Statistics 101 for the oncologist

 

Episode 12:

David L. Streiner, PhD, of McMaster University and the University of Toronto, joins Blood & Cancer host David Henry, MD, of Pennsylvania Hospital, Philadelphia, to break down the basic statistics knowledge that doctors need – but may not have – to understand and contextualize research findings.

In Clinical Correlation, Ilana Yurkiewicz, MD, of Stanford (Calif.) University talks about reclaiming the term “sorry.” Dr. Yurkiewicz writes a regular column for Hematology News, which you can find by clicking here.

 

Show notes

By Emily Bryer, DO, resident in the department of internal medicine, University of Pennsylvania, Philadelphia

• Statistics provide objective evidence for whether we are seeing a true effect of a medicine or intervention, or if that outcome occurs because of chance alone.
• Clinical trials include a selective population (usually from tertiary care hospitals) and tend to involve the most severe cases.
• When interpreting data reported from a trial, it is essential to ask how similar the population in the trial is to the population you see in your clinical practice.

 

• Drug Approval Stages
  • Phase 1: Assess drug efficacy and toxicity
  • Phase 2: Identify correct dose; assess pharmacodynamics and pharmacokinetics
  • Phase 3: Randomized, controlled trial necessary for confirming drug effectiveness and providing a comparison to current standard of care treatments
• How to design a trial
  • Ensure that the surrogate endpoint is highly correlated to the outcome of interest

 

 

Additional reading by Dr. Streiner:

• Health Measurement Scales: A practical guide to their development and use (5th edition)
• PDQ Epidemiology (3rd edition)