ORLANDO, FL USA (UroToday.com) - One of the early sessions in the Society for Pediatric Urology (SPU) meeting was a concise review of statistical concepts and methods of reporting data. This was a panel discussion led by Bruce Schlomer, MD from Texas Children’s Hospital designed to familiarize clinicians with the concepts that will allow them to critically appraise the modern scientific literature.
Hillary Copp, MD from the University of California, San Francisco discussed secondary data analysis and outlined 4 steps to produce a publication of this nature. She warned of the dangers of “data-mining” and emphasized that papers of this nature should be subject to the same rules as other research, meaning that it should be hypothesis-driven and original. She discussed the importance of having thorough knowledge of one’s dataset, including a complete understanding of its structure and limitations. Dr. Copp also suggested that detailed study planning is necessary at the outset, and creating shell tables will facilitate study design. Finally she emphasized the need for appropriate analysis and interpretation since this data is subject to multiple limitations.
Luis Braga, MD from McMaster University in Ontario outlined the proper way to conduct a systemic review and meta-analysis, as well as how to critically appraise the data of a publication. He pointed out that studies may have a large amount of heterogeneity, and thus may not appropriate to group within an analysis. He suggested readers should use a checklist to determine the quality of the systematic review (i.e., CEBM). Authors should seek literature from multiple sources when personally conducting a review, such as MEDLINE, EMBASE, Cochrane reviews, CINAHL and the grey literature (unpublished results).
Armando Lorenzo, MD from the Hospital for Sick Children reviewed how guidelines are created, and Caleb Nelson, MD from the Children’s Hospital Boston reviewed core statistical concepts to complete the session.
The SPU is placing enhanced emphasis on the creation and publication of high quality data as well as improving our ability to analytically review it. Once again, a call for randomized controlled trials was put forth. As the literature becomes more complex, there is a suggestion that physicians need to play a more active, critical role as part of the audience. It is not enough to read the published papers; one must be able to digest them and be able to recognize how the findings may or may not be relevant to a general population.
Moderated by Bruce Schlomer, MD at the Society for Pediatric Urology (SPU) Annual Meeting held concurrently with the American Urological Association (AUA) Annual Meeting - May 16 - 21, 2014 - Orlando, Florida USA
Texas Children’s Hospital, Houston, Texas USA
Written by Michaella Prasad, MD of the Medical University of South Carolina (Charleston), and medical writer for UroToday.com
