Commentary on Krans <i>et al</i>.: Outcomes associated with the use of medications for opioid use disorder during pregnancy

Medications for opioid use disorder (MOUD) during pregnancy substantially reduce the use of unregulated opioids and are associated with improved fetal and maternal outcomes [1]. Krans et al. [2] assessed the impact of MOUD on five perinatal outcomes and classified MOUD as the total number of exposed weeks to account for non-continuous use. While this method avoided misclassifying MOUD duration for longer than actually present, understanding longitudinal behaviours such as initiation timing and adherence is essential to fully understand the impact of MOUD. The timing of the exposure, not just the duration, may be both developmentally important (i.e. potential risks for perinatal complications differ across critical and sensitive periods of fetal development in utero) and clinically relevant (i.e. timing of MOUD is associated with other maternal behaviours such as adequate prenatal care and illicit opioid use, which also have implications for perinatal outcomes). MOUD are associated with increased health-care utilization and, conversely, later initiation of MOUD in pregnancy is associated with fewer prenatal visits [3]. Inadequate prenatal care is associated with increased risk for preterm birth, low birthweight and neonatal intensive care unit admission [4]. Although not available in their data, by not accounting for the number and timing of prenatal visits the authors may have conflated the impact of prenatal care and the effects of MOUD. While it may be assumed that women with the least MOUD exposure started treatment in the third trimester, this is not necessarily true. Although not recommended in clinical guidelines, there is increasing interest in detoxification to reduce the risk of neonatal abstinence syndrome [5]. More than half of their study population (51%) with 1–10 weeks’ MOUD exposure started treatment before conception or during the first trimester. Such people may have stabilized prior to pregnancy and intentionally tapered or engaged in medically supervised withdrawal in the first few weeks after conception. These patients would have probably maintained contact with prenatal care, which would have a positive impact on prenatal outcomes, despite the short exposure to MOUD. As a result, their findings may have underestimated the effect of late initiation of MOUD on perinatal outcomes. The use of variable-centred analyses, such as multivariable regression, which predict outcomes based on relationships between variables, has dominated much of the literature assessing prenatal exposure to MOUD [6]. These methods do not account for unobserved heterogeneity in study populations [6]. In contrast, person-centred analysis techniques such as cluster analysis and latent class modelling can be used to identify important subgroups within a larger population [7]. In recent years the field of prenatal alcohol and tobacco use have greatly benefited from such techniques to identify trajectories of exposure [8-11], and recently they have been applied to MOUD [12]. Person-centred methods may overcome some of the existing limitations in administrative data, as acknowledged as an issue by Krans et al. [2]. By defining trajectories of MOUD, we may be better able to assess the specific impact of MOUD on perinatal outcomes accounting for the nuances of this complex exposure. None. I wish to acknowledge Dr Hilary Brown for her support and mentorship. Rose A. Schmidt: Conceptualization; writing-original draft preparation; writing-reviewing and editing.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.