To the Editor—We acknowledge the Letter to the Editor submitted by Rankin et al^{Reference Rankin, Rebeiro, Haddadin, Halasa and Greevy1} and appreciate its unique perspective. We thank Rankin et al for their feedback and emphasis on methodological integrity when making causal inferences.

In the original paper, the intention was to explore and describe associative or correlational factors of a physician’s decision to prescribe antibiotics with a viral respiratory infection confirmed by PCR.^{Reference Li, Kang-Birken, Mathews, Kenner and Fitzgibbons2} A “table 2 fallacy” occurs when causal modeling is intended but does not account for confounding variables appropriately in the model,^{Reference Westreich and Greenland3} which does not apply to our study since our study was purely descriptive, with the limitations acknowledged.^{Reference Li, Kang-Birken, Mathews, Kenner and Fitzgibbons2} We thank Rankin et al again for the comments, and we suggest additional research in the future to investigate causation.