emif_pharmaceutical it conference london_24 september 2013

EMIF European Medical Information Framework

Bart Vannieuwenhuyse 24 September 2013

Pharmaceutical IT Conference London

10-Mar-14 2

The IMI is a unique Public-Private Partnership (PPP) between the pharmaceutical industry represented by the European Federation of

Pharmaceutical Industries and Associations (EFPIA) and the European Union represented by the European Commission.

Pharmaceutical IT Conference 2013

Imagine a world where…

10-Mar-14 3

Researchers can access available patient-level data to gain new insights around disease etiologies and define new treatment targets…

We can optimize Clinical R&D by 50% using existing patient level data…

New treatments are effective in 95% of patients…

You can reduce treatment failure by 60% with timely prediction and intervention…

Where we can continuously monitor the risk/benefit profile of new therapies using naturalistic data…

All health care providers have access to up-to-date information on existing and new treatments… Pharmaceutical IT Conference 2013

Framework for RWE

10-Mar-14 4

Clinical Care Clinical Research


Challenges with re-use of patient level data

10-Mar-14 5

Data gaps

• Missing data elements (e.g. outcomes)

• Studies require details that may not be routinely collected

• Coding often only at first level (e.g. ICD-9) therefore missing granularity

• 80% of info stored as unstructured data

Data quality

• Longitudinal coherence • Coding for administrative reasons (up

– down coding) • Coding often months after patient

encounter • Data provenance – who entered the

data?

“Semantics”

• Many standards – many versions • Complex care – many HCP’s involved

– many hand-overs • Need to pool data cross sites and

cross different countries • Pharma focused on CDISC

Privacy

• Clearly a top priority • Different interpretations by country, by

region – complex • TRUST


Successful example of data re-use for research

RWE of ACHeI > 2500 patient years of therapy

documented in EHR-system > 8 fold dataset compared to

Cochrane

Cost effective Text mining derivation of service

utilisation and costs. Created in one month

Resembles cognitive decline curve derived from clinical trials

Cholinesterase inhibitors and Alzheimer’s disease


-1 0 1 2 3

1618

2022

24

time(years)

MM

SE

sco

re

EMIF Vision

10-Mar-14 7

To be the trusted European hub for health care data intelligence

enabling new insights into diseases and treatments


Project objectives EMIF: one project – three topics 1. EMIF-Platform: Develop a framework for evaluating,

enhancing and providing access to human health data across Europe, to support the two specific topics below as well as research using human health data in general

2. EMIF-Metabolic: Identify predictors of metabolic complications in obesity, with the support of EMIF-Platform

3. EMIF-AD: Identify predictors of Alzheimer’s Disease (AD) in the pre-clinical and prodromal phase, with the support of EMIF-Platform

10-Mar-14 8 Pharmaceutical IT Conference 2013

EMIF – platform for modular extension

10-Mar-14 9

EMIF

- M

etab

olic

EMIF

- AD

Data Privacy

Analytical tools

Semantic Integration

Information standards

Data access / mgmt

IMI Structure and Network

Research Topics

EMIF governance

Prev

entio

n al

gorit

hms

Pred

ictiv

e sc

reen

ing

Ris

k st

ratif

icat

ion

Call x Call x

Ris

k fa

ctor

ana

lysi

s

Patie

nt g

ener

ated

dat

a

TBD

EMIF

- Pl

atfo

rm

Metabolic CNS


Data types in platform: Identified categories or “types” of data Types:

– Primary care data sets – Hospital data – Administrative – Regional record-linkage systems – Disease-specific registries – Biobanks

Selected representative examples – Combined around 48MM subjects from 7 EU countries

And of course the cohorts from the Research Topics


Data bases “typology”


EMIF systems view

10-Mar-14 12

User admin

User admin

Remote user 1

Remote user 2

Site 1

Site 2

Data access Module

extract

Data access Module

extract Com

mon

ont

olog

y / D

e-id

entif

icat

ion

EMIF platform solution

Governance

Data owners Researchers


EMIF roadmap


EMIF catalogue - prototype

10-Mar-14 14 [Meeting name]

Re-using existing solutions


Data being loaded in tranSMART

10-Mar-14 16

4 times the number of subjects in ADNI

Data available for analysis end Oct 2013 Data structure aligned with CDISC

AD MCI Subjective Memory impairment CTL/Controls Total

Cohort 1 259 225 232 716

Cohort 2 160 100 50 80 390

ADNI 1 188 405 229 822

Cohort 3 881 881

Cohort 4 20 14 42 76

Cohort 5 420 420

Total 607 1631 64 1003 3305


Framework for RWE

10-Mar-14 17



Framework for RWE

3/10/2014 18

System biology

Biomarker definition

Lead identification

Clinical trial Execution

Market Access

Ongoing safety tracking

incident monitoring & detection

retrieval of similar patient history

outcome analysis

care management patients at risk

re-admission prevention

diagnosis & treatment assistance



More info

EMIF general – Bart Vannieuwenhuyse ([email protected]) – Simon Lovestone ([email protected]) – Johan van der Lei ([email protected])

EMIF-Platform – Johan van der Lei ([email protected]) – Patrick Genyn ([email protected])

EMIF-Metabolics – Ulf Smith ([email protected]) – Dawn Waterworth

([email protected])

EMIF-AD – Pieter Jelle Visser

([email protected]) – Johannes Streffer ([email protected])


www.emif.eu

emif_pharmaceutical it conference london_24 september 2013

Health & Medicine

data structure

naturalistic data

data elements

data gaps

types of data types

human health data

adni data available

data cross sites