QuintilesIMS RWE Data Management Training Class TitleBroad, Nationally Relevant Databases (e.g., IMS...

Copyright © 2017 IQVIA. All rights reserved.

Saturday 15 June 2019, 13:00-16:30

Maritim Hotel Cologne

Massoud Toussi, MD, PhD, MBA

Senior Principal, Pharmacoepidemiology and Drug Safety Lead, IQVIA

Chair, Regulatory Interactions and Conditional Coverage IG, HTAi

HTAi Workshop

A Primer on RICC:

How to Generate Real

World Evidence? A

Modern Approach

1HTAi RICC IG Workshop - Massoud Toussi - 2019

A paradigm change is under way!

Diseaseorientedcare

EBMClinical trials

Patient oriented care

• RWE

• Observationalstudies

2

Increasing focus on RWE is associated with the greater supply of electronic patient-level data.

HTAi RICC IG Workshop - Massoud Toussi - 2019

Cumulative publications; some of the research output is not related to medicines

Source: PubMed

0

500

1 000

1 500

04012000 02 03 0605 07 2010 20121108 09

UK Germany USACanadaFrance

3

The volume of « unused » data is massively growing.


Source: IDC Digitila Universe 2020 Study

RWD supply push causes a seismic shift in our approach.

Few evaluators at launch, mostly

regulators and large payers

Many groups over time

including clinical and

small payers

THE PASTTHE PRESENT

RCT

Few

Many

Efficacy and Safety Almost everything

Initial view of

benefit-risk

Insights on environment,

outcomes, costs, comparative

effectiveness

Controlled trials,

manufacturer led

Shift to secondary patient-level

data across sources

RCT and RWE


5

Social media

Test results,

lab values,

pathology results

Consumer

data

Claims

databases

Meaningful

questions

Fit for purpose

data

REAL-WORLD EVIDENCE (RWE)

Data sources

Primary data collection

(excluding RCTs)

Mortality,

other registries

Hospital visits, service

details

Pharmacy

data

Electronic medical and health

records

Generating evidence from real world data (RWD)

Appropriate

Analyses


6

Step 1: Asking the right questions

Regulatory HTA

Exposure

Epidemiology of the indication(s)

Prescribing conditions

Characteristics of patients who actually receive the drug

New safety concerns, known ones, risk factors

Efficacy in real life / in specific populations

Effectiveness of risk minimization measures

Signal detection

Burden of target disease (mortality, morbidity

prevalence, incidence, DALYs, QALYs)

Conditions of use

Expected benefit of the technology

- On burden of disease

- On management of disease

- Economical

- Organisational

- Social

Confirmation of the expected benefit versus risk

Potential to cover unmet medical needs or to improve

covered needs


7

Step 2: Finding fit-for-purpose data

Fit-for-purpose

Relevance (appropriate data)

Accuracy (lesserrors)

Timeliness (data stillusefull)

Linkability(identifying fields)

Completeness (lessmissing records /

variables)

Accessibility(available easily)

Representativeness


8

Cost effective approach to data collection

The data shall be collected entirely (primary)

The data shall be collected partially (enriched, mosaic)

The data resides in a database (secondary)

We don’t know if the data is available in a database (landscaping)

Begin with the research question.

Rele

vance

Co

st

Tim

e

Desig

n fre

edom


9

Claims data

Initial purpose

• Economic management

• Reimbursement

Content

• Demographics

• Diagnoses

• Diagnostic related groups

• Procedures

• Reimbursed drugs/devices

Settings

• Mainly Hospitals

• Increasingly linked to outpatient claims

Good for

• Economic and resource utilization

• Epidemiology

• Healthcare system

0

0,5

1

1,5

2

2,5

3

3,5

4

Accuracy

Timeliness

Linkability

Completeness

Accessibility

Representativeness

0

0,5

1

1,5

2

2,5

3

3,5

Accuracy

Timeliness

Linkability

Completeness

Accessibility

Representativeness

Claims


10

EMR data

0

0,5

1

1,5

2

2,5

3

3,5

4

Accuracy

Timeliness

Linkability

Completeness

Accessibility

Representativeness

Initial purpose

• Clinical management

• Patient follow up

Content

• Demographics

• Diagnoses

• Signs and symptoms, allergies, smoking

• Lab values

• Drugs and to a less extent procedures

Settings

• Mainly primary care

• Increasingly secondary care and hospitals

Good for

• Exposure evaluation

• Drug utilization

• Disease epidemiology

• Benefit-risk assessment

• Unmet needs, burden, adherence

0

0,5

1

1,5

2

2,5

3

3,5

4

Accuracy

Timeliness

Linkability

Completeness

Accessibility

Representativeness

Claims EMR


11

Pharmacy records / sales data

Initial purpose

• Sales management

• Benchmarking

Content

• Demographics

• Drugs (packages sold)

Settings

• Retail pharmacies

• Wholesales

• Company outputs

Good for

• Exposure

• Treatment dynamics (switch, discontinuation…)

• Population movements

• Linkage

0

0,5

1

1,5

2

2,5

3

3,5

4

Accuracy

Timeliness

Linkability

Completeness

Accessibility

Representativeness

0

0,5

1

1,5

2

2,5

3

3,5

4

Accuracy

Timeliness

Linkability

Completeness

Accessibility

Representativeness

Claims EMR Pharmacy/sales


12

Registries

Initial purpose

• Research

Content

• Demographics

• Clinical details

• Procedures

• Drugs

• Lab values

• Relevant markers, genetic data, tests, etc

Settings

• Disease or drug oriented

• Mostly secondary care

• Mostly research intensive areas (oncology, ...)

Good for


• Benefit-risk assessment

• Treatment pathways

0

0,5

1

1,5

2

2,5

3

3,5

4

Accuracy

Timeliness

Linkability

Completeness

Accessibility

Representativeness

0

0,5

1

1,5

2

2,5

3

3,5

4

Accuracy

Timeliness

Linkability

Completeness

Accessibility

Representativeness

Claims EMR Pharmacy/sales Registries


13

Social media, wearables, connected devices, etc

Initial purpose

• Networking

• Follow up

• Experimental

Content

• Demographics

• Narrow and specific data

Settings

• Everyday life

• Smoothly entering the healthcare system

• Telemedicine programs

Good for

• Hypothesis generation

• Signal detection / monitoring

• Population behavior

• Public health intervention evaluation

0

0,5

1

1,5

2

2,5

3

3,5

4

Accuracy

Timeliness

Linkability

Completeness

Accessibility

Representativeness

0

0,5

1

1,5

2

2,5

3

3,5

4

Accuracy

Timeliness

Linkability

Completeness

Accessibility

Representativeness

Claims EMR Pharmacy/sales Registries Social media


14

Syndicated surveys

Initial purpose

• Market Research

Content

• Demographics

• Clinical data

• Procedures

• Labs & specialized tests

• Drugs

• Outcome measures

Settings

• Inpatient & outpatient

• Focused on pathologies with high demand for data

Good for


• Treatment dynamics (split per indication, off label use,…)

• Clinician behavior and understanding

0

0,5

1

1,5

2

2,5

3

3,5

4

Accuracy

Timeliness

Linkability

Completeness

Accessibility

Representativeness

0

0,5

1

1,5

2

2,5

3

3,5

4

Accuracy

Timeliness

Linkability

Completeness

Accessibility

Representativeness

Claims EMR Pharmacy/sales Registries Social media Syndicated survey


15

Linking different types of secondary data may be needed.

