Rasmussen 2011a.

Methods	Parallel‐RCT to determine if women who were overweight or obese who received additional breastfeeding support via telephone would breastfeed for longer than those receiving usual care
Participants	Setting: a hospital in a rural part of New York, USA, which has a history of promoting and supporting breastfeeding. Recruitment: infants born May 2006‐Februrary 2007 Inclusion criteria: women with a pre‐pregnancy BMI > 29 kg/m², who were intending to breastfeed, had no history of breast surgery, who were at least 19 years old and had a singleton fetus and were ≤ 35 weeks' gestation at enrolment to the study. Women were recruited if they resided near to the recruiting healthcare centre. Exclusion criteria: after enrolment women were excluded if they gave birth outside of the study data collection period, if their infant was not born at term, if the infant was never put to the breast, if the infant was injured during delivery, if the infant was placed into foster care or cared for elsewhere or if the mother was no longer in telephone contact during the postnatal period.
Interventions	Breastfeeding support intervention n = 25 Usual care n = 25 In both cases 5 women excluded after enrolment leaving 20 women in each group. Women receiving the intervention received usual care (rooming in with the infant post‐delivery, observed using the Mother‐Baby Assessment tool during at least 1 breastfeed in each 8‐h shift and a peripartum call from 1 of 3 lactation consultants asking them if they had any questions) and targeted breastfeeding support via telephone from lactation consultants. This included a more detailed peripartum call from a lactation consultant, which asked questions on knowledge, expectations and perceptions, answered questions the women had and reviewed practical points about breastfeeding pre‐delivery. After delivery nurses encouraged women to get up and move post‐delivery, nurses asked visitors to leave if they had been there > 2 h or were too numerous for privacy with breastfeeding or for bonding. An additional call from a lactation consultant at 24 h and 72 h after discharge was provided, which followed a script to standardise the assistance given, but also asked questions and addressed issues as necessary. The lactation consultant was able to order a visit if deemed necessary during the call.
Outcomes	Data were collected via telephone. This included a questionnaire pre‐delivery, collecting demographics, the woman’s goal for the duration of breastfeeding, prior experience with breastfeeding and pumping, and participation in the Special Supplemental Nutrition Program for Women, Infant and Children or the Prenatal Care Assistance Program for low‐income women. Women were contacted daily over the 1st 7 days asking questions to determine timing of lactogenesis and feeding methods. At 30 and 90 days postpartum data around infant feeding and breastfeeding support were collected. Successful breastfeeding initiation was defined as breastfeeding on day 4 after delivery. Duration of any breastfeeding was defined as the difference between the infant's date of birth and the last date they were offered the breast. Duration of exclusive breastfeeding was the date after discharge from hospital when the infant was given anything other than breast milk.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Participants were randomised at enrolment. Additional information from trial author stated that block randomisation was used (blocks of 10). Blocks were created using a random number table.
Allocation concealment (selection bias)	Unclear risk	Allocation concealment not mentioned in report. Correspondence with trial author stated that people recruiting were blinded to group allocation, but that they could not remember where group allocation was kept.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding of participants and lactation consultants was not possible due to the nature of the study. This could have led to performance bias and affected outcomes as breastfeeding outcomes were self‐reported.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Research assistants collecting data did not know the women's group allocation.
Incomplete outcome data (attrition bias) All outcomes	High risk	78% of those enrolled in the trial were included in the final analysis at 3 months postpartum, therefore attrition was > 20% therefore at high risk of bias.
Selective reporting (reporting bias)	Unclear risk	This study was assessed from a published report without access to the protocol, therefore we cannot be certain whether all pre‐specified outcomes were reported.
Other bias	High risk	Intervention fidelity: only 11/20 women in each group received the prespecified intervention calls as the intervention was not implemented as planned. Only 24 out of the 40 participants received the scheduled antenatal phone call and only 10 out of 20 postpartum intervention phone calls were completed.