A REALIST EVALUATION OF THE PROCESSES AND OUTCOMES OF INFANT MASSAGE PROGRAMS

Abstract

ABSTRACTThe sensitivity of early interactions conveyed through eye contact, voice tone, facial expression, and gentle touch plays a crucial role in healthy development. Recognition of the importance of parent–infant interaction for later attachment has underpinned the search for innovative ways of promoting sensitive parent–infant interaction. One such early intervention is infant massage, but a recent review found no evidence of its effectiveness in population samples. This points to the need to identify which parent–infant dyads are potentially able to benefit from such an intervention. This study aimed to identify the contextual factors and program mechanisms that were associated with different outcomes for a group of mother–infant dyads who attended infant massage programs. A Realist mixed‐methods research design was used to identify the context, mechanism, and outcome (CMO) patterns across a sample of 39 mother–infant dyads attending infant massage programs. A range of quantitative measures (Working Model of the Child Interview, Edinburgh Postnatal Depression Scale, and video recordings of mother–infant interaction coded with the CARE‐Index) were administered pre‐ and post‐intervention, alongside qualitative observations and interviews with a range of stakeholders. Three key CMO patterns were identified. Women categorized as “low” risk on the whole showed no change in parent–infant interaction (i.e., because they were already in the adequate range) and limited change in levels of depression, irrespective of the quality of the program attended. Mothers who were categorized as being at “moderate” risk (i.e., they had one to two risk factors over and above their demographic risk) appeared to require “good” quality programs (seven or more program mechanisms) for change to occur. Mothers categorized as being at “high risk” showed no benefits irrespective of the quality of the program, and there was evidence of unresponsive mothers becoming more intrusive. These findings suggest that infant massage programs should be targeted at parents experiencing moderate problems in terms of parent–infant interaction, and that primary care professionals working in a range of settings need the skills to identify such compromised parenting to do this. High‐risk mothers appear unlikely to benefit from infant massage alone. Rigorous randomized controlled trial (RCT) evidence is needed to test the hypotheses raised by this study.