To the Editor,

We thank the authors for their engagement with our recent article, “Maternal Prenatal Use of Tobacco, Alcohol, and Illicit Drugs and Associations with Childhood Cancer Subtypes” (1). We wish to address their concerns and clarify the ways in which case-only studies can supplement the epidemiologic literature in the context of childhood cancer.

Padda and colleagues note that retrospective epidemiologic studies most frequently use a case–control design. Although “no etiologic investigation […] need describe a population,” the validity of case–control studies necessitates controls arising from the same source population as the cases (2). Unaffected participants in biomedical research are often socioeconomically advantaged compared with affected participants. Furthermore, childhood cancer has profound and lasting psychologic impacts on parents (3), exacerbating the potential for recall bias. Our article cites many traditional case–control studies of prenatal substance use and childhood cancer and also one registry-based...

You do not currently have access to this content.