The Alberta Neonatal Abstinence Syndrome Mother-Baby Care ImprovEmeNT Program

The Alberta Neonatal Abstinence Syndrome Mother-Baby Care ImprovEmeNT (NASCENT) Program: A Stepped Wedge Cluster Randomized Trial of a Hospital-level Neonatal Abstinence Syndrome Intervention

Neonatal Abstinence Syndrome (NAS), is a common and costly problem in Alberta that affects approximately 250 babies per year exposed to drugs during pregnancy. Unfortunately, this has become more common in the last 10 years. Babies with NAS can be very difficult to care for with poor feeding, diarrhea, and extreme irritability. These babies often receive specialized care and medications in the Neonatal Intensive Care Unit (NICU), which leads to separation of mothers and babies at a time when it is most important that they be together. This separation is traumatic for families and expensive for the health and foster care systems, as babies often end up being cared for by governmental agencies. Recent research has shown that keeping mothers and babies together in a quiet, supportive environment in hospital, called 'rooming in', leads to a decreased need for NICU admission, decreased amount of time spent in the NICU, increased rates of breastfeeding, and an increase in babies going home with their mothers. This project will systematically introduce a program of 'rooming-in' to hospitals in Alberta to determine if the investigators can improve NAS care provided to babies and mothers. The goal is to decrease NICU admission and length of stay, increase the number of babies going home with mothers, increase breastfeeding rates, and increase the number of women enrolled in supportive programs for substance use. The investigators will also determine if this rooming-in model of care decreases health and societal costs associated with caring for babies with NAS.

This is a Stepped Wedge-Cluster Randomized Controlled Trial (SW-cRCT) of 8 hospitals with NICUs and post-partum/pediatric units. The SW-cRCT is ideally suited to interventions that require implementation by multiple team members in which there is an impact on workflow and the structure of care delivery. There is randomized sequential roll-out of the intervention to all sites over time. Baseline data acts as the pre-intervention control comparisons and in addition, there is some ability to account for secular trends due to concurrent control groups for most periods. The analysis can also assess whether the impact of the intervention changes over time.

Primary Research Question: Does an evidence-based implementation project to apply a NAS care bundle for infants born at >36 weeks gestation to mothers who report opiate use during pregnancy and are admitted to hospital lead to a reduction in NICU Length of Stay (LOS)? Secondary outcomes that will be assessed include:

• rates of child apprehension, NAS pharmacological management, breastfeeding at discharge, and maternal participation in an Opioid Dependancy Program/ Virtual Opioid Dependancy Program (ODP/VODP);
• satisfaction of stakeholders;
• costs associated with intervention versus current care;
• maternal mental health, parenting self-efficacy, and quality of life at discharge and at 6 months follow-up; and
• short-term developmental outcomes as determined at 6 months using the Ages and Stages Questionnaire: Social-Emotional Second Edition (ASQ:SE-2).

Hospitals will implement the intervention following a computer-generated stratified random allocation sequence created by the independent study statistician. Stratification will be based on 'readiness'. The goal is to recruit participants prior to hospital admission for delivery but potential participants will also be approached if their delivery presentation is the first recruitment opportunity.

Data will be collected from consenting participants using standardized data collection forms. Anonymous LOS data will also be collected from administrative data for NAS admissions at all sites pre-initiation and at each step for patients that do not consent. The following databases will be used to collect baseline LOS, perinatal characteristics, and data to inform the economic analysis: Alberta Perinatal Health Program; Canadian Institute for Human Information (CIHI) Costing Database; Discharge Abstract Database; ConnectCare; and Alberta Health Services (AHS) Finance and Data Enterprise Warehouse. Data linkage will be through unique identifiers and probabilistic linkage.

Sample Size/Feasibility - Conservatively, mean NICU LOS was set at 15 days and standard deviation (SD) at 3 days based on local and national data to provide adequate power to detect small differences in LOS. Intra-cluster correlation (ICC) was conservatively estimated at 0.1. With 8 centres, 8 initiation steps and 1 centre initiated per step over the course of 3 years, a mean LOS of 15 days (SD 3 days), and an ICC of 0.1 the study will have at least 80% power with a two-sided 5% significance level to detect a 19% difference (2.75 days) in NICU LOS.

Statistical Analysis - The primary outcome of NICU LOS was chosen as NICU admission may occur for several medical reasons unrelated to NAS and a difference in LOS may be more meaningful in understanding NAS severity. Cases in which there is no NICU admission will be assigned a NICU LOS value of 0 days. Generalized Linear Mixed Models and Estimating Equations will be used in analyzing the primary and secondary outcomes.

Barriers and Facilitators The investigators have connected directly with patients and families in identifying gaps and solutions. The MMUNA and EMBRACE programs receive feedback from families and staff related to the care provided and iteratively improve the programs based on lessons learned. The drive to change current practice is directly informed by our ongoing interactions with these organizations and their clients, as well as by our patients.

NASCENT Team Members completed the Comprehensive Accessible care for Infants with Neonatal abstinence (CAIN) study funded by the Addiction and Mental Health (AMH) (Strategic Clinical Network) SCN. This project was a first step in identifying appropriate and safe care of NAS newborns across all Alberta regions. The strategies used in the NASCENT study will be directly informed by themes identified in the CAIN study. Through focus groups and individual interviews with care-providers in hospital and community settings as well as with parents who had experienced care of an infant with NAS the investigators explored:

• existing gaps in the care of babies with NAS in hospital and in the community;
• how to improve communication between existing hospital and community networks;
• attitudes, perceptions and experiences regarding NAS;
• facilitators and barriers to changing NAS care;
• challenges and experiences of providing care to newborns with NAS; and
• strategies and existing resources to enhance support program interconnectedness to improve health services for newborns with NAS.

