electronic health records for clinical research 1 the 9 th nri-conference in promoting health...

Electronic Health Records for Clinical Research 1

THE 9TH NRI-CONFERENCE IN PROMOTING HEALTH RESEARCH AND INNOVATION

MAY 13TH 2014 BERGEN, NORWAYMats Sundgren

AstraZeneca

Innovation in Re-using EHRs for Clinical Trials The Case of the EHR4CR Project


Objectives

Setting the Scene: the Opportunities for Efficient Clinical Research

Overview of the EHR4CR Project

The EHR4CR Platform & Services

Governance & the EHR4CR Institute

Concluding Remarks


SETTING THE SCENE: THE OPPORTUNITIES FOR EFFICIENT CLINICAL RESEARCH


This need has driven Life Science innovation

A large number of medicines are in development in order to…

leverage new science

expand treatment options

improve quality of life

provide value for money

Sequencing of the Human Genome

personalised medicine

Source: PhRMA 2012 Profile of the Pharmaceutical Industry

Medicines in Development in 2012


Parallel industry-centric growth in ICT

The inefficiencies become obvious at the clinical trial interface

Physician/ Investigator

57% of R&D investment is within Clinical Development1

In some countries nearly 90% of all

healthcare records are digital

Patienthealth recordsClinical trial

research data

Electronic data capture

of Clinical Trial data

Patient Care Data

Over 40% of clinical trial

data are entered into

health record and EDC1

1. Integrating Electronic Health Records and Clinical Trials: An Examination of Pragmatic Issues, Michael Kahn, University of Colorado.


The clinical trial journey today is fragmented with many hurdles

TRIAL DESIGN

FEASIBILITY/SITE SELECTION

PATIENT RECRUITMENT

Protocol design Early feasibility Site input Country and early

site selection

Detailed feasibility Site selection Contracting Site training

Patient recruitment/ screening

Consent Randomisation

First patient, first visit

Visits/follow-up Safety Reporting Data management Last patient enter

treatment End of study

TRIAL EXECUTION

Data lock End of study report Study publication Study outcome

REPORTING

Key data for Pharma: Previous trial performance in this area Available patients with inclusion/exclusion

criteria listed Ability to access patients through referrals or

other means Evidence of good quality and operations

Data for Hospitals: Number of Eligible and accessible patients Knowledge of previous trial performance

Key data for Pharma: Screening/recruitment rate Numbers of screening dropouts Numbers of patients randomised

Data for Hospitals: Identification of eligible patients coming

through clinic & elsewhere Numbers that have consented

& randomised

Key data for Pharma: Country standard of care for disease area Ethics and local regulatory knowledge Previous trial performance in this area Operational/scientific expertise if area is new

Data for Hospitals: Upcoming studies that may fit Hospital patient population and expertise


There is a need to bridge the gap

We have imagined an environment where de-identified patient data can be re-used within healthcare and research for clinical research purposes…

Across countries

Across systems

Across sites

…to speed up protocol design, patient recruitment, data capture, safety reporting…

Patienthealth records

De-identified data for Clinical

Research


This can create value for hospitals

Better patient careImproved route to inclusion in clinical trials. Enhances treatment options, giving patients access to trial drugs and care pathways with no cost to the Trust

Improved clinical researchImproved efficiencies and interconnectivity with other hospitals facilitates, streamlines and enriches clinical research

Income streamBetter placed to generate income from clinical research.At a time of squeezed budgets, income from research can help drive innovation and efficiency with better outcomes for patients

Better quality EHR dataImproved monitoring, performance benchmarking, reporting and management (e.g. reimbursement coding) Drives optimisation of patient care and improved efficiencies

Enhanced reputation Greater visibility of hospital/clinicians in scientific community. Improved ability to participate in/conduct clinical trials


