How to Review a Large Database Study - Reviewer Workshop from July 2022 AOSSM Annual Meeting-How to Review a Large Database Study

Video Transcription

Video Summary

In this video, Jim is speaking at a conference and thanking Dr. Ryder and Donna for the invitation and assistance. He then goes on to discuss large database studies, their definition, strengths, and limitations. He explains that large database studies involve studying a collection of data that encompasses records at a state, multi-state, or national level. He mentions that these studies allow for reporting on rare patient groups or complications that would be difficult to study at a single institution. However, he notes that the quality of the data may be a limitation, as administrative or billing records were not intended for research purposes. He also mentions that these studies may lack detailed information, such as laterality, and may have limitations in follow-up beyond a certain time period. Jim explains the importance of understanding numerators and denominators when analyzing counts and rates in large database studies. He discusses the difference between rates and proportions and the use of confidence intervals. He also talks about statistical considerations, such as the need for adjusted p-values when multiple hypotheses are being tested. Jim concludes by highlighting the importance of considering the role of chance in both prospective and retrospective studies and the need to evaluate the statistics used to analyze rates and proportions.

Asset Caption

James L. Carey, MD, MPH

Keywords

large database studies

limitations

rates

proportions

confidence intervals

adjusted p-values