IMS AccessPoint 6 - May 2013

IMS REAL-WORLD EVIDENCE SOLUTIONS AND HEALTH ECONOMICS & OUTCOMES RESEARCH

VOLUME 3, ISSUE 6 MAY 2013

AccessPoint News, views and insights from leading experts in RWE and HEOR

RWE market impact on medicines: A lens for pharma

Where next for RWE? Demystifying propensity scoring

Focus is key to unlocking the value of real-world, evidence-based data

The path to value-based healthcare in Latin America

Stefan Plantรถr reveals the critical patient endpoints PageOUTCOMES 1 - Issue 1 for AMNOG Page 50

Oncology calls for Massoud Toussi convincing evidence explains how of comparative creative coding 1 IMS HEALTH ECONOMICS AND OUTCOMES RESEARCH effectiveness says doublesPage available Karin Berger patient-level data Page 24 Page 30

VOLUME 3, ISSUE 6 MAY 2013

AccessPoint News, views and insights from leading experts in RWE and HEOR An RWE lens for pharma Unique IMS research brings actionable insights into RWE use and market impact around the world. page 12

New innovations in outcomes research How information technology is transforming the future of real-world data. page 18

Study designs in observational research Increased scrutiny demands rigorous methodology. page 40

All change for medical devices in Germany New regulations bring new potential but what do early insights show? page 56

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

“The trends demand a very different mindset centered on better managing real-world data and focusing its use to maximize the value.”

Welcome

Contents

Welcome to the latest issue of AccessPoint – a chance to take stock of the evolving landscape and set a course to meet new needs in a market that is more than ever reliant on our community for success.

NEWS

In 2013, a noteworthy new direction is emerging as we observe several pharma companies separating HEOR into two complementary functional groups. While structures differ, there are two general themes: HTA excellence – through accelerated scale and focus on HTA support, including modeling and submissions; and real-world evidence (RWE) excellence – by transforming evidence development into a coordinated, at-scale and highly-skilled lifecycle process. This is just the latest in a two-decade search to balance the specialist functional, geographic and technical needs for generating and communicating customer-centric evidence. These activities clearly remain interdependent but the signs suggest an enduring organizational trend.

2 4 6 7

IMS SUPPORTS HTA IN ASIA PACIFIC DATA LINKAGE ACCELERATES RESEARCH IMS INSTITUTE ANALYZES US CARE TRENDS PARTNERSHIP COMPLETES PATIENT PATH

INSIGHTS 8 12 18

RWE PERSPECTIVE Why focus is key RWE MARKET IMPACT International lens for pharma IMS SYMPOSIUM New age for outcomes research ONCOLOGY REAL-WORLD DATA Identifying valid sources in Europe NATURAL LANGUAGE PROCESSING Optimizing EMR value PROPENSITY SCORING Enabling trust in RWE ObSERVATIONAL STUDY DESIGNS Ensuring credible research HTA IN LATIN AMERICA Fulfilling evidence demands AMNOG ENDPOINTS Proving patient relevance MEDICAL DEVICES IN GERMANY Understanding new potential

We are also seeing the real market impact of RWE and its influence on decisions about the use of medicines (page 12), as well as the challenges of implementing systems for HTA (page 44) and addressing the growing demands for RWE (page 24). At the same time, tremendous strides are being made in the scope and depth of available real-world information.

Sophisticated database linkage and integration (page 4), fuelled by new applications of information technology (page 18), has enabled unprecedented volumes of data for healthcare decision making, just as novel techniques are extending its quality and utility for outcomes research (page 30).

24

These trends demand a very different mindset - one that moves beyond more and deeper data towards better managing that data and focusing its use to maximize the value. In building a foundation for future growth, companies need new ways to collect and organize information, and tools to help them turn this into a relevant body of evidence.

50

With over 230 multi-disciplinary experts focused on RWE solutions and HEOR globally, IMS is now, more than ever, finding innovative ways to help our clients succeed in this changing environment: we have centered our approaches to data delivery on new client needs, leveraging the largest collection of patient-level data and data-agnostic sourcing and acquisition to provide customized therapy area views; we have invested in technology to create true platforms that integrate data into common, linked data models; we have designed novel technology solutions to help clients build, interrogate and analyze cohorts and visualize the results; and we continue to expand our expertise and capabilities as partners in research that creates the strongest evidence base for strategic decisions, value demonstration and engagement with a growing range of stakeholders.

PROJECT FOCUS

I hope you will enjoy this issue of AccessPoint.

Jon Resnick Vice President and General Manager Real-World Evidence Solutions IMS Health Jresnick@imshealth.com

ACCESSPOINT • VOLUME 3 ISSUE 6

30 36 40 44

56

60 CHRONIC HEPATITIS b Demonstrating cost-effectiveness in Italy 63 RHEUMATOID ARTHRITIS Treatment pathway decision support

IMS RWES & HEOR OVERVIEW 66 ENAbLING YOUR REAL-WORLD SUCCESS Solutions, locations and expertise

AccessPoint is published twice yearly by the IMS Real-World Evidence (RWE) Solutions and Health Economics & Outcomes Research (HEOR) team.

VOLUME 3, ISSUE 6. PUbLISHED MAY 2013.

IMS HEALTH 210 Pentonville Road, London N1 9JY, UK Tel: +44 (0) 20 3075 4800 • www.imshealth.com/rwe RWEinfo@imshealth.com

©2013 IMS Health Incorporated and its affiliates.

All rights reserved.

Trademarks are registered in the United States and in various other countries.

PAGE 1

NEWS | HTA IN ASIA PACIFIC International experts bring new momentum to China’s move towards evidence-based medicine at 3rd CORE Summit in Shanghai.

IMS AsiaPac supports HTA in China at cornerstone regional meeting Internationally, healthcare community interest in real-world outcomes research and integrated healthcare management has grown rapidly as an effective measure of the clinical- and costeffectiveness of medical interventions in large patient population settings.

Specifically in China, the following trends are observed: Clinical experts are demanding additional data on comparative effectiveness and patient-reported outcomes Social health insurance agencies are increasingly requesting evidence of treatment pathways as part of the reimbursement process China Ministry of Health and SFDA are strengthening post-market surveillance of medical products Healthcare professionals across China have accepted the concept of evidence-based medicine and are increasingly demanding data on burden of disease and product use in the Chinese patient population.

• • • •

STIMULATING EVIDENCE-BASED MEDICINE IN CHINA As the only event dedicated to outcomes research, health technology assessment (HTA), value-based healthcare management and eHealth in China, CORE (China Outcomes Research & Evidence-based Medicine) Summit has a key role to play in stimulating and reinforcing evidence-based medicine in this country.

INTERNATIONAL CONTRIBUTIONS As one of the main event sponsors, IMS contributed to a preconference workshop on ‘Challenges in Retrospective Database Acquisition, Analysis & Reporting’, as well as giving a plenary talk on ‘Real-World Evidence: The Next Step’. This is the second year running that IMS has been invited to speak at the meeting.

Day 1: Clinical outcomes measurement and outcomes research Key topics covered during the first day were data quality; methodological issues and best practice; incentivization for both patients and clinicians/researchers; and stakeholder engagement and accountability.

Methodology and key findings were presented from studies and registries around the world, including the National Cardiovascular Data Registry (NCDR) run by the American College of Cardiology, and China’s own Cardiometabolic Registry (CCMR), supported by the Chinese Ministry of Health.

Dr Carolyn Clancy, Director of the US Agency for Healthcare Research and Quality (AHRQ), described the Patient Centered Outcomes Research Initiative (PCORI) and how putting the patient at the center of chronic disease management could be the blockbuster drug of the 21st century – with the right education and engagement.

IMS contributed to a workshop at the 3rd CORE Summit in Shanghai

Organized jointly by China Medical Doctor Association – which has a reach of 2 million licensed medical physicians in China – and Vital Strategic Research Institute, CORE serves as a critical platform for leaders from government, medical societies, academia and industry to share their vision, policies and knowledge in this area. In April 2013, more than 30 experts from around the world were invited to the 3rd CORE Summit in Shanghai to offer their perspective on the impact of real-world evidence (RWE), healthcare value definition, challenges and opportunities in chronic disease management, HTA, eHealth and big data.

PAGE 2

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

HTA IN ASIA PACIFIC | NEWS

Another distinguished speaker, Mr John brooks, President, CEO of Joslin Diabetes Center, Harvard Medical School, outlined the use of smart phone technology in providing information that empowers patients to make the right lifestyle choices for managing their diabetes.

Day 2: Application of outcomes research and evidence-based medicine The second day brought topics ranging from post-marketing surveillance in the regulatory context to the use of real-world data for health policy and HTA in China. Some overriding themes were the sheer size of China’s task in incorporating evidence-based decision making as part of its reform program, as well uncertainty around the best way of implementing HTA to the optimum benefit of the Chinese health system. It was suggested that scientific evidence and guidelines alone are not enough for efficient clinical practice, and that the importance of communication and dissemination of evidence should not be underestimated.

Highlights included a keynote presentation by Jianying Guo, Deputy Director Pricing, China National Development and Reform Commission, on the use of RWE as part of China’s pricing and reimbursement policy development, and insights from Professor Gordon Liu, Professor of Economics at Peking University, on developments in health economic research policy in China.

Joe Caputo from IMS AsiaPac overviewed current applications of RWE and the reasons why it has gained such global traction,

including examples from Asia. He also considered how big data, which will bring together electronic medical records (EMRs) with many other sources of medical, social and behavioral data, will require organizations to think differently about developing and implementing infrastructure and data management frameworks so that information is readily available to all stakeholders in the healthcare system.

CONTINUING THE MOMENTUM CORE has attracted increasing numbers of experts, speakers and delegates from within and outside China each year. This was another successful meeting with excellent input from leading experts showcasing examples of RWE generation/chronic disease management initiatives whilst addressing the practical issues and challenges of collecting and utilizing such data. Availability and quality of data in China may still be variable but it is clear that this country is ready to progress the evidence-based healthcare agenda.

That said, there are many practical challenges; by its very nature, the data reflect real-world practice and it would be naïve to expect the perfect dataset at this stage of the evolution. Rather, all stakeholders should be encouraged by the interest and enthusiasm around the methodology and use of such data to help improve healthcare quality and efficiency. based on the success of this and previous CORE meetings, the 2014 Summit will place more weight on the topics of eHealth, eClinical and the practice of using EMRs for measuring clinical outcomes, healthcare quality and patient engagement.

•

First Health Policy Decision Makers Forum Asia Pacific, in Singapore IMS sponsors independent platform for high-level dialogue to help disseminate innovative solutions in developed and developing countries.

In November, 2012, the Institute of Health Economics and Management (IHEM), ESSEC business School gathered experts and delegates from Asia, Europe and US to address the challenges of making good healthcare available to all in the rapidly evolving landscape of Asia Pacific. Topics included an assessment of needs, resource issues, affordability and funding options, case study reviews, and access to efficient treatments. IMS will also be supporting a forthcoming ESSEC executive training program in Singapore to help participants develop the knowledge and skills for customizing economic analyses to their countries.

For further information on IMS Health RWE Solutions & HEOR capabilities in Asia Pacific, please contact Joe Caputo at Jcaputo@sg.imshealth.com

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 3

NEWS | IMS PHARMETRICS PLUS™ Data linkage offers tremendous potential to study long-term outcomes of treatment choices in important diseases.

Integrated IMS PharMetrics Plus™ powers significant, wide-ranging new research opportunities IMS Health is working with leading life science organizations to build patient-centric integrated datasets in therapy areas such as oncology, rheumatoid arthritis, multiple sclerosis, diabetes and rare diseases.

The ability to link disparate data sources brings the power to generate richer and more detailed clinical information across the entire continuum of patient care.

Through its recent collaboration with Health Intelligence Company, operating as blue Health Intelligence, IMS biopharmaceutical clients have gained access to one of the largest US health plan claims databases, adding to the company’s market leading PharMetrics Plus™ database.

The aggregated dataset comprises adjudicated claims for more than 150 million unique enrollees across the United States. Enrollees with both medical and pharmacy coverage represent more than 42 million unique lives in 2011 and mirror the US age distribution of the US population under 65 years. In sample disease areas for 2011 this patient coverage includes: Hypertension: >4.5 million Migraine: >550,000 breast cancer: 250,000

• • •

Among key information enabled by PharMetrics Plus are greater patient segmentation, availability of 3-digit zip-code, patient out-of-pocket payment, hospital discharge status as well as additional inpatient and provider-level detail.

Data can be linked with all IMS US patient-level data assets, including elecronic medical records (EMR), laboratory, hospital charge master and consumer data as well as clients’ own data (eg, from registries or clinical trials) (Figure 1). We are currently approaching a total of 50 million lives available for linking with internal and external data, and this number is scheduled to grow.

1

EXPANDED RESEARCH POTENTIAL PharMetrics Plus paves the way for significantly expanded health economics and outcomes research opportunities including:

• • • • • • • •

burden of illness across settings of care Cost and resource utilization studies Medication adherence, persistence and compliance studies Comparative effectiveness analyses Patient treatment patterns and treatment flows Patient and provider segmentation Epidemiological prevalence and incidence analyses Pharmacovigilance and safety studies

The database is supported by a unique and proprietary algorithm for data linkage based on de-identification of patients, ensuring compliance with HIPAA1 regulations, as well as an algorithm for the creation of unique patient IDs and deterministic matching.

fIGURE 1: INTEGRATING PATIENT-LEVEL DATA WITH PHARMETRICS PLUS

Consumer OTC Medication

Hospital

Ambulatory EMR

Oncology/Specialty EMR

Lab Data

PharMetrics Plus 150m US total patient lives

Consumer Behavioral & Demographic Data

Open-Source Medical & Pharmacy Claims

Customer Data

Health Insurance Portability and Accountability Act

PAGE 4

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

IMS PHARMETRICS PLUS™ | NEWS

fIGURE 2: HOW DOES IMS LINK DATA AND MAINTAIN HIPAA COMPLIANCE? Hospital Ambulatory EMR Oncology/Specialty EMR HIPAA-compliant de-identification process

Lab Data Patient

Consumer OTC Medication

Patient

Consumer Behavioral and Demographic Data Open-Source Medical and Pharmacy Claims Customer Data

PATIENT

MULTIPLE CHANNELS

DE-IDENTIFICATION

PATIENT

A single patient interacts with the healthcare system across multiple entry points

Each interaction is captured at the de-identified patient level in the same manner across channels

All interactions are combined upon the common algorithm in place across channels

A singular view of the patient experience is represented

OVERVIEW Of IMS US PATIENT-LEVEL DATABASES

EXTENDED DATA LINKAGE OPPORTUNITIES beyond the data linkage possibilities within IMS, there is further potential for customers to link their own patient-level data – from clinical trials, prospective observational studies and disease registries – to IMS data assets, subject to patient privacy agreements. The data feed is processed from source by a trusted third-party contractor, leveraging a HIPAAcompliant de-identification process, to provide a singular view of the patient experience across multiple channels (Figure 2).

Patient-centric data warehouse IMS data can also be linked to external sources of patient-level data such as EMRs from specialty physician practices.

A patient-centric data warehouse has been created linking IMS US data to external claims and registries. The resulting platform contains more than 500,000 patient records with enhanced client value gained from links to death records, tumor registries and medical claims. Work on this platform has formed the basis of 30 publications in the 5 years since its inception. The knowledge gained from this process is being applied to the integration of all PharMetrics Plus datasets.

•

For further information about IMS PharMetrics Plus or for data inquiries, please contact Shibani Pokras at Spokras@us.imshealth.com

ACCESSPOINT • VOLUME 3 ISSUE 6

IMS Health US APLD data source

Linkable

Size*

Health Plan Claims

✓

Adjudicated claims for over 150M unique enrollees in the US. Enrollees with both medical and pharmacy coverage represent more than 42M unique lives in 2011

Medical Claims

✓

Over 150M patients, 1B claims and 3B service records obtained annually from more than 800,000 providers

Pharmacy Claims

✓

Over 120M patients, 1.6B prescriptions, capturing ~40% of all US prescriptions claims

Ambulatory EMRs Oncology EMRs Hospital Charge Data Master Laboratory Test Results Consumer Behavioral and Demographic Data Consumer Over-theCounter Medications

✓

✓

✓

✓

✓

✓

23M patients with data from 40,000 physicians 650,000 patients across 370 locations of care 113M encounters annually from more than 650 hospitals 40% of all outpatient laboratory tests conducted in the US Consumer information on ~130M patients, 1M physicians and 150,000 physician practices Data collected from ~40M households

*Size indicative of stand-alone dataset; matching across datasets varies. For profiling inquire at IMS Health.

PAGE 5

NEWS | IMS INSTITUTE INSIGHTS IMS Institute for Healthcare Informatics expands understanding with a meaningful perspective for optimized performance of medical care.

New studies analyze dynamic trends in key areas of US health system The IMS Institute for Healthcare Informatics continues to support its research agenda with insights to help improve the quality and cost of healthcare delivered.

Findings from three of its latest publications identify the dynamics driving change in the use of medicines, Managed Medicaid, and e-Prescribing prevalence.

DECLINING MEDICINE USE AND COSTS: FOR BETTER OR WORSE? This new review of the use of medicines in the US in 2012 shows a market in a state of flux. Trends are marked by a decline in drug spending to US$325.7 billion for the first time ever – suggesting that patients are self-rationing due to the bad economy – and explosive growth of generics reflecting double the level of patent expiries from 2011. The analysis also highlights the highest number of new medicine approvals in 15 years with many breakthroughs available for the first time that will potentially affect over 20 million patients. As the Affordable Care Act fundamentally changes access to healthcare in the US, the IMS Institute considers why the landscape will continue to change, and why spending on medicines is expected to remain below overall healthcare expenditure levels in the next 5 years (Figure 1).

fIGURE 1: REAL PER CAPITA SPENDING GROWTH 2002-2017

SHIFT FROM FEE-FOR-SERVICE TO MANAGED MEDICAID: WHAT IS THE IMPACT ON PATIENT CARE? Managed Medicaid is seen by many States in the US as a way to deliver better care at lower cost; recent actions to reduce use of Fee-for-Service plans have been significant.

This new review explores early trends in prescription drug utilization in four States that have switched their pharmacy benefit to Managed Medicaid. Comparing changes in the use of antipsychotic, respiratory and diabetes medications, it identifies some areas of impact, including increased cost savings when pharmacy and medical benefits are offered in the same plan, but wide differences in the magnitude of the changes between States and disease areas. Highlighting the substantial and growing importance of Medicaid, the findings also point to the value of care coordination services to monitor and encourage the appropriate use of medicines.

E-PRESCRIBING PREVALENCE: HOW EXTENSIVE AND HOW INTENSE? Electronic prescribing is recognized to be an increasingly significant driver of pharmaceutical prescriptions. Focusing on the cholesterol-lowering market, this analysis leverages retail prescribing data across both 'traditional only' prescribers and e-Prescribers to consider the extent and intensity of e-Prescribing within the US healthcare provider population between 2011 and 2012. It offers insights into changes in the volume of e-Prescribing and in the number of e-Prescribers, together with a perspective on meaningful use levels and implications for growth in e-Prescribing.

•

Growth 10% 5% 0% -5% 2002

2007

Healthcare Spending

2012

2017

Medicines Spending

Source: CMS National Health Expenditures Jul 2012; IMS Health, National Sales Perspectives, Dec 2012; US Census Bureau Jan 2013; Economic Intelligence Unit Nov 2012; IMS Market Prognosis Apr 2013

PAGE 6

Further information on these reports and other insights from the IMS Institute can be accessed from the Institute website at www.theimsinstitute.org, together with information on its extensive range of research activities.

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

NEW LINKED DATASET | NEWS Unique linkage of retrospective longitudinal datasets completes patient pathway in England for robust decision support.

HTI-CPRD unlocks research potential across GP and secondary care Responding to market demand to understand the value of medicines using real-world evidence (RWE), IMS Health has announced the availability of HTI-CPRD. This retrospective longitudinal patient dataset enables the entire patient treatment pathway in England to be mapped across both major sites of care.

The development follows the June 2012 launch of HTI (Hospital Treatment Insights) and the expansion of a strategic partnership between IMS and the CPRD (Clinical Practice Research Datalink) in September 2012. HTI links hospital activity with drugs dispensed to create a longitudinal patient record in the hospital setting in England.

Providing a window on the true standard of care, the HTICPRD retrospective dataset can be interrogated to answer critical questions around patient diagnosis, treatment and outcomes. It is anticipated to have value for decision makers both within and outside of England, enabling more robust support of value messages, informing the cost vs benefit debate, aiding investment decisions for new indications, responses to safety concerns from regulators, and preparation of research to share with payers.

Today, coverage within the linked primary and secondary care dataset has reached ~208,000 unique patients who can be followed longitudinally back to January 2010, and this number continues to grow.

•

Secondary Care

Hospital Treatment Insights

Drug dispensed

Hospital Pharmacy Audit Patient drug record

FASTER, MORE ACCURATE DECISION SUPPORT

Diagnosis

Discharged

Change in drug regimen

Drug dispensed Readmission

Outpatient visit

Operation

Hospital Episode Statistics Patient outcomes

Clinical Practice Research Datalink Primary Care Patient drug outcomes

Initial GP consultation

GP consultation Drug repeat

Drug repeat

Drug repeat

Repeat prescription

Primary Care

For further information on HTI-CPRD and its ability to inform the complete patient journey in England, please contact Joshua Hiller at Jhiller@imshealth.com

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 7

INSIGHTS | REAL-WORLD EVIDENCE PERSPECTIVE

Increased evidence demands for market access have driven major advancements in patient-level datasets and the technologies to increase value to the growing number of data users. but how can this vast foundation of data be leveraged to improve eďŹƒciency and value? We ask IMS expert Jon Resnick for his perspective on what it will take to realize the true potential of a growing body of real-world evidence (RWE).

The interviewee Jon Resnick, MBA is Vice President and General Manager RWE Solutions, IMS Health Jresnick@imshealth.com

PAGE 8

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

REAL-WORLD EVIDENCE PERSPECTIVE | INSIGHTS

Where next for RWE? Why focus is the key to unlocking the value of real-world, evidence-based data Q: What has intensified the need for RWE? JR: We are hearing a lot about RWE at the moment but it has actually been around in some form or another for a long period of time. Health economic and outcomes researchers have been generating this type of information for decades to support prescriber and payer decision making. What has changed in the last several years is the timeframe over which RWE is being used to determine the value of a product. In Europe, we have seen much greater emphasis on current agents as health systems try to deter the rapid uptake of many innovative medicines. Similarly in the US, RWE – the data collected outside of clinical trials – is already being leveraged by payers and providers to better inform treatment decisions. What was once a race to achieve pricing and market access has now become a process of continued scrutiny by all healthcare stakeholders across the product lifecycle, creating the need for new and expanded real-world data and different types of study endpoints.

Q: How have data sources been evolving to meet this new demand for RWE? JR: There have been significant developments, both in the magnitude and reach of commercial patient-level databases as well as the availability of the data. We have started to move beyond siloed datasets to sophisticated linkage and integration of multiple data sources, allowing a more holistic picture of patient treatment across different sites of care.

At the same time, initiatives such as the OMOP (Observational Medical Outcomes Partnership) in the US, and EU-ADR (Exploring and Understanding Adverse Drug Reactions) in Europe are making patient-level data more publically available, for example, to monitor drug safety and for signal detection. For pharma, the resultant growth in data means both new opportunity and complexity. These data assets, when coupled with technology and the skilled resources to execute the analytics, offer a new mechanism to make business decisions, articulate product value and engage stakeholders through new service offerings. On the other hand, these datasets and required analytics are incredibly complex. At IMS, for instance, we have patient-level data in more than 260 million patients in the US alone across more than eight linkable assets. The range of analytic questions that these start to address across pharma stakeholders can be dizzying.

Q: How far has the market come in using RWE? JR: It is quite surprising how extensively RWE is being used today. IMS has spent the last few months compiling a universe of studies and we have been amazed by the number of examples we have found. Just using literature searches and informal discussions, we identified more than 100 products globally where RWE has had a measurable impact on either initial or sustained market access. For the most part, these were high-profile products from large drug classes which had generic or low-cost alternatives that were being traded against branded products in the comparisons. There are some fairly prominent examples from companies and payers across Europe and the US, as well as a complete mix of studies. These include payer-sponsored analyses examining specific drug classes, pharma/payer collaborative initiatives to look at particular products and disease areas, and sometimes pharma-sponsored work that has been presented. So, already, we are starting to see tangible results from its use.

