Health Organization, 1998, World Health Organization, 2002; Binns and Davidson, 2003; Gartner et al., 2005). There is evidence that human milk has many nutritional, immunological, psychological and maternal advantages that benefit both disadvantaged socio-economic communities as well as relatively advantaged communities (Raisler et al., 1999; Gartner et al., 2005; Coulibaly et al., 2006).
Australian guidelines currently aim for a breast-feeding initiation rate of 90%, and for 80% of babies to be breast fed at six months of age (50% exclusively) (Binns and Davidson, 2003). While the majority of Australian centres are approaching the goal for initiation, continuation of breast feeding after discharge from hospital remains below desired levels. Recent Australian studies have found rates of breast feeding, whether exclusive (no artificial milk substitute is introduced to the baby's diet) or partial (supplementary artificial milk feeds given in addition to breast feeds), at six months to be between 50% and 60% (Scott et al., 1999; Henderson A.M. et al., 2001; Henderson J.J. et al., 2003; Forster et al., 2006). In contrast, there is a wide variation in the duration of breast feeding in other developed countries. Rates of any breast feeding at six months have been reported recently as 32–40% in the USA (Ryan et al., 2002; Bonuck et al., 2005), 62% in New Zealand (only three per cent exclusively) (Heath et al., 2002), 80% in Norway (only seven per cent exclusively) (Lande et al., 2003), and 13% in the UK (Morrell et al., 2000). A more recent UK survey of 9416 mothers indicated that 25% were breast feeding at six months. While this represents an increase from earlier reports, the rates of exclusive breast feeding remain negligible with less than one per cent of mothers choosing to feed with breast milk alone (Bolling et al., 2007).
The factors that influence early cessation of breast feeding have been studied extensively. Sociodemographic factors such as social disadvantage, lower levels of education and younger maternal age have been shown to be strongly associated with reduced duration of breast feeding (Redman et al., 1992; Ford et al., 1994; Michaelsen et al., 1994; Piper and Parks, 1996; Barber et al., 1997; Bourgoin et al., 1997; Evers et al., 1998; Scott et al., 1999; Donath and Amir, 2000). Insufficient support for breast feeding, particularly from women's social networks, also reduces breast-feeding success (Barber et al., 1997; Raj and Plichta, 1998; Tarkka et al., 1999).
A large variety of interventions to increase the duration of breast feeding have been tested in different settings with varied effectiveness. The Baby-Friendly Hospitals Initiative (BFHI) is an internationally recognised initiative that utilises ‘10 Steps to Successful Breastfeeding’ (World Health Organization/United Nations Children's Fund Joint Statement, 1989; Hofvander, 2005, p. 1013) to guide specific practices in maternity wards aimed at supporting and promoting breast feeding (Luzia et al., 2003; Hofvander, 2005). The BFHI has been instrumental in increasing breast-feeding initiation rates, but has been shown to be inconsistent in increasing the duration of breast feeding. A large UK prospective study showed no benefits of BFHI accreditation for breast feeding at one month postpartum after statistical adjustment for social factors (Bartington et al., 2006). In contrast, a large cluster randomised controlled trial (RCT) in Belarus found significant improvements in both exclusive and any breast feeding in sites randomised for BFHI implementation (Kramer et al., 2001).
Single interventions, either antenatal or postnatal, have generally been ineffective in populations with high initiation rates (Britton et al., 2007). Interventions tested vary in timing (antenatal or postnatal), provider (lactation consultant, midwife, community nurse or peer counsellor), and whether their principal objective is to increase knowledge or provide support.
Antenatal education sessions have been shown to improve breast-feeding initiation rates and duration in settings where few women initiate breast feeding (Mattar et al., 2007) but not where initiation rates are relatively high (Redman et al., 1995; Forster et al., 2004). A large Australian RCT compared two different mid-pregnancy education programmes, one skill-based and the other addressing attitudes towards breast feeding, with usual care. Initiation rates were uniformly high (96%); however, no differences were found between groups for breast feeding at six months postpartum (Forster et al., 2004).
Single postnatal breast-feeding educational sessions were found to have no effect on the duration of breast feeding (Schy et al., 1996; Henderson A.M. et al., 2001; Labere et al., 2003). However, a small RCT conducted by Schy et al. in 1996 indicated that the duration of breast feeding was statistically correlated with mothers’ perceived level of satisfaction, education level and expected duration of breast feeding. Schy et al. (1996) recommended that further research is required into women's individual breast-feeding experiences and their determinants for breast-feeding continuation.
