Transcript Document 15242

SP02
Using Data and Text
Mining to drive Innovation
Dave Smith, SAS UK
Copyright © 2010 SAS Institute Inc. All rights reserved.
Agenda
Context
• The state of the European pharmaceutical industry
What is analytics?
• An introduction to the main elements and why they
matter
The promise of analytics
• How analytics can revolutionise the pharmaceutical
industry
2
Copyright © 2010, SAS Institute Inc. All rights reserved.
Dark Days for the Pharmaceutical Industry in
Europe
3
Copyright © 2010, SAS Institute Inc. All rights reserved.
Why Analytics?
4
Copyright © 2010, SAS Institute Inc. All rights reserved.
5
Copyright © 2010, SAS Institute Inc. All rights reserved.
Decision Making
“It would appear, Hopkins, that your gut feel was only indigestion.”
6
Copyright © 2010, SAS Institute Inc. All rights reserved.
7
Copyright © 2010, SAS Institute Inc. All rights reserved.
8
Copyright © 2010, SAS Institute Inc. All rights reserved.
Competitive
Advantage
Optimise
What is the best that can happen?
What will happen next?
Predict
What if these trends continue?
Why is this happening?
Alerts
Query
drill down
Raw
data
Clean
data
Standard
reports
Forecast
Statistical
Analysis
What actions are needed?
Where exactly is the problem?
Ad hoc
reports
How many, how often, where?
What happened?
Degree of Intelligence
Copyright © 2010, SAS Institute Inc. All rights reserved.
9
Visualisation & Statistical Analysis
“Quickly find related
patterns within a set of
data.”
10
Copyright © 2010, SAS Institute Inc. All rights reserved.
Associations
“Identifying events
that occur together,
potentially with a
sequence or order.”
11
Copyright © 2010, SAS Institute Inc. All rights reserved.
“We’ve doubled our output in terms of the number of reports we produce for
peer review publications, and the Ministry of Health and Long-Term Care, with
only a 10 percent increase in our frontline analytic staff”
 Dr. Jan Hux, Chief Operating Officer
Challenges:
SAS Solution:
Business Impact:
 Sifting through 17 years of data
 Grid enabled platform for Data
 Able to evaluate chronic care
on every hospitalization,
Mining, Text Mining and
and map chronic diseases to
physician appointment and
Optimisation
improve the lives of Ontarians
publicly funded drug claim in
Ontario to guide provincial
healthcare decision making
12
Copyright © 2010, SAS Institute Inc. All rights reserved.
Variable Identification
“What’s important to answer
a specific question?”
Masses of information…
Age
No. of staff
County
Income Band
Practice Size
Brick
Gender
Last Training
Existing Prescriber
Attended Meeting
GI Specialist
GI Specialist
Contact History
Other Products
Attended Meeting
Brick
Web customer
Marital status
Existing Prescriber
Value
Status
Time as GP
Behaviour Explained
13
Copyright © 2010, SAS Institute Inc. All rights reserved.
Top 10 Pharmaceutical Company
Challenges:
SAS Solution:
Business Impact:
 Determining the causes of
 Data integration, data
 Batch Failures almost
manufacturing batch failures
analysis and web portal
eliminated from
manufacturing line for best
selling product, saving £2M
per annum
 Other products
subsequently included from
across the globe
14
Copyright © 2010, SAS Institute Inc. All rights reserved.
Segmentation
“Create groups that have
similar characteristics. Also a
measure of how different each
group is from the others.”
15
Copyright © 2010, SAS Institute Inc. All rights reserved.
Segmentation
Physician Group A
Large practice, Strong formulary, Hard to see
“Create groups that have
similar characteristics. Also a
measure of how different each
group is from the others.”
Physician Group B
Small Practice, Weak formulary, Easy to see
Physician Group C
GI Specialists, existing prescribers of Brand B
Copyright © 2010, SAS Institute Inc. All rights reserved.
16
Top 10 Pharmaceutical Company
Challenges:
SAS Solution:
Business Impact:
 Improving sales force
 SAS Physician Targeting
 Able to reduce the number
targeting to get the most out
of the sales reps
solution
of physicians targeted from
62,000 to 39,000 while
maintaining the same level
of prescribing potential,
saving $18M
 Competitive advantage from
calling on physicians not
targeted by other companies
17
Copyright © 2010, SAS Institute Inc. All rights reserved.
G
Q
C
A
O
V
S
A
F
Q
T
W
M
Z
P
H
D
L
E
P
E
E
N
C
H
J
W
K
W
S
W
T
X
F
Y
U
V
J
I
F
V
D
H
T
R
Y
B
T
Y
T
G
F
G
E
M
T
U
M
N
E
H
G
A
R
W
I
I
A
H
M
U
J
L
T
S
P
N
P
K
X
Copyright © 2010, SAS Institute Inc. All rights reserved.
