Carryover Effect in Sibling Analysis of Acetaminophen Use in Pregnancy
The carryover effect in sibling analysis refers to the potential for acetaminophen exposure during one pregnancy to biologically or environmentally affect subsequent pregnancies, thereby violating a key assumption of the sibling-comparison design that siblings serve as independent controls for one another. 1
Understanding the Sibling-Comparison Design
The sibling-comparison design is a powerful epidemiological method used to control for unmeasured confounding in studies of prenatal exposures:
Siblings share approximately 50% of their segregating genes and often have similar early environments, making them ideal natural controls for genetic and family-level confounding factors. 1
The design compares siblings who are discordant for prenatal acetaminophen exposure (one exposed, one unexposed) to account for all unmeasured genetic and environmental factors that make siblings similar, including maternal characteristics and other factors that remain stable across pregnancies. 1
If differentially exposed siblings show similar risk for neurodevelopmental outcomes, this suggests no causal effect of the exposure, as familial confounding would be controlled. 1
The Carryover Effect Problem
A critical assumption of sibling-comparison studies is that there are no carryover effects from the exposed sibling to the unexposed sibling. 1 This means:
Acetaminophen use during one pregnancy should not biologically or environmentally affect subsequent pregnancies. 1
If carryover effects exist, the "unexposed" sibling may actually be indirectly affected by the exposure during their sibling's pregnancy, contaminating the comparison and potentially attenuating observed associations. 1
This would make the sibling-comparison design unable to properly isolate the causal effect of acetaminophen exposure, as the control group would no longer be truly unexposed. 1
Implications for Acetaminophen Research
In the context of acetaminophen use during pregnancy and neurodevelopmental outcomes:
One notable sibling-controlled study found that children exposed to prenatal paracetamol for more than 28 days had poorer gross motor development, communication, externalizing behavior, internalizing behavior, and higher activity levels even after controlling for familial factors. 2
The persistence of associations in sibling-comparison analyses suggests that familial confounding alone does not fully explain the observed relationships, though carryover effects remain a theoretical concern that cannot be definitively ruled out. 1, 2
The sibling-comparison design cannot account for unmeasured confounding by factors that vary across a woman's pregnancies, though researchers can use measured covariates to adjust for time-varying factors. 1
Additional Methodological Considerations
Beyond carryover effects, sibling-comparison studies face other limitations:
Random measurement error or mediators shared by siblings could cause observed attenuation of associations in sibling-comparison studies, making interpretation complex. 1
Large sample sizes are required to obtain adequate numbers of discordant siblings, limiting the feasibility of this design. 1
The design is particularly helpful for testing genetic and family-level confounding given that genetic confounding is likely to exist in studies of prenatal exposures. 1