The relationship between collapsibility and confounding has been subject to an extensive and ongoing discussion in the methodological literature. We discuss two subtly different definitions of collapsibility, and show that by considering causal effect measures based on counterfactual variables (rather than measures of association based on observed variables) it is possible to separate out the component of non-collapsibility which is due to the mathematical properties of the effect measure, from the components that are due to structural bias such as confounding. We provide new weights such that the causal risk ratio is collapsible over arbitrary baseline covariates. In the absence of confounding, these weights may be used for standardization of the risk ratio.

Introduction:Because a comparison of noninitiators and initiators of treatment may be hopelessly confounded, guidelines for the conduct of observational research often recommend using an “active” comparator group consisting of people who initiate a treatment other than the medication of interest. In this paper, we discuss the conditions under which this approach is valid if the goal is to emulate a trial with an inactive comparator.Identification of Effects:We provide conditions under which a target trial in a subpopulation can be validly emulated from observational data, using an active comparator that is known or believed to be inactive for the outcome of interest. The average treatment effect in the population as a whole is not identified, but under certain conditions this approach can be used to emulate a trial in the subset of individuals who were treated with the treatment of interest, in the subset of individuals who were treated with the treatment of interest but not with the comparator, or in the subset of individuals who were treated with both the treatment of interest and the active comparator.The Plausibility of the Comparability Conditions:We discuss whether the required conditions can be expected to hold in pharmacoepidemiologic research, with a particular focus on whether the conditions are plausible in situations where the standard analysis fails due to unmeasured confounding by access to health care or health seeking behaviors.Discussion:The conditions discussed in this paper may at best be approximately true. Investigators using active comparator designs to emulate trials with inactive comparators should exercise caution.

IMPORTANCE Understanding the safety profile of medications used in pregnancy is crucial for clinical decision-making. Few studies exist on the associations of exposure to benzodiazepines and benzodiazepine-like hypnotic drugs (z-hypnotics) in pregnancy with pregnancy outcomes. OBJECTIVE To determine whether exposure to benzodiazepines or z-hypnotics in pregnancy is associated with greater risk of negative immediate pregnancy outcomes compared with nonexposure. DESIGN, SETTING, AND PARTICIPANTS This questionnaire-based cohort study used data from the Norwegian Mother, Father and Child cohort study (MoBa), which also includes data from the Medical Birth Registry of Norway. Pregnant women were recruited from all over Norway from 1999 and 2008. The first child was born in October 1999 and the last in July 2009. This analysis included women who completed 3 questionnaires, twice during pregnancy and once 6 months after delivery. Data analyses were conducted from September to November 2019. EXPOSURES Self-reported exposure to benzodiazepines or z-hypnotics during pregnancy, characterized in terms of any exposure, timing (ie, early, middle, or late), and duration of exposure. MAIN OUTCOMES AND MEASURES The main outcomes were gestational age at delivery, risk of preterm delivery, birth weight, birth weight relative to gestational age and sex, risk of being small for gestational age, head circumference, Apgar score less than 7 at 5 minutes, and risk of neonatal respiratory distress. Continuous outcomes are reported using effect estimates as mean differences, and binary outcomes are reported using risk ratios. RESULTS The MoBa study included 114 234 mother-child dyads. This analysis of MoBa data includes 82 038 singleton pregnancies among 69 434 unique women. Mean (SD) maternal age was 30.2 (4.5) years, and 37 641 pregnancies (45.9%) were in primiparous women. Exposure to benzodiazepines or z-hypnotics was reported in 679 pregnancies (0.8%). After adjusting for all measured baseline and postbaseline confounders, benzodiazepine or z-hypnotic use during pregnancy was associated with lower birth weight (mean difference, −79.3 [95% CI, −126.7 to −31.9] g), lower gestational age at birth (mean difference, −2.1 [95% CI, −3.3 to −0.9] days), and higher risk of preterm birth (risk ratio, 1.41 [95% CI, 1.03 to 1.94]). We found no significant association of exposure to benzodiazepines or z-hypnotics with the child's birth weight relative to gestational age and sex (z score), or any of the other immediate birth outcomes. CONCLUSIONS AND RELEVANCE These findings suggest that the magnitude of the association of exposure to benzodiazepines or z-hypnotics with gestational age is not necessarily clinically significant. The absence of an association of exposure to benzodiazepines or z-hypnotics with z score for birth weight relative to gestational age and sex suggests that association of exposure to (continued) Key Points Question Is there an association of prenatal exposure to benzodiazepines or benzodiazepine-like hypnotics with immediate...

Anders Huitfeldt argues that the answer depends on your definition of “risk factor” and calls for greater clarity in research