What was once a race to achieve pricing and market access has now become a process of continued scrutiny by all healthcare stakeholders across the product lifecycle. continued on next page

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 9

INSIGHTS | REAL-WORLD EVIDENCE PERSPECTIVE

Q: How are different countries performing in collecting real-world data?

JR: We’ve looked at this globally at IMS and compared markets across several dimensions: availability and access to data; standards and approaches to using the data that are systematic to a market; and applications and decision making. What we’ve found first and foremost is that if you look across markets on a 20-point scale, no market scores above 11. This means we are only about halfway there on a scale of sophistication as to how RWE can be used. What we found on a relative basis is that several markets are doing well against some of those dimensions. For example, in the UK, there is phenomenal access to data. The NHS has done a very good job in terms of making that information available and although the systematic frameworks aren’t quite there yet, there is now a strong foundation for conducting good research in this market. The US scores well in having a lot of commercial availability of data, but the fragmented nature of the market makes it challenging to integrate this in a very seamless way. but the challenges across countries are very different.

Q: What is your view about the progress being made?

JR: I think these are still early days in terms of the types of analysis that can be done in linking, for example, aspects of treatment like adherence to cost-effectiveness. There is much more that can be achieved going forward as RWE expands in the domain. but already, it is a practice that everyone is using in some way to evaluate products, and the more we scrutinize definitions around it and develop standards and consistent methodologies, the better it will be understood and utilized to support good decision making.

Q: What are the hurdles to achieving this objective?

JR: There is a need for scientific methods to improve understanding of how to work with the data and raise trust in and acceptance of real-world information. A major challenge, too, is navigating privacy regulations. Although personal data can be leveraged in a fully anonymized manner for the greater good, specific markets are less advanced in working through the detailed issues to make this a reality. One of the biggest challenges, though, is stakeholder alignment and willingness to share the data across different settings. Collaboration is going to be key to ensuring that technologies and data pockets can connect. From our perspective, the only way to encourage this discourse is open, honest discussion around strong, rich data with good methodologies. This is where we enter the marketplace, aggregating data, linking different datasets and bringing things together in a safe and secure way to facilitate that stakeholder trust and collaboration.

Q: How can the full potential of real-world data be realized?

JR: I think this will take effort on several fronts. Firstly, through continuing to work with the data and make meaningful connections by pulling together fragmented information and joining individual data streams. Physicians, providers, payers and regulators are all looking for information to help them make better decisions to improve health outcomes. but no one is accessing a solid foundation of evidence. It is by linking all this information into an evidence base that we can drive greater efficiency and value.

The other area that will be increasingly important is more focused and efficient use of the data to support changing needs over time. We now have a huge wealth of information which continues to expand in both scope and volume.

While this significantly improves our ability to demonstrate and articulate the value of products, it also calls for new skills and capabilities in pulling from this broad reach of data to ensure that the right evidence is created for the right stakeholders. Real-world data has the power to drive discussions across the entire healthcare setting – with regulators to inform market access decisions, with physicians on which patients are likely to benefit most from different types of treatment, and with managed care to demonstrate how companies can add value to the ongoing care of their patient populations. The ability to support this dialogue with relevant, focused evidence will be essential to improving efficiency and quality in healthcare.

PAGE 10

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

REAL-WORLD EVIDENCE PERSPECTIVE | INSIGHTS

I am sure that, with industry commitment, five years from now the role of RWE will be secure, methods will be further advanced and trust will have grown.

Q: How is IMS helping to move this agenda forward?

JR: As a third party that is independent of product creation and the decision-making process, we are committed to supporting the advancement of evidence-based healthcare and promoting dialogue across decision makers through the use of real-world research. We are working with our customers, as well as key stakeholders, to understand their strategic needs and help them build the tools and capabilities to leverage the information to assess value, ensure safety and drive healthcare efficiency.

We have been expanding and linking our real-world data assets – the recent launch of IMS PharMetrics Plus™ has created the most comprehensive patient-level database in the US for a complete picture of the patient pathway across settings of care. This has brought tremendous new analytic possibilities for outcomes research. In Europe, we are collaborating with payers and providers to integrate IMS assets with external data sources to provide the scope and depth of information needed. The recent linkage of IMS hospital data to the CPRD primary-care dataset, for example, has provided the means for following patients across the full treatment journey in England.

As companies face the challenge of handling ever increasing amounts of data, we have also been building technical platforms to help them organize and leverage this information in a cohesive way. These include data warehouses hosting customized datasets, catalogues of profiled data to optimize searches, and advanced analytics libraries allowing users to analyze disparate data using standardized statistical methods. New applications and visualization tools, such as IMS Real-World Explorer, are enabling the creation of population cohorts through a user-friendly interface. And engagement platforms are providing the means to share and communicate the research. Together, these solutions are enabling customers to more strategically manage and better use their data assets for RWE generation.

Q: What is the outlook for RWE? JR: We are already starting to see a move towards shared responsibility in not only conducting good analytics around the information base that’s there but also helping that to impact patient care. Our own collaboration with AstraZeneca reflects a shared perspective on the transformative power of RWE on global health systems.

The incentives are coming around, people understand the role we all need to play and things are moving forward.

I am sure that, with industry commitment, five years from now the role of RWE will be secure, methods will be further advanced and trust will have grown.

•

RWE lens for pharma See our Insights feature overleaf for new IMS findings on the market impact of RWE.

ACCESSPOINT • VOLUME 3 ISSUE 6

IMS RWE TECHNOLOGY SOLUTIONS Real-world reporting Applications to standardize, share and communicate research Analytic workstation User-friendly RWE research tools and disease models Real-World Explorer Cohort builder and visualization to simplify and govern analytics Advanced analytics libraries Powerful analytical libraries for deep insights and advanced and standardized statistical methods Data catalogue Library of profiled datasets to optimize search Data harmonization Linked de-identified data warehouses to host customized datasets

Client RCT data

Client observ. data

IMS claims, EMR, LRx data

Third-party assets

PAGE 11

INSIGHTS | REAL-WORLD EVIDENCE IMPACT

Despite the growing importance of real-world evidence (RWE) as a basis for stakeholder decision making, pharmaceutical manufacturers have struggled to act on this trend. New research from IMS oers quantifiable and actionable insights for informed and focused RWE investments within the context of prevailing dynamics.

The authors Ben Hughes, PHD, MBA, MRES, MSC is Senior Principal RWE Solutions, IMS Health Bhughes@uk.imshealth.com

Marla Kessler, MBA, BS is Vice President and EMEA Leader, IMS Consulting Group, IMS Health Mkessler@imsg.com

Page PAGE112

IMS REAL-WORLD HEALTH ECONOMICS AND OUTCOMES RESEARCH IMS EVIDENCE SOLUTIONS & HEOR

REAL-WORLD EVIDENCE IMPACT | INSIGHTS

RWE market impact on medicines: A lens for pharma An international comparison of the use and impact of real-world evidence The pharmaceutical industry’s increasing focus on RWE reflects the greater supply of electronic patient-level data and higher stakeholder demand for RWE-based decisions. So why have manufacturers struggled to act? They have been constrained by a limited fact base and isolated case examples, prompting this research detailing data supply dynamics and over 100 cases studies of actual RWE influence on product decisions.

These quantifiable insights debunk a number of common beliefs:

• • • •

RWE’s influence on decisions about medicines has increased in magnitude and scale in western markets: more than 100 observed case studies illustrate its evolution beyond pharmacovigilance (PV) Payers have applied RWE in assessing value in a variety of ways, including expanding medicines use where warranted: cost containment is not the sole objective Although payers are a powerful stakeholder in setting the RWE agenda, proactive pharma engagement matters: manufacturer-generated RWE influenced over 25% of observed decisions RWE strategies need a local context but four fundamental market archetypes can focus pharma efforts: pharma does not need a unique approach in every country

This study provides a detailed understanding of market dynamics, consolidating them into four dominant archetypes. It enables manufacturers to focus RWE investments via improved internal alignment and gain greater value from stakeholder engagement. The insights it provides are also relevant to policy makers and payers seeking value from RWE.

THE STRUGGLE FOR DECISIVE ACTION The increasing need to obtain better value for healthcare spend has elevated RWE’s importance as a decision-making tool. This is particularly true for medicines. In separate research, IMS estimated that improved medicines’ use could avoid USD300-650 billion of cost worldwide.

Even stakeholders who see the RWE opportunity and its increased use struggle to act on it. Limited, isolated public case studies create a narrow picture of how RWE has affected decisions, and misjudge the complexities in the underlying drivers of RWE across markets. The debate is muddied further by different stakeholder perspectives – industry versus payers, health economics versus pricing and market access (PMA). continued on next page

OUTCOMES - Issue 1 ACCESSPOINT • VOLUME 3 ISSUE 6

Page13 1 PAGE

INSIGHTS | REAL-WORLD EVIDENCE IMPACT

The result is confusion, misalignment and even organizational paralysis over what to do about RWE in pharmaceutical companies. Some see RWE narrowly – supporting safety or mandatory submissions – while others see a broad lever to engage stakeholders. While RWE evangelists clash with skeptics wanting proof that RWE matters, they themselves are split between those who see many positive opportunities and those focused on using it to mitigate risks.

FACT-BASED INSIGHT To forward the debate, IMS sought an objective demand and supply lens on RWE. This focused on licensed medicines use rather than innovation, PV, or broader payer and provider use such as patient pathway evaluation, and engaged payers, health technology assessment (HTA) experts and clients in over 50 interviews.

To characterize demand, more than 100 non-safety examples were identified in which RWE impacted medicines (Figure 1).

Mainly driven by payers, RWE has influenced license (label), access, pricing, and use across countries and therapeutic areas (TAs). Approximately 25% of these decisions reflect industry-generated RWE, demonstrating that pharmaceutical companies influence this evolving landscape.

In addition to demand, real-world data (RWD) supply was examined, focusing on database use rather than (costly) prospective data generation. A proxy for supply, RWE output through peer-reviewed research varied from the thousands of publications in the US and UK to only a few hundred in Germany and France. This difference reflects varying usefulness of electronic data and different stakeholders’ ability to access it. Useful data would have extensive coverage, illustrate the full patient journey and have high clinical depth and quality. While only selected actors might need this level of data to create value, in practice widespread appropriate access generates more research output (Figure 2).

Overall, no country has an ideal data supply, with usefulness of or access to data constraining supply to different extents. In access-led countries (above the arrow) improving supply focuses on usefulness, such as the UK’s national CPRD1 linkage program. Companies play a role, too, such as IMS’ US strategy of developing sophisticated HIPAAcompliant2 linkage technology. In usefulness-led countries (below the arrow), debates are ongoing to improve access, such as to payer data in France. Meanwhile, individual companies are engaging directly with physicians and patients for consent to lever data for research.

fIGURE 1: RWE APPLICATION CASE STUDIES

fIGURE 2: RWE SUPPLY fROM DATABASES

Approximately 25% of these decisions reflect industry-generated RWE, demonstrating that pharmaceutical companies influence this evolving landscape.

PAGE 14

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

REAL-WORLD EVIDENCE IMPACT | INSIGHTS

This new demand-supply fact base is not complete without market-making mechanisms or frameworks.

Enabling demand, these detail how decision making includes RWE. Examples include evaluation mechanisms (eg, HTA, reimbursement processes, clinical guideline development), dissemination, and measurement (eg, prescriber incentives, payment-for-performance, quality indicators). This complete framing can help pharmaceutical companies, payers and policy makers alike derive the fuller potential of RWE.

fIGURE 3: RWE MARKET IMPACT SCORES

RELATIVE MARKET COMPARISON To highlight markets’ individual characteristics, the aforementioned drivers were translated into an RWE assessment scale – data supply and demand frameworks were each scored out of five, and application was scored out of ten reflecting the importance of RWE demand in practice.

This reveals major differences in RWE impact, with countries scoring between 2 and 11 of a potential 20 (Figure 3).

The maximum score of 11 reflects that no country has the ideal conditions for RWE use in a scalable manner and highlights RWE’s infancy. Lower scores indicate that RWE is relatively less available or more costly to generate with less consistent or transparent use in decision making. but even in markets with lower scores, RWE is still relevant.

In terms of data supply, the US scored highest, with a commercial market ensuring data availability for research needs, enabling research output greatly surpassing other countries. The US did not score a maximum five as ongoing linkage efforts are yet to achieve their potential, and underlying Electronic Medical Record (EMR) data capture is lower relative to other countries. Conversely, countries such as Spain score low given limited pockets of usable data.

On frameworks, the UK is closest to the ideal because RWE is used in systematic review for most evaluation processes (HTA, reimbursement, clinical guidelines). Stakeholders can disseminate RWE directly to prescribers, and RWE-enabled payment-for-performance contracts encourage appropriate prescribing.

Even the UK can go further: for example, RWE-enabled prescribing indicators are still limited. Conversely, countries like Denmark and Spain lack clearly defined roles for RWE in decision frameworks.

In terms of application – where RWE has informed decisions – all countries are distant from the ideal, with little consistent use of RWE in transparent decision making across TAs or patient populations. Case studies from the UK suggest the most extensive application, given the number, variety and breadth of resulting decisions relative to the entire health system.

Conversely, in countries such as Germany public case studies of RWE application are rare.

MARKET CLUSTERS AND STRATEGIES The analysis explains markets through RWD supply and RWE demand (clear frameworks and application to decisions).

These dimensions and scores segment markets into four groups: Pioneers, Traders, Explorers, and Laggards (Figure 4 overleaf).

Pioneers Stakeholders in Pioneer markets use their relatively notable RWD supply to inform drug decisions. Countries in this group – the Netherlands, Sweden and the UK – all have strong national HTAs, suggesting an impact from concentrated decision making. In these markets, pharmaceutical companies should set high ambitions for RWE plans, demonstrating value and engaging stakeholders based on a variety of real-life views (eg, disease, product, class, cross care settings, long-term outcomes, payer-relevant quality of care indicators). They must fully exploit RWE beyond traditional evaluations to enable commercial strategy, leverage outcomes-based marketing, and use innovative evidence tools with local health systems. continued on next page

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 15

INSIGHTS | REAL-WORLD EVIDENCE IMPACT

For example, one ground-breaking manufacturer developed a mobile (iPad) evidence platform to support its diabetes drug launch in the UK. It used RCTs and various RWD sources to build models of prescribing patterns, cost and outcomes for general practices. Trained sales representatives engaged prescribers by adjusting the preloaded parameters of the model (eg, patient numbers, prescribing profile, cardiovascular risk factors) to discuss prescribing from the clinician’s perspective. As RWE becomes an accepted dialogue with payers and clinicians in Pioneer markets, companies without these capabilities will be disadvantaged against or unable to respond to more innovative competitors.

fIGURE 4: MARKET SEGMENTATION fOR PHARMACEUTICAL INDUSTRY STRATEGIES

Traders The US is the only country representing the Traders, though other countries not in scope could have a similar model. Owners of RWD share it without stipulating how it should be used beyond ensuring individual privacy.

Most US insurance companies and providers sell data and only use it to support specific analyses about their own populations. Thus, pharmaceutical companies have broad data access to drive performance, from trial design through commercial support. Successful US strategies involve evidence platforms and tools that support multiple internal stakeholders. However, without clear interpretative frameworks, the channels for external engagement are more nuanced. Only selected payers engage readily on RWE, and FDA regulations on RWE dissemination are more restrictive than in Pioneer countries. A differentiated engagement approach is needed, requiring creative thought and investment.

One leading company, for example, developed a rich platform in one priority TA rather than a ubiquitous platform across TAs. Over several years it linked all relevant datasets (Rx, Claims, EMR, registries, RCTs, observational studies, biobanks) and developed different internal customer tools to exploit it. This asset supports traditional RWE and multiple peer-reviewed publications. More impressively it generates hundreds of internal standard reports, even improving sales forecasting accuracy. This capability enhances external engagement too, as the manufacturer is now a reference for local prevalence estimates or for characterizing local unmet patient needs.

Explorers Explorers, such as France and Italy, have a significant demand-side vision but limited RWD supply. France’s bold vision includes using RWE for cost-effectiveness assessments and regular class reviews without detailing how extensive data in the health system can be accessed or levered. In Italy, there is widespread use of payment-for-outcomes or coverage with evidence development, but how these schemes inform coverage or pricing decisions based on the captured data remains unclear. While manufacturers can react to these limited demands for RWE, the more innovative ones might place a bet that the markets will expand RWE use over time. There is no crystal ball, but this RWE demand could signal an evolution to pioneer-style markets. Either way, pharmaceutical companies must develop some RWE capabilities for payers’ current focus areas.

As RWE becomes an accepted dialogue with payers and clinicians in Pioneer markets, companies without these capabilities will be disadvantaged against or unable to respond to more innovative competitors.

PAGE 16

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

REAL-WORLD EVIDENCE IMPACT | INSIGHTS

For example, one inventive company has “gambled” on developing a high quality evidence platform in France. With no access to payer data, it has partnered with commercial vendors to use innovative and cost-effective approaches to maximize the value of existing EMR and Ministry of Health datasets. Using these as an initial platform, the company is gathering supplementary data to develop a high quality reference cohort in a chronic disease. In addition to classic RWE, this generates process-of-care indicators, setting a benchmark for understanding patient outcomes.

Laggards Finally, there are the Laggards who may use RWE more in future, but face significant hurdles today. The Laggards in this study are Canada, Denmark, Germany and Spain, all of which illustrate different challenges (eg, strong data privacy, fragmented healthcare landscapes). In these markets, pharmaceutical companies benefit from engaging directly with selected stakeholders willing to lead on RWE. Markets with strong regional payers may see that leadership from those regions, such as Cataluña in Spain or Ontario in Canada. In Germany, some sick funds have expressed willingness to partner on RWE. Given the limited resources of these stakeholders and the large number of manufacturers, developing a clear value proposition and local RWE capabilities are essential to becoming a preferred partner. Innovative companies have long been in dialogue with regional stakeholders, quietly making co-investments in research capabilities to further all parties’ goals.

FROM INSIGHT TO ACTION How can senior executives lever these insights for actionable RWE strategies? The emphasis and insights can be used to engage their teams to determine:

• • •

Where will additional investments in RWE create most value for our portfolio (eg, market types, TAs, stakeholders, phases of the lifecycle)?

What changes to brand evidence plans or stakeholder engagement approaches on evidence can capture the RWE potential in each of the four market types?

How should our RWE-generation capabilities be reinforced, such as scalable platforms, targeted stakeholder engagement, or deployment of innovative channel tools?

While franchise and brand teams naturally drive questions on the where, increasing leadership from PMA is required on the what, as is leadership from HEOR, epidemiology and other evidence functions on the how. Senior executives may need to personally champion cross-functional RWE discussions given the strategic issues involved and given organizational hesitancy around perceived RWE risks even at the expense of potential gains.

•

Acknowledgment This work has been a collaborative effort from many individuals across a range of backgrounds and settings. The authors sincerely thank the contributions from numerous payers, clients, members of the global RWES, HEOR and Consulting teams at IMS and others for their knowledge and expertise, without which these insights would not have been possible.

1 2

Clinical Practice Research Database Health Insurance Portability and Accountability Act

For a full view of country-by-country dynamics, detailed case studies and methodologies, please request the extended white paper from the authors.

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 17

INSIGHTS | IMS SYMPOSIUM

As demand for real-world evidence (RWE) continues to grow, new innovations in information technology are enabling more sophisticated use of real-world data through linkage and harmonization. An IMS Symposium at the ISPOR 15th Annual European Congress in berlin explored the tangible value for outcomes research and progress towards the ultimate goal of international interoperability.

The authors Jacco Keja, PHD is Senior Principal RWE Solutions & HEOR, IMS Health Jkeja@nl.imshealth.com

Ian Bonzani, PHD is Engagement Manager RWE Solutions, IMS Health Ibonzani@uk.imshealth.com

Page PAGE118

IMS REAL-WORLD HEALTH ECONOMICS AND OUTCOMES RESEARCH IMS EVIDENCE SOLUTIONS & HEOR

IMS SYMPOSIUM | INSIGHTS

Powering a new age of outcomes research The driving force of information technology Several factors are fuelling the need for real-world evidence (RWE) on a much larger scale, not least the growing demand for more accurate and timelier drug safety evidence information.

In the US, the Observational Medical Outcomes Partnership (OMOP) is leading the development of infrastructure and methods to support the FDA’s Sentinel safety initiative, with a focus on the linkage of multi-source observational data.

Similarly, in Europe, the EU-ADR project is developing an innovative system for the earlier detection of adverse drug reactions by pooling clinical data from the electronic medical records of millions of patients in the region – leveraging modern biomedical informatics technologies.

Other factors driving requirements for bigger, deeper, connected real-world data include the need for more efficacious medicines, based on better understanding of patient segmentation and the link between outcomes and genetic variability, as well as increasing use of payment-for-performance arrangements and value-based pricing. These demand a greater level of granularity than that afforded by traditional retrospective research focused on claims and medical records, with a population-based approach being the ultimate goal.

The move away from single datasets towards interoperable data and analysis platforms integrating and linking data from multiple sources has many benefits for health outcomes research and healthcare delivery, enabling the creation of a clearer, more comprehensive longitudinal patient journey across disease and treatment pathways of interest.

THE VISION Looking to the future of healthcare delivery in 2020, it is possible to envision a world with access to truly integrated clinical trial and real-world data, supported by the technologies to facilitate its regular use in the healthcare decisionmaking process (Figure 1 overleaf ).

This is a world of broader engagements between healthcare stakeholders where information is the currency of these exchanges. A world where, in real time, appropriate care options can be examined and assessed across a variety of different disease and patient characteristics; where optimal treatment options can be identified across these different metrics and translated effectively into more integrated and holistic healthcare choices tailored to individual patient and disease area needs; a world where the right patients have access to the right treatments at the right time with the best opportunity to maximize outcomes and minimize disease severity and risk.

In this world, healthcare stakeholders can evolve, becoming collaborators in a system focused on individual patient care and shared decision making. Patients can move from being largely passive information receivers to more proactive decision supporters; providers can change from being implementers of guidelines and policies to more empowered vehicles for delivering quality and efficiency in healthcare; payers can evolve from being holistic population risk-benefit managers to being more active around patient and disease areas and point of care interventions; and pharma can transform from the role of medicine providers to being true collaborators in the healthcare delivery space.

The move away from single datasets towards interoperable data and analysis platforms integrating and linking data from multiple sources has many benefits for health outcomes research. continued on next page

OUTCOMES - Issue 1 ACCESSPOINT • VOLUME 3 ISSUE 6

Page19 1 PAGE

INSIGHTS | IMS SYMPOSIUM

fIGURE 1: HEALTHCARE 2020

Prescribing decision support of the future

E-detailing forum Sp Specialist ecialist M anufacturer support supp ort Manufacturer

Network GPss Network of GP

At their heart is information technology (IT) and the applications that fall within it. IT is the engine that will drive the success, powering, for example, the building of integrated research platforms based on the best practice methods and standards being developed by data consortia such as OMOP and C-DISC, and guidelines from bodies such as ENCePP and the PCORI. It will also enable the development of infrastructure firewalls and governance layers wrapped around patient data to address and mitigate some of the concerns over legalities and patient privacy.

KEY AREAS OF INNOVATION

Appropriate care options

Patient 1 C haracteristics Characteristics Epidemiology, Risk Risk Epidemiology, Factor, Disease stage, stage, Factor, Genomic Genomic profile profile X Optimal Optimal treatment treatment Standard Standard of care care

Patient 2 Characteristics Char Epidemiology , Risk R Factor, Factor, Disease SStage, tage G enomic pr ofile Y Genomic profile O ptimal tr eatment Optimal treatment Rx A ccoupled oupled tto o ser vices B services

A deeper dive reveals six core areas of innovation that are progressing quite quickly. These include emerging applications of natural language processing and the ability to extract more useful information from patient notes or discharge letters; federated research approaches such as SHRINE, MAELSTROM/DATASHIELD, and deidentification/linkage engines to address nuances around data access, governance and integration; predictive modeling and smart learning systems, including use of artificial neural networks in oncology clinics enabling the faster, more accurate diagnosis and prediction of patient outcomes across various tumor types; systems to power the processing speed and engines behind all the developments; and visualization tools to package information in more useful and valuable ways (Figure 2).

fIGURE 2: INNOVATION COMES fROM KEY TECHNOLOGIES THAT IMPROVE THE CREATION & UNDERSTANDING Of EVIDENCE

Patient-tailored healthcare delivery Most appropriate: t Drug treatment & dosing regimen t Integrated supportive care services (eg, home care/monitoring, nurse outreach, remote link-ups to GPs/specialists) t Patient education and empowerment programs

New range of data EHR, genomics, biobank, social media/sentiment, PROs… Visualization tools Care and patient flow diagrams, patient P&L…

Structure to unstructured Natural language processing, sentiment analysis, signal processing, association rule learning…

DRIVERS OF EVOLUTION The shape and pace of these evolutions are being driven by several factors which are increasing the ability to process, access and link real-world data. These include the establishment of methods and analytical standards, the legal and ethical framework for governance of data usage, and activities to build stakeholder acceptance and trust based on evidence transparency.

