Big-Data-Driven Medicine We need it now David Madigan Department of Statistics Columbia University
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Big-Data-Driven Medicine
We need it now
David Madigan Department of Statistics
Columbia University
“Evidence-Based Medicine” as against ???
How does the combination of Clinical Judgment and Evidence-Based
Medicine work in practice?
Should John have an angiogram?
John went to see a world-class cardiologist
48 years old
LDL = 70
university professor
no diabetes
HDL = 59
triglycerides = 106
calcium score in 2003 = 19 calcium score in 2008 = 42
father died of heart disease (47)
mother died of cancer (83) stress test normal in 2007
lipitor
CRP normal
exercise red wine
aspirin
arrhythmia in 2008
non-smoker
BMI = 21.6
normal heart ultrasound (2008) genotyping
EKG unusual in 2009
calcium score in 2010 = 70
48 years old
LDL = 70
university professor
no diabetes
HDL = 59
triglycerides = 106
calcium score in 2003 = 19 calcium score in 2008 = 42
father died of heart disease (47)
mother died of cancer (83) stress test normal in 2007
lipitor
CRP normal
exercise red wine
aspirin
arrhythmia in 2008
non-smoker
BMI = 21.6
normal heart ultrasound (2008)
Should John have an angiogram?
Clinical judgment?
Who are we kidding? genotyping
EKG unusual in 2009
calcium score in 2010 = 70
Big-Data-Driven Medicine
• Multiple years of medical records for 200+ million people • Largest collection of medical records in the world? • 32,430 patients just like John
Observational Medical Outcomes Partnership
Predictive Modeling a Core Methodological Challenge
Bias is the Enemy. What Causes it?
Nicola Orsini
Profound Data Quality Issues
We pay lip service to measurement error…
…and then do nothing about it.
Often acknowledge selection bias and confounding…
…and then do nothing about it.
Cohort
SCCS
Two Pieces of the Solution
1. Sensitivity Analysis
2. Establish Operating Characteristics
- How well do observational studies work? - How big is the bias in practice? - How often do 95% intervals contain the true RR? - What does this mean?
Related Documents