Interviews were with leaders, care-providers, and parents with experience with the MMUNA program. The overarching theme identified in the CAIN study was 'hope'. Fostering hope was key for families and care-providers in working with mother-infant dyads with NAS. In addition, patients who went through the MMUNA program and care-providers who are directly involved with MMUNA have provided feedback through the CAIN study as well as through direct feedback to the MMUNA program. The success of both MMUNA and EMBRACE is an indication of their engagement with families.

Despite the successes seen with the MMUNA program there are still significant challenges related to cultural barriers in providing care to families impacted by NAS. From previous work in CAIN it is evident that there is trepidation, discomfort, and judgement related to providing care to infants and families who require support for NAS. Care-providers experience moral distress related to uncertainty and discomfort with addictions. Tailored implementation strategies targeting physician, nursing, and social work barriers and leveraging existing facilitators can be effective in changing practice. The NASCENT Project will be designed to bring about provider and unit-level behaviour and culture change. This will be explored during Pre-Implementation at each site using focus groups and a theory-based survey and addressed through the use of a behaviour change management framework; the Capability-Opportunity-Motivation-Behaviour Framework and Behaviour Change Wheel . Part of the planned intervention is to access existing resources developed to assist caregivers to examine their behaviors and identify unconscious bias. The NASCENT model of intervention is ideally suited to understanding and changing hospital culture and individual behaviour related to NAS care. This behaviour change framework is already being used successfully in Alberta by NASCENT co-investigators in the REDUCED Trial.

Implementation Study Outcomes Through prospective data collection, including surveys of stakeholder satisfaction, during each phase of the project the NASCENT team will be able to respond to risks as they arise. The experiences and lessons learned from each Site Implementation Team will be shared across sites to inform risk-mitigation strategies. Sites that have successfully implemented the intervention will share their stories with sites that are getting ready for implementation.

There will be three phases at each site during which Implementation, Service, and Client outcomes will be measured including Pre-Implementation, Implementation and Post-Implementation. During all phases, stakeholder satisfaction with current infant care for NAS will be assessed. Stakeholders to be surveyed include hospital and community health care providers, social workers, managers, families, and clinical and operational leaders. Surveys will have common elements across sites but will also include site-specific items that address identified facilitators, barriers, and contextual influences. Data related to costs, safety, and adverse events will be collected during all phases.

NASCENT is as an evidence-based intervention to bring about provider, unit, and community-level behaviour and culture change. Physician, nursing, social work, patient, and management facilitators and barriers as well as contextual influences on implementation will be explored using theory-based surveys, stakeholder interviews, and focus groups at participant sites prior to site initiation. Transcripts from surveys, interviews, and focus groups will be analyzed thematically by the investigative team with representation from leadership, nursing, physicians, social work, patients, and families to identify site-specific facilitators and barriers as well as contextual influences within and across sites. Findings from this process will be mapped to the Capability-Opportunity-Motivation-Behaviour Framework (COM-B) and then used to generate Behaviour Change Wheel interventions to introduce the 'rooming in' model of care. This methodological approach is already being used successfully in Alberta by NASCENT co-investigators in the REDUCED Trial which introduced new obstetrical guidelines related to identifying poor progress in labor. The NASCENT team will guide evidence- and local experience-based identification of appropriate adaptations to the core NAS-interventions including, recent national and provincial guidelines, and lessons learned from existing programs like MMUNA and EMBRACE.

With NASCENT there will be sequential roll-out of the intervention to all participant hospitals, with a new hospital initiated every 3 months following an initial minimum 6 months of baseline data collection.

There will be three phases during implementation: Pre-Implementation, Implementation and Post-Implementation. During all phases, stakeholder satisfaction with current infant care for NAS will be assessed including site-specific items that address identified facilitators, barriers, and contextual influences. The focus of each implementation phase is as follows:

A) Pre-implementation

• Identification of site-specific facilitators and barriers;
• Identification of site operational, medical, nursing, and social work champions and formation of Site Implementation Team;
• Creation of site-specific interventions and plan to prepare for NAS care, e.g., identification of units and commencement dates where rooming-in will take place and contingency plans for when census reaches 100% occupancy; development of site-specific mechanisms to refer to the ODP/VODP program; development of additional staff resources to address specific learning needs such as (Frequently Asked Questions) FAQs; development of mechanisms for prenatal recruitment of opioid dependent women through community agencies; and
• Completion of staff education on MyLearningLink related to NASCENT interventions.

B) Implementation

-With each step (3-month period) in the trial, data related to the intervention will be collected. These data will be shared with stakeholders in feedback sessions as a mechanism to iteratively explore facilitators and barriers, modify interventions based on feedback, and provide ongoing staff education .

C) Post-Implementation

• NASCENT results, experience at each site and between sites, lessons learned, satisfaction from stakeholder groups, and economic analysis will be shared in feedback sessions with stakeholders at each site; and
• The NASCENT Project is being designed to become the 'new normal' standard of care for infants with NAS and their mothers. The elements related to stakeholder education will become part of routine training and orientation for care providers.

Inclusion Criteria:

• Infants born at >36 weeks gestation to mothers who report opiate use during pregnancy (or who are in an ODP/VODP program) and who are admitted to a participating implementation project hospital.

Exclusion Criteria:

• Infants born at less than 36 weeks and/or birth weight less than 2000 grams and infants with congenital anomalies will be excluded as these infants would be expected to be admitted to NICU.