PAGE 20

Predictive modeling and smart systems New and learning algorithms to better diagnose patients and predict patient outcomes…

System processing Distributed computing, open-source software development…

Data integration (Linkage, ETL, Harmonization) Patient De-ID engines, harmonization/validation rules, data marts…

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

IMS SYMPOSIUM | INSIGHTS

These technologies are not only improving the ability to collect, curate, link and use patient-level information but are also then enabling that data to be translated into a holistic package of evidence that can be moved closer to the decision-making front line to enable more informed choices. However, it is the way in which they converge and integrate that their true value becomes apparent in facilitating the outcomes research of tomorrow:

• • • • •

Driving RWE value to wider groups of business and healthcare decision makers Enabling more accurate health economic models based on real-world populations and evidence Translating short-term endpoints to long-term outcomes Allowing deep patient segmentation/characterization of patients Avoiding under/over-estimations of appropriate use of Rx, events, signals

Figure 3 shows some of the applications that are emerging to support the change process. Externally, these include tools to aid HTA or better simulate real-world use and risk-benefit; more advanced point-of-care management to help physicians and pharmacists make more accurate and informed decisions regarding individual patient care; real-time risk-sharing/payment-for-performance platforms; and the ability to design more continuity in stakeholder engagement. Underlying these is a range of integrated and mechanized applications for optimizing internal working space, in terms of real-time benchmarking and KPI tracking against quality indicators and being able to drive efficiency and incentivize those indicators. fIGURE 3: TECHNOLOGICAL INNOVATION IS ALSO POWERING APPLICATIONS fOR OPERATIONAL AND CLINICAL CHANGE Prescribing decision support of the future

INTERNAL CHANGE SUPPORT

EXTERNAL CHANGE SUPPORT Consultative reimbursement/value support tools HTA validation/support CER/H2H simulation dashboards Point-of-care management Advanced CPOE to support care decisions Health surveillance & response Real-time risk sharing P4P systems Engagement platforms Provider/Plan selection tools (“Trip Advisor”)

• • • • • •

Integrated & mechanized research platforms KPI/QI tracking & benchmarking R&D optimization: Advanced CTO tools/Signal validation/Asset investment validation

•

ULTIMATELy BETTER PATIENT OUTCOMES & MORE COST-EFFECTIVE HEALTHCARE DELIVERy

IMPLICATIONS FOR INDUSTRY

E-detailing forum Sp ecialist M Specialist anufacturer supp ort Manufacturer support

GPss Network of GP Network

Appropriate care options

Patient 1 Characteristics Characteristics Epidemiology, Epidemiology, Risk Risk Factor, stage,, Factor, Disease stage ofile X Genomic pr Genomic profile O ptimal treatment treatment Optimal Standard of care care Standard

Patient 2 Char Characteristics R Epidemiology , Risk Factor, Disease SStage, tage Factor, Genomic profile profile Y Genomic Optimal tr eatment treatment Optimal coupled to to Rx A coupled services B services

Patient-tailored healthcare delivery Most appropriate: t Drug treatment & dosing regimen t Integrated supportive care services (eg, home care/monitoring, nurse outreach, remote link-ups to GPs/specialists) t Patient education and empowerment programs

For pharmaceutical companies, the evolution of RWE has placed new demands on the information generated during research and development. Increasingly sophisticated types of RWE are required to demonstrate how new drugs will perform in specific healthcare settings, with evidence requirements now extending right through to lifecycle management. RWE has a critical role to play in improving clinical development through understanding treatment and outcome diversity, in minimizing decision uncertainty, and in creating a “learning healthcare system” through performance indicators, information and incentives. It will also benefit patients long-term by enabling the industry to develop more targeted, value-added medicines. However, the creation of effective RWE requires the right data and the right technology to access, integrate and analyze these data. Innovative, integrated ‘big data’ networks, such as those being developed by AstraZeneca and IMS through their RWE collaboration, will enable a much clearer understanding of unmet need and better articulation of the value of medicines. continued on next page

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 21

INSIGHTS | IMS SYMPOSIUM

DATA HARMONIZATION EFFORTS UNDERWAY Harmonizing and linking data across multiple cohorts bring the potential to expand scientific knowledge based on very large sample sizes and synthesized information. Many new cohorts are collecting high-quality data, and a broader range of data is being made available through linkage to health registries or environmental exposure registries.

In the US, for example, the NUgene gene-banking project at Northwestern University in Chicago is targeting the collection and storage of DNA samples and longitudinal medical information from more than 100,000 patients to help understand the genetic mechanisms behind common diseases. In Europe, bbMRI-ERIC (biobanking and biomolecular Resources Research Infrastructure-European Research Infrastructure Consortium) is working to improve the accessibility and interoperability of bio-samples from various populations across the region. And research centers such as KORA in Germany and CARTaGENE in Canada are generating novel scientific knowledge through research platforms based on the pooling of data. At the same time, organizations such as Obiba and LifeLines are developing methods, software and expertise to support the synthesis of information in different research areas as well as harmonization and linkage of data. Projects such as ENGAGE (European Network of Genetic and Genomic Epidemiology) and HALCYon (Healthy Ageing across the Life Course) and bioSHaRE (biobank Standardisation and Harmonisation for Research Excellence in the European Union) will make use of the data collected and the tools developed in order to generate that new scientific knowledge.

BioSHaRE Work conducted in the bioSHaRE project illustrates some of the practical issues in harmonizing data. Led by Professor Ronald Stolk from the Netherlands, this aims to develop specific tools in collaboration with the Maelstrom Research Platform to create harmonization and standardization in the use of pooled data from different cohort and biobank studies. Together they are developing a series of software applications to cover all the building blocks for pooling and sharing data across cohorts, through a series of sample projects.

The success and value of harmonized analysis of collaborative research across heterogeneous studies depends not only on access to high-quality data but also on rigorous methodological approaches and on the specific documentation of the data. Knowing exactly which harmonization variable has been used and where the data have come from is key to enabling reproducibility of the results that are generated.

PROGRESS TOWARDS POSITIVE CHANGE Few would question that the path towards data interoperability is a challenging one – not least in managing the idiosyncrasies of real-world datasets, in improving internal efficiencies and resource allocation decisions, and in understanding the implications for achieving credibility with decision makers. However, there is growing recognition that the path exists and that the challenges are not insurmountable. It is by overcoming these through the use of technology, that the potential to drive operational and clinical change across healthcare settings can be realized.

•

This article draws on presentations from the IMS Symposium “Information technology: Powering the next generation of outcomes research and healthcare delivery”, held during the ISPOR 15th Annual European Congress in Berlin, Germany in November, 2012. Chair: Jacco Keja, PhD, Senior Principal RWE Solutions & HEOR, IMS Health. Speakers: Ian Bonzani, PhD, Engagement Manager RWE Solutions, IMS Health; Andrew Gaughan, MSc, Global Director, Payer and Real World Evidence Informatics, AstraZeneca Pharmaceuticals; and Isabel Fortier, PhD, Researcher, Research Institute of the McGill University Health Center and Director of Research and Development for the P3G Consortium. A copy of the full proceedings can be obtained by emailing Angelika Boucsein at Aboucsein@de.imshealth.com

PAGE 22

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

PATIENT-LEVEL INSIGHTS | IMS

Real-world patient insights. Real impact.

IMS LifeLink™ is a global source of comprehensive real-world patient information and solutions.

LifeLink delivers more powerful insights into the patient experience that will help healthcare stakeholders improve business performance, patient outcomes and quality of care.

LifeLink offers a powerful way to discover real-world patient insights: Information assets that track hundreds of millions of anonymized patients around the world

• • •

Proven real-world patient analytics and insights to address scientific and commercial applications enterprise-wide Delivery tools and customized consulting services for clients’ unique business needs

LifeLink provides an extensive library of powerful analytics for game-changing real-world results: Industry-leading methodologies and technologies ensure accurate and reliable linking of disparate datasets

• • •

Consistent and comparable views across time Highest level of patient data privacy

IMS LifeLink PharMetrics Plus™ is the largest, most diverse integrated US health plan database available. With the broadest coverage of geographies, care settings and industries, it provides data on more than 150 million US patients and allows for data integration with IMS patient-level data as well as clients’ own and external data.

GLOBAL FOOTPRINT INFORMATION ON HUNDREDS OF MILLIONS OF ANONYMIZED PATIENTS CANADA

EUROPE

ASIA PACIFIC

• Longitudinal Rx

• Longitudinal Rx (belgium, France, Germany, Italy, Netherlands, Sweden, UK) • Electronic Medical Records (France, Germany, Italy, UK) • Oncology Analyzer (France, Germany, Italy, Netherlands, Spain, Turkey, UK) • Hospital Disease Database (belgium) • Hospital Treatment Insights (UK) • Longitudinal Patient Database (Sweden)

• Oncology Analyzer (China, Japan, South Korea, Taiwan)

• Drug Plan Claims (Oncology, hospital)

UNITED STATES • Longitudinal Rx • Health Plan Claims • Electronic Medical Records (Oncology, ambulatory) • PharMetrics Plus • Hospital Charge Detail Master

ACCESSPOINT • VOLUME 3 ISSUE 6

• Longitudinal Rx (Japan, South Korea)

AUSTRALIA • Longitudinal Rx

PAGE 23

INSIGHTS | ONCOLOGY REAL-WORLD DATA

Data sources suitable for oncology comparative eectiveness research are limited and heterogeneous in Europe. As reliance on observational studies continues to grow, a basic prerequisite for a successful real-world evidence (RWE) strategy is the systematic identification, evaluation and prioritization of accessible real-world data.

The authors Saeed Noibi, MPH is Consultant RWE Solutions & HEOR, IMS Health SNoibi@uk.imshealth.com

David Bertwistle, PHD is Senior Consultant RWE Solutions & HEOR, IMS Health DBertwistle@uk.imshealth.com

Karin Berger, MBA Principal RWE Solutions & HEOR, IMS Health Kberger@de.imshealth.com

PAGE 24

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

ONCOLOGY REAL-WORLD DATA | INSIGHTS

Oncology real-world data in Europe Identifying the most appropriate sources for specific research needs Ensuring future reimbursement for oncology drugs is dependent on high-quality clinical data to generate convincing evidence of comparative effectiveness for alternative treatment options.

The particular demands on evidence requirements for oncology drugs have been documented in previous IMS research.1 Relative to other disease areas, randomized controlled trials (RCTs) in oncology do not always demonstrate real clinical endpoints and adverse drug reaction profiles. These evidence gaps are most profound in rare oncology indications, due to the very limited number of cases combined with the frequent use of personalized treatment plans.

Observational studies using patient-level clinical practice data with large sample sizes and long-term follow-up are increasingly relied on to meet data and evidence needs.

TRENDS TOWARDS OBSERVATIONAL RESEARCH In the US, population-based data sources, such as administrative claims data, are already used to generate evidence in outcomes research. In Europe, evidence requirements from a manufacturer’s perspective include characterization of unmet medical need across different countries, estimations of the size of target study populations, and determination of disease pathways and relevant endpoints.

Recent advances in data technology enabling the linkage of different data sources to create complete patient pathways are becoming an important real-world data (RWD) capability. The reliability of this approach depends on data attributes that allow for the most precise linkages based on unique patient identifiers across multiple data sources (deterministic linkage). Therefore, data sources with these attributes are especially valuable for real-world evidence (RWE). For example, linkage of pharmacoepidemiology databases and cancer registry data sources with unique patient identifiers in the respective sources facilitates high-quality research which is not achievable in the individual data sources. The huge surge in healthcare data generated for varied purposes has enabled the evolution of platforms for RWD and evidence generation. International research collaborations and pan-European legislation provide a real opportunity to shape the landscape for evaluating, accessing and utilizing these data sources for health technology assessments (HTAs) in particular, and health policy research in general. beyond Europe, the benefits of harnessing ‘big data’ generated in the clinical care of cancer patients has started to yield fruitful discussions in the wider oncology community. The American Society of Clinical Oncology (ASCO) has recently launched an initiative to link the multitude of electronic medical records (EMR) from different clinical practices and malignancies with the purpose of generating aggregate data to inform clinical practice and inform physicians on realworld patterns of care.2 This initiative exemplifies the need for credible RWD and sets the pace for the European clinical oncology community.

Gaining access to large retrospective data sources in Europe for pharmacoepidemiology research is still challenging when compared to the US. Until recently, population-based cancer registries were the major source of cancer data for healthcare policy research and cancer surveillance in Europe. However, due to their primary purpose of estimating cancer incidence and prevalence, these registries have limited usefulness for outcomes research. Furthermore, the issue of patient confidentiality continues to pose restrictions in accessing the data source.3 Discussions on the use of retrospective observational studies have now started in earnest in European countries.

continued on next page

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 25

INSIGHTS | ONCOLOGY REAL-WORLD DATA

CATEGORIZING ONCOLOGY RWD SOURCES IN EUROPE To better understand and characterize the European landscape of patient-level data in oncology, IMS has undertaken a comprehensive evaluation of RWD sources in the region. When mapped against the needs of different stakeholders, this reveals potential gaps where future investments and initiatives could optimize the RWD environment for oncology.

The process followed is outlined below.

1. Systematic data source identification The study began with a detailed literature search of peer-reviewed papers, published in the MedLine and EMbASE databases, to identify RWD sources used in epidemiology and outcomes research. In order to increase search sensitivity, search strings included therapy area of interest and studies conducted in Europe while excluding interventional studies such as RCTs. Since many RWD sources of emerging interest extend beyond those traditionally used for research studies, an extensive search of the grey literature was also conducted (Figure 1). The efficiency of this search was facilitated by an understanding of which literature types would likely contain the data sources of interest. Databases cataloguing different data sources were also reviewed to fill any potential gaps. These included the “ISPOR Outcomes Research Digest”4, b.R.I.D.G.E To Data5, and the ENCePP repository.6 Finally, to supplement the desk research and identify any further sources of data, an interview program was undertaken with key opinion leaders (KOLs) who had published works on outcomes research.

fIGURE 1: AN EffICIENT GREY LITERATURE SEARCH PROCEEDS WITH CAREfUL CATEGORIZATION Of LITERATURE TYPES fOR RWD SOURCE IDENTIfICATION

broad-base grey literature categories

International networks/ collaborations

Grey literature search strategy

Search terms included: • “Cancer” • “Registries” • “Research databases” • “EMRs” • “Longitudinal research” • “Prospective databases” • “Retrospective databases”

PAGE 26

Country-specific data sources

Reference to primary/secondary research data/evidence

Multi-country prospective data collection Retrospective/data re-use initiatives

Regional/country-wide data initiatives Patient groups/non-governmental organizations data

Proceedings/ Abstracts and other research sources

‘Ice-breaker’ in the use of non-conventional sources Local research data initiatives

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

ONCOLOGY REAL-WORLD DATA | INSIGHTS

2. Screening for relevant data sources Due to the variety of RWD sources, screening criteria were adopted to ensure that only data sources of relevance were considered for further evaluation. The most important criterion was that sources must contain patient-level data rather than population-based aggregate data. Furthermore, previous experience has shown that some patient-level data initiatives may have been discontinued due to inadequate funds, terminated projects or logistic reasons. While data sources that are no longer accruing data may still be useful in evidence generation for descriptive- and hypothesis-generating studies, contemporaneous data sources provide insight to more recent health technologies and advancements in clinical practice. For this reason, sources where data accrual had stopped over a period of time were not included for further evaluation.

Another benefit of including contemporaneous data concerns the duration of disease cycle and clinical pathway. Indications with longer disease pathways or with the likelihood of recurrence after a long period of remission will require data sources with continuous and longitudinal data. A critical element of the evaluative process was to understand the data variables (metadata) and the values that constitute the data. This description is an integral part of well-maintained data source dictionaries. While established data sources have defined data dictionaries or some form of metadata repository, many other sources do not have these dictionaries readily available to external researchers or stakeholders.

3. Evaluation of screened data sources Data sources that were screened into the analysis, which represented just under half of those originally identified, were subsequently evaluated using criteria which take into account the particular indications of interest. This involved reviewing data dictionaries where available, as well as interviewing data owners to elucidate the types of data collected within the data sources. Table 1 shows the complete list of the evaluation criteria that were applied. continued on next page

TABLE 1: EVALUATION CRITERIA APPLIED IN THE SCREENING PROCESS Database General Name Weblink Administrator contact details Scope (Regional, National, International) Scope (Country/countries) Brief description of database Owner of database Database Access Conditions for access Data fees Access to full database or restricted to specific data cuts Possibility of licensing data or requirement for collaborative research group Characteristics of Data Source Data type (eg Audit, Biobank, Claims data, Electronic health records,Epidemiology database, Registry) Coverage (Indications, eg general health records, oncology, hematological malignancies, etc) Patient population size Status (Ongoing or complete) First data available Most recent update Frequency of data accrual Linked to other data sources Linkage capability Patient Demographics Age Ethnicity Gender Clinical Data Diagnoses Date of diagnoses Details of diagnosis Symptoms at diagnosis Comorbidities Diagnoses recurrence Date of recurrence Details of recurrence (eg, Ipsilateral/contralateral, local/regional, etc) Treatment Data Dates of treatments Treatment line Treatment regimen Treatment response Treatment duration Adverse events Lab Data Lab tests recorded Dates of lab tests recorded Results of lab tests recorded Resource Use & Unit Costs Resource use Cost Outcome Mortality Date of mortality Cause-specific mortality

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 27

INSIGHTS | ONCOLOGY REAL-WORLD DATA

fIGURE 2: DATA PRIORITIZATION fRAMEWORK TO INfORM INVESTMENT DECISIONS fOLLOWING EVALUATION Of DATA SOURCES Optimal: Data sources with potential quick-wins for which business cases can be developed

Data attributes covered Moderate Significant Limited

4. Prioritization of data sources In order to identify optimal data sources with data attributes that would facilitate the fulfillment of different evidence needs, a prioritization framework was applied (Figure 2). Evaluated data sources were plotted on the framework based on the level of attributes collected and the conditions for access (access models). Consideration of data-linkage capabilities and the privacy laws in different European countries was a critical part of this process. Through the use of the framework, only 12% of the original data sources were identified for prioritization.

Potential: Data sources that require further analysis and prioritization which can be included in the long-term business plan

Deprioritized: Data sources that can be dropped from any medium-long-term business plan Limited

Moderate Assessment of access

Significant

LANDSCAPE OF ONCOLOGY DATA SOURCES IN EUROPE The evaluation revealed a paucity of data sources valid for the purposes of oncology RWD research in Europe. As shown in Table 2, no more than one fifth of those identified contained relevant information in terms of treatment-based attributes (20%), outcomes data attributes (19%) and linkage capabilities (17%).

Overall, the rigorous end-to-end process of identification, screening, evaluation and prioritization found that only 5% of the original pool of data sources had the potential to become accessible pan-European real-world oncology data sources. Furthermore, the study highlighted six countries as being best positioned to drive the oncology RWD landscape in the region: Denmark, Finland, Germany, the Netherlands, Sweden and the UK (Figure 3).

TABLE 2: CURRENT ESTIMATE Of ONCOLOGY RWD LANDSCAPE IN EUROPE Real-world data research capability

Percentage of data sources*

Number of countries

Data sources with treatment data attributes

20%

10

Data sources with outcomes data attributes

19%

14

Data sources with linkage capability

17%

12

*Not cumulative percentage as some data sources fall under more than one category.

Identifying the most valid data sources for individual research requirements is an essential part of evidence generation but demands both time and resources.

PAGE 28

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

ONCOLOGY REAL-WORLD DATA | INSIGHTS

fIGURE 3: END-TO-END ONCOLOGY REAL-WORLD DATA REALIZATION PROCESS Data sources screened (83%)

Data sources identified (100%)

Data sources evaluated (45%)

Data sources prioritized (12%)

Pan-European realization of accessible data sources (5%)

Health surveys National data linkage EMR

National data linkage

Surveillance

Systematic and structured grey literature search

Hospital registry

National data linkage

National data linkage

Hospital registry

Biobank

EMR

Biobank

Audit

Audit

EMR EMR

Hospital registry Registries

Data harmonization

Biobank

Biobank

Data mart

Research databases

Research databases

Research databases

Research databases

IMS perspective on European countries with potential to drive oncology RWD landscape in Europe

19

17

The end-to-end process involves a diverse IMS RWE Solutions team that includes a wide range of deep, in-country expertise

17

10

6 Types of data sources (Not exhaustive)

CONCLUSIONS Identifying the most valid data sources for individual research requirements is an essential part of evidence generation but demands both time and resources. These are critical factors to consider when planning RWE strategy. The variation in healthcare systems and confidentiality laws across Europe is perhaps the biggest challenge to RWD and observational research, alongside variation in data standards and quality in the region. At a broad level, much will depend on the success of ongoing EU and national initiatives to drive the creation of a harmonized framework for data collection and the implementation of real-world studies.

Notwithstanding these challenges, multi-factorial solutions do exist to shape the RWD landscape in Europe. These include a flexible and pragmatic approach to address the variety of data source types available. Sophisticated data platforms and data marts are already enabling more efficient integration of multiple datasets and providing the means to interrogate and harness much larger banks of real-world data more effectively and efficiently. Ultimately, transparent communication of mutual values led by credible researchers and KOLs will be important to fostering collaboration and a comprehensive data governance platform to access and generate oncology evidence from RWD in Europe.

•

1

berger K. Oncology growth drives new evidence needs: The special demands on cancer treatment. IMS Health AccessPoint, 2012; 3(5): 12-15 Transforming Cancer Care through big Data: ASCO Unveils CancerLinQ Prototype. Accessed 27 April 2013 at http://www.asco.org/transforming-cancer-care-through-bigdata-asco-unveils-cancerlinq-prototype 3 Izquierdo JN, Schoenback VJ. The potential and limitations of data from population-based state cancer registries. Am J Public Health. 2000 May; 90(5): 695-698 4 International Society of Pharmacoeconomics and Outcomes Research (ISPOR) Outcomes Research Digest. Available at http://www.ispor.org/research_study_digest/index.asp 5 The search engine for healthcare databases. Available at http://www.bridgetodata.org/ 6 European Network of Centres for Pharmacoepidemiology and Pharmacoviligilance (ENCePP) Database of Research Resources. Available at www.encepp.eu 2

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 29

INSIGHTS | NATURAL LANGUAGE PROCESSING

As the use of patient-level databases continues to expand, creative approaches to overcoming their limitations are significantly extending their quality and utility for real-world research. Here we describe a novel coding technique, using natural language processing (NLP), which has nearly doubled available data on key measures reported in IMS LifeLink™ EMR Disease Analyzer France.

The Author Massoud Toussi, MD, MSC, PHD, MBA is Principal and Medical Director, RWES & HEOR, IMS Health Mtoussi@fr.imshealth.com

PAGE 30

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

NATURAL LANGUAGE PROCESSING | INSIGHTS

Optimizing EMR database value using natural language processing Practical insights from IMS LifeLink™ EMR Disease Analyzer France Healthcare databases are unique in their ability to capture relevant and timely information on drug prescribing and effectiveness in conditions of real-world practice across large sample patient populations.

Varying in value and application, these datasets are considered to be of high quality if they are fit for use in their intended operational, business and scientific role.1 When a gold standard exists, “fit for use” can be interpreted as compliance of the data with that standard in a number of ways. As shown in Figure 1, data properties such as relevance, accuracy, timeliness, comparability and completeness can be used to define the quality of a database.

One of the major limitations of electronic medical records (EMR) is lack of completeness. In the case of a primary-care EMR database, this can arise for several reasons:

• • • •

Physicians are often short of time and fail to record in the EMR all the information they receive on a patient.