But a win for all stakeholders is critical

Pharma, academia, CROs

Clinical trial development will

become more efficient by reducing the time it

takes to bring new drugs to market, thus generating substantial

value

HospitalsAble to participate in

more clinical research programmes,

benefiting their patients

Health authorities

Access to new and better evidence to underpin health

policy, strategy and resource planning

Health community/

governmentsAble to offer improved quality of healthcare

with reduced healthcare costs

EUMore attractive for R&D investment

PatientsFaster access to safe

and effective medicines, improving

health outcomes across Europe


Public Private Partnerships can achieve something that individual groups cannot realise alone

Political support

Engage multiple stakeholder communities

biopharmaceutical companies, patient organisations, academia, hospitals, small- and medium-sized enterprises (SMEs) and public authorities

Transfer of knowledge

Public deliverables

Consensus and synergy

Moving towards

deployment of a sustainable

ecosystemOvercoming barriers that

limit access to EHRs for research

Developing a platform and services to re-use EHR data

Offering a new paradigm for

clinical research in Europe

Supported by


THE EHR4CR PROJECT

11


A unique initiative

Mandated by IMI

One of the largest European public/private partnership projects in this area

4-year project (2011-2015)

Budget of € >16m For further information see www.ehr4cr.eu or

contactGeert Thienpont

(EuroRec) [email protected]


Brings together key stakeholders35 participants

including pharmaceutical

industry, academia , hospitals, small and

medium-sized enterprises, patient

associations and public authorities

11 hospital sites

Advisory boards and other experts


InteroperabilityEHRs generated by single institutions (the doctor has a set of information for each patient; if the patient goes to another doctor there is another set of information) Separate and disparate systems Incompatible EHR systems Different models Variable quality, uniformity and

organisation of the data Different coding and content standards Structured (e.g. prescriptions) versus

unstructured (e.g. clinical narrative) Different languages across Europe

Data security, privacy & ethics Complying with ethical, legal and privacy requirements that differ from country to country is critical to gain acceptance with the general public, patients and medical professionals

Scalability and sustainabilitySolutions need to be adaptable and reusable and governed within a sustainable ecosystem

To address key challenges to enable the re-use of EHR for clinical research

Confidence in data All data has to be complete and accurate at the point of capture. A single error presents risk that can be magnified as data transmits downstream


1. A self-sustaining economic model

2. A roadmap for pan-European adoption

A SET OF TOOLSAND SERVICES

Project deliverables

BUSINESS MODEL

A set of tools and services

Validated through pilots

Different therapeutic areas (e.g. oncology, neuroscience, diabetes, CVS…)

Several countries (under different legal frameworks)

1. Protocol Feasibility

2. Patient Identification and Recruitment

3. Clinical Trial Data Exchange

TECHNICAL PLATFORM


Protocol Feasibility

With no, or limited, access to actual patient data, trial design is based on discussions with expert clinicians

Increased amendments, slower than expected enrolment, costly changes to add new sites and countries, even failed trials

A third of protocol

amendments are avoidable1 ,

at a cost of $0.5m per

amendment.2

How long will the trial take?

Do the inclusion/ exclusion criteria

make sense?

Will we find sufficient

numbers of the right patients?

1. Drug Information Journal, Vol 45, 20112. Industry Standard Research, 2010

Protocol design based on estimates and not optimised


Patient Recruitment

With no searchable patient database, identifying and recruiting suitable patients and trial sites are principal causes of trial delays

Delayed trials increase the burden for sites, waste costly resources and slow access to new drugs

Almost

half of all trial

delays caused by patient recruitment problems2

1. State of the Clinical Trials Industry: A Sourcebook of Charts and Statistics, Center Watch, 2008.2. Study Participant Recruitment and Retention in Clinical Trials: Emerging strategies in Europe, the US and Asia, Business Insights, June 2007.3. Beasley, “Recruiting” 20084. Tufts -http://clinicalperformancepartners.com/wp-content/uploads/2012/07/Fixing-Feasibility-Final-Jan-2012.pdf

