Interpreting Results from Clinical Research

Introduction

Practicing Evidence Based Medicine (EBM) has become the norm in Clinical Medicine for past few decades. Generating evidences by carrying out and publishing clinical research of various types (clinical trials, observational studies) have become a norm. It has been observed that around 80% of the clinical research findings and interpretations are not replicable, implicating that wrong clinical decisions are made based on the false findings. Major factors contributing to false results are biases created by researcher, editorial boards of journals and mis-interpretation of statistical terminologies. Hypothesis testing framework of inferential statistics and p value have been shown to be the most prevalent source of such mis-understandings as shown by G Cumming and Regina Nuzzo.

Estimation of effect size (ES) with confidence intervals (CI) of ES are much easy to understand and carry much more informations with them making them a preferable way to present results in clinical research.

Recently I delivered a talk on the above subject and this post provides beamer presentation for the same.

Hope you like the post.

Feel free to post any comment.

comments powered by Disqus