Physicians who do register clinical information in the EMR tend to do so in their own free-text wording rather than by using coding systems such as the ICD 10 (WHO International Classification of Diseases) or ICPC (WONCA International Classification of Primary Care).

Even among physicians who are willing and trained to use coding systems, in some areas, such as laboratory observations, there is lack of harmonization in terms of human usable coding systems.

A significant amount of information on a patient, such as hospitalization reports and referral letters, is by its nature composed in free text and is not exploited in databases due to technical or data privacy issues.

fIGURE 1: QUALITY INDICATORS Of A DATABASE

Timeliness (data still useful) Comparability (identifying fields)

Accuracy (fewer errors)

Relevance (appropriate data)

Gold Standard

Completeness (no missing records/variables)

Thus, one of the ways of overcoming the limitation of incompleteness of the data is to code the free-text information and transform it into structured, exploitable information, thereby improving its value.

NATURAL LANGUAGE PROCESSING Major EMR databases often contain millions of patient records, representing tens of millions of lines. To manually code all the free-text information contained within these records would require significant resources. Natural language processing (NLP) is an approach that allows automatic coding of free-text information in a fraction of the time it would take to achieve by hand. Moreover, NLP is a sustainable solution: once an NLP engine is developed and plugged into an EMR database it can be re-run periodically, for almost no incremental cost, to code any newly added free text.

continued on next page

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 31

INSIGHTS | NATURAL LANGUAGE PROCESSING

fIGURE 2: A TYPICAL NLP PROJECT PROCESS

6.

Evaluation

5.

Modeling

2.

Understanding of free text

3.

Preparation and cleaning of free-text data

4.

Knowledge base development

1.

Coding process understanding

NLP is a field of computer science, artificial intelligence and linguistics concerned with the interactions between computers and human (natural) languages. The history of NLP dates back as far as the 1950s and today it is widely used in the processing and routing of letters (script recognition), speech recognition applications, mobile phones and search engines. It encompasses a wide range of techniques and can be implemented using various commercial and free, open-source tools.

The principle behind NLP is that a human understandable text or speech is analyzed by computer against a knowledge base. This knowledge base can be a dictionary of terms with codes related to each term, with the objective of mapping free text to those codes (eg, ICD Codes).

As shown in Figure 2, an NLP project for an EMR database typically involves the following steps: 1. Understanding the process of transforming free text to codes 2. Understanding the free-text data to be coded 3. Preparing the data to be coded: Text data preparation Sampling of the text data into two text corpuses: one for testing and one for implementation

• • 4. Developing a knowledge base: (or lexicon) of ordinary terms with their variations • Dictionary Lexicon of technical information for use in coding containing all linguistic variations of the coded information • 5. Modeling: of spelling and grammar errors • Correction of text pieces (tokens) which could be considered as valuable • Identification • Identification of codes which correspond to the above tokens (bridging) 6. Evaluation: • Performance measures and repetition of iterations until achievement of acceptable accuracy Through the following project example, the different steps of NLP for coding data from an EMR database are discussed, together with the challenges that can arise, especially in the case of a non English-language EMR.

A CASE IN PRACTICE: IMS LIFELINK EMR DISEASE ANALYZER DATABASE FRANCE The hurdles that may be encountered in practical implementation of NLP have been demonstrated in a project using IMS LifeLink™ EMR Disease Analyzer Database France (DA). This is a longitudinal database containing the electronic health records of patients treated and followed-up by a representative panel of about 1200 GPs in France since the early 2000s. The database contains 4.5 million patient records comprising patient demographics, prescriptions and diagnoses.

At the commencement of the project, one of the limitations of DA was its lack of laboratory values due to the unavailability of coded data. In order to address this issue, 49 million lines of free text recorded in the field “Laboratory exam label” and 53 million lines of free text in the field “Laboratory exam value” were identified in the database. About two thirds of this data had been already bridged by traditional methods but lacked preciseness and validity. It was therefore important to re-bridge the entire corpus.

PAGE 32

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

NATURAL LANGUAGE PROCESSING | INSIGHTS

PRACTICAL CHALLENGES 1. Text language One of the main challenges of NLP projects is that they are very much language dependent. In the DA project, although several commercial and free open-source solutions existed for the standard English and French languages, no tool could be identified for use in retrieving laboratory values from GP texts in French. Thus, it was necessary to create one.

The tool was developed using Python programming language. This is one of the most commonly applied computer languages for text processing due to its various libraries and features (most of the Google search engine is written in Python). It also has the advantage of being under a wide open-source license which makes it usable and distributable even for commercial applications.2 2. Text corpus Unlike ordinary text corpuses, such as web blogs, forums and even twitter messages, where text is accompanied by a minimum of context and background, the information of interest in EMRs is not always in the form of complete, understandable sentences. In DA, the text was written in short-hand form, probably due to physicians’ lack of time for transcribing laboratory values or other observations in complete phrases or even words. More specifically, each physician used his or her own short-hand language and layout for recording information. For example, for “cholesterol LDL to total cholesterol ratio”, a variety of writings could be found, such as “LDL/tot”, “LDL/chol”, “LDL to chol”, “LDL chol rat”, etc, none of which are common expressions that can be found in a standard lexicon.

In addition, the shorthand used by the doctor was in many cases very personalized, making it difficult to understand even by other colleagues. For example, the letter “p” was variously used by different physicians for “patient”, “plasma”, “pulse rate”, “pain”, “purulent”, “polymorphonuclear”, “perimeter”, “pathogen”, etc. Thus, it was decided to develop a rulebased engine on the top of the lexicon to deduce from the context the right intended meaning (see below).

3. Encoding problems fIGURE 3: EXAMPLE Of ENCODING PROBLEMS A further challenge of NLP projects is the encoding of characters in the text. Indeed, in DA, when data is sent from the physician’s software to the core database, the free text can be encoded differently based on the operating system and the software used by the physician. Some characters are transformed into non alphanumerical characters, making it impossible even for advanced spellchecker modules to work correctly (Figure 3). During the project, this phenomenon happened even more frequently because of the French letters and the high number of Macintosh users in the DA panel. It was therefore necessary to develop a specific module to correct these encoding problems before implementing the spelling checker module.

4. Lexicons Dictionaries and lexicons of relevant domains are not always available in all languages. For the DA project, a lexicon of general and medical French language words was created from different resources found on the internet. The result was a dictionary of more than 500,000 French words, rich in medical terminology, which is available for use in future projects and even commercial products.

However, a comprehensive commercial or non-commercial lexicon of laboratory observations can be difficult to find, even through medical dictionaries. In this instance, it was even more challenging to identify one for French laboratory observations. In the absence of such a resource, a book of laboratory values with about 1700 pages was implemented and almost the entire book coded to create a dictionary of coded terms. continued on next page

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 33

INSIGHTS | NATURAL LANGUAGE PROCESSING

NLP can bring tremendous benefits to a database and increase its value by providing new information for a very low cost.

5. Spelling correction Spell checking is another key task in NLP. before any tentative bridging can commence, it is essential to make sure that the words to be checked in the database are correctly spelled – particularly given the spelling skills of many doctors!

The purpose of the spelling correction process is to identify words that do not exist in a dictionary and to guess the word which was intended. The objective is to correct lexical "non-word errors", ie, those where the word is misspelled.

Omitting this step can mean the loss of useful information which potentially can be retrieved from the database.

Several studies have shown that misspelled words are generally close to their correct form.3 Typically, the DamerauLevenshtein distance is applied to guess the closest correct word for one that has been misspelled. In this case, the minimum “cost” necessary for transforming word “a” to word “b” is calculated using the four following operations described by Damerau:4

• • • •

Insertion: "cholesterool" Deletion: "cholestero" Substitution: cholezterol " Inversion: "cholestreol"

(+ transposition or substitution pairs) For the purpose of the DA project, a module was developed to calculate this distance for all of the words of the corpus that did not exist in the dictionary, to find the appropriate word guessed. Each word had to be compared to the word in the dictionary. In each case, this involved several thousand comparisons – a very processor-consuming procedure.

6. Bridging Once these steps had been completed, bridging could begin. bridging consists of finding words expressed by physicians in the lexicon and suggesting the appropriate code for them (here a unique form of the word). While this process can be the simplest step in a standard NLP project, in the case of an EMR database it poses a particular challenge due to the extremely private nature of medical records. Indeed, physicians often consider the free text they enter into the EMRs to be personal notes needing only to be understood and de-coded by them. Consequently, the expressions and layout forms used tend to be unique to each physician.

To resolve this, contextual intelligence was used in the bridging algorithm which applied the rule-based engine previously described. This engine allows the same word or phrase to be interpreted using evidence from the context of that word. For example, to identify whether ‘p’ in “blurred p”, stands for “plasma”, “patient” or “pulse rate”, both the adjacent words to “p” and the other uses of “p” by the same physician should be considered.

7. Evaluation In an NLP project, not all the database is used for bridging. Generally, a random sample of a few thousand lines is prepared as the learning corpus and another sample is prepared for testing. All the processes described above are thus repeated over and over until the required level of sensitivity and accuracy is obtained. Then, the NLP is implemented on the overall database and, if necessary, enhanced again through iterations.

What should be the size of the initial sample? Nobody knows exactly. It depends on the variety of the text that is being processed and also the capacity of the computers used during the development phase. In the DA project, it was decided to sample 30,000 lines of laboratory observations and as many for values. The evaluation of the right coding of these lines was done by human intervention through inspection of the codes and the programs were adjusted after each evaluation cycle.

PAGE 34

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

NATURAL LANGUAGE PROCESSING | INSIGHTS

The preparation of data, spelling correction, lexicon development, bridging and contextual intelligence was thus worked on by iteration until 97-98% of precision in the bridging was reached and it was certain that non-bridged text was not bridgeable, even manually. Once the entire algorithm was considered acceptable in terms of sensibility and accuracy, it was applied to the whole database and further enhanced.

It is important to define metrics to evaluate the gain obtained through the NLP project and to be able to quantify its added value. In the DA project, a total of 47 million lines of free-text laboratory exam labels (out of 49 million) and 53 million lines of free text laboratory exam values (out of 53 million) were bridged (Figure 4). In practice, the overall gain is always more than this plain increase in the number of bridged records. In this case, more than 20 million lines of laboratory values which were already bridged were not linked to identifiable laboratory exam labels, and were thus not usable. With the NLP project, those values were linked to their corresponding newly bridged labels, forming groups of “laboratory exam label + value + unit” which can actually be used in research.

fIGURE 4: OVERALL GAIN OBTAINED THROUGH THE NLP PROJECT

Million

before NLP

After NLP

60 50 Number of bridged (coded) data items

MEASURING THE VALUE OF NLP

40 30 20 10 0 Laboratory exam labels

Laboratory exam values

Nb: Although the laboratory exam values (the numbers and units) were already well bridged, for 17 million records they were useless as not related to a laboratory exam label.

As a practical illustration in one disease area, immediately prior to implementation of the project, HbA1C was available for 8.6% of patients with diabetes mellitus. After completion of the NLP project, this percentage was as high as 16.5%, thus almost twice the previous rate of available information.

COST-EFFECTIVENESS AND SUSTAINABILITY NLP can bring tremendous benefits to a database and increase its value by providing new information for a very low cost. The DA project was conducted entirely within the framework of the six-month internship of a student studying for a Master’s degree in Computer Science and Linguistics.

Finally, a further advantage of an NLP project is its sustainability. In the case of DA, the whole cycle of the NLP was automated, so that the chain of the programs is run automatically every month to bridge the newly entered data into the database.

•

After completion of the NLP project, HbA1C was available for 16.5% of patients with diabetes, almost twice the previous rate of available information.

Acknowledgment We would like to thank the following for their support of this project: Aude Robert, intern at IMS Health; Grégory Coulthard, IT engineer at IMS Health; Alain Venot, Professor of Medical Informatics and Thierry Hamon, Professor of NLP, both at the University of Paris 13.

1

Herzog TN, Scheuren FJ, Winkler WE. Data quality and record linkage techniques. Springer-Verlag New York Inc.; 2007 bird S, Klein E, Loper E. Natural language processing with Python. 1st ed. O’Reilly Media, Inc, USA; 2009 3 Kukich K. Techniques for automatically correcting words in text. ACM Comput. Surv. 1992 Dec; 24(4):377–439 4 Damerau FJ. A technique for computer detection and correction of spelling errors. Commun ACM, 1964 Mar;7(3):171–6 2

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 35

INSIGHTS | PROPENSITY SCORING

Alongside growing demands for real-world evidence (RWE) is the imperative to ensure its reliable and confident interpretation. Propensity scoring has a key role to play in ensuring stakeholder trust in RWE insights but requires careful consideration of its relevance and application to preserve its value.

The Authors Saeed Noibi, MPH is Consultant RWE Solutions & HEOR, IMS Health SNoibi@uk.imshealth.com

Vernon Schabert, PHD is Senior Principal RWE Solutions & HEOR, IMS Health Vschabert@us.imshealth.com

PAGE 36

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

PROPENSITY SCORING | INSIGHTS

Demystifying the propensity scoring method in real-world evidence generation Practical considerations for appropriate use The need for real-world evidence (RWE) by healthcare payers and providers to demonstrate the effectiveness of medical interventions has been well established.

However, concerns over the reliability of inferences from such evidence have been the subject of ongoing methodological advances in research study design and analytics.1 Unlike randomized controlled trials (RCTs) in which the ‘ideal’ experimental environment is created, RWE relies on data generated from prevailing standards of care without any intended intervention. As much as RWE addresses questions of intervention effectiveness in community settings, which is not the traditional focus of clinical trials, a real concern still lies in the reliability of any ensuing association between healthcare interventions and outcomes.

PROPENSITY SCORING METHOD Rosenbaum and Rubin’s seminal paper in 1983 introduced the medical research community to propensity scoring method (PSM).2 Initially developed as a method to match patients on different exposures in order to minimize systematic selection of patients for treatment (selection bias), PSM is a probabilistic model that collapses all measured covariates to generate one continuous variable that is predictive of the exposure of interest, eg, treatment.

The statistical models used to develop propensity scores were not new to biostatisticians or outcome researchers, but the use of the output from one model to control the inputs of a separate outcome model was less familiar in outcomes research than it was in other social sciences. by the early 2000s, use of PSM had become established practice in outcomes research. Since then, perhaps due to the popularity and varied applications of the methodology as well as the passage of time, it has attained both a mythical status and a carelessness of application that undermines its value. The reliability of PSM over other methods of covariate adjustment depends heavily on its appropriate application and an understanding of its suitability given the study scenario (Figure 1). fIGURE 1: APPLICATIONS Of PROPENSITY SCORING METHOD

Matching on propensity scores

I

Dependent on robust matching patient numbers

STUDy DESIGN PROPENSITy SCORING METHOD (PSM) Pre-agreed variables in the real-world data source: 1. Predictive of intervention in the prevailing standard of care 2. Proxies for unmeasured covariates Validation of selected variables through external information eg, expert opinion and sensitivity analysis

Decision to proceed with design informed by sound methodological rationale

Restricting patients by excluding outliers in the propensity

Stratification by propensity score

II ANALyTICS Careful model fitting

Dependent on: 1. Sufficient patient numbers 2. Presence of outliers

Potential to identify treatment effect modifiers

Multivariate analysis of propensity scores

Applied where it potentially confers advantage to ordinary modeling

Weighting

Weighting by inverse of propensity scores to create a quasi-sample

continued on next page

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 37

INSIGHTS | PROPENSITY SCORING

Selecting the right covariates PSM lends itself, at least intuitively, to interpretation of the way in which selection bias occurs and therefore how the benefits of randomization may be approximated in an observational study. However, this impression needs to be carefully considered in view of the fact that, from the outset, PSM is underpinned by an assumption that relevant covariates to treatment selection are at least correlated with variables that are observable and have been included in the propensity model.

PSM is a method for creating a multi-dimensional proxy variable for bias that may influence treatment effect estimates.

It is therefore imperative that covariates included in the propensity model are those that are most predictive of the exposure treatment and the outcome.3,4 For example, covariates that are known, a priori, not to relate to the outcome should be removed from the propensity model from the outset. In establishing the degree of representation of this final list of variables to be included, external information should be used to validate the selection of covariates.5,6 This may include expert opinions and sensitivity analyses.7 Ensuring good model-fitting Propensity scores are estimated using logistic regression. As with all models, careful model-fitting is essential to assure the reliability of the resultant propensity score estimates. This includes attention to co-linearity among the covariates, examination of residuals to confirm whether effects are linear or non-linear, and consideration of estimates with large standard errors. These are remedial steps for an attentive statistician but, unfortunately, the ease with which statistical models may be fit in modern software packages unintentionally discourages these very steps. Insisting on careful model diagnostics is critical in building a propensity model, as a poor-fitting model can actually obscure treatment differences rather than clarify them.

Sometimes, propensity models do not fit well. This may be true when selection bias exists but is not measured directly or indirectly by observed variables. It may also be true when selection bias does not exist. It is rarely possible to determine which of these cases is true when a propensity model explains little variation in treatment selection.

However, a propensity score should not be retained simply because stakeholders expect to see one; neither should predictors that fail to explain variance in treatment selection be kept in a propensity model. Insistence on careful model-fitting gives researchers greater confidence in rejecting a propensity model when it does not add value to the analysis.8 fIGURE 2: GRAPHICAL PLOTTING Of PROPENSITY SCORES DEMONSTRATES OVERLAP BETWEEN DIffERENT TREATMENT GROUPS Hypothetical propensity score distribution

35 30 Study Population (1 X103)

Understanding the implications of score distribution A very important factor in deciding the suitability of applying PSM in study design or analysis is the ability to decipher the messages conveyed by the distribution of the propensity scores.

Graphical plots of propensity scores of comparison groups help to show the level of overlap based on the baseline covariates (Figure 2).

25 20 15 10 5 0

0

0.1

0.2

0.3

0.4

0.5

0.6

Treatment A

0.7

0.8

0.9

1

Treatment b

Minimal overlap of propensities score distribution in dierent treatment groups may be due to extensive selection of treatment, or suggest critical variables are not used to estimate propensity scores.

PAGE 38

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

PROPENSITY SCORING | INSIGHTS

PSM is a critical tool for strengthening stakeholder trust in analyses of RWE but attentive and responsible use is critical to maintain the perceived value of this technique.

A propensity score graph with minimal overlap of the treatment groups suggests extensive selection of treatment; such a strong selection bias may limit the ability to control for this bias in the ultimate outcomes models. However, such distribution may also suggest that the PSM does not have the most critical variables required.

While nearest-neighbor matching (pairing or matching observations with very similar propensity scores) remains a popular application of propensity scores,9 other applications may also control for selection bias. Regression adjustment by using the propensity score in the outcome model has the advantage of allowing all study observations to be used rather than discarding some without a clear nearest-neighbor match. In addition, propensity score matching within caliper would stratify analyses across the range of observed propensity scores which can be useful when a clear hypothesis exists on how treatment propensity may influence outcomes across the study population. Matching techniques may be helpful, but they also risk lending a sense of overconfidence in the degree of control in making study populations look subjectively more like those from a clinical trial. Matching techniques also discard substantial covariance in the original population when strong selection biases are detected. As with model-fitting, the application of control techniques should be driven as much as possible by a priori hypotheses and careful consideration of the likely influence of selection bias on outcomes.

CONCLUSION While PSM is a simple and intuitive approach to adjusting for confounding, an unobservant researcher may err in the application of the method or indeed interpretation of the evidence. The reliability of PSM is dependent on the appropriate selection of covariates and specification for the estimation of propensity score by regression model.

Furthermore, the way in which the PSM is used in a regression model of the outcome should be selected with careful attention to the level of precision it implies. PSM is a critical tool for strengthening stakeholder trust in analyses of RWE but attentive and responsible use is critical to maintain the perceived value of this technique and of the data to which it is applied.

•

1

brunelli SM, Rassen JA. Emerging analytical techniques for comparative effectiveness research. Am J Kidney Dis. 2013 Jan; 61(1):13-7 Rosenbaum PR, Rubin Db. The central role of the propensity score in observational studies for causal effects. biometrika. 1983; 70(1):41–55 3 Patrick AR, Schneeweiss S, brookhart MA,Glynn RJ,Rothman KJ, Stürmer T. The implications of propensity score variable selection strategies in pharmacoepdemiology: An empirical solution. Pharmacoepidemiol Drug Saf. 2011 Jun; 20(6):551-9 4 brooks JM, Ohsfeldt RL. Squeezing the balloon: Propensity scores and unmeasured covariate balance. Health Serv Res. 2012 Dec 6. doi: 10.1111/1475-6773.12020 5 Seeger JD, Kurth T, Walker AM. Use of propensity score technique to account for exposure-related covariates. An example and lesson. Medical Care. 2007 Oct; 45 (10 Suppl 2):S143-8 6 Westreich D, Cole SR,Funk MJ, brookhart MA, Til Stürmer. The role of the c-statistics for propensity score models. Pharmacoepidemiol Drug Saf. 2011 March; 20(3):317-320 7 Li L, Shen C, Wu, Li X. Propensity score-based sensitivity analysis method for uncontrolled confounding. Am J Epidemiol 2011 Aug 1; 174(3):345-53 8 Faries D, Peng X, Pawaskar M, Price K, Stamey JD, Seaman JW Jr. Evaluating the impact of unmeasured confounding with internal validation data: an example cost evaluation in type 2 diabetes. Value Health 2013 Mar-Apr; 16(2):259-66 9 Austin PC. A critical appraisal of propensity-score matching in the medical literature between 1996 and 2003. Stat Med. 2008 May 30; 27(12):2037-49 2

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 39

INSIGHTS | ObSERVATIONAL STUDY DESIGNS

As new realities in the healthcare landscape redefine the information required for market access, the use of observational research is increasingly relevant. Understanding observational study designs, their respective benefits and biases is essential to ensuring robust and rigorous data collection and real-world analyses.

The authors NĂşria Lara Surinach, MD, MSC is Principal RWE Solutions & HEOR, IMS Health Nlara@es.imshealth.com

Charles Makin, MS, MBA, MM, BS is Principal RWE Solutions & HEOR, IMS Health Cmakin@us.imshealth.com

PAGE 40

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

ObSERVATIONAL STUDY DESIGNS | INSIGHTS

Study designs in observational research Ensuring credible and efficient evidence generation Randomized controlled trials (RCTs) have traditionally remained the gold standard for safety and efficacy information, thus having the greatest impact on market access.

However, recent developments have accelerated the demand for greater transparency around the real-world performance of pharmaceutical products in a non-controlled environment.

The Affordable Care Act of 2010 (ACA) in the US, the implementation of the AMNOG law in Germany, and the move towards value-based pricing in the UK, have all indicated new models for payment and delivery of care, necessitating the generation and synthesis of better evidence on effectiveness and comparative efficiency.

This new direction has seen most pharmaceutical manufacturers rethinking their go-to-market strategy, with economic and patient-reported outcomes assuming greater significance in addition to clinical endpoints. At the same time, there is an increased push towards personalized medicine and targeting patient subpopulations that would benefit most from a drug (rather than a one-size-fits-all approach). With a stronger focus on external validity and real-world usage, this research – variously referred to as real-world evidence (RWE), health technology assessment (HTA) or observational research – presents its own set of challenges that are distinct from traditional clinical research.

In order to better collect and interpret increasingly important observational data as the basis for efficient decisionmaking, it is imperative to understand the various ways it may be collected in a scientifically rigorous manner. Some of the more commonly-used study designs, their associated selection biases, and approaches to dealing with that bias, are discussed below.

CASE-CONTROL STUDIES A case-control study is a type of observational analytic epidemiological investigation where subjects are selected on the basis of whether they do (cases) or do not (controls) have a particular outcome of interest. The groups are then compared on the proportion having a history of an exposure (eg, risk factor, preventive treatment, etc) or a characteristic of interest. As shown in Table 1, this type of study has the benefit of enabling relatively fast and inexpensive investigation of multiple exposures. However, it also has a number of disadvantages, including a tendency towards selection, observer and recall bias.

