Five Authors Reply

We appreciate the interest of Smits et al. (1) in our study (2). They raised an interesting point regarding possible “index event bias,” which is essentially a form of collider stratification bias that becomes possible when studying recurrent conditions. This can result in bias toward the null in crude associations between the risk factor of interest and disease outcome. One of the strengths of our study was our ability to adjust for multiple possible risk factors; however, the ability to adjust for unknown risk factors is a limitation in all epidemiologic studies and not only in this setting. Further, we agree with a point made by the authors in a prior publication (3); namely, the methods currently available to address index event bias in research related to the causal etiology of recurrent events are lacking, and additional methodological work is still needed to adequately address this issue.