This research was approved by the Ethics Committee of Shahroud University of Medical Sciences (code: 930/24). This experimental study was conducted in NICUs of two hospitals of Tehran; Mahdieh and Shahid Akbar-Abadi, from Feb 14, 2016 to May 14, 2016. Both are educational and referral centers including three levels of NICU. Unit of random allocation was hospitals. Avoiding the contamination between parents is the reason of this type of randomization. In the long run Mahdieh assigned to intervention site and Shahid Akbar-Abadi assigned as control site. The study population of this study was the mothers of premature newborns. The Sampling was done after obtaining written consent.
All premature newborns and their mothers with inclusion/exclusion criteria were recruited in the study during three months period (15). The inclusion criteria were: having a preterm infant with gestational age <37 weeks, birth weight <2500 gr, High probability of survival, Declaration of consent to participate in the study, Iranian nationality and ability to communicate verbally. Exclusion criteria were: preterm infants without abnormality or disabling conditions such as intraventricular hemorrhage (IVH) grade 3 or 4.
In the intervention group, 75 infants and mothers were included and 68 infants and mothers were included in the control group during the study period.
Intervention: The designed intervention was conducted based on the support system pattern of mothers with premature infants in the experimental group. In this pattern, parents and babies are at the center of the supportive model that need to be supported at critical transitional periods and consisted of preconception, prenatal, neonatal unit, the transition to home, and at home. Different interventions in support of mothers in NICU were planned for the program, which was gradually implemented within 3 months of the intervention. At the first stage, mothers were prepared for the first visit of their infants in the neonatal ward as described in another paper. At the same time, the mothers were encouraged to a constant presence in the NICU to gradually become empowered in the maternal process by the constant presence in the ward, observation, and guided participation in the care of the infants (appraisal and Instrumental support). The researcher, who also had the experience of having a premature baby, was present in the NICU six days a week from 9 am to 4 pm and contacted the neonatologist and nurses for the infants to inform about details of the infants' conditions, and then conveyed the appropriate information to the mothers with simple and fluent words. Furthermore, the information about the NICU, appearance, and behavior of preterm infants, and the way of maternal roles were provided for every mother according to their understanding, literacy, and needs, and responded to any questions and, if necessary, repeated the training (informational support). The researcher examined the mother's relationship with the infants, and accordingly, guided the mothers to interact with the infants, and identify their behavioral symptoms (appraisal support). She guided the mothers to carry out baby calming interactions through the exchange of smell (encouraging the mother to put milk pad under the infant's head and picking up the previous pad to keep it on her side during milking and for the separation times from the baby), touch slow and steady, quiet whispering and eye contact with the baby (emotional support). She provided breastfeeding training in the neonatal ward in counseling and practical assistance, and followed up by phone (informational support) after the baby was discharged., She was provided encouragement, practical assistance, training, counseling for the mothers by various ways to start kangaroo care and perform it as soon as possible (appraisal and instrumental support). In cases of intra-ward or inter-ward transfer of the infants, she accompanied the mothers and explained the new conditions to the mothers (informational support). The mothers became members of the "MATIN mothers" group in the social network to express their concerns in that space (emotional support) and gain practical information by talking to mothers who had experienced having premature infants. In the group, which was under the constant supervision of the researcher, there were new mothers and mothers who had successful experience of having premature infants in previous years. In the group, the researcher uploaded different educational contents about breastfeeding, kangaroo care, massage, infant follow-up, etc. according to the mothers' needs, and answered possible uploaded questions, and monitored mothers' interactions to prevent any misguided guidance and advice (informational support).
A session was held often once a week with the researcher as a leader/facilitator in the mothers' restroom by inviting an experienced volunteer mother who had previously been trained about supportive boundaries in a four-hour workshop. In the session, the mothers' needs and demands were discussed and the practical experiences of the peer mother were used (informational and appraisal support).
An attempt was made to provide spiritual support for mothers by emphasizing trust, recourse, patience, and submission to divine destiny. The telephone number was provided for mothers who could call and ask any questions they might have throughout the day, whether during hospitalization or discharge, up to three months after the infant's birth (informational support). Before discharge, the mothers were taught about checking the temperature, method of bathe, giving the baby possible medications, etc. (appraisal support) and the way of pursuing recommended follow-ups for the premature infant, including vision, hearing and etc. and they were reminded until three months after birth (informational support).
Measurements: Data collection tools included: demographic questionnaire and maternal and infant clinical data, the Nurse Parent Support Tool (NPST). Demographic questionnaire including: mother's age, mother's education level, mother's employment status, insurance status, family income, ethnicity and mother's clinical data including: parity, abortion, live child, multiple pregnancy status, type of delivery and neonatal clinical data included: sex of infant, gestational age, birth weight, number of days of invasive and non-invasive ventilation.
We utilized the Nurse Parent Support Tool (NPST) to assess the mother's perception of the perceived support in the four fields, instrumental, emotional, appraisal, and informational support. The Miles Questionnaire (1998) is a 21-item self-report tool that assesses the mother's perception of perceived support in four fields: instrumental, emotional, appraisal, and informational. Answers are ranked on a Likert scale. A score 1 means the minimum support and 5 means the maximum support. The validity and reliability of the English and Persian versions of the questionnaire have already been determined (16, 17).
Statistical Analysis: The obtained data were analyzed by Stata 13 (StataCorp, College Station, Texas). T-test used for comparison of means between the two groups and chi-square test was used to assess the independence between categorical variables. For control of pre-treatment variable due to cluster randomization (not random allocation at the individual level), we used Inverse Probability Treatment Weights (IPTW) to compute weighted averages of the outcomes for each intervention level (18). The IPTW method was utilized to control the heterogeneity in basic variables between two groups. In this regard, variable entitled disease intensity was initially generated based on the infant’s requirements and a lack of need for invasive and non-invasive ventilation. Then, the treatment variable was modelled by logistic regression on basic variables of infant’s disease intensity, age, birth weight, the reception-time status of anxiety and reception-time personality anxiety which is known as the propensity score (PS). The weights are 1/PS for the intervention group participants and 1/ (1−PS) for the control participants. The average treatment effect (ATE) is an adjusted estimate of difference scores between two groups using weights.