Potential for selection bias in case-control studies Case-control studies have a control group which serves to represent the reference population, where cases come from. According to Schlesselman, this control group “is intended to provide an estimate of the exposure rate that would be expected to occur in the cases if there were no association between the study disease and exposure”.1

TABLE 1: ADVANTAGES AND DISADVANTAGES Of CASE-CONTROL STUDIES

Advantages

• Can investigate multiple exposures • Can be relatively quick and cheap • Good for rare outcomes • Prone to selection, observer and recall bias

Disadvantages

• Time sequence of events difficult to ascertain

• Not good for rate exposures • Limited to one outcome • Cannot estimate incidence

continued on next page

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 41

INSIGHTS | ObSERVATIONAL STUDY DESIGNS

The ideal case control study with no selection bias is one where: 1. There is a clearly defined population (reference population) 2. All cases in that population are included in the study 3. Controls are a random sample of that population The potential for selection bias in case-control studies is particularly great in the common situation where cases and controls are drawn exclusively from hospitals or clinics. The main factors that drive the use of hospital rather than community controls are: Logistical reasons (it is easier and cheaper) To reduce recall bias To define population as “hospital users”

• • •

The problem with using hospital controls is that individuals in this setting typically do not represent the population of potential hospital users, but rather those who are sick. As such, they tend to be poorer, heavier smokers and drinkers, and living in worse conditions than the population of potential hospital users. Thus, should one of the factors that is over-represented in hospital controls be the exposure of interest, selection bias is introduced.

Dealing with selection bias in case-control studies As already noted, selection bias is likely to be less of a problem in population-based case-control studies where the cases are sampled from all incident cases in a defined population – the controls being sampled at random from the same population. The defined population is often, but not necessarily, a geographic area and period of time.

If cases are identified through a hospital or clinic, the use of neighborhood controls may be preferable to controls drawn from other patient groups in the hospital or clinic. However, this would not be the case if the probability of being hospitalized with the disease of interest was related to the exposure of interest. In this situation, the most appropriate source of controls may be individuals who are hospitalized with other diagnoses, where the probability of being admitted to hospital was similarly related to exposure.

Rather than trying to identify the perfect control group, some researchers choose to select controls from more than one source. The logic of this strategy is the unlikelihood that the effects of selection bias resulting from the use of two separate control groups would be identical. On this basis, it is reasonable to feel relatively confident that a major selection bias has been avoided if the estimated relative risks are the same using the two different groups.

However, the conclusion reached is less clear if the estimates differ.

COHORT STUDIES Cohort studies begin with defining a group of people according to their exposure (eg, treatment, risk factor, intervention, etc) status. These groups are then followed up over time to see who develops the outcome of interest.

As shown in Table 2, while these studies can be time consuming and expensive, if appropriately designed they can allow for the examination of multiple exposures.

Potential for selection bias in cohort studies Selection bias tends to be less of a problem in cohort studies. This is mainly because it is usually easier to see when bias may occur, which is typically when the exposed and unexposed groups are drawn from different populations.

The simplest form of selection bias in a cohort study is when completeness of follow-up or case ascertainment differs between exposure categories. In general, this problem is dealt with pragmatically by treating study findings with extreme caution if follow-up in any group is below some arbitrary level (eg, 80%), when differential case ascertainment could seriously bias results.

Another form of selection bias in cohort studies occurs when comparisons are made between disease rates in the study cohort and disease rates in some external standard population. Selection bias is introduced if membership of the exposed cohort is partly dependent upon health – which in itself may be related to the presence or absence of the disease being studied. This type of health-related selection is frequently encountered in occupational studies where the people in the study (exposed) cohort (eg, miners) are, by definition, relatively healthy because they are in employment.

PAGE 42

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

ObSERVATIONAL STUDY DESIGNS | INSIGHTS

Thus, the mortality or morbidity rates in the occupational cohort are (initially at least) lower than those seen in the population as a whole which includes those people too ill to work. This bias is known as the healthy worker effect.

Two approaches may be used to address the healthy worker effect: either make all comparisons internal to the employed study population, or make comparisons with an external standard which is composed of employed people.

TABLE 2: ADVANTAGES AND DISADVANTAGES Of COHORT STUDIES

• Incidence can be measured • Time sequence ascertained (exposure Advantages

can be measured before disease onset)

• Rare exposures can be investigated if cohort groups appropriately selected

• Multiple outcomes can be studied CASE CROSSOVER DESIGN In the case crossover design, each case acts as its own control, being the control period that was previous to the exposure of interest.2 As this type of study is self controlled, confounding factors that are stable over time, such as genetics, are removed. This design is appropriate for assessing acute effects of transient exposures.

As in a case-control study, the first step is to identify all cases (those with the outcome of interest) and assess the prevalence of exposure before the outcome occurred.

With each case serving as its own control, this creates a separate observation period containing the same variables except for the exposure of interest. It is important that the control time period is the same length as the case period.

• Time consuming and expensive • Losses to follow-up can cause Disadvantages

serious bias

• Ascertainment of outcome may be influenced by knowledge of exposure status • Classification of individuals (exposure or outcome status) can be affected by changes in diagnostic procedures, natural changes over time, etc • Not good for rare outcomes

Clearly, the main advantages of this study design are the fact that there is no need to select controls, and the ability to assess short-term reversible effects. However, there are some limitations which can reduce its efficiency. These are mainly due to the fact that only cases with discrepant exposure history can contribute information to the analysis. It is also important to take into account that, while the design avoids confounding by factors that are stable over time, it can still be confounded by factors that vary over time.

COMBINING PROSPECTIVE OBSERVATIONAL DATA COLLECTION WITH RETROSPECTIVE DATA Rather than bucketing data-collection methods as prospective or retrospective, an efficient way of addressing research and business problems is to combine both data streams when possible. Data collected prospectively are often the most comprehensive, being collected with a specific research question in mind. However, there is an attendant time and resource investment. by supplementing prospective data with the correct retrospective data, it is possible to reduce collection timelines and cost without compromising the integrity of the research.

IMS has often combined retrospective medical record abstraction with patient surveys. While medical records provide accurate and up-to-date clinical information on the patient, the surveys provide insights into patient-reported outcomes (PROs) such as health-related quality of life (HRQoL), activities of daily life (ADL), preferences, and the indirect burden of a health condition. More recently, IMS has also combined primary data collection with information from its databases, linking PROs with, for example, data on adherence and persistence from prescription claims databases. With research demonstrating that self-reported medication adherence generally overstates actual adherence, obtaining a measure of refill adherence from prescription records provides a more accurate means to assess refill rates and understand the impact of adherence interventions.

As the importance and frequency of observational research increase, so will scrutiny on the appropriateness of the methods chosen. Using the most suitable study design for a research question and accounting for its inherent biases will not only make the study more credible with the intended audience, it will also address the research and business need more efficiently.

•

1 2

Schlesselman JJ, 1982. Case-control studies: Design, conduct, analysis. New York: Oxford University Press Maclure M. The case-crossover design: A method for studying transient effects on the risk of acute events. Am J Epidemiol. 1991; 133:144-53

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 43

INSIGHTS | HTA IN LATIN AMERICA

Growing HTA requirements in Latin America (LATAM) are increasing the need for real-world data to support the inclusion of drugs and devices in national and local formularies. Along with new challenges for HEOR are opportunities to facilitate use of this evidence in the fast evolving markets of the region.

The author RenĂŠe JG Arnold, PHARMD, RPH Principal RWE Solutions & HEOR, IMS Health renee.arnold@us.imshealth.com

Page PAGE144

IMS REAL-WORLD HEALTH ECONOMICS AND OUTCOMES RESEARCH IMS EVIDENCE SOLUTIONS & HEOR

HTA IN LATIN AMERICA | INSIGHTS

Evolution of HTA and pharmacoeconomic analyses in Latin America The path to value-based healthcare In common with most developed and developing countries around the world, the healthcare systems of Latin America (LATAM) face the challenge of balancing escalating costs with the need to achieve quality improvements. At the same time, an increasingly chronic disease burden and growing demand for new, often more expensive, medicines is adding a further layer of complexity.

Reflecting these trends, the evolution of the pharmaceutical market has been characterized in recent years by more pronounced use of generics, often first-line, as a tool for cost containment, the establishment of national and local drug formularies, and greater employment of price references for trading internationally.

Throughout the region, a focus on approaches that enable more efficient, value-driven healthcare, while broadening access and encouraging innovation, has seen more countries turning to the use of health technology assessment (HTA) as part of the strong push for healthcare reform. Supporting this move is the recent launch of RELACSIS (Latin American and Caribbean Network for the Strengthening of Health Information Systems) by the Pan American Health Organization (PAHO), which is actively promoting HTA in the region. Among the key stated goals of this leading initiative are the proposal of standards for producing higher quality, more reliable and timely information, the generation and sharing of practices, lessons and knowledge, promotion of the monitoring and evaluation of national information systems, and cooperation between countries.1

HEALTH TECHNOLOGY ASSESSMENT HTA has been defined as “a form of policy research that examines short- and long-term consequences of the application of a healthcare technology. Properties assessed include evidence of safety and efficacy, patient-reported outcomes, real-world effectiveness, cost and cost-effectiveness as well as social, legal, ethical and political impacts”.2 The adoption of HTA principles, including incorporation of economic information into these evaluations, differs significantly by country.

Various factors influence the use of HTA, both in LATAM and globally. Surprisingly, perhaps, market size is not necessarily associated with greater use of HTA and health economic (HE) information. Three of the largest global economies in terms of GDP – the US, China and Japan – do not possess the most sophisticated HTA processes. Nor do countries with the highest healthcare costs per capita (eg, Switzerland, Norway, US). However, greater use of HE information does appear to correlate with the presence of guidelines for submission, HTA requirements and centralized decision making. Factors which appear to have a negative association are complexity of the healthcare system (both high and low complexity may reduce use of HE), poor data accessibility and limited availability of trained individuals to conduct and interpret studies.

In most emerging markets, mandatory and recommended HE evidence requirements are increasing (Figure 1). In LATAM, a growing number of countries have been moving towards some form of HTA process or actively institutionalizing HTA. Argentina, brazil, Chile and Mexico now have formal HTA agencies in place which are part of INAHTA (International Network of Agencies for Health Technology Assessment). Colombia, Costa Rica, Cuba, Paraguay, Peru and Uruguay are also following in a similar direction and are among a number of nations in the region that have joined REDETSA – the Health Technology Assessment Network of the Americas, established in 2011 through PAHO to promote and strengthen HTA processes across the Americas. continued on next page

OUTCOMES - Issue 1 ACCESSPOINT • VOLUME 3 ISSUE 6

Page45 1 PAGE

INSIGHTS | HTA IN LATIN AMERICA

MARKERS OF PROGRESS An indication of the progress being made towards HTA in LATAM can be seen in the growing use of pharmacoeconomic evaluations as a decision making tool in a number of markets. In Mexico, for example, they are mandatory for listing in the national formulary.

In brazil, they are now part of the pricing process for innovative patented drugs, reflecting the country’s much stricter approach to reimbursement following the recent creation of its new HTA agency, CONITEC (National Commission for Incorporation of Technologies in the Unified Healthcare System).

In Argentina, economic studies are recommended for all active ingredients included in the national formulary and hospital listings. And in Colombia, after many years in the making, the establishment of a national HTA agency (IETS) is finally now coming to fruition.

fIGURE 1: MANDATORY AND RECOMMENDED HEALTH ECONOMIC EVIDENCE REQUIREMENTS ARE INCREASING IN EMERGING MARKETS

Submission guidelines Thailand

Israel Poland Colombia

Taiwan

PE guidelines Cuba

Mexico brazil S. Africa PE recommends

Russia China

YEAR

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

Sources: www.ispor.org; IMS Consulting analysis PE: Pharmacoeconomic

Within each of these countries, different payers have different perspectives and needs: national formularies are typically focused on cost-effectiveness analysis and adaptations; those at the local level are principally concerned with the immediate financial implications of treatment as demonstrated through budget impact analyses.

HTA DEVELOPMENT In the drive to achieve more efficient use of healthcare resources, governments typically seek tools to help them determine the value of innovations and reimburse accordingly, often through the use of national, independent HTA agencies, eg, CONITEC and NICE (National Institute for Health and Care Excellence) in England and Wales. As they shift from a position of covering drugs as and when approved, to questioning the value of innovative new technologies and ultimately managing treatment algorithms and patient access to therapies, they are increasingly reliant on real-world data. Commensurate with requirements and emerging market value growth, HTA evidence in LATAM has been consistently increasing in recent years (Figure 2).

CHALLENGES OF HTA IN DEVELOPING COUNTRIES Many challenges exist in fulfilling the requirements of HTA processes in LATAM, not least: the lack of reliable data and information sources; the scarcity of theory studies to conduct decision making; the shortage of skills and conceptual knowledge to appraise the results of analyses; poor dissemination of findings; and the high degree of institutional fragmentation. However, by their very nature, these hurdles also present considerable opportunities to optimize the use of HE evidence in the region. They are by no means unique to LATAM and as the HTA agenda moves forward, examples of innovative practices and solutions employed in other selected markets can offer useful direction for overcoming the challenges of developing value evidence throughout the product lifecycle while supporting efforts at the country level to reach evidence-based decisions in accordance with the local situation.

The ability to communicate research findings to relevant stakeholders and the broader healthcare community is essential to raising awareness and strengthening the local evidence base for decision making.

Page PAGE PAGE146 46

IMS IMS HEALTH AND EVIDENCE OUTCOMES RESEARCH IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR HEORECONOMICS & REAL-WORLD SOLUTIONS

HTA IN LATIN AMERICA | INSIGHTS

fIGURE 2: COMMENSURATE WITH REQUIREMENTS AND EMERGING MARKET VALUE GROWTH HTA EVIDENCE HAS INCREASED IN LATIN AMERICA

Development and implementation of pharmacoeconomic (PE) guidelines in Latin America Argentina Colombia Guatemala Uruguay Venezuela

Clinical and cost-effectiveness – through the application of PE guidelines being considered to be, and could well be, incorporated within the existing set of policy instruments. For example, Colombia established HTA Institute (IETS) in 2011.

Mexico

A set of PE guidelines has been developed and a law established requiring a PE dossier before inclusion of a technology in the national formulary (NF). After inclusion in the NF, each healthcare institution can then decide whether or not to purchase based on its cost, budget availability, previous experience with the technology at the institution concerned, and priority of disease.

Chile

Clinical guidelines have been established in several different pathologies based on a prioritization process.

Brazil

The Latin American country with the most experience in implementing PE guidelines.

Implementation is the responsibility of ANVISA and the Ministry of Health (MoH). PE is applied in pricing decisions for new drugs. CONITEC, the HTA assessment arm of the MoH ,requires manufacturers to provide cost-effectiveness and HTA evaluations in support of pricing.

Source: Augustovski f, Melendez G, Lemgruber A, Drummond M. Implementing pharmacoeconomic guidelines in Latin America: Lessons learned. Value Health 2011; 14 (Suppl. 1): S3-7

OVERCOMING THE CHALLENGES: EVIDENCE GENERATION 1. Developing credible data: Saudi Arabia and US Approaches to overcoming the lack of robust, reliable data have seen the use of partnership collaborations by many organizations to generate an evidence base. These include several recent examples in the Middle East and US:

• GE Healthcare partnership with Saudi Ministry of Health on Healthymagination Launched in 2009 with the goal of providing greater access to healthcare services in Saudi Arabia, this initiative included a consumer campaign of health and wellness research to raise awareness of healthy living and early diagnosis.

• National Headache Foundation/Ortho McNeil Neurologics AMPP study The American Migraine Prevalence and Prevention (AMPP) Study – the largest ever analysis of headache sufferers – is based on data compiled from 2004 through 2009 examining nearly 163,000 Americans ages 12 and older, selected to be representative of the US population. It has yielded extensive data on symptoms and treatment patterns.

• AstraZeneca/HealthCore real-world evidence (RWE) data collaboration (US) Announced in 2011, this collaborative agreement between AstraZeneca and HealthCore has the goal of conducting real-world studies designed to determine how to most effectively and economically treat disease, with a special emphasis on chronic illnesses. continued on next page

OUTCOMES - Issue 1 ACCESSPOINT VOLUME ISSUE ACCESSPOINT • •VOLUME 3 3ISSUE 56

Page47 1 PAGE PAGE

INSIGHTS | HTA IN LATIN AMERICA

• AstraZeneca/IMS Health collaboration to expand RWE (Europe) Announced in 2012, this collaborative agreement between AstraZeneca and IMS Health has the goal of advancing the use of RWE, based on observational and retrospective studies throughout Europe, to inform the delivery of effective and cost-efficient healthcare.

2. Improving conceptual knowledge: Thailand In the absence of the established capabilities and means for determining optimal healthcare resource allocation, external expertise can play a key role in helping healthcare systems create the evidence they need to make locallyrelevant decisions.

An example can be seen in the case of cervical cancer vaccine coverage in Thailand. Under pressure from the multinational drug industry, international agencies, and patient and professional groups, the Thai FDA licensed two HPV vaccines. However, when it commissioned an independent, expert body, HITAP (Health Intervention and Technology Assessment Program), to “generate reliable and relevant information” using local Thai data, it was able to ascertain that the vaccine’s ICER (incremental cost-effectiveness ratio) was three times Thailand’s GDP per capita.

Although this metric (3x GDP/capita) may be considered cost-effective in emerging markets, in this case, based on this transparent, consultative scientific approach, the vaccine was not included in the universal coverage program.

OVERCOMING THE CHALLENGES: LESSONS LEARNED 1. Broadening dissemination of findings The ability to communicate research findings to relevant stakeholders and the broader healthcare community is essential to raising awareness and strengthening the local evidence base for decision making. Experience has shown that engaging stakeholders earlier in the product lifecycle is critical, with opportunities to create pull-through using non-branded modeling tools.

Examples include, in early development, the use of exploratory analyses to determine the value of a hypothetical new intervention or change in practice; and during the pre-launch phase, raising disease awareness through the use of, eg, disease state models, adherence modeling and demonstration of economic burden. Analyses of patient flow in a disease can help to identify areas of opportunity and unmet need in a local treatment path; the use of longitudinal databases can serve to inform treatment patterns and resource-use issues in clinical practice; and patient-reported outcomes (PROs) can be used to define influences of a disease on patient quality of life, thereby giving a baseline to measure drug effects in orphan or other illnesses.

The capacity to conduct rigorous outcomes research studies has been dramatically expanded in recent years with the development of RWE platforms for data exchange, such as IMS’ PharMetrics Plus™. by allowing the seamless integration of multiple patient-level data assets, these can enable a complete and holistic picture of the patient journey across settings of care.

2. Building institutional cohesion Efforts to improve decision making and healthcare at the local level have encompassed various partnerships and collaborative efforts in brazil, Germany, Spain, UK and US. Examples include the establishment of Primary Care Trust alliances for sub-regional access decision making in the UK, industry partnerships with private insurers to demonstrate cost savings from outpatient drug coverage in brazil, and the development of interactive hospital/clinic perspective tools to quantify treatment-related costs associated with current and future practice in the US. 3. Raising the quality of theory studies In markets where cost per QALY is not a standard measure, analyses can be adapted to inform current decision making (Figure 3). For example, in emerging markets, such as China, the metric used to identify cost-effectiveness is <3x GDP. As the move towards HTA continues to gather momentum in the markets of LATAM, companies looking to participate in their evolution towards evidence-based decision making can benefit from the lessons learned in adapting global models to other emerging markets.

PAGE 48

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

HTA IN LATIN AMERICA | INSIGHTS

fIGURE 3: ANALYSES CAN BE ADAPTED TO INfORM DECISION MAKING IN MARKETS WHERE COST/QALY IS NOT STANDARD WHAT IS COST IMPACT?

• Show budget impact for national payer

UK

• Show budget impact for national payer

S. KOREA

trial analysis of healthcare • Within resource use adapted in US

US

WHAT IS RELATIVE VALUE? CEA model developed for NICE • submission in UK. Leveraged within trial analysis of economic data. benchmark ICER 30K/QALY in UK CEA model adapted in S. Korea with • Korean-specific parameters where possible.

No official benchmark ICER in S. Korea; anecdotal range from 10M -30M KRW, depending on perceived urgency to treat

• No CEA model requirement in US

These provide a number of important pointers in the steps to developing robust and relevant real-world data in the region: 1. Create as sophisticated an analysis as country level, budget and time frame will allow, taking into consideration: Level: Sophistication of country-specific pharmacoeconomic/HTA guideline Budget: Consider using local labor Time frame: Prepare key opinion leader (KOL) expectations in advance

• • •

2. Collect local literature (both in English and translated from local language if not a native speaker) about practice patterns in country of interest 3. Retrieve and translate international disease treatment guidelines, if available 4. Use local cost sources (depending on model perspective: country, region, large hospital, local KOLs) 5. For budgetary impact models, use local market access (market share) figures for baseline

CONCLUSIONS In summary, health expenditure, particularly in the public sector, is increasing and governments have the challenge to improve health outcomes with a limited budget. As such, the requirements for HTA are increasing and evolving in LATAM in terms of their sophistication and RWE components, to enable governments to meet this challenge using objective evidence. Indeed, LATAM countries are taking their cues from more seasoned HTA authorities such as NICE and PAHO is actively promoting HTA in the region. In several countries, cost-effectiveness/HE analysis is already an essential part of the access process; in others, it is increasing in importance. Outcomes research, using RWE from the region/country, is also particularly important, given the lack of information regarding treatment paradigms, burden of illness and unmet medical need.

•

1 2

Strengthening Health Information Systems - RELACSIS. Accessed 21 April at http://new.paho.org International Society for Pharmacoeconomics & Outcomes Research (ISPOR). Healthcare Cost, Quality and Outcomes: ISPOR book of Terms, Laurenceville NJ: 2003

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 49

INSIGHTS | AMNOG PATIENT-RELEVANT ENDPOINTS

Two years after AMNOG brought new stipulations for drug assessment in Germany, pointers are emerging on the practical implications for demonstrating additional benefit. Analysis of dossiers reviewed to date in two significant therapeutic areas provides important insights into the hard patient endpoints that are key to successful market access in this country.

The Authors Dirk Eheberg, MPH is Senior Consultant RWE Solutions & HEOR, IMS Health Deheberg@de.imshealth.com

Doreen Bonduelle, FH is Director RWE Solutions & HEOR, IMS Health Dbonduelle@de.imshealth.com

Stefan Plantรถr, PHD, MBA, MSC is Director RWE Solutions & HEOR, IMS Health Splantoer@de.imshealth.com

PAGE 50

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

AMNOG PATIENT-RELEVANT ENDPOINTS | INSIGHTS

AMNOG: Additional benefit demands patient relevance Why the right study endpoints are key Since the introduction of AMNOG (Arzneimittelmarktneuordnungsgesetz; Law on the Reorganization of the Pharmaceutical Market) in January 2011, manufacturers in Germany have been required to submit a ‘benefit dossier’ when launching a new active ingredient. Ensuring the right patient-relevant endpoints can be decisive for the Federal Joint Committee (G-bA) resolution on the additional benefits conferred.

Preparation for the AMNOG process (see Figure 1, page 55) should therefore start as early as possible. Incorporating patient-relevant endpoints into the study design while planning the clinical program is the most promising approach.

However, this option is not always straightforward as international clinical studies must meet the requirements of not only marketing authorization agencies but also a rapidly growing number of health technology assessment (HTA) organizations from different countries. The degree to which these HTA bodies overlap in their requirements, especially with regard to endpoint design, is relatively small,1 making selection of the right study design even more complex.

ENSURING PATIENT-RELEVANCE One of the most important arguments to bring forward within the AMNOG dossier is the patient relevance of the clinical study endpoints. The G-bA clearly states in §3 of Chapter 5 of the Code of Procedure (G-bA Verfahrensordnung) that patient-relevant endpoints are only those that can be classified into one of the following categories: Improvement in the state of health Shortening of duration of illness Extension of survival Reduction of side effects Improvement in quality of life (§3 par 1)

• • • • •

The Institute for Quality and Efficiency in Healthcare (IQWiG) defines patient-relevant endpoints in its paper on methodology (IQWiG Methodenpapier 4.0) as any endpoint that directly relates to how a patient feels, is able to function or survives.

However, there remains uncertainty around the term ‘patient-relevance’ and the criteria that have to be met in order for IQWiG to accept an endpoint as patient-relevant. Perhaps the greatest public misconception is that therapeuticallyrelevant equals patient-relevant. The G-bA and IQWiG have demonstrated quite clearly that therapeutic relevance needs to be translated into patient relevance and that this translation is not given by concept. A therapeuticallyrelevant endpoint (eg, controlling blood pressure in patients with coronary heart disease), is not per se patient-relevant.

In order to appreciate this difference, it is necessary to understand the concept of a surrogate parameter.