Postnatal telephone-based support alone provided inconsistent findings. This was demonstrated by an Australian study of rural women receiving telephone support from a lactation consultant. It showed a small reduction in early cessation of exclusive breast feeding in women who attended a private hospital compared with similar women who gave birth prior to the introduction of this support service. However, no such difference was in evidence for women who attended a public hospital (Fallon et al., 2005).
The recently updated Cochrane systematic review of RCTs found no differences in the duration of breast feeding for either telephone support alone or combined with home visiting (Britton et al., 2007). However, home visiting without other measures was found to be beneficial for breast-feeding continuation.
Overall, the Cochrane systematic review of all forms of extra postnatal support found a decrease in breast-feeding cessation before six months, although there was considerable variation in the subgroup analyses (Britton et al., 2007). The review was inconclusive in relation to support given by professional providers (midwives or lactation consultants). There were slight beneficial effects for exclusive breast feeding but no differences for any breast feeding up to six months. Furthermore, no beneficial effects of postnatal support were demonstrated in settings where there were high breast-feeding initiation rates (over 80%) compared with areas of moderate initiation (Britton et al., 2007).
Hence, there remains uncertainty about the most effective ‘package’ of interventions to promote breast feeding. Previous research has shown that individual interventions, such as telephone contact and home visiting, provide some success for breast-feeding continuation. However, such support programmes have been time limited and largely restricted to either antenatal or postnatal interventions. No research has endeavoured to encourage continuity by combining promising interventions into one package of support.
The aim of the present study was to develop and test a package of postnatal support measures that would prolong breast feeding effectively in a setting where initiation is high. This package consisted of a one-on-one postnatal education session and up to six weeks of midwifery home visiting with telephone contact.
Section snippets
Methods
The aims of this study were:
- •
to test, by means of an RCT, whether an extended midwifery support (EMS) programme, consisting of up to six weeks of midwifery home visiting, would increase the proportion of women who breast feed fully to six months;
- •
to evaluate the psychological well-being of women; and
- •
to undertake an economic evaluation of the programme.
This paper only reports on the first aim; the evaluation of the midwifery support programme. Other outcomes from this study will be the subject of
Findings
In total, 849 women were recruited with 425 allocated to the EMS group and 424 allocated to the SMS group. Participant flow through the study is shown in Figure 1. The majority of women who declined to take part tended to do so for reasons of feeling unable to commit to the study requirements. Ninety-three per cent of the women in the EMS group received the in-hospital education session. The majority of women (74%) received between two and four home visits, 8% only received one visit and 7%
Discussion
The present study is the largest to compare the effectiveness of additional professional support with standard midwifery care on breast feeding. The assumption that additional postnatal support, in the form of telephone calls and home visits by a midwife, would be beneficial for the duration of breast feeding was tested. This was based on the belief that early identification of breast-feeding problems and early intervention would prevent early cessation of breast feeding. It was hypothesised
Conclusion
In conclusion, the EMS programme did not succeed in increasing the proportion of women performing full breast feeding or any breast feeding at six months postpartum in a setting with high initiation rates. Future research into programmes designed to promote breast feeding continues to be imperative in view of the advantages of breast feeding for all mothers and babies.
Acknowledgements
The authors wish to thank research midwives Jocelyn Bristol, Jenny Francis, Colleen Ball and Sheena McClery, and all the women who participated in this trial. This research was supported by grants from Healthway, Women and Infants Research Foundation, and King Edward Memorial Hospital, Perth, Western Australia.
References (52)
- et al.
A longitudinal study of breastfeeding and weaning practices during the first year of life in Dunedin, New Zealand
Journal of the American Dietetic Association
(2002) - et al.
Postnatal home visits in teenage mothers: a randomised controlled trial
The Lancet
(2003) - et al.
Preventive health behaviours among parents of infants aged four months
Australian Journal of Public Health
(1992) - et al.
Using a breastfeeding prevalence survey to identify a population for targeted programs
Canadian Journal of Public Health
(1997) - et al.
Are breastfeeding rates higher among mothers delivering in baby friendly accredited maternity units in the UK?
International Journal of Epidemiology
(2006) - et al.
Maternal education and the incidence and duration of breast feeding: a prospective study
Journal of Pediatric Gastroenterology and Nutrition
(2003) - et al.
Infant feeding guidelines for health workers. Dietary guidelines for children and adolescents in Australia
National Health and Medical Research Council
(2003) - et al.
Factors affecting the continuance of breastfeeding
Acta Paediatrica Scandinavica Supplement
(1982) - et al.
Infant Feeding Survey 2005
(2007) - et al.
Randomized, controlled trial of a prenatal and postnatal lactation consultant intervention on duration and intensity of breastfeeding up to 12 months
Pediatrics
(2005)