Each day a drug is delayed from market, sponsors lose up to

$8m3

The percentage of studies that complete enrolment on time:

18% in Europe,

7% in the US1

A major cause of trial delay


Model: a free market ecosystem

An organisation that uses EHR4CR services such as a pharmaceutical company or an academic institution

An organisation that provides EHR4CR services to Service Users

An organisation that contributes data for EHR4CR e.g. hospital

Service User

Service Provider

Data Provider

EHR4CR INSTITUTE

Data aggregation and access services

through standard EHR4CR interfaces

Applications access and deliver

EHR4CR services

DATA PROVIDER

SERVICE PROVIDER

APPLICATION PROVIDERS


SERVICE USER

SERVICE USER

SERVICE USER

NETWORK PROVIDER

DATA PROVIDER

SERVICE PROVIDER



SERVICE PROVIDER



SERVICE PROVIDER



UK DATA ONCOLOGY DATA

PROTOCOL DESIGN &

FEASIBILITY

1 2 3

NETWORK PROVIDERINTERCONNECT


THE EHR4CR PLATFORM AND SERVICES


The EHR4CR platform – an open architecture

Infrastructure

Semantics

(end-user) Tools

Hospital Connectors

Open platform Service Oriented Architecture

Standards based

Maximal service decoupling

Objectives Avoid vendor lock-in

Stimulate alternative tool development

Open to different semantic interoperability

approaches

Create added-value by enabling service re-use

beyond EHR4CR


EHR4CR Overview – data flow

RESEARCH CENTRE e.g. pharmaceutical company

EHR4CR PLATFORM HOSPITAL/DATA PROVIDER

QU

ER

Y W

OR

KB

EN

CH

RECRUITMENT PROGRESS

RECRUITMENT PROGRESSRECRUITMENT

PROGRESSRECRUITMENT PROGRESS

RECRUITMENT

PROGRESS

RECRUITMENT

PROGRESS

#COUNT

EHR4CRSERVICES

RECRUITMENT

PROGRESS

RECRUITMENT

PROGRESS

QUERY

RECRUITMENT

PROGRESS

RECRUITMENT

PROGRESS

#COUNT

RECRUITMENT

PROGRESS

RECRUITMENT

PROGRESS

QUERYSTUDY CRITERIA

RECRUITMENT WORKBENCH

CLINICAL DATA WAREHOUSE EHR

ELECTRONIC CASE REPORT FORM

EXTRACT TRANSFORM

LOAD

QUERY CANDIDATE LIST

FEASIBILITY

RECRUITMENT

QUERIES AND RESULTS ACROSS MULTIPLE SITES


Security & Privacy

EHR4CR Security Framework

Advanced SOA security framework (cross-organisation Web Services)

Key points

All operations fully (end-user) authenticated through credential delegation

Seamless SSO between websites & services

Central Identity & Access Management (IAM)

Enforces platform-wide policies (e.g. contractual agreements)

Fully complementary to local access control (federates)

Unified platform audit trail

SSO Access Control

Audit


Protocol Feasibility Service pilot

Tested viability and performance of EHR4CR platform to support protocol feasibility service

o 11 major hospitals in five countries

o EHR4CR-compliant data warehouses were established

at all pilot sites

o Large set of eligibility criteria from EFPIA trials analysed

to identify commonly used data elements (75

EHR data elements)o De-identified data from >five million

patients was loaded for these elements into

local EHR4CR-compliant data warehouses as far as

available at the sites

o 12 clinical studies evaluated, technical

testing of four clinical studies

Germany (WWU, FAU) France (AP-HP, U936) UK (UoD, UoG, UoM, UCL, KCL) Switzerland (HUG) Poland (MuW)


Protocol Feasibility

Select sites of interest Launch queries Analyse results

Accept Execute

Edit Eligibility Criteria


User acceptance and functionality testing

A subset of the previously selected 12 studies were used for two User Acceptance Test pilots