There remains uncertainty around the term ‘patient-relevance’ and the criteria that have to be met. continued on next page

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 51

INSIGHTS | AMNOG PATIENT-RELEVANT ENDPOINTS SURROGATE PARAMETERS Surrogate parameters measure directly a parameter that is an indicator for a (patient-relevant) event which is either hard to track (eg, rare events) or is a long-term consequence. As such, surrogates indirectly measure patient-relevant endpoints and are commonly used as mediators between therapeutically-relevant endpoints and patient-relevant endpoints. However, surrogate parameters are only to be considered if they are validated for the indication and the substance in question. Furthermore, surrogate parameters will be ignored by the G-bA and IQWiG if the endpoint they replace can be derived from the clinical study reports.

A well-known illustration of this concept is in the area of overall survival/progression-free survival. IQWiG and G-bA consider progression-free survival as a non-validated surrogate for overall survival. In its report on the validity of surrogate endpoints in oncology2, IQWiG states on the one hand that “a mere correlation between a surrogate and patient-relevant endpoint is not sufficient for validation”, but on the other hand leaves it open in stating that “there is no universal measure nor common estimate nor threshold which is to be exceeded to gain validity for surrogates”. For example, the question of whether endpoints relating to tumor-progression are surrogates has been under discussion.

One way to argue for their patient-relevance can be seen in the IQWiG report on allogeneic stem-cell transplantation in Hodgkin’s lymphoma3. Here tumor progression is reported as a derivate of survival time, with the endpoints “progression-free survival” or – quoting IQWiG – “comparable endpoints” (eg, disease-free survival). Possible reasons could be that patients are younger, in the early stage of disease or have longer life expectancy.

In the benefit assessment for new products in oncology, overall survival is the only widely accepted patient-relevant endpoint.

EVIDENCE TO DATE The practical implications of guidance regarding patient-relevant endpoints can best be considered in the light of evidence from AMNOG evaluations to date. A recent IMS study, drawing on the IMS AMNOG database, analyzed all submitted (as of 15 March 2013) benefit dossiers in the indications of infectious disease and oncology, for which at least the first written statement from IQWiG for complete dossiers, and from the G-bA for orphan drug dossiers, were available. The analysis (shown in Table 1) demonstrates that apart from overall survival there is little overlap between the manufacturer, IQWiG and G-bA assessment of endpoints.

Infectious disease Dossiers in the indications HIV (rilpivirin) and chronic hepatitis C (telaprevir; boceprevir) included the endpoints ‘viral response’ and ‘sustained viral response’. IQWiG regarded both endpoints either as sufficient validated surrogates (rilpivirin, telaprevir) or as not validated surrogates (boceprevir). The G-bA accepted viral response as a sufficient validated endpoint and agreed with the pharmaceutical manufacturer that sustained viral response is a patientrelevant endpoint (without clarifying whether it is a validated surrogate endpoint or not a surrogate endpoint). All other endpoints, such as viral failure, relapse rate, rapid response or immunological response were marked by IQWiG as surrogate parameters or as redundant to the (sustained) viral response and were consequently not considered by the G-bA for the benefit assessment.

Oncology In oncology, the benefit assessment for new products follows the same pattern. Overall survival is the only widely accepted patient-relevant endpoint. In addition, endpoints regarding symptoms, pain (response and progression) or time to tumor-specific event (eg, skeletal events for abirateronacetat) were considered patient-relevant by IQWiG and the G-bA consecutively. Progression-free survival, tumor response, objective response and tumor progression were always marked as not validated surrogates for overall survival by IQWiG and were not considered for the benefit assessment by the G-bA.

PAGE 52

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

AMNOG PATIENT-RELEVANT ENDPOINTS | INSIGHTS TABLE 1: ANALYSIS Of AMNOG DOSSIER SUBMISSIONS IN INfECTIOUS DISEASE AND ONCOLOGY TO 15 MARCH 2013

Substance

Dossier type

Endpoint

Infectious Diseases (HIV; Chronic hepatitis C) Overall survival Full Boceprevir dossier Sustained viral response (SVR)

Rilpivirin

Full dossier

Statement from pharmaceutical manufacturer

Telaprevir

Full dossier

Full dossier

G-BA decision

No surrogate

No surrogate

No surrogate; patient-relevant

No surrogate

Surrogate; not formally validated

No surrogate; patient-relevant

Overall survival

No surrogate

No surrogate

No surrogate; patient-relevant

Viral response (viral load)

Patient-relevant surrogate

Sufficient validated surrogate; not necessarily Formally validated surrogate; patientpatient-relevant relevant for the present indication

Viral failure (effectiveness) No surrogate

Redundant; endpoint already considered by viral response

Not considered

Viral failure (resistance)

Redundant; endpoint already considered by viral response

Not considered

No IQWiG assessment: Incomplete dossier

No surrogate

Overall survival Rilpivirin triple combination

Written statement from IQWiG or G-BA

No surrogate Viral response (viral load) Patient-relevant surrogate Viral failure (effectiveness) No surrogate

No IQWiG assessment: Incomplete dossier

No surrogate; patient-relevant Formally validated surrogate; patientrelevant for the present indication Not considered

Immunological response: Surrogate CD-4-cell count

No IQWiG assessment: Incomplete dossier

Not patient-relevant

Overall survival

No surrogate

No IQWiG assessment: Incomplete dossier

No surrogate; patient-relevant

Sustained viral response (SVR)

Patient-relevant surrogate

Sufficient validated surrogate; not necessarily No surrogate; patient-relevant patient-relevant

Relapse rate

No surrogate

Surrogate; no detailed information on validation by PC; adequately taken into account by SVR

---

Rapid virologic response

No surrogate?

Possibly surrogate; not validated; not necessarily patient-relevant

---

Extended rapid virologic response

No surrogate?

Possibly surrogate; not validated; not necessarily patient-relevant

---

Fatigue

No surrogate

No surrogate

---

Overall survival Radiographic progressionfree survival Prostate-specific antigen response Time to PSA progression Time to first skeletal event Time to pain progression Overall survival Symptomatic

No surrogate

No Surrogate

No surrogate; patient-relevant

Surrogate

Surrogate; no sufficient explanation/validation Not considered

Surrogate

Surrogate; no sufficient explanation/validation Not considered

Surrogate No surrogate No surrogate No surrogate No surrogate

Surrogate; no sufficient explanation/validation No surrogate No surrogate No surrogate No surrogate

Not considered No surrogate; patient-relevant No surrogate; patient-relevant Not completed yet Not completed yet

Progression-free survival

No surrogate

Surrogate; not valid; not patient-relevant for morbidity or life quality

Not completed yet

Objective response rate

No surrogate

Surrogate; not valid; not patient-relevant for morbidity or life quality

Not completed yet

Overall survival

No surrogate

No surrogate

Not completed yet, but written statement represents G-BA point of view (Orphan drug)

Event-free survival

Surrogate

Surrogate; no valid surrogate for patientrelevant endpoint ‘overall mortality’

Not completed yet, but written statement represents G-BA point of view (Orphan drug)

Progression-free survival

Surrogate

Surrogate; no valid surrogate for patientrelevant endpoint ‘overall mortality’

Not completed yet, but written statement represents G-BA point of view (Orphan drug)

Objective response rate

Surrogate

Surrogate; unclear validity

Not completed yet, but written statement represents G-BA point of view (Orphan drug)

Remission rate for Bsymptomatic

No surrogate

No surrogate

Not completed yet, but written statement represents G-BA point of view (Orphan drug)

Proportion of patients with stem cell transplantation No surrogate after treatment

No surrogate (assessment uncertain, as endpoint was not included in the final assessment by the pc)

Not completed yet, but written statement represents G-BA point of view (Orphan drug)

Complete remission

No surrogate (partly accepted)

Not completed yet, but written statement represents G-BA point of view (Orphan drug)

No IQWiG assessment: Incomplete dossier

Oncology

Abirateronacetat

Axitinib

Brentuximab

Full dossier

Full dossier

Orphan drug

No surrogate

continued on next page

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 53

INSIGHTS | AMNOG PATIENT-RELEVANT ENDPOINTS TABLE 1: ANALYSIS Of AMNOG DOSSIER SUBMISSIONS IN INfECTIOUS DISEASE AND ONCOLOGY TO 15 MARCH 2013 continued Substance

Dossier type

Endpoint

Statement from pharmaceutical manufacturer

Written statement from IQWiG or G-BA

G-BA decision

Oncology ctd

Cabazitaxel

Crizotinib

Decitabin

Full dossier

Full dossier

Orphan drug

Eribulin

Full dossier

Ipilimumab

Full dossier

Pixantron

Full dossier

Overall survival

No surrogate

No surrogate

No surrogate; patient-relevant

Change in pain score

No surrogate

No surrogate

No surrogate; patient-relevant

Pain response

No surrogate

No clear surrogate, but redundant

Not considered

Pain progression

No surrogate

No clear surrogate, but redundant

Not considered

Mean AUC for pain score

No surrogate

No clear surrogate, but redundant

Not considered

Progression-free survival

Surrogate

Surrogate without details on validity

Not considered

Tumor progression

No surrogate

Surrogate without details on validity

Not considered

Tumor response rate

No surrogate

Surrogate without details on validity

Not considered

PSA progression

No surrogate

Surrogate without details on validity

Not considered

PSA response rate

Surrogate

Surrogate without details on validity

Not considered

Overall survival

No surrogate

No surrogate

Not completed yet

Progression-free survival

No surrogate

Surrogate; not validated; not patient-relevant Not completed yet

Objective response rate and associated endpoints (TTR (time to tumor response); DR (duration of No surrogate response); DCR (disease control rate))

Surrogate; not validated; not patient-relevant Not completed yet

Symptomatic

No surrogate

No surrogate

Not completed yet

Time to impairment

No surrogate

No surrogate

Not completed yet

Overall survival

No surrogate

No surrogate

Not completed yet, but written statement represents G-BA point of view (Orphan drug)

Response-related endpoints

No surrogate

Unclear wording by the G-BA; validity and patient-relevance of CR+CRp (+duration) conclusively not assessable; CR important prognostic factor

Not completed yet, but written statement represents G-BA point of view (Orphan drug)

Event-free survival

No surrogate

Surrogate; not validated

Not completed yet, but written statement represents G-BA point of view (Orphan drug)

Recurrence-free survival

No surrogate

Surrogate; not validated

Not completed yet, but written statement represents G-BA point of view (Orphan drug)

Hospitalization

No surrogate

Not completed yet, but written statement No surrogate; not necessarily patient-relevant represents G-BA point of view (Orphan drug)

Transfusions

No surrogate

No surrogate; not necessarily patient-relevant Not completed yet, but written statement represents G-BA point of view (Orphan drug)

Overall survival

No surrogate

No surrogate

No surrogate; patient-relevant

Overall survival

No surrogate

No surrogate

No surrogate; patient-relevant

Complications

No surrogate

No surrogate, but wrong category (belongs to Considered among side-eďŹ&#x20AC;ects side eďŹ&#x20AC;ects)

Overall survival

No surrogate

No IQWiG assessment: Incomplete dossier

No G-BA assessment: Incomplete dossier

Complete remission

No surrogate

No IQWiG assessment: Incomplete dossier

No G-BA assessment: Incomplete dossier

No IQWiG assessment: Incomplete dossier

No G-BA assessment: Incomplete dossier

Progression-free survival

Vandetanib

Vemurafenib

Full dossier

Full dossier

Overall survival

No surrogate

No IQWiG assessment: Incomplete dossier

Not completed yet

Biochemical response

Surrogate

No IQWiG assessment: Incomplete dossier

Not completed yet

Progression-free survival

No surrogate

No IQWiG assessment: Incomplete dossier

Not completed yet

Objective response rate

No surrogate

No IQWiG assessment: Incomplete dossier

Not completed yet

Time to pain progression

No surrogate

No IQWiG assessment: Incomplete dossier

Not completed yet

Overall survival

No Surrogate

No surrogate

No surrogate; patient-relevant

Progression-free survival

No surrogate

Surrogate; not validated

Not considered

Tumor response

No surrogate

Surrogate; not validated

Not considered

Change in pain score

No Surrogate

No surrogate

No surrogate; patient-relevant

Source: IMS Health AMNOG database

PAGE 54

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

AMNOG PATIENT-RELEVANT ENDPOINTS | INSIGHTS As the validation of a surrogate endpoint is a very technical process and IQWiG sets extremely high standards for quality, IQWiG and G-bA rarely differ. The analysis shows that, for infectious diseases, the G-bA was more likely to overthrow the IQWiG assessment, which stated in most cases that the (sustained) viral response either was a not validated or sufficient validated endpoint, but not necessarily a patient-relevant surrogate. No such difference in interpretation of the available data is discernible in the benefit assessments for oncology.

KEY LEARNINGS Several important messages emerge from these findings: pharmaceutical manufacturers should assess the patient relevance of endpoints in clinical studies in light of the HTA requirements; preference should be given to hard endpoints instead of surrogates; and when dealing with surrogate endpoints, special consideration should be given to the following dimensions:

• Accordance with an accepted methodology • Validation in the target indication in a population with comparable severity of the disease of robustness and the basis for generalization concerning correlation of surrogate and • Verification patient-relevant outcome Going forward, companies should continue to look for consistency in the decisions of the G-bA and reflect these decisions in their clinical study design as well as in the presentation of these endpoints within the benefit dossiers.

•

fIGURE 1: THE AMNOG PROCESS IN GERMANY Act on the Reform of the Market for Medicinal Products (AMNOG) introduced a • The benefit assessment process in 2011 Act created requirements for comparator-driven evidence and altered the pricing • The process to include discount negotiations AMNOG covers products already on the market, opening up the possibility • ofIn addition, price cuts to existing products

Institute for Quality and Efficiency in Health Care (IQWiG) Benefit Assessment

Commission possible

Report

Dossier

Hearing

Manufacturer

Market Launch

Additional benefit

Federal Joint Commitee (G-BA)

Federal Joint Commitee (G-BA)

Benefit Assessment

Benefit Assessment

(internet publication)

(Decision)

No additional benefit

Manufacturer’s price (set freely)

Market Launch

Manufacturer

6 months

Not accepted

Head association of the SHI scheme (GKV)

Arbitration panel

Decision

(eg, based on international prices)

Rebate Negotiations

Reference price not possible

FRP Reference price

3 months

No agreement

Agreement

Discounted ‘net’ price

Cost/benefit assessment

Decision

Retroactive

12 months

Institute for Quality and Efficiency in Health Care (IQWiG)

Discounted ‘net’ price

Valid until the end of the process

15 months

FRP = Fixed reference price

1

Neumann PJ, Drummond MF, Jonsson b, Luce bR, Schwartz JS, Siebert U, Sullivan SD. Are key principles for improved health technology assessment supported and used by health technology assessment organizations? International Journal of Technology Assessment in Health Care, 2010; 26(1): 71-78 2 IQWiG. Aussagekraft von surrogatendpunkten in der onkologie [Validity of surrogate endpoints in oncology]. IQWiG Rapid Reports – Commission No. A10-105, 2011 3 IQWiG Allogene stammzelltransplantation mit nicht verwandtem spender bei der indikation hodgkin-lymphom [Unrelated donor allogeneic stem cell transplantation for Hodgkin's lymphoma] Report N05-03F, 2010

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 55

INSIGHTS | MEDICAL DEVICES IN GERMANY

In size alone, Germany is one of the most attractive markets for medical devices. Recent reforms have extended the criteria for achieving market access in this innovative sector but understanding their dynamics is essential to harnessing the new potential. Lessons can be drawn from early experience of the regulatory changes.

The authors Roger-Axel Greiner, PHD is Senior Consultant RWE Solutions & HEOR, IMS Health Ragreiner@de.imshealth.com

Stefan Plantรถr, PHD, MBA, MSC is Director RWE Solutions & HEOR, IMS Health Splantoer@de.imshealth.com

Page PAGE156

IMS REAL-WORLD HEALTH ECONOMICS AND OUTCOMES RESEARCH IMS EVIDENCE SOLUTIONS & HEOR

MEDICAL DEVICES IN GERMANY | INSIGHTS

A new era for medical devices in Germany?

Understanding the impact of regulatory change Long established as a principal producer of high-quality medical equipment and diagnostics, Germany’s medical devices market reached an estimated US$23.2 billion in 2012 making it the largest in Europe and the third largest in the world next to the US and Japan. Although the local manufacturing industry is strong, about 75% of this market is accounted for by imports.1 Despite the appeal of the medical devices market in Germany, there are sectoral differences specific to the German healthcare system that directly impact the reimbursement of innovative medical procedures and products by the statutory health insurance (SHI) funds. For the vast majority of the German population, these funds are key to determining how the money is allocated and how services are provided in the healthcare system. With new regulations now coming into play, what do these dynamics mean for this highly innovative sector?

CURRENT REIMBURSEMENT OF MEDICAL DEVICES: A TALE OF TWO SETTINGS The peculiarity of the German healthcare system in its very clear division of inpatient and outpatient care in the SHI is reflected in two quite opposite approaches to reimbursing medical devices.

Inpatient sector The principle of “permission unless explicitly banned” (Erlaubnis mit Verbotsvorbehalt) refers to the provision of new methods in the inpatient sector and is defined in Social Code book V (SGb V) § 137c. This covers the assessment of diagnostic and treatment methods in the hospital. The decision to use an innovative method rests mainly with the hospital doctor. After uptake in the inpatient sector, the innovation is in most cases reimbursed according to one of the German diagnosis-related groups. Only if the Federal Joint Committee (FJC, Gemeinsamer bundesausschuss G-bA) has explicitly excluded the method following an evaluation is it not covered by SHI funds.

Outpatient sector Here, the principle is “banned until explicitly permitted” (Verbot mit Erlaubnisvorbehalt) according to § 135 SGb V where diagnostic and treatment methods are required to be of benefit, medical necessity and efficiency. Therefore, new methods must be assessed before they can be admitted to the catalogue of services. Medical procedures are only reimbursed in the outpatient sector if the FJC has made a positive decision.

In the case of a positive decision, the FJC publishes a directive recommending admission of the method to the Physicians' Fee Schedule (PFS, Einheitlicher bewertungsmaßstab EbM) catalogue, and the Valuation Committee (bewertungsausschuss) then determines the value of the new method relative to other methods already listed in the PFS. In the opposite case, if the FJC disclaims the innovative method, a directive is published adding the procedure to the exclusion list or “negative list” (in Annex II of the Guidelines on SHI-accredited Outpatient Methods) and the method is not covered by SHI funds.

NEW EXTENSION OF REIMBURSEMENT CRITERIA The law on SHI Care Structure (GKV-Versorgungsstrukturgesetz, GKV-VStG), introduced in § 137e SGb V in 2012, extends the reimbursement criteria of new methods in both the inpatient and outpatient sectors. This means that for methods of diagnosis and treatment whose benefits have not yet been sufficiently demonstrated but which reveal the potential for a required alternative treatment, the Federal Joint Committee (FJC, Gemeinsamer bundesausschuss G-bA) can, in future, decide on a directive to test. continued on next page

OUTCOMES - Issue 1 ACCESSPOINT • VOLUME 3 ISSUE 6

Page57 1 PAGE

INSIGHTS | MEDICAL DEVICES IN GERMANY Conditional coverage for inpatient sector briefly, the new § 137e SGb V grants the possibility that an application for testing a new method for diagnosis or treatment can be made to the FJC independently of the advisory procedure according to § 135 or § 137c SGb V. In future, in the inpatient sector, a new method without proven evidence will not be excluded from inpatient care until its potential has been evaluated. Only new methods that are recognizably harmful or inappropriate will be excluded without evaluation. The SHI will ensure coverage during the assessment period. In the outpatient sector, the admission of a new method in the ambulatory reimbursement system will continue to depend solely on the FJA decision after evaluation of benefit, medical necessity and efficiency.

Trial phase for evidence development What else is new with § 137e SGb V? In accordance with § 135 and § 137c SbG V, only impartial members of the FJC or organizations representing patients in the FJC were eligible to apply for the examination and assessment of a method, apart from the top associations of the care providers and the health insurances.

With § 137e SGb V, eligible applicants include, for the first time, manufacturers of a medical device and companies that have in any other way an economic interest in providing a new technical application at the expense of SHI. Moreover, the manufacturers have the opportunity to conduct a study which is in accordance with the FJC’s requirements, the costs of which will be partially funded on the basis of company size and revenues.

During the trial phase, the new method is already covered by the SHI funds and as soon as the scientific evaluation is completed, the FJC decides whether the method is definitely reimbursed and publishes a directive in the federal bulletin accordingly (bundesanzeiger).

From the perspective of the FJC, the new regulations for testing will improve the conditions for generating evidence which is needed for its decisions. Testing means temporary suspension of the benefit assessment and limited permission for providing the method under the directive for testing.

The FJC is responsible for the formal examination of the application for testing and commissions the impartial Institute for Quality and Efficiency in Health Care (IQWiG) to evaluate the content.

In summary, the important change (§ 137c in connection with § 137e SGb V) is the provision of a directive for testing before exclusion of an inpatient service. The new § 137e SGb V considers only innovative methods that show a potential benefit, whereas methods without potential are excluded (Figure 1).

fIGURE 1: SCHEMATIC fLOW Of METHOD EVALUATION fOR INPATIENT SECTOR EVALUATION CRITERIA: BENEFIT – MEDICAL NECESSITy – EFFICIENCy

In compliance with criteria

No sufficient evidence for benefit but studies ongoing

No sufficient evidence but showing potential

No evidence for benefit and no potential

Inclusion

Suspension

Testing for evidence due to § 137e SGB V

Exclusion

PRACTICAL APPLICATION: BENEFIT ASSESSMENTS OF PET AND PET/CT One interesting example of how the evaluation process actually maps for public reimbursement is IQWiG’s benefit assessments of positron emission tomography (PET), alone or in combination with computed tomography (CT).

IQWiG investigated the benefit of PET and PET/CT in 10 indications. According to the health technologies assessment methodology, the level of evidence of clinical studies has been proven, presenting randomized controlled trials (RCTs) and meta-analyses of RCTs with the highest level of evidence. RCTs have to be designed with the standard therapy as comparator in order to create the potential to directly demonstrate the additional benefit. The additional benefit results from the impact of the medical therapy on patient-relevant endpoints that are specified in the IQWiG Methodology paper 4.0: Improvement in the state of health Shorter duration of illness Extension of life Reduction of side effects Improvement in quality of life

• • • • •

PAGE 58

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

MEDICAL DEVICES IN GERMANY | INSIGHTS The results of the IQWiG benefit dossiers are sobering (Table 1). In all 10 assessments, IQWiG stated that the clinical data did not allow robust conclusions related to the advantages and disadvantages of using PET or PET/CT, for example, there was no proven benefit. This is because none of the studies directly compared the benefit of these imaging techniques with conventional diagnostics: in 10 of 10 indications, RCTs – the standard requirement to prove benefit – were missing. Furthermore, IQWiG’s principal criticism was that patient-relevant endpoints were not included in the studies (Table 2).

The three key learnings from the IQWiG PET assessment for the clinical study design concern RCTs, standard of care as comparator and patient-relevant outcomes (Table 3). For manufacturers of medical devices, this underscores the increasing importance of reviewing and verifying the evidence for products and starting early to close the gaps.

CONCLUSION Overall, with the introduction of § 137e in the Social Code book V, the FJC has expanded its scope of action. The new testing option allows for evidence to be generated on the basis of the potential in new methods. As all stakeholders break new ground, the reform might be regarded as a learning system. It is still too early to estimate the effects of the regulations but their close and continued evaluation will be important.

•

TABLE 1: RESULTS Of IQWIG BENEfIT ASSESSMENTS Of PET IN 10 INDICATIONS Part of assessment

IQWiG conclusion

Number of reports

Proof of benefit.

0

No evidence of patient-relevant benefit.

10

Improved diagnostic accuracy in comparison to standard diagnostic procedures.

2

Patient-relevant benefit

Diagnostic (and prognostic) Improved diagnostic accuracy only for subgroup, in either restaging or recidive staging, or in comparison against accuracy simple/cheap comparator.

No evidence of improved diagnostic accuracy.

3 5

TABLE 2: MAJOR CRITICISMS fROM IQWIG BENEfIT ASSESSMENTS Of PET IN 10 INDICATIONS Critical

Number of indications (% of all 10 indications)

No randomized controlled trials (RCTs)

RCTs are the standard requirement to prove benefit. An RCT for PET should be planned in an indication where PET has shown improved diagnostic accuracy and where a good treatment therapy is available. In special cases the study must not mandatorily be a standard RCT.

