Need help in deciding whether a study finding is clinically important?

A critical step in evaluating a quantitative medical study is to make a judgment about the size of the reported effect.  That requires an understanding of the outcome measure as well as some familiarity with interpreting small numbers, ratio measures, absolute vs. relative changes, and 95% confidence intervals.  In this series of short videos and assessments, H. Gilbert Welch MD, MPH, of the Center for Surgery and Public Health at Brigham and Women’s Hospital, walks learners through the basic skills required to distinguish clinical importance from statistical significance.

Learning Objectives:

By the end of this course, you will be able to:

• Recognize the fundamental components of a study: exposure, outcome and effect size – and that outcome measures fall on a spectrum of variable importance.
• Apply basic numeracy skills: work with small numbers, convert ratio measures to percentage change.  Learn how to use a spreadsheet.
• Compare two competing expressions for the effect size: absolute vs. relative change. Understand how relative changes tend to exaggerate both the perception of risk from harmful exposures and the perception of benefit from medical interventions.
• Contrast the role of chance in the observed effect (p value) from its precision (95% CI). Distinguish the magnitude of the p value (statistical significance) from the range of plausible effect sizes (clinical importance).
  Contrast the role of chance in the observed effect (p value) from its precision (95% CI).
  Distinguish the magnitude of the p value (statistical significance) from the range of plausible effect sizes (clinical importance).