Evaluation of Baby Friendly Spaces in Cox's Bazar, Bangladesh.

Action contre la Faim (ACF) currently provides psychosocial support services to Rohingya mothers of malnourished children who are receiving nutrition support at Integrated Nutrition Centers (INCs) within refugee camps in Cox's Bazar, Bangladesh. At the INCs, the Baby Friendly Spaces (BFS) program is offered as a part of ACF's psychosocial and care practices support strategy. The BFS program is a holistic psychosocial support program designed to enhance mothers' wellbeing in order to buffer against the deleterious health and developmental impacts of conflict and disaster on children. BFS is designed to be flexible to meet the unique needs of different conflict-affected populations using a community-based approach, but always focuses on two domains: psychosocial support that targets maternal wellbeing and childcare practices that target caregiver functioning to also address child wellbeing and development. In so doing, BFS strengthens mothers' internal resources and skills in caring for their children to positively impact the nutritional status and wellbeing of their children during humanitarian emergencies. The purpose of this study is to evaluate the effectiveness of the BFS program for improving the psychosocial wellbeing of Rohingya refugee mothers of young (ages 0-2) children experiencing malnutrition in Cox's Bazar, Bangladesh.

As they are currently functioning, BFS activities in Cox's Bazar are not standardized as intended. Thus, this study will use a pre-post, paired randomized study design to compare two conditions being delivered within Integrated Nutrition Centres (INCs) in Cox's Bazar-based refugee camps: 1) treatment-as-usual BFS mental health and psychosocial support prevention and promotion activities (TAU BFS) as they are currently being offered; and 2) implementation-enhanced BFS activities (enhanced-BFS) that includes intervention standardization, re-training, and implementation supports. The primary outcomes being evaluated are symptoms of psychological distress and functional impairment, and secondary outcomes include subjective maternal wellbeing and coping skills. Although a randomized controlled trial (RCT) comparing BFS against a non-active control condition would provide the clearest evidence, such a design is not feasible in this context because the services are already being offered as a part of existing programming, and removal of any existing supports would be unethical. As such, the proposed study design uses current BFS practices occurring in Integrated Nutrition Centers (INCs) that have not been manipulated as part of the research (treatment as usual). These services are not manualized and have been affected by natural drifts in fidelity over time since initial training. In the intervention condition, psychosocial workers and psychologists at select INCs will be re-trained in a manualized version of the BFS program, and will continue to receive additional implementation support during the study through newly established supervision structures (enhanced-BFS).

As the experimental manipulation is occurring at the INC-level, INCs were paired by overseeing psychologist (as in some cases one psychologist oversees two sites) and randomize within the pair to intervention (BFS-enhanced) or treatment as usual control (BFS TAU). In addition to incorporating site matching in the design phase to help try to minimize bias in group comparisons while assessing the intervention as implemented in the real world, propensity score adjustments in the analysis approach will be leveraged. After completing the BFS intake, eligible and interested mothers are referred to a data collector to obtain informed consent and administer a baseline interview. Mothers can then participate in BFS to whatever extent they choose. The same data collector will follow up with each mother 8-10 weeks later to complete a follow-up interview. Because of logistical constraints, each INC has been assigned a single data collector to carry out study activities at that site. All sites will also record detailed information (as part of regular program monitoring) about activities delivered and received by women enrolled in BFS at the INCs. Two trained observers will rotate through the INCs to record fidelity data. This real-world comparison of an intervention implemented "as is" versus an enhanced program implemented with effort towards greater standardization and fidelity will provide insight about effectiveness of the program model and structures and resources that need to be in place to effectively implement the program.