Although breast milk has numerous benefits for infants' development, with heightened effects in those born preterm (at < 37 gestational weeks), mothers of preterm infants have shorter breastfeeding duration than mothers of term infants. One of the explanations proposed is the difficulties in the transition from a Neonatal Intensive Care Unit (NICU) to the home environment. A person-centred proactive telephone support to breastfeeding mothers after discharge from NICU is expected to promote mothers' sense of trust in their own capacity and thereby facilitate breastfeeding. We hypothesize that proactive (health service initiated) telephone breastfeeding support offered to mothers of preterm infants after hospital discharge is more effective than reactive (mother initiated, and defined as usual care) telephone support at increasing the proportion of mothers who are exclusively breastfeeding 8 weeks after discharge.

A multicentre randomized controlled blinded trial has been designed to evaluate the effectiveness and cost-effectiveness of person-centred proactive telephone support on breastfeeding to mothers of preterm infants. Mothers will be informed about the study before discharge. Participating mothers will be randomized to either a control group or intervention group, immediately after discharge. Mothers will be notified to what group they have been randomized to through phone call or sms, depending on mother's preferences.

• Control group: person-centred reactive telephone support where mothers can phone the breastfeeding support team up to day 14 after hospital discharge.
• Intervention group: reactive support AND person-centred proactive telephone support in which the breastfeeding support team phones the mother daily for up to 14 days after hospital discharge.

A stratified block randomization will be used; group allocation will be done on high or low SES (i.e. educational level) and by NICU. Recruitment will be performed continuously until 1116 mothers (I: 558 C: 558) have been included. The data will be collected at eight weeks after discharge and at six months of infant's postnatal age using telephone interviews and questionnaires. Primary outcome is exclusive breastfeeding at eight weeks after discharge from the NICU. Secondary outcomes are breastfeeding (i.e. exclusive, partial, none and method), mothers satisfaction with breastfeeding, attachment, stress and quality of life in mothers/partners at eight weeks after hospital discharge and at six months postnatal age. A qualitative evaluation of experiences of providing/receiving the intervention will also be performed with mothers and staff respectively.

Aim and hypotheses This is a multi-centre randomized controlled trial (RCT), blinded for research team. The primary aim of the RCT is to evaluate the effectiveness of proactive telephone support on breastfeeding to mothers of preterm infants for up to 14 days after hospital discharge from NICUs. The secondary aim is to evaluate the effectiveness and cost-effectiveness of proactive telephone breastfeeding support on breastfeeding (exclusive, partial, none and method), mothers satisfaction with breastfeeding, attachment, parental stress and quality of life in mothers/partners at 8 weeks after hospital discharge and at six months postnatal age. We hypothesize that breastfeeding (i.e. duration and satisfaction), attachment, parental stress, and quality in life will be improved in mothers and partners who receive proactive telephone breastfeeding support. In addition, a qualitative evaluation will be performed, in which mothers and staff will be interviewed on their experiences of delivering and receiving proactive breastfeeding telephone support respectively.

Monitoring the study To evaluate the level of person-centeredness throughout the study and between groups (reactive vs. proactive), all phone-calls made by each team-member on three specified dates during the study period (beginning, middle, end) will be recorded and analyzed with the help of a coding-scheme. Telephone calls will only be taped if the mother consents to it. This procedure is important in order to: 1) monitor that support given is person-centered and regardless of group (I or C) and timing (beginning or end of the study period).

In addition, during the study period, members of the research team will sit in on feeding support team meetings (not when specific mothers are discussed) and during staff meetings, and will conduct short informal interviews with NICU managers - all with the aim to detect possible major changes (e.g. in care/staff-infant ratio/environment) or problems occurring.

Data dictionary

Following data will be collected from mothers:

Educational level (compulsory school or less; upper secondary school, higher education) Parity (primi or multi) Mode of delivery (C-section or vaginal deliver) Ethnicity (born in Sweden or not) Smoking habits (smoking in the past 2 months, yes or no) Gender of the infant(male, female, unknown) Single or multiple birth (singleton, twin, triplet, quadruplet) Gestational age (GA) at birth (weeks and days) Weight at birth (grams)

Following data will be collected from partners:

Educational level (compulsory school or less; upper secondary school, higher education) Ethnicity (born in Sweden or not) Smoking habits (smoking in the past 2 months, yes or no)

Data collected from medical records:

Infant's days on ventilator (number of days) Infant's days on Continuous positive airway pressure (CPAP)(number of days) Infant's length of hospital stay (number of days) Gestational age at discharge (grams) Weight at discharge (grams) Neonatal sequelae at discharge (boxes ticked and degree of severity if infant has: bronchopulmonary dysplasia, intraventricular haemorrhage, retinopathy of the preterm, necrotising enterocolitis, periventricular leucomalacia) Breast milk at discharge(exclusive, partial, non) Method of feeding at discharge(breastfeeding, bottle, other)

Standard Operating Procedures Each unit will keep a Log-book, the BST will record data on infants enrolled in the NICU, enrollment date, gestation week, eligibility for inclusion, assessed for eligibility and if mothers that decline participation have been asked to voluntary answer a few questions (i.e. age, parity, gestational age, delivery and educational level). Data on mothers/partners and infants participating in the study (i.e. name, phone number, address, identification code, group-allocation and infant's date of birth) will be recorded in the Log-book. This information, except group allocation, will be forwarded to researcher in charge, together with data on date of discharge and maternal educational level. Only the BST in each unit will know the allocated group for each mother. The identification code will be used to identify all distributed questionnaires and link data to the correct study subject. Established data protocols on demographics, infant health and breastfeeding are filled in by a BST member at inclusion and at discharge. All data collection at eight weeks after discharge and six months of infant's postnatal age will be sent out by the researcher in charge. Data will be transferred to SPSS and analyzed.

Sample size assessment A priori power analysis has been done to determine adequate sample size for the study. This is a study of independent cases and controls with 1 control per case. Prior data indicate that the exclusive breastfeeding rate among controls is 0.53 (at two month in preterm infants). If the true exclusive breastfeeding rate for intervention mothers is 0.615 (effect size 0.085), we will need to study 531 intervention mothers and 531 control mothers to be able to reject the null hypothesis that the exclusive breastfeeding rates for intervention and control mothers are equal with a probability (power) of 0.8. The Type I error probability associated with this test of this null hypothesis is 0.05. We estimate that the drop-out is 5 %, thus require additional 54 mothers. In total we need a sample size of at least 1116 mothers (I: 558, C: 558). The trial is also powdered in the subgroup, low SES mothers. We will use an uncorrected chi-squared statistic to evaluate this null hypothesis.

Statistical analyses All members of the research team will be blind to the group allocation (I, C) throughout the study period and during analyses of the primary outcome. Intention-to-treat will be used. This means that analyses will be include all randomized mothers in the groups to which they were randomly assigned, regardless of their adherence with the entry criteria, regardless of the treatment they actually received, and regardless of subsequent withdrawal from treatment or deviation from the protocol. The specific statistical analysis used with each outcome variable will be determined by the distribution of the outcome variables. Subgroup analyses will be conducted on SES (low vs. high), parity (primipara vs. multipara), and on GA (very preterm vs. preterm).

Inclusion Criteria:

• mothers with preterm infants (< 37 gestational weeks)
• infant admitted to one of the four selected NICUs for at least 48 hours
• mother who breastfeed or express breast milk

Exclusion Criteria:

• serious maternal medical or psychiatric problems at discharge
• language problems that cannot be resolved
• infant is transferred to another hospital/unit after discharge
• infant that is terminally ill