Home > Non‐pharmacological care for opioid withdrawal in newborns.

Pahl, Adrienne and Young, Leslie and Buus-Frank, Madge E and Marcellus, Leonora and Soll, Roger (2020) Non‐pharmacological care for opioid withdrawal in newborns. Cochrane Database of Systematic Reviews, 12, Art. no.: CD013217. DOI: 10.1002/14651858.CD013217.pub2.

External website: https://www.cochranelibrary.com/cdsr/doi/10.1002/1...


Implications for practice

Although non‐pharmacological care for opioid withdrawal in newborns has biologic plausibility, is low cost, and has potentially low risk of harm, limited evidence informs specific practices. We found six randomized controlled trials (RCTs) that provide very low‐ to low‐certainty evidence for all reported outcomes. Based on this evidence, it is difficult to definitively state an association of specific interventions with outcomes. Given the limited evidence to guide practice, other factors, such as the risk associated with the interventions or the cost to implement the intervention, may guide practitioners to determine optimal non‐pharmacological care practices for their setting while awaiting further evidence. For example, prone positioning is associated with measurable risk of infant death and is not a recommended position for safe sleep in infants (Moon 2016). It is notable that there is often a greater difference in outcomes between sites than there is between comparison groups within a study, suggesting that there are many unmeasured variables that impact care. These may include variability in quality and quantity of non‐pharmacological interventions. We do not have sufficient evidence to guide clinical practice about the use of non‐pharmacological care for opioid withdrawal in newborns.

Implications for research

Well‐designed appropriately powered studies are needed to determine the effect of non‐pharmacological care for opioid withdrawal in newborns. Studies should use clear and consistent definitions to reduce heterogeneity across sites and studies. To assess heterogeneity in population, it is important to record details about perinatal exposures and maternal medical history, particularly psychiatric history. Strategies to ensure representative enrollment of this vulnerable population are also important to ensure generalizability of studies. Clear definition and categorization of non‐pharmacological care practices is important in the framework of future studies. Better definition of timing and duration or 'dose' of non‐pharmacological interventions, cointerventions, and research outcomes will allow meaningful interpretation of interventions in a systematic review and meta‐analysis. Cointerventions may include other non‐pharmacological interventions, pharmacological interventions, and site of care among other interventions. Use of a core outcome set such as that proposed by Kelly 2020 could facilitate study of meaningful outcomes. Standardized definitions of outcome measures will facilitate meaningful combination of studies in systematic review and meta‐analysis. Future studies should consistently assess both important short‐term outcomes and longer‐term outcomes such as maternal mood and bonding, long‐term neurodevelopment, and social outcomes such as custody status.

The body of literature identified for this systematic review consists of few RCTs and a larger body of non‐randomized study. This is likely partially due to the practical challenges in designing randomized or blinded studies of point‐of‐care practices. Many practices, such as breastfeeding and rooming in, are difficult to randomize and impossible to blind. Others, such as systematic support of the maternal‐infant dyad, consist of cultural changes that will affect care of all patients in a care setting and may require broader community wide interventions. These cultural changes range from hospital practices such as staff ratios and site of care to deeper social issues such as stigma, racism, and poverty. Well‐designed larger studies to determine the most efficacious non‐pharmacological care practices will likely require creative design, such as a stepped wedge cluster randomized trial. To accomplish these important research goals, resources are needed to study point‐of‐care non‐pharmacological interventions and the effects of hospital practices such as site of care, staff ratios, and casualization of the workforce.

Repository Staff Only: item control page