October 2012: Initial pilot with four studies, test scripts with 52 steps

January 2013: Extended pilot with 10 studies

User functionality testing

October 2013: Three independent testers, validation of patient counts


End user view of the application:Protocol Feasibility Service

query workbench


End user view of the application:Results


Protocol Feasibility Service pilot outcome

Conclusion of defined pilot success criteria:

Retrieving information from hospital sites:

Timely response but endpoints without data halt query execution

User functionality testing: patient counts validated for test dataset:

EHR chart review to validate precision of patient counts by EHR4CR platform (currently ongoing)

Query modification and re-running of queries:

Transnational platform across systems and hospitals:


Validated solutions

Developed pilots to validate the solutions:

For different scenarios (e.g. protocol feasibility)

Across different therapeutic areas (oncology, inflammatory diseases, neuroscience, diabetes, cardiovascular diseases, respiratory diseases)

Across several countries (under different legal frameworks)


GOVERNANCE AND THE EHR4CR INSTITUTE


The EHR4CR Institute

Critical to the sustainability of EHR4CR

Provides environment for EHR4CR ecosystem to develop

Not-for-profit, formally registered company providing services to registered members (e.g. data providers, data users, service providers)

Funded by license fee, subscriptions, certification, membership (cover cost of operations and providing services to registered members)

Incomes reinvested to improve services & fund public interest research

EHR4CR Institute

Specifications and standards

Promotion of EHR4CR

services

Accreditation and

Certification

Oversight, governance, auditing

Guardian of shared

Intellectual Property

Build EHR4CR

community and best

practices


Develop resources

Software Requirements Specifications

EHR4CR projectEHR4CR Institute

Software components: platform and

tools

Governance instruments

Maintain specifications and standards

Maintain conformance

criteria, testing tools

Maintain a reference

implementation

Publish policies and

Standard Operating

Rules


EHR4CR SERVICE PROVIDER


SERVICE USER NETWORK

PROVIDER

SERVICE USER

OTHER DATA AGGREGATORS

Govern the ecosystem

DATA PROVIDER

RESEARCH SPONSOR

RESEARCH SPONSOR

DATA PROVIDER

Educate and train research and ICT

staff

ICT SOLUTION PROVIDERS

Accredit staff and

organisations

Certify service providers

and EHR systems

Oversee and audit governance &

security

RESEARCH ORGANISATIONS

The EHR4CR Institute will partner EuroRec, ECRIN and UKCHIP to:



Privacy protection delivered on multiple levels

Personal data only processed by original data controller (treating physician)

EHR4CR only handles aggregate data with additional protections

Clinical Data Warehouse holds only pseudonymous data

Role-based access controls to limit access to aggregate data

Extensive audit trails with reporting

Integration of Operating Procedures with system functionality


CONCLUDING REMARKS


The EHR4CR project is an important initiative

Bringing together multiple stakeholders

Overcoming barriers that limit access to EHRs for research

Developing a platform and services for trustworthy re-use of EHR data

Patienthealth records

Clinicalresearcher

De-identified data for Clinical

Research


Within an environment for trustworthy re-use

Segregationof EHR4CR data from EHR

De-identification individual patient anonymous

Controllock/unlock access by hospital

Consolidation only aggregated patient numbers leave the hospital

Governanceindependent institute ensures data are accessed in a trustworthy way

EHR4CR Institute

n= 16Patient

health recordsDe-identified

data for Clinical Research


THANK YOU


Video links

Platform demo

http://devworkbench-ehr4cr.custodix.com

EHR4CR Institute

http://youtu.be/fcRoy8jlu4A

http://devworkbench-ehr4cr.custodix.com/

http://youtu.be/fcRoy8jlu4A

electronic health records for clinical research 1 the 9 th nri-conference in promoting health...

Documents

health research

clinical development

electronic health records

clinical trials

clinical trial journey

new data

operations data

randomised key data