10 (100%)

No patient-relevant outcomes (PROs)

GAP analysis

PROs according to G-BA/IQWiG are mortality, morbidity, quality of life, safety/reduction of side-effects and must be implemented in a study to show benefit.

9 (90%)

TABLE 3: KEY LEARNINGS fROM IQWIG PET ASSESSMENT fOR CLINICAL STUDY DESIGN Best practice

Description

Recommended methodology

Patients, after assessment of eligibility and recruitment, but before the Randomized controlled trial intervention to be studied begins, are randomly allocated to receive one or other CONSORT (Consolidated Standards of Reporting Trials) of the alternative treatments under study.

Comparator

The comparator has to be identified according to benchmarks derived from the international standards of evidence-based medicine (standard of care diagnostic Systematic review of treatment guidelines with high level procedure). If there are several alternatives, the more economic of evidence. therapy/diagnostic procedure is selected.

Patient-relevant outcome

The benefit of a medical device is the patient-relevant therapeutic effect, in particular in respect of: the improvement in state of health; reduction of duration of the disease; longer survival; reduction in side-effects; or improvement in quality of life.

1 Espicom. The

IQWIG Methodenpapier 4.0

Medical Device Market: Germany, 2012. www.espicom.com/germany-medical-device-market. Accessed on April 19, 2013

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 59

PROJECT FOCUS | CHRONIC HEPATITIS b Sophisticated simulation modeling reveals important cost savings for product repositioning leveraging emerging new clinical evidence

The authors Sergio Iannazzo, MEng, MBA is Director RWE Solutions & HEOR, IMS Health Siannazzo@it.imshealth.com Maria De Francesco, MSC is Analyst RWE Solutions & HEOR, IMS Health Mdefrancesco@it.imshealth.com

Demonstrating cost-effectiveness of an individualized approach to chronic hepatitis b treatment in Italy Today, more than ever, pharmaceutical manufacturers are dependent on insights from real-world evidence to support the value of their medicines. Especially for mature products facing growing competition, these may be key to affirming their continued and appropriate use in recommended protocols of care. Robust and validated economic models can play a critical role in this process, affording a deeper understanding of cost and outcomes in the local clinical setting. For one leading innovative company with a specialty in personalized healthcare, the ability to develop a compelling value case leveraging new observational evidence was pivotal to the repositioning of an inmarket brand for chronic hepatitis b (CHb) in Italy, based on its proven cost-effectiveness versus alternative available treatments.

OPTIMIZING HEPATITIS B TREATMENT Hepatitis b virus (HbV) infection is a serious global health concern, with two billion infected individuals and more than 350 million having active disease. Although its incidence has been drastically reduced through vaccination, CHb still presents a large burden due to the high associated risk of cirrhosis and liver cancer.

In CHb, the progression of liver disease is essentially due to ongoing viral replication. Inhibiting this replication is thus the main goal of treatment. The two available pharmacological approaches are a finite 48-week course of pegylated interferon (PEG-IFN) or continuous administration of nucleoside analogues (NAs). PEG-IFN works by stimulating host immunity and can induce the sustained immune control of HbV infection in responsive patients after the end of treatment. Conversely, therapy with NAs centers on the direct inhibition of viral replication; patients must continue treatment indefinitely since they are unable to achieve sustained immune control of HbV infection following withdrawal, even after years of continuous administration. The two most novel and more effective NAs are entecavir (ETV) and tenofovir (TDF).

In Europe, PEG-IFN is licensed for the treatment of CHb and chronic hepatitis C and within current European guidelines1 it is the recommended first-line pharmacotherapy for both variants of CHb (HbeAg-positive and -negative). However, this approach appears to have only limited application in Italian clinical practice. This is possibly due to the relatively low antiviral effect of PEG-IFN at the end of the 48-week course, and the inferior tolerability that may lead clinicians to commence treatment directly with NAs (Figure 1).

PAGE 60

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

CHRONIC HEPATITIS b | PROJECT FOCUS LEVERAGING NEW CLINICAL EVIDENCE It was against this background that the company approached IMS to help improve the visibility of its mature PEG-IFN in Italy with a view to re-launching the product in this market. The move was triggered by the emergence of new evidence from a number of cohort studies,2,3,4,5 which demonstrated a correlation between specific virologic/serologic markers at week 12 of treatment with PEG-IFN and the absence of response at the end of the 48-week course. Despite the observational nature of these findings, they nevertheless paved the way for early identification of non-responders to the drug. fIGURE 1: KEY ATTRIBUTES Of PEG-IfN AND NUCELOSIDE ANALOGUES IN CHB

(PEG-)IFN

Advantages

• Finite duration • Absence of resistance • Higher rates of antiHbe and anti-Hbs seroconversion with 12 months of therapy

• Moderate antiviral effect • Inferior tolerability Disadvantages • Risk of adverse events • Subcutaneous injections

fIGURE 2: APPLICATION Of THE STOPPING RULE IN PEG-IfN TREATMENT

Start PEG-IFN Serologic/virologic markers

Stopping Rule

Discontinue PEG-IFN

Complete PEG-IFN course

Wk12

Wk48

NAs • Potential antiviral effect

• Good tolerance • Oral adminstration

• Indefinite duration • Risk of resistance • Unknown long-term safety

Source: European Association for the Study of the Liver. EASL Clinical Practice Guidelines: Management of chronic hepatitis B virus infection. J Hepatology, 2012; 57: 167-185

Adoption of the stopping rule at week 12 thus created the opportunity to improve the efficient allocation of resources, allowing non-responders to discontinue PEGIFN early without the need to complete the full course (Figure 2).

Working in collaboration with the company’s market access and marketing teams, IMS was tasked with identifying the potential of the 12-week stopping rule to reduce the cost per responsive patient and positively impact the value of PEG-IFN. This was particularly valid given growing payer concerns about the expense of NAs for CHb and the appeal of newer drugs in this class that could eliminate the viral load. National payers were well aware that respect of the clinical guidelines would easily reduce current expenditure for CHb. The pharmacoeconomic evaluation could thus serve as a useful tool to demonstrate the extent of this saving.

MODELING THE IMPACT OF ALTERNATIVE STRATEGIES For the purpose of the economic analysis, IMS developed a Markov model to compare, over the lifetime horizon, HbeAg-negative CHb treatment strategies, consisting either of first-line PEG-IFN treatment with the stopping rule and potential switch to the current most effective NAs (eg, ETV, TDF), or NA treatment first-line. The costeffectiveness analysis focused on HbeAg-negative CHb, the most common form of the disease in Italy.

The approach involved first a systematic literature review of published randomized controlled trials (RCTs) and new observational evidence to identify the most reliable and current clinical data for populating the model. Measured outcomes were average survival, quality-adjusted life years and costs, calculated from the Italian National Health Service (SSN) perspective. To account for all the most common stages of disease progression, ten health states were included in the model: active CHb; virologist response; s-antigen clearance; compensated cirrhosis with active CHb; compensated cirrhosis with virologist response; decompensated cirrhosis; hepatocellular carcinoma; liver transplant; post-liver transplant; and death (Figure 3 overleaf ). Probabilities associated with treatment efficacy were extrapolated from related RCTs and long-term observational studies. Response probabilities for PEG-IFN were adjusted to account for adoption of the stopping rule.

continued on next page

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 61

PROJECT FOCUS | CHRONIC HEPATITIS b DEMONSTRATING AND COMMUNICATING COST-EFFECTIVENESS base-case results from the analysis showed that treatment with first-line PEG-IFN using the stopping rule was highly cost-effective compared to the use of NAs first line.

Through its work with IMS in the development of a simulation model for the economic evaluation of CHb treatment strategies, the company thus gained compelling proof that the cost-effectiveness of HbeAgnegative CHb therapy in Italy could be improved significantly using first-line PEG-IFN with the adoption of the week-12 stopping rule. This has enabled the development and dissemination of key value messages to support the re-positioning of the product in Italy. The evidence, together with the model methodology, has been presented at two major conferences,6,7 and accepted for publication in a leading international journal in virology and infectious diseases.8 IMS has also supported the development of a dynamic, simplified version of the cost-effectiveness model for use on an App for iPad which will further enhance visibility of the drug across a broad audience of stakeholders.

•

fIGURE 3: COST-EffECTIVENESS MODEL STRUCTURE8

sCL

CHb-A

CHb-R

CC-A

CC-R

DCC

HCC

LT

Post LT

Death

All-cause mortality is also taken into consideration in every health state (not shown), defined as death for any cause which is not directly attributable to CHb. Key: CHb-A: Active Chronic Hepatitis b; CHb-R: Chronic Hepatitis b with virologic response; sCL: Clearance of HbsAg; CC-A: Compensated Cirrhosis with active CHb; CC-R : Compensated Cirrhosis with virologic response; DCC: Decompensated Cirrhosis; HCC: Hepatocellular Carcinoma; LT: Liver Transplant.

1

European Association for the Study of the Liver. EASL Clinical Practice Guidelines: Management of chronic hepatitis b virus infection. J Hepatology, 2012; 57: 167-185 Rijckborst V, Hansen bE, Cakaloglu Y, Ferenci P, Tabak F, Akdogan M, et al. Early on-treatment prediction of response to peginterferon alfa-2a for HbeAg-negative chronic hepatitis b using HbsAg and HbV DNA levels. Hepatology, 2010; 52:454–461 3 brunetto MR, Moriconi F, bonino F, Lau GK, Farci P, Yurdaydin C, et al. Hepatitis b virus surface antigen levels: a guide to sustained response to peginterferon alfa-2a in HbeAg-negative chronic hepatitis b. Hepatology, 2009 Apr; 49(4):1141-50 4 Marcellin P, Piratvisuth T, brunetto M, bonino F, Farci P, Yurdaydin C, Gurel S, Kapprell H-P, Messinger D, Popescu M. On-treatment decline in serum HbsAg levels predicts sustained immune control 1 year post-treatment and subsequent HbsAg clearance in HbeAg-negative hepatitis b virus-infected patients treated with peginterferon Alfa2a [40KD] (Pegasys). 20th Conference of the Asian Pacific Association for the Study of the Liver (APASL), beijing, China, 25-28 March, 2010 5 Rijckborst V, Hansen bE, Ferenci P, brunetto MR, Tabak F, Cakaloglu Y, et al. Validation of a stopping rule at week 12 using HbsAg and HbV DNA for HbeAg-negative patients treated with peginterferon alfa-2a. J Hepatol, 2012 May; 56(5):1006-11 6 Iannazzo S, Espinós b, Coco b, brunetto M, Rossetti F, Caputo A, bonino F. Cost-effectiveness of an individualized approach in the treatment of HbeAG-negative CHb patients with peginterferon alfa-2A in Italy. ISPOR 15th Annual European Congress, berlin, Germany, 3-7 November, 2012 7 Iannazzo S, Coco b, brunetto MR, Rossetti F, Caputo A, bonino F. Cost-effectiveness of peg-interferon alfa-2a therapy of HbeAg-negative chronic hepatitis b in Italy using a personalized approach based on week 12 HbV-DNA and HbsAG stopping rule. 63rd Annual Meeting of the American Association for the Study of Liver Diseases (AASLD), boston, MA, USA, 9–13 November, 2012 8 Iannazzo S, Coco b, brunetto MR, Rossetti F, Caputo A, Latour A, Espinos b, bonino F. Individualized treatment of HbeAg-negative CHb using peg-interferon alfa-2a as firstline and week 12 HbV-DNA\HbsAg stopping rule. A cost-effectiveness analysis. Antivir Ther 2013, Mar 13. doi: 10.3851/IMP2555. [Epub ahead of print] 2

PAGE 62

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

RHEUMATOID ARTHRITIS | PROJECT FOCUS The development of validated proxy measures for outcomes, leveraging large healthcare datasets, can enable evidence-based decisions in therapy areas where measurable endpoints are not easily accessible

The authors Vernon Schabert, PHD is Senior Principal RWE Solutions & HEOR, IMS Health Vschabert@us.imshealth.com Jason yeaw, MS is Director RWE Solutions & HEOR, IMS Health Jyeaw@us.imshealth.com Jon Korn, BS is Statistical Programmer RWE Solutions & HEOR, IMS Health Jkorn@us.imshealth.com

Treatment pathway decision support using real-world evidence from large populations A number of specific therapy areas continue to see new market entrants that obscure real-world comparative effectiveness. As a result, some payers and HTAs are avoiding effectiveness-based decisions altogether, passing responsibility for treatment decisions directly to physicians and patients. The patient access scheme approved by NICE for UCb’s Cimzia (certolizumab), a biologic disease-modifying anti-rheumatic drug (DMARD) for the treatment of rheumatoid arthritis (RA), is a case in point.1 The scheme is based on findings from UCb’s phase III trials that response at 12 weeks of therapy is predictive of response at 52 weeks of therapy. UCb has agreed to bear the cost of therapy for the first three months, with the UK National Health Service assuming responsibility thereafter. While the scheme was explicitly proposed as not requiring clinician input on continuing therapy past three months, it is reasonable to expect that most physicians who are aware of it will evaluate whether their patients will benefit from continued therapy at the point where financial responsibility shifts from manufacturer to payer. Under this precedent-setting scheme, the HTA market access decision is essentially repeated for every patient rather than for the payer’s budget as a whole. What evidence will physicians and patients use to predict the chances of future benefit in light of the patient’s limited experience on treatment?

MEASURING TREATMENT OUTCOMES Some therapy areas, for example diabetes, have effectiveness endpoints such as HbA1c that are accessible and systematically measurable. Where crowded therapy areas have such clearly measurable outcomes, physicians and patients can make effectiveness decisions based on unambiguous clinical evidence.

In addition, payers could verify whether treatments were being chosen consistent with clinical evidence and guidelines, because they could more easily access real-world evidence (RWE) to compare those treatment choices against outcomes.

In RA, however, the situation is less clear. RA is a progressive autoimmune disease, where the primary measures of outcome are either patient-reported or physician-rated markers of pain, joint tenderness and disability. Diagnostics do exist, but in the form of joint imaging studies that must still be interpreted by a clinician for signs of progressive joint damage. This reduces the likelihood that physicians and patients will have truly objective ratings of outcome on which to base treatment effectiveness decisions. It also limits their ability to benchmark performance against that of other patients, or for payers to review the consistency of treatment decisions against clinical evidence. continued on next page

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 63

PROJECT FOCUS | RHEUMATOID ARTHRITIS If payers and HTAs delegate treatment decisions to physicians and patients in these or other therapy areas, they are also ceding influence on the evidence base used by those stakeholders to make treatment decisions. When the outcomes of care are measured only with substantial effort, how can payers and HTAs trust that treatment decisions will be consistent with the evidence-based decision making that the payer might prefer? And in therapy areas with many available treatments, how can payers assure that adequate observational data are available to provide an adequate evidence base?

RHEUMATOID ARTHRITIS This is precisely the challenge that manufacturers of RA therapies face in today’s market. After many years of the biologic DMARD market being shared by etanercept, infliximab and then adalimumab, six new products have received FDA or EMA approval since 2006 (abatacept, certolizumab, golimumab, rituximab, tocilizumab, tofacitinib). Another product has been approved for related autoimmune conditions also treated by the first three agents in market (ustekinumab, in 2009).2 Outcomes in RA are frequently measured by counting swollen and tender joints. The DAS-28 is an often-used clinician rating instrument for this purpose and is a critical component of other commonly-cited outcome measures.3 Effective treatments reduce the number of swollen and tender joints experienced by patients. Initial response can occur in a few weeks, but treatments vary in how long patients perceive continued benefit. Some courses of treatment are characterized by early discontinuation or switching of therapy, increased dosing, or supplementation with traditional DMARDs or corticosteroids in order to maintain the level of response.

Professor Jeff Curtis of the University of Alabamabirmingham validated a proxy measure for estimating the effectiveness of these biologic DMARDs over a one-year period.4 In his algorithm, treatment changes such as dosing titration, inadequate persistence or adherence, and the addition of other therapies were signs of noneffective treatment. These treatment changes correlated strongly with inadequate response as measured by the DAS-28 among VA patients enrolled in an RA registry.

Application of an algorithm such as this can empower multiple stakeholders to estimate outcomes using more accessible data and measures. DAS-28 scores are time consuming to calculate and many clinicians and patients rely on quicker assessments in routine clinical practice.

This means that few payers, physicians or patients would have routine access to real-world outcomes for a wide population of RA patients. However, establishing the value of changing treatment patterns as a predictor of outcomes enables all stakeholders to judge effectiveness with widely-accessible treatment pattern data, and can guide a physician and patient through treatment assessment based on the patient’s own treatment history.

EXTENDING THE EVIDENCE BASE IMS applied the Curtis algorithm to evaluate treatment effectiveness for a wide range of biologics using IMS LifeLink PharMetrics Plus™. Two of the analyses have formed the basis of research presentations at ISPOR,5,6 with others scheduled for submission to journals and presentation at other conferences.

The use of such a large payer database enabled the development of effectiveness estimates for many of the newer biologic DMARDs. Given the dominance of adalimumab, etanercept and infliximab for many years, far fewer real-world studies have examined the treatment patterns and fIGURE 1: SUPPORT fOR MULTI-LINE THERAPY COHORTS IN CROWDED MARKETS WITH effectiveness of the more recent IMS PHARMETRICS PLUS™ treatment alternatives. Starting from over 220,000 patients with evidence of a biologic DMARD between 2007 and 2010, it was possible to construct meaningful treatment cohorts for many agents that had not been widely studied before (Figure 1).

First year Cohort Abatacept (n=1,160) Adalimumab (n=4,991) Certolizumab (n=138) Golimumab (n=261) Infliximab (n=2,352) Etanercept (n=7,247)

PAGE 64

Second year Cohort Abatacept (n=1,156) Adalimumab (n=620) Infliximab (n=229) Etanercept (n=994)

Switch Cohort Abatacept (n=196) Adalimumab (n=527) Certolizumab (n=22) Golimumab (n=60) Infliximab (n=202) Etanercept (n=318)

The findings confirmed those in the original validation study; about 28% of patients avoided the dosing changes, poor adherence, or therapy modifications that were considered signs of failed effectiveness in the algorithm.

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

RHEUMATOID ARTHRITIS | PROJECT FOCUS The application of this algorithm was then extended one step further. In a typical HEOR cohort analysis, measuring an outcome over a fixed time period is the end of the analysis. However, physicians and patients will continue to struggle with treatment choices for the duration of the patient’s condition. A follow-on study thus set out to determine what happens among patients who (a) continue on therapy a second year after a first year of successful therapy, or (b) switch to a second therapy some time in their first year.

These second-line analyses yielded ample evidence that patient experience with initial therapy would predict experience on future therapies. First year “effectiveness”, as defined by the algorithm, was about 28% across all treatments. but of those who continued effective therapy (and who, by definition, were 100% effective in year one), nearly 47% experienced effectiveness for the entire second year. In contrast, among those who switched from their initial therapy within the first year to a second biologic DMARD, only 20% experienced effective therapy for a year following the switch to a new agent.

The predictability of future therapy outcomes extended to the individual behaviors that made up the effectiveness algorithm. As shown in Figure 2, 17% of patients switched fIGURE 2: BIOLOGIC DMARD SWITCH RATES OVER ONE YEAR, BASED ON LINE Of THERAPY 28%

30%

20%

17%

10%

therapies in their first year. Among those who experienced effectiveness over the first year, second-year switch rates dropped to 7%. Among those who switched once, switch rates to a third biologic increased to 28%.

Drug-specific characteristics also seemed to follow patients to future therapies. Several studies have previously shown that infliximab is associated with high rates of dose increase during the first year of therapy.7,8 This analysis showed the same, with 46% of infliximab patients increasing their dose during first-year therapy, compared to only 10% of first-year abatacept patients. but most patients who switched to abatacept for second-line therapy had started infliximab as their first-line therapy.

After the switch, the rate of dose increase among abatacept patients climbed to 19%, as if the prior experience with infliximab was a marker for future risk of dose increases.

CONCLUSIONS Market pressures such as personalized medicine, paymentfor-performance and value-based contracting will continue to increase pressure on stakeholders other than payers to make evidenced-based treatment decisions. Payers have grown accustomed to using RWE to benchmark and guide those decisions although this has occurred mainly where therapy areas have measurable and widely accessible outcomes in RWE sources. These analyses demonstrate how such decision making can be extended to therapy areas without accessible outcome measures, through the development of validated proxy measures for outcome.

The insights gained suggest that alternative framing of analytics from large RWE sources could ensure greater access to evidence-based insights for non-payer stakeholders at various milestones in the patient’s course of illness.

•

7%

0% First year (All)

Second Year (Effective)

Switched from prior DMARD

1

Certolizumab pegol (CIMZIA®) for the treatment of Rheumatoid Arthritis. Patient access scheme (pas) submission to NICE, 23 July 2009 (approved 21 January 2010). http://www.nice.org.uk/nicemedia/pdf/Cimzia%20PAS%20submission.pdf; accessed May 2013 2 IMS KnowledgeLink, May 2013 3 Prevoo MLL, Hof van ‘t MA, Kuper HH, Leeuwen van MA, Putte van de LbA, Riel van PLCM Modified disease activity scores that include twenty-eight-joint counts: Development and validation in a prospective longitudinal study of patients with rheumatoid arthritis. Arthritis Rheum. 1995, 38:44-48 4 Curtis JR, et al. Derivation and preliminary validation of an administrative claims-based algorithm for the effectiveness of medications for rheumatoid arthritis. Arthritis Res Ther. 2011;13(5): R155 5 Schabert VF, Yeaw J, Korn JR, Quach C, Harrison DJ, Yun H, Joseph G, Collier D. Algorithm-based estimation of biologic effectiveness for rheumatoid arthritis. International Society for Pharmacoeconomics and Outcomes Research 18th Annual International Meeting, New Orleans, LA, USA, 18-22 May, 2013 (Poster) 6 Schabert VF, Yeaw J, Korn JR, Quach C, Harrison DJ, Yun H, Joseph G, Collier D, Curtis JR. Effectiveness of second line biologic use in rheumatoid arthritis after switching from first-line biologics. International Society for Pharmacoeconomics and Outcomes Research 18th Annual International Meeting, New Orleans, LA, USA, 18-22 May, 2013 (Poster) 7 Schabert VF, bruce b, Ferrufino CF, Globe DR, Harrison DJ, Lingala b, Fries JF. Disability outcomes and dose escalation with etanercept, adalimumab, and infliximab in rheumatoid arthritis patients: A US-based retrospective comparative effectiveness study. Curr Med Res Opin. 2012 Apr; 28(4):569-80 8 Schabert VF, Watson C, Gandra SR, Goodman S, Fox KM, Harrison DJ. Annual costs of tumor necrosis factor inhibitors using real-world data in a commercially insured population in the United States. J Med Econ. 2012; 15(2):264-75

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 65

IMS RWE SOLUTIONS & HEOR | OVERVIEW

Enabling your real-world success IMS offers a spectrum of world-class expertise in real-world evidence (RWE) and health economics & outcomes research (HEOR) to deliver the local excellence you need.

In a future where healthcare efficiency and quality are measured through the lens of ‘real-world’ insights, external validity demands a focus on the right data sources, scientifically credible research and actionable communication.

IMS is committed to helping you succeed:

• Largest multi-disciplinary team of RWE and HEOR experts, based in 18 countries worldwide • Credible scientific voice and deep therapy area knowledge, captured in over 2400 publications • Market leadership in developing and adapting robust economic models • Most advanced capabilities in RWE management and analysis, leveraging relevant IMS proprietary and other key external, third-party data assets

• Proven expertise in generating and communicating RWE to advance stakeholder engagement at all levels IMS is a leading independent provider of RWE, outcomes research, economic modeling and market access solutions, and value communication.

Our unique, data-agnostic market position can help you develop and support the evidence required to engage global and local healthcare stakeholders, with deep insights into product safety, efficacy, cost, value for money and affordability.

