We conducted a prospective observational cohort study to evaluate exclusive breastfeeding during the first three months of life in mother–child dyads in two Neonatology Units in the same geographical area, Apulia region, Southern Italy but with different hospital settings: Ente Ecclesiastico Ospedale Generale Miulli of Acquaviva delle Fonti, with on-site nurse h24 in the Nurse Unit (on site group) and Policlinico of Bari, with nurse not to be present in Nurse Unit but available on call h24 from Neonatology Unit which was on a different floor of the hospital (on call group).

Both Hospitals follow the steps 2, 5, 6, 7, 8, 9 of the Baby-Friendly Hospital Initiative. [28] The ten Steps to Successful Breastfeeding are as follows:

1.

Have a written breastfeeding policy that is routinely communicated to all healthcare staff

2.

Train all health-care staff in the skills necessary to implement this policy

3.

Inform all pregnant women about the benefits and management of breastfeeding

4.

Help mothers initiate breastfeeding within a half-hour of birth

5.

Show mothers how to breastfeed and how to maintain lactation, even if they are separated from their infants

6.

Give newborn infants no food or drink other than breast milk, unless medically indicated

7.

Practise rooming-in – allow mothers and infants to remain together – 24 h a day

8.

Encourage breastfeeding on demand

9.

Give no artificial teats or pacifiers (also called dummies or soothers) to breastfeeding infants

10.

Foster the establishment of breastfeeding support groups and refer mothers to them on discharge from the hospital or clinic.

In particular, step 2, which requires the training of all healthcare staff, represents an important intervention that improves breastfeeding rates. A recent systematic review, which included studies in 5 countries with 390 subjects belonging to the category of healthcare professionals, showed the usefulness of the action of healthcare personnel. [1] Provision of educational interventions aimed at increasing knowledge and practice of Baby-Friendly Hospital Initiative and support was found to improve health worker's knowledge, attitude, and compliance with the optimal breastfeeding practices. [1]

Both local Ethics Committees approved the study protocol.

Consecutive mother-baby dyads admitted to the units from 3 January to 31 of March 2018 were considered.

Healthy term newborns with gestational age ≥ 37 weeks in “rooming in” (baby in the same room of the mother all day) from birth to discharge and never hospitalized in Neonatal Intensive or Sub-Intensive Care Unit were included.

Exclusion criteria were all maternal and/or neonatal conditions that could interfere with breastfeeding (maternal HIV or active tuberculosis infection, herpes simplex lesions on both breasts, use of therapeutic radioactive isotopes, or exposure to radioactive materials, the use of drugs contraindicated in breastfeeding, galactosemia of the infant) or women not speaking Italian to ensure a full understanding of the questionnaire. Informed consent was obtained from both parents.

At discharge, a structured interview was performed, and a questionnaire was administered to the mother. The variables investigated with the questionnaire included sociodemographic features (maternal age and education), previous experiences (participation to a prenatal class and previous pregnancy), type of delivery, use of pacifier, nipple fissures, and satisfaction of nurse support. Variables subjected to changes during the timeframe of the study were collected by phone interview at 30 and 90 days of newborn’s life. The mode of breastfeeding was defined according to World Health Organization (WHO). The definition of breastfed infant was “an infant receives only breastmilk, no other liquids or solids are given”. [29]

During the study period, 400 infants were born in Ente Ecclesiastico Miulli Hospital: 43 preterm newborns with gestational age < 37 weeks and 51 newborns not exclusively assisted in rooming were excluded. A total of 303 dyads met the eligibility criteria. Among those, 7 were excluded (2 declined to participate and 5 did not speak Italian). In this hospital the Rooming-in Unit was continuously staffed by nurses that have additional training as lactation consultants, and who were physically present in the ward (on site group).

425 infants were born in Policlinico Hospital: 109 preterm newborns with gestational age < 37 weeks and 48 newborns not assisted in rooming-in were excluded. A total of 268 dyads met the eligibility criteria. Among those, 3 were excluded because did not speak Italian. In Policlinico Hospital a nurse with additional training as lactation consultant was available on call when a mother needed help for breastfeeding or to answer to her questions but was not physically present in the mother’s hospital ward (Rooming in Unit); they guaranteed a round in the Rooming in Unit two times a day. For the day long, a relative was present in the same room of the mother-infant dyads to take care of mother and newborn (on call group).

So, a total of 564 mother-baby dyads were examined, 299 in the on-site group and 265 in the on-call group.

In both groups no mothers reported post-partum depression, flare-up of previous breast pathologies (excluding fissures) or problems related to the family or external environment. 13/299 mothers (4,3%) in the on-site group and 35/265 (13,2%) in the on-call group reported little and/or discordant information about breastfeeding received during hospitalization. Of these, 6/13 (46,1%) in the on-site group and 17/35 (48,5%) in the on-call group were not exclusively breastfeeding at 1 and 3 months.

Data were collected by a healthcare professional at discharge and/or extracted from infants’ computerized medical charts (Neocare, I&T Informatica e Tecnologia Srl, Italy) and by phone interview at 30 and 90 days of newborn’s life. One mother did not answer at 30 days interview and 46 at 90 days and were excluded in the analysis of 30 and 90 days respectively.

Data were reported as mean ± standard deviation or percentage for categorical variables. The Student's t-test was used to compare continuous variables. Associations between categorical data were evaluated by using Chi-squared test or Fisher Exact test as appropriate. Multivariate logistic regression model was used to adjust for possible confounder factors. For all data a p-value of 0.05 or less was considered statistically significant. All analyses were conducted using STATA software, version 16 (Stata-Corp LP, College Station, Texas, USA).