In Reply Although we agree with Wisner and Yonkers that our findings1 can potentially be explained by residual confounding by indication, we have performed several additional sensitivity analyses to validate our results. First, the impact of the diseases for which benzodiazepines are prescribed was addressed in the main analysis where statistical adjustments were made for the diagnoses of the 2 main indications (mood and anxiety disorders and insomnia) and documented proxies of these diseases (concomitant exposure to antidepressants and/or antipsychotics, psychiatrist visits, comorbidities, and emergency department visits and/or hospitalizations in the year before pregnancy). Second, additional sensitivity analyses were performed restricting the study population to pregnant women with history of mood and anxiety disorders, where incident benzodiazepine use during early pregnancy was associated with an even higher risk of spontaneous abortion (adjusted odds ratio [aOR], 2.85; P < .05). These findings suggest that incident use of benzodiazepines in early pregnancy is associated with an increased risk of spontaneous abortion above and beyond the risk for maternal stress or depression.