Adverse impact of hospitalisation on infant breastfeeding practices: a prospective cohort study

Master Thesis

2019

Permanent link to this Item
Authors
Journal Title
Link to Journal
Journal ISSN
Volume Title
Publisher
Publisher
License
Series
Abstract
Background: In South Africa, the exclusive breastfeeding prevalence at six months is low at 24% and the under-5 mortality rate remains high. Improving breastfeeding rates is the most cost-effective intervention to reduce under-5 mortality and morbidity. Data on the effect of infant hospitalisation on breastfeeding may inform facility-based interventions to protect and support exclusive and prolonged breastfeeding. Aim: To assess the impact of hospitalisation on breastfeeding and explore reasons for stopping or continuing breastfeeding. Methods: We conducted a prospective cohort study of infant feeding practices among mother-infant dyads admitted to general paediatric wards at a tertiary children’s hospital in Cape Town, South Africa. Medical, demographic and feeding practice data were collected through semi-structured interviews on admission, again during hospitalisation and a third interview was conducted telephonically post discharge. Logistic regression analysis was used to assess factors associated with different feeding practices. Results: Between January and April 2018, 119 mothers (median age 26 years, IQR 22-32; 28% HIV-positive) were interviewed at admission; 39% (46/119) breastfed exclusively (EBF) and 28 (24%) reported no breastfeeding. Most infants (median age 1.8 months, IQR 1.0-3.2; 34% preterm) were admitted for lower respiratory tract infection (59%) or diarrhoea (14%). EBF at admission was associated with younger infant age (per month increase, aOR 0.18, 95% CI 0.07-0.43); none of the children admitted for diarrhoea had been EBF. A second in-hospital interview occurred at median 4 days (IQR 2-6) after admission. The overall prevalence of any breastfeeding declined from 77% at admission to 61% in-hospital. Risk factors for in-hospital breastfeeding cessation included low birth weight (<2500g; OR 3.81, 95% CI 1.35-10.74) and feeding via either bottle/tube (OR 51.00, 95% CI 6.38-407.71). Maternal expression of breastmilk (vs no expression in-hospital) was protective against in hospital breastfeeding cessation (OR 0.07, 95% CI 0.01-0.33). Post-discharge telephonic interviews (median 5 months after discharge) were available for 92 mother-infant dyads; 21 infants were ≤ six months of age, of whom 24% (5/21) were still exclusively breastfeeding. Breastfeeding cessation at any time after admission and before post-discharge telephonic interview was associated with maternal HIV infection (OR 2.82, 95% CI 0.84-9.40), full time employment (OR 4.95, 95% CI 1.40-17.46) and preterm birth (OR 3.53, 95% CI 1.27-9.81). Conclusion: Prevalence of both any and exclusive breastfeeding was low at admission to hospital, and lack of breastfeeding strongly correlated with increased risk of an infectious morbidity diagnosis. In addition, hospitalisation substantially reduced the probability of continued breastfeeding. In-hospital breastfeeding support and facilitation of breastmilk expression while infants are unable to breastfeed should be increased. Implementation research may define effective in-hospital breastfeeding support interventions.
Description
Keywords

Reference:

Collections