I
K
O
N
D
Q
S
I
D
T
B
O
J
J
F
A
W
O
N
R
C
I
U
H
M
P
B
I
Q
G
X
U
T
Y
N
G
C
U
U
E
A
T
Q
U
B
F
Z
X
P
O
18
Top 20 Pharmaceutical Company
Challenges:
SAS Solution:
Business Impact:
 Identifying key opinion
 SAS Text Analytics
 Able to identify KOLs from
leaders in the Rheumatoid
Arthritis market
mining of publications
 As a side benefit identified
Vioxx signals
19
Copyright © 2010, SAS Institute Inc. All rights reserved.
Time Series Forecasting
“Observations repeated over
time, with past values and
other factors being used to
predict future values.”
20
Copyright © 2010, SAS Institute Inc. All rights reserved.
.
“SAS is the linchpin of our supply chain management strategy.”
 Custom Pak Materials Manager
Challenges:
SAS Solution:
Business Impact:
 Ability to track customer usage
 Analytics for accurate and
 Saved $6 million in inventory
and demand patterns to
unique forecast for each
accurately predict inventory
Custom Pak and consumables
levels and meet customer
product.
demand.
costs.
 Customer service levels have
been improved year on year.
 Able to negotiate lower
purchasing costs and maintain
excellent relationships with
suppliers.
21
Copyright © 2010, SAS Institute Inc. All rights reserved.
Predictive Modelling
100,000 observations
YES = 40%
NO= 60%
“What are the chances of an event
Age > 35
occurring some future point in time?”
Age < 35
70,000
YES = 53%
NO= 47%
30,000
YES = 10%
NO= 90%
Income > 30k
“Should an event occur, what is the impact
or amount involved?”
Income < 30k
45,000
YES = 78%
NO= 22%
25,000
YES = 8%
NO= 92%
Activation
α+ βage+ γincome
tanh(α+ βage+ γincome)=A
Input
Layer
Combination
ε + δA + ζB + πC
22
Copyright © 2010, SAS Institute Inc. All rights reserved.
US Healthcare Provider
.
Challenges:
SAS Solution:
Business Impact:
 High costs and poor outcomes
 Data Mining
 Built a predictive model to
for patients with Asthma and
identify Asthma and cardiac
Cardiac conditions
patients at risk of hospitalisation
 Reduced hospitalisations by
80%
23
Copyright © 2010, SAS Institute Inc. All rights reserved.
Simulation
“Understand the impact of
changing conditions based on
different scenarios.”
24
Copyright © 2010, SAS Institute Inc. All rights reserved.
US Biotechnology Company
.
Challenges:
SAS Solution:
Business Impact:
 Need to eliminate costly trials
 Grid enabled simulation
 Massive increase in throughput
with inconclusive outcomes
application on 120 processor
of simulations (1,000
Linux Cluster
replicates/day  10,000
replicates/hour)
 Generally available SAS
application created
25
Copyright © 2010, SAS Institute Inc. All rights reserved.
Constraint Based Optimisation
How quickly can you get from the start to the end…
“What is the right thing to do to
maximise or minimise an
objective given a number of
limitations.”
… given you can only turn left nine times and back once?
26
Copyright © 2010, SAS Institute Inc. All rights reserved.
Model Management
“Make accurate, fact-based decisions
using the right method at the right time.”
27
Copyright © 2010, SAS Institute Inc. All rights reserved.
Competitive
Advantage
Optimise
Predict
Forecast
Statistical
Analysis
Alerts
Query
drill down
Raw
data
Clean
data
Standard
reports
Ad hoc
reports
Degree of Intelligence
Copyright © 2010, SAS Institute Inc. All rights reserved.
28
It’s Not Magic!
29
Copyright © 2010, SAS Institute Inc. All rights reserved.
Where will Prescriptions be Made?
Pharmaceutical
sales data, rep activity
Segmentation
Visualisation
Associations
Prescriber
Profiles
Determine what is
prescriber’s normal
behaviour.
Decision
Visualisation Identify Variables
Determine which elements are useful
in predicting whether a particular
brand will be prescribed
Predictive Modelling
Build a method to determine the
optimum marketing activity and
deployment of sales resources
.
Copyright © 2010, SAS Institute Inc. All rights reserved.
Deployment
?
Monitor
Ensure the predicted Rx occur and
that the model is consistently
monitored and improved.
30
31
Copyright © 2010, SAS Institute Inc. All rights reserved.
32
Copyright © 2010, SAS Institute Inc. All rights reserved.
The Value of Analytics
Predictive modelling
• Healthcare provider reduces hospitalisations by 80%
• The average cost of an asthma hospitalisation is $11,671
Variable Reduction
• Large Pharma saves £2m per annum on one product by reducing batch failures
Simulation
• Trial Simulation avoids inconclusive trials – typically $10-30M saving each time
Forecasting
• Alcon save $600M in inventory reduction
33
Copyright © 2010, SAS Institute Inc. All rights reserved.
The Options for the European
Pharmaceutical Industry
34
Copyright © 2010, SAS Institute Inc. All rights reserved.
Copyright © 2010 SAS Institute Inc. All rights reserved.