Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

経済研Evidence-based Policy(2016Feb)

No description
by

Ryo Yamada

on 1 February 2016

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of 経済研Evidence-based Policy(2016Feb)

Decisions
in
Medicine

1st International Conference on Evidence-based Policy
Feb. 5, 2016

Ryo Yamada
Unit of Statistical Genetics
Graduate School of Medicine
Kyoto University

Diagnosis
Evidence-based Medicine
Multi-armed Bandit Problem
Population decisions
&
Multi-cellular Systems
Multiple-testings
James-Stein Estimation
DNA identification
Parentage
Mass Disaster
Chief complaint
Signs and Symptoms
Lab data and Images
Differential diagnosis
Clinical diagnosis
Diagnosed, then
Treatment:
TrA vs. TrB
https://en.wikipedia.org/wiki/Hemoptysis
Bayes' theorem
" TrA is Better than TrB for disease X"
Evidence from
Clinical Trials
Cancer Treatment - Conventional and Innovative Approaches DOI: 10.5772/55966
Glioblastoma
" TrA is Better than TrB for disease X"
Better Survival
Glioblastoma
Treatmen A
Treatmen B
If pts in Tr A are younger than pts in Tr B?
If pts in Tr A have worse kidney functions than pts in Tr B?
How about liver, heart... ?
Disease X = Glioma

Disease X = Glioma
Within a limited age range
With organs betther than cutoffs
Confounding factors
Precision medicine
The Grading of Recommendations Assessment, Development and Evaluation
" TrA is recommended for disease X"
"But TrB can also be taken"
NON-STRICT RULE
Difference
between
STRICT vs. NON-STRICT RULE?
Fixation
Right-side fixation
Wrong-side fixation
Multiple testing problem
When NO difference between TrA and TrB,
1 out of 100 trials give you p < 0.01
If ABSOLUTELY true,
follow it.

But...
Evidence based on statistical tests:
p-value based evidences
are still reliable, but...
Contexts not ready for
large-scale clinical trials
p-value-based evidence
Based on genetic background of individuals,
Choose the best actions for each.

No same person in the world.
Every patient's disease is its own and different from others'.

We have to decide with limited information.
How to make a decision
when data are limited?

TrA
TrB
Success
Failure
Personal optimization
Population optimization
vs.
Heterogeneity in selections avoids wrong-side fixation and save the population as a whole.
" TrA is Better than TrB for disease X"
Better Survival
Glioblastoma
No evidence for
Disease X
out of age range
not-satisfying organ conditions
...
Various types of evidences
with heterogeneous reliability
Integration of them
Personalized medicine
TrA
TrB
Success
Failure
0
0
0
0
TrA
TrB
Success
Failure
1
0
0
0
TrA
TrB
Success
Failure
1
1
0
0
TrA
TrB
Success
Failure
8
4
4
2
What to be optimized for a population ?
Big data-based science
Genome 3,000,000,000 base pairs
25,000 gene
Sometimes Clear
Sometimes Difficult
Disease X = Glioma

Disease X = Glioma
Within a limited age range
With organs better than cutoffs
Too young, too old
Organ dysfunction

Non-central majorities
Non-majority
Full transcript