Changes in Hospitalizations, Treatment Patterns, and Outcomes During Major Cardiovascular Meetings--Reply.

Changes in Hospitalizations, Treatment Patterns, and Outcomes During Major Cardiovascular Meetings To the Editor We read with interest the article by Jena et al,1 and we commend them for their thoughtful analyses. Many readers will undoubtedly cite decreased mortality during cardiovascular meetings as the study’s key finding, but we favor a more conservative conclusion. The investigators set out to test the reasonable hypothesis that mortality would increase during meetings. For this reason, we believe a 1-sided statistical test would have been more appropriate for the primary analysis rather than the 2-sided test employed.2 By negating their primary hypothesis, the authors showed that mortality did not increase during meetings, which we view as the study’s key finding. The observation that mortality rates decreased during meetings merits further investigation and may be explained by factors beyond those raised by the authors. Differences in case mix may still partially account for the observed outcomes. Neither information about hemodynamics nor details and sequences of therapeutics that could reflect disease severity are provided by available Medicare data. Furthermore, at many centers, complex procedures, such as staged percutaneous coronary interventions for postmyocardial infarction cases, are generally performed by experts likely to attend meetings. Less frequent performance of such high-risk procedures during conferences could lower mortality rates. Most importantly, we wonder if patient care during the period immediately surrounding meetings could have created the appearance of decreased mortality during meetings. Is it possible that pressure to complete work before leaving, or trying to catch up on return, led to worse outcomes?3 To address this question, werecommendcomparingoutcomesimmediatelyprecedingand followingmeetingswithdatesunrelatedtomeetings.Distinguishing between a rise in mortality before and after meetings, as opposedtothedropduringmeetingssuggestedbythisstudy,would constitute an essential first step toward explaining the authors’ potentially important observation.