Deep, flexible

therapy area specific data including

primary data collection

Broad

Deep

Population coverage

Ra

ng

e o

f d

ata

typ

es

National claims, dispensing data

Electronic Medical Records

Labs, registries, biobanks

T-Shaped model for

better efficiency in

database studies


16

Step 3: Conducting the appropriate analyses

BETTER TRADITIONAL

STUDIES

Collect data from clinicians

and/or directly from patients;

combine with existing data for

broader stakeholder value

Reusable, scalable approaches

to evidence generation

Improved execution of

traditional studies, more precise

selection of sites, reduced

timelines and errors

INNOVATIVE STUDY DESIGN SMARTER EVIDENCE

GENERATION


17

Data and technology make innovative designs possible

Primary data

collection

Secondary data

collection

Copyright © 2017 IQVIA. All rights reserved.

Pragmatic

clinical trialsProspective

research (registries,

RCTs)

Mosaic

studies &

Enriched

studies

Site less

studies

Database

studiesExtension

studies with

direct to

patient

follow-up

Evidence

platformsAugmentation

studies

Tra

dit

ion

al

New


18

What are Mosaic studies?

Database analysis

Prospective study

Enriched EMR based panel

driven approach

Enriched claims based

approach

Case Study: A Global PASS

Challenge:

• Client wanted a cost effective and innovative solution for

a PASS in US and EU (~5000 patients globally over 10

year time-frame)

Enriched opportunity:

• Bring together a segmented solution for the best-fit

design for the US and each EU market, identified by the

use of secondary data for feasibility and planning

• Use data from claims + primary site network to identify

optimal US sites for recruitment & data collection

• Utilize network of registries to collect, analyse and pool

relevant safety information

Enriched value points:

• Huge cost efficiencies (~$25M) through avoidance of

unnecessary data collection in certain markets

• Early indication of improved site and patient recruitment

timelines using IQVIA data–driven approach

Mosaic studies identify the best-fit data in each country

• Due to the differences in secondary data availability across

countries, one method for data collection cannot be always

used in all countries

• Mosaic studies use multiple data collection approaches within

a single study - countries are grouped according to the method

for data collection – to provide an optimised study design to

the client.


19

Enriched studies combine primary and secondary data

STUDY PLANNING RECRUITMENT STUDY EXECUTION STUDY

DATABASE

• Linkage and de-ID patient

information

• Final study database

linking all data sources

EMR “backbone”• Aids patient

recruitment

• Provides core

patient informationEMR data Other datasets

(e.g. claims)

E-CRF and

PRO provide

supplementary data

on variables not in

EMR, including QoL Data collected directly

from MD (e-CRF)

Patient reported

outcomes (PRO)


20

• Evidence platforms are built on a foundation of real world data that supports clinical and commercial needs.

• It embeds a layer of technology to extract and analyze the data in a consistent way across the organization, with

appropriate governance and privacy protections.

• Applications designed to help teams use those insights appropriately for their needs.

Evidence platforms: the future of evidence generation

R&D

Predictive

Analytics Tools

Platform

Core

Regulatory

Studies /

Publications

HTA / HEOR

Forecasting & KOL

Engagement

Commercial

Patient Analytics

Software Application

(e.g., E360™, SAS, R, Tableau

• Data integration

• Data Hosting

• Business Rules & Analytical Methods

• Ongoing Data Governance

Broad, Nationally Relevant Databases (e.g., IMS RWD Claims Data)

Augmented Datasets (e.g., Oncology EMR)

Deep

Clinical Data

(e.g., Genomics

Data)

Enriched Data

(e.g., NLP,

Linkage, Local

Registry)

T-Shaped portfolio

of Data Sources

Services &

Engagement

Technology-

Enabled

Analytics

Real-World

Data


21

Trends in real-world data, evidence, and insights

Expanding

application of

RWD in

clinical

development

Increasing use

and acceptance

of innovative

study designs to

generate RWE

Scalable

approaches to

generate real

world insights

(RWI)


Thank you!

Massoud Toussi, MD, PhD, MBA

Senior principal, Pharmacoepidemiology and

Drug Safety Lead, IQVIA

[email protected]

Date post:	31-May-2020
Category:	Documents
Upload:	others
View:	1 times
Download:	0 times

QuintilesIMS RWE Data Management Training Class TitleBroad, Nationally Relevant Databases (e.g., IMS...

Documents