We offer a wide spectrum of solutions Real-World Evidence Solutions

Outcomes Research

• • • • • •

Data sourcing & validation Data integration & linking Data management & curation Platform development Customized analytics & reporting Evidence planning

• • • • • • •

Technology

Health Economic Modeling

• • • • •

• Health economic evaluations • Core models & local adaptations • Budget impact • Meta-analyses • Indirect comparisons • IMS CORE Diabetes Model

Platform engines Data warehouse/data marts Encryption systems & linking technology Meta-data repository User interface & sophisticated analytics library • Electronic data capture

IMS LifelinkTM

Market Access

Largest collection of scientifically validated, anonymized patient-level data assets: • Health plan claims • PharMetrics Plus • Longitudinal Rx • Electronic medical records • Hospital disease • Oncology • Diabetes

• Value development planning

TM

PAGE 66

Evidence generation Late-phase studies Patient-Reported Outcomes (PROs) Database studies Mixed methods Epidemiology Risk management

• Market access strategy • Core value dossiers & local adaptations • HTA readiness • Value communication • Reimbursement submissions

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

LOCATIONS | IMS RWE SOLUTIONS & HEOR

Global scope, local expertise IMS RWE Solutions & HEOR experts are located in 18 countries worldwide and they have published on projects completed in more than 50 countries on all continents. yOUR PRIMARy CONTACTS Jon Resnick Vice President and General Manager IMS Health 1725 Duke Street, Suite 510 Alexandria, VA 22314 USA Tel: +1 703 837 5150 Jresnick@imshealth.com

Dr. Michael Nelson Senior Principal IMS Health 1725 Duke Street, Suite 510 Alexandria, VA 22314 USA Tel: +1 703 837 5150 Mnelson@us.imshealth.com

Dr. Jacco Keja Senior Principal IMS Health 210 Pentonville Road London N1 9JY UK Tel: +31 (0) 631 693 939 Jkeja@nl.imshealth.com

Adam Lloyd Senior Principal IMS Health 210 Pentonville Road London N1 9JY UK Tel: +44 (0) 20 3075 4800 Alloyd@uk.imshealth.com

SPAIN Dr Ferran, 25-27 08034 Barcelona Spain Tel: +34 93 749 63 00

AUSTRALIA Level 5, Charter Grove 29-57 Christie Street St Leonards, NSW 2065 Australia Telephone: +61 2 9805 6800

IMS RWE Solutions & HEOR office locations NORTH AMERICA REGIONAL HEADQUARTERS 11 Waterview Boulevard Parsippany, NJ 07054 USA Tel: +1 973 316 4000

EUROPE REGIONAL HEADQUARTERS 210 Pentonville Road London N1 9JY United Kingdom Tel: +44 (0) 20 3075 4800

UNITED STATES 1725 Duke Street Suite 510 Alexandria, VA 22314 USA Tel: +1 703 837 5150

BELGIUM Medialaan 38 1800 Vilvoorde Belgium Tel: +32 2 627 3211

One IMS Drive Plymouth Meeting PA 19462 USA Tel: +1 610 834 0800 CANADA 16720 Route Transcanadienne Kirkland, Québec H9H 5M3 Canada Tel: +1 514 428 6000 LATIN AMERICA REGIONAL HEADQUARTERS Insurgentes Sur # 2375 5th Floor, Col. Tizapan Mexico City D.F. - C.P. 01090 Mexico Tel: +52 (55) 5062 5239 or +1 917 542 5844

FRANCE 29ème Etage Tour Ariane 5-7 Place de la Pyramide 92088 La Défense Cedex France Tel: +33 1 41 35 1000 GERMANY Erika-Mann-Str. 5 80636 München Germany Tel: +49 89 457912 6400 ITALY Viale Certosa 2 20155 Milano Italy Tel: +39 02 69 78 6721

SWEDEN Sveavägen 155/Plan9 11346 Stockholm Sweden Tel: +46 8 508 842 00 SWITZERLAND Theaterstr. 4 4051 Basle Switzerland Tel: +41 61 204 5071 UNITED KINGDOM 210 Pentonville Road London N1 9JY United Kingdom Tel: +44 (0) 20 3075 4800 ASIA PACIFIC REGIONAL HEADQUARTERS 8 Cross Street #21-01/02/03 PWC Building Singapore 048424 Tel: +65 6412 7365

CHINA 7/F Central Tower China Overseas Plaza Jianguomenwai Avenue, Chaoyang District Beijing 100001 China Tel: +86 10 8567 4255 SOUTH KOREA 9F Handok Building 735 Yeoksam1-dong Kangnam-ku Seoul 135-755 S. Korea Tel: +82 2 3459 7307 TAIWAN 8th Floor No 2, Tun Hwa South Road Section 1 Taipei 10506 Taiwan ROC Tel: +886 2 2721 5337

FOR FURTHER INFORMATION: email RWEinfo@imshealth.com or visit www.imshealth.com/rwe

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 67

IMS RWE SOLUTIONS & HEOR | EXPERTISE

Expertise in depth IMS has one of the largest global teams of experts in RWE, HEOR and market access of any organization in the world.

We apply unrivalled experience and specialist expertise to help our clients meet the demands of an increasingly complex global, regional and local pharmaceutical landscape. Our highly qualified, multi-disciplinary consultants and researchers have proven skills and capabilities across all key therapy areas. Spanning industry, consulting, government and academia, their expertise reflects a global grasp, local experience and a unique, inside market perspective.

Our senior team Renée J. G. Arnold, PHARMD • Dr. Renée Arnold is Principal RWE Solutions & HEOR at IMS Health with particular expertise in the use of technology to collect and/or model real-world data to facilitate rational decision making by healthcare practitioners and policy makers.

• Renée was previously President and CEO, Arnold Consultancy & Technology where she developed and oversaw outcomes research for the pharmaceutical, biotech and device industries as well as federal government programs. Her distinguished career in health economics and outcomes research includes roles as President and co-founder of Pharmacon International, Center for Health Outcomes Excellence and Senior VP and Medical Director at William J Bologna International.

• Founding member and former Chair of the Education Committee of ISPOR, Renée has several adjunct appointments and is the author of numerous articles on pharmacoeconomics and cost-containment strategies. She holds a Doctor of Pharmacy degree from the University of Southern California in Los Angeles.

yumiko Asukai, MSC • Yumi Asukai is Principal RWE Solutions & HEOR at IMS Health, specializing in the development of economic models across the product lifecycle and the interpretation of model outputs for strategic market access and value demonstration. Her expertise in this field spans from early strategic modeling through to global core cost-utility models.

• Yumi’s background includes roles at Fourth Hurdle Consulting and in healthcare and business consulting in San Francisco and Tokyo, where she focused on comparative studies of health policies between Japan and the US complemented by analyses of primary data. Yumi has worked extensively in the cardiovascular, oncology and respiratory disease areas and she is part of a global modeling taskforce for COPD composed of academic and industry members.

• Yumi holds a Bachelor's degree in Political Science from Stanford University and a Master's degree in Health Policy, Planning and Financing from the London School of Hygiene & Tropical Medicine and the London School of Economics.

Karin Berger, MBA • Karin Berger is Principal RWE Solutions & HEOR at IMS Health with a particular focus on RWE, patient- reported outcomes, and cost-effectiveness evaluation analyses at a national and international level.

• Formerly Managing Director of MERG (Medical Economics Research Group), an independent German organization providing health economics services to the pharmaceutical industry, university hospitals and European Commission, Karin has more than 15 years experience in the health economics arena. She lectures at several universities, has published extensively in peer-reviewed journals, and regularly presents at economic and medical conferences around the world.

• Karin graduated as Diplom-Kaufmann (German MBA equivalent) from the Bayreuth University, Germany, with a special focus on health economics.

Nevzeta Bosnic, BA • Nevzeta Bosnic is Principal at IMS Brogan, where she manages projects to meet the broad spectrum of client needs in the Canadian pharmaceutical market.

• Formerly Director of Economic Consulting at Brogan Inc, Nev has led many strategic consulting, policy and data analyses for pharmaceutical clients, government bodies and academic institutions in Canada. She has extensive knowledge of public and private drug plans across the country and in-depth expertise and experience on the drug reimbursement process.

• Nev holds a Bachelor’s degree in Business Economics from the School of Economics and Business at the University of Sarajevo, Bosnia-Herzegovina.

PAGE 68

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

EXPERTISE | IMS RWE SOLUTIONS & HEOR Joe Caputo, BSC • Joe Caputo is Regional Principal RWE Solutions & HEOR, Asia Pacific at IMS Health leveraging more than 20 years experience in the pharmaceutical sector to help clients address the challenges of global reimbursement and market access throughout the drug development program. He has led numerous projects involving payer research, value dossiers, local market access models and HTA submissions.

• Joe's background includes industry roles in drug development, sales and marketing, and UK and global health outcomes, as well as consulting in health economics. He has wide-ranging knowledge of the drug development process at both local and international level and a unique understanding of evidence gaps in light of reimbursement and market access requirements.

• Joe holds a BSc in Applied Statistics and Operational Research from Sheffield Hallam University, UK.

Frank-Ulrich Fricke, PHD, MSC • Dr. Frank-Ulrich Fricke is Principal RWE Solutions & HEOR at IMS Health and Professor for Health Economics, GeorgSimon-Ohm University of Applied Sciences, Nuremberg in Germany, with a focus on health economic evaluations, market access strategies and health policy.

• Formerly a Managing Director of Fricke & Pirk GmbH, and previously Head of Health Economics at Novartis Pharmaceuticals, Frank-Ulrich has conducted health economic evaluations across a wide range of therapeutic areas, developing a wealth of experience in pricing, health affairs and health policy. As a co-founder of the NIG 21 association, he has forged strong relationships with health economists, physicians and related researchers working in the German healthcare system.

• Frank-Ulrich holds a PhD in Economics from the Bayreuth University, and an MBA equivalent from the ChristianAlbrechts-University, Kiel.

David Grant, MBA • David Grant is Senior Principal RWE Solutions & HEOR at IMS Health, specializing in reimbursement and market access, environmental analysis, prospective and retrospective data collection and communications for product support.

• A co-founder and former Director of Fourth Hurdle, David’s experience spans more than 10 years in health economics and outcomes research consulting, and 15 years in the pharmaceutical industry, including roles in clinical research, new product marketing and health economics in the UK and Japan.

• David holds a degree in Microbiology and an MBA from the London Business School.

Joshua Hiller, MBA • Joshua Hiller is Senior Principal RWE Solutions at IMS Health, supporting the strategic planning and development of IMS capabilities for data sourcing, integration, analytics and studies. He is also currently serving as Alliance Director in the IMS collaboration with AstraZeneca for the advancement of RWE. • During a career that includes roles in market analytics, government and healthcare consulting in both the US and UK, Joshua has led a wide range of projects for clients in the pharmaceutical and biotech sector as well as industry associations. He has extensive experience in pharmaceutical pricing, contracting, market landscape development, supply management, cross border trade, lifecycle management, competitive defense, generics market drivers and account management, with expertise across US and European markets.

• Joshua holds an MBA (Beta Gamma Sigma) from Columbia Business School, New York, and a BS in Mathematics from James Madison University, Virginia.

Benjamin Hughes, PHD, MBA, MRES, MSC • Dr. Ben Hughes is Senior Principal RWE Solutions at IMS Health, leading the development of the IMS RWE service offering.

He has helped many clients in the pharmaceutical industry to articulate and implement their RWE strategies, through definition of RWE vision, business cases for RWE investments, capability roadmaps, partnerships, brand evidence reviews, HEOR function design, RWE training programs and related clinical IT strategies.

• Previously head of the European RWE service line at McKinsey & Co, Ben has extensive experience advising healthcare stakeholders on health informatics and RWE-related topics. This includes work on France’s electronic health record strategy, EMR adoption strategy for governments across Europe and Asia, data releases to support the UK’s transparency agenda, and the development of payer health analytics and RWE capabilities across countries in Europe.

• A widely published author on health informatics, Ben holds Masters’ degrees in Research from ESADE Barcelona and in Physics from University College, London, an MBA from HEC Paris, and a PhD in Medical Informatics from ESADE Barcelona.

continued on next page

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 69

IMS RWE SOLUTIONS & HEOR | EXPERTISE Jacco Keja, PHD • Dr. Jacco Keja is Senior Principal RWE Solutions & HEOR at IMS Health, drawing on deep expertise in global market access, operational and strategic pricing, and health economics and outcomes research.

• Jacco’s background includes four years as global head of pricing, reimbursement, health outcomes and market access consulting services at a large clinical research organization and more than 13 years experience in the pharmaceutical industry, including senior-level international and global roles in strategic marketing, pricing and reimbursement and health economics.

• Jacco holds a PhD in Biology (Neurophysiology) from Vrije Universiteit in Amsterdam, a Masters in Medical Biology, and an undergraduate degree in Biology, both from Utrecht. He is also visiting Professor at the Institute of Health Policy & Management at Erasmus University, Rotterdam.

Tim Kelly, MSC, BS • Tim Kelly is Vice President RWE Solutions at IMS Health, with responsibility for the company’s RWE data assets and data architecture backbone, and for overseeing platform delivery infrastructure and engagements to ensure at-scale, high-quality data mart deployment. He also leads the client services team supporting data and technology applications.

• Tim’s background includes two decades of life-science experience managing large-scale data warehousing, technology, and analytic applications and engagements. He has worked with many clients in the pharmaceutical and biotech sectors, leveraging deep expertise in information management and modeling, commercial operations and analytics, advanced analytics, business intelligence, data warehousing and longitudinal analytics.

• Tim holds a Bachelor’s degree in Quantitative Business Analysis from Penn State University and a Master’s degree in Management Science from Temple University, Philadelphia.

Mark Lamotte, MD • Dr. Mark Lamotte is Principal RWE Solutions & HEOR at IMS Health with responsibility for the content and quality of all health economic evaluations conducted by his team.

• A physician by training (cardiology), Mark spent a number of years in medical practice before joining Rhône-Poulenc Rorer as Cardiovascular Medical Advisor and later becoming Scientific Director at the Belgian research organization, HEDM. He has since worked on more than 150 projects, involving expert interviews, patient record reviews, extensive modeling and report writing across a wide range of therapy areas, and authored many peer-reviewed publications.

• Mark holds an MD from the Free University of Brussels (Vrije Univeristeit Brussel, Belgium) and is fluent in Dutch, French, English and Spanish.

Won Chan Lee, PHD • Dr. Won Chan Lee is Principal RWE Solutions & HEOR at IMS Health, specializing in prospective and retrospective health economics research.

• Over the course of his career, Won has completed numerous international economic evaluations employing a variety of analytical methods across a range of diseases and geographies. His expertise includes econometric database analysis, quality of life assessment and advanced economic modeling to establish the economic and humanistic value of new and existing therapeutic interventions.

• Won holds a Master’s degree in Economics from the University of Grenoble II, and a PhD in Economics from the Graduate Center of the City University of New York.

Claude Le Pen, PHD • Dr. Claude Le Pen is a member of the strategic committee of IMS Health and Professor of Health Economics at Paris-Dauphine University, providing expert economic advisory services to the consulting practice.

• A renowned economist, leading academic, and respected public commentator, Claude has served as an appointed senior member of several state commissions in the French Ministry of Health and is an expert for a number of parliamentary bodies, bringing a unique perspective and unparalleled insights into the economic evaluation of pharmaceutical technologies at the highest level.

• Claude studied Business Administration in HEC Business School in Paris and holds a PhD in Economics from Panthéon-Sorbonne University.

Adam Lloyd, MPHIL, BA • Adam Lloyd is Senior Principal RWE Solutions & HEOR at IMS Health, with a particular focus on economic modeling and the global application of economic tools to support the needs of local markets.

• A former founder and Director of Fourth Hurdle, and previously Senior Manager of Global Health Outcomes at GlaxoWellcome, Adam has extensive experience leading economic evaluations of pre-launched and marketed products, developing submissions to NICE and the SMC, decision-analytic and Markov modeling, and in the use of health economics in reimbursement and marketing in continental Europe.

• Adam holds an MPhil in Economics, and a BA (Hons) in Philosophy, Politics and Economics from the University of Oxford.

PAGE 70

IMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

EXPERTISE | IMS RWE SOLUTIONS & HEOR Charles Makin, MS, MBA, MM, BS • Charles Makin is Principal RWE Solutions & HEOR at IMS Health, leading value strategy development, economic evaluations and health outcomes research studies and providing direction on the use of observational research for evidence-based healthcare decisions. He has served as principal investigator on numerous US-based and global database analyses, economic modeling, multi-country patient registries, adherence interventions, systematic literature reviews, value development plans and PRO research.

• A recognized leader and widely published author, Charles has deep insight into best practices in global research and market access. He was previously Global Head of Research Design and Proposal Development at UnitedHealth Group where he directed and managed research design activities for all health outcomes, economics and drug safety research.

He also worked as Research Operations Manager at WellPoint, leading project teams to execute HEOR projects.

• Charles holds a Bachelor’s degree in Pharmacy from the University of Pune, India, a Master’s degree in Pharmacy Administration from Purdue University, Indiana, and an MBA (summa cum laude) in Marketing Management and a Master’s degree in Management (summa cum laude), both from Goldey Beacom College, Delaware.

Frédérique Maurel, MS, MPH • Frédérique Maurel is Principal RWE Solutions & HEOR at IMS Health, with a particular focus on observational research and health economics studies.

• A skilled consultant and project manager, Frédérique has extensive experience in the economic evaluation of medical technologies gained in roles at ANDEM, Medicoeconomie, and AREMIS Consultants.

• Frédérique holds a Master’s degree in Economics – equivalent to an MS – and completed a post-graduate degree equivalent to an MPH with a specialization in Health Economics at the University of Paris-Dauphine (Paris IX) as well as a degree in Industrial Strategies at the Pantheon-Sorbonne University (Paris I).

Joan McCormick, MBA • Joan McCormick is Principal at IMS Brogan, leading a team providing strategic advice to companies with new products coming to market and ongoing consultation on the rules for existing drugs post launch.

• Formerly Head of Price Regulation Consulting at Brogan Inc, Joan has supported many major pharmaceutical companies with the preparation of pricing submissions to the Patented Medicine Prices Review Board (PMPRB), gaining extensive insights into the operation of the Canadian pharmaceutical market.

• Joan holds a Bachelor’s degree in Life Sciences from Queen’s University in Kingston, Canada, and an MBA from the University of Ottawa, Canada.

Julie Munakata, MS • Julie Munakata is Principal RWE Solutions & HEOR at IMS Health, with a particular focus on global economic modeling, value development planning, and survey data analysis.

• An accomplished researcher and author of more than 25 original articles, Julie has extensive experience in managing clinical trials, health economic studies and decision analytic modeling work, gained in senior roles at ValueMedics Research LLC, the VA Health Economics Resource Center and Stanford Center for Primary Care & Outcomes Research, and Wyeth Pharmaceuticals.

• Julie holds an MS in Health Policy and Management from the Harvard School of Public Health and a BS in Psychobiology from the University of California, Los Angeles.

Michael Nelson, PHARMD • Dr. Michael Nelson is Senior Principal RWE Solutions & HEOR at IMS Health, with particular expertise in retrospective database research, prospective observational research, health program evaluation, and cost-effectiveness analysis.

• During a career that includes leadership roles in HEOR at PharmaNet, i3 Innovus, SmithKline Beecham, and DPS/UnitedHealth Group, Mike has gained extensive experience in health information-based product development, formulary design, drug use evaluation, and disease management program design and implementation.

• A thought leader in health economics for more than 20 years, Mike holds a doctorate in Pharmacy and a Bachelor of Science degree, both from the University of Minnesota College of Pharmacy. He also served as an adjunct clinical faculty member at the University of Minnesota whilst in clinical pharmacy practice.

continued on next page

ACCESSPOINT • VOLUME 3 ISSUE 6

PAGE 71

IMS RWE SOLUTIONS & HEOR | EXPERTISE Amy O’Sullivan, PHD • Dr. Amy O’Sullivan is Principal RWE Solutions & HEOR at IMS Health, with a particular focus on global economic modeling to support product reimbursement in jurisdictions around the world.

• Highly experienced in the economic evaluation of medical technologies, Amy’s background includes roles at Policy Analysis Inc. (PAI) and most recently in a senior capacity at OptumInsight (formerly i3 Innovus). She has led numerous pharmacoeconomic and outcomes research studies including cost-effectiveness analyses, budgetary impact analyses, burden-of-illness studies and piggyback economic evaluations. Her research spans a wide range of therapeutic areas, including autoimmune conditions, CV disease, CNS and behavioral health disorders, metabolic disorders, musculoskeletal conditions, oncology, respiratory disease and women’s health.

• Amy holds a PhD in Health Economics from the Johns Hopkins University Bloomberg School of Public Health, Baltimore, and a BA in Economics and English from Boston College.

Carme Piñol, MD, MSC • Dr. Carme Piñol is Principal RWE Solutions & HEOR at IMS Health, with more than 20 years experience in the pharmaceutical industry spanning clinical research, health economics and market access.

• Previously Head of Market Access for Spain at Bayer, a role that included pricing, HEOR, advocacy and institutional relations with the Regions, Carme is a Board member of the Spanish Association of Health Economics as well as the ISPOR Spain Regional Chapter and coordinator of the Pharmacoeconomics Interest Group of the Spanish Association of Medicine of the Pharmaceutical Industry (AMIFE). She has authored more than 60 communications in international and national congresses and more than 20 papers in peer-reviewed journals.

• Carme holds an MD from the Autonomous University of Barcelona, an MSc in Pharmacoeconomics and Health Economics from Pompeu Fabra University, an MSc in Health Research from Castilla-La Mancha University (UCLM), and an Executive Program Degree from ESADE Business School, Barcelona, Spain. She is currently completing a PhD in Health Research at UCLM.

Jon Resnick, MBA • Jon Resnick is Vice President and General Manager RWE Solutions at IMS Health, leading the company’s global RWE & HEOR business, including the development of RWE strategy, offerings, collaborations and foundational technologies to meet the RWE needs of healthcare stakeholders.

• A former Legislative Research Assistant in Washington DC and member of the Professional Health and Social Security staff for the US Senate Committee on Finance, Jon has 10 years consulting experience at IMS. He was most recently responsible for leading the European management consulting team and global HEOR business teams of 300 colleagues, advising clients on a wide range of strategic, pricing and market access issues.

• Jon holds an MBA from the Kellogg School of Management, Northwestern University, with majors in Management and Strategy, Finance, Health Industry Management, and Biotechnology.

Vernon Schabert, PHD • Dr. Vernon Schabert is Senior Principal RWE Solutions & HEOR at IMS Health, with a particular focus on the assessment and validation of patient-reported outcomes (PRO) instruments, retrospective analyses of claims and survey databases, and primary data collection surveys.

• A founder and former President of Integral Health Decisions, Inc, Vernon has extensive experience in conducting claims analyses, creating custom administrative databases, developing business intelligence software, and leading national quality research projects, gained in roles with Thomson Reuters, Strategic Healthcare Programs LLC, and CIGNA HealthCare. His expertise spans numerous disease areas and diverse topics including medication adherence, inpatient safety and outcomes in post-acute care.

• Vernon holds a PhD in Personality and Social Psychology from Stanford University and a BA in Psychology from Princeton University.

Núria Lara Surinach, MD, MSC • Dr. Núria Lara is Principal RWE Solutions & HEOR at IMS Health, with a particular focus on the design and coordination of local and international observational and patient-reported outcomes studies.

• A former practicing GP and clinical researcher, Núria’s experience spans roles in outcomes research at the Institute of Public Health in Barcelona and in Catalan Health Authorities, and consulting positions within the pharmaceutical and medical device industries focusing on medical regulatory and pricing affairs, pharmacoeconomics and market access strategies.

• Núria holds an MD (specializing in Family and Community Medicine in Barcelona), and a Master’s degree in Public Health from the London School of Hygiene and Tropical Medicine and London School of Economics.

Page PAGE72 72

HEALTH ECONOMICS & OUTCOMES RESEARCH IMSIMS REAL-WORLD EVIDENCE SOLUTIONS & HEOR

EXPERTISE | IMS RWE SOLUTIONS & HEOR Massoud Toussi, MD, MSC, PHD, MBA • Dr. Massoud Toussi is Principal and Medical Director RWE Solutions & HEOR at IMS Health with particular expertise in methodological and operational aspects of clinical research to assure the quality of interventional, observational and database studies.

• Previously Head of Global Clinical Research Operations at Cegedim, Massoud has also worked at the French High Authority for Health and various CROs as Project Lead, Scientific Manager and Operations Director. His experience includes defining and elaborating a new service process in drug safety signal detection and transmission.

• Massoud holds an MD from Mashad University in Iran, an MSc in Medical Informatics and Communication Technology from Paris IV, a PhD in Medical Informatics from Paris XIII University and a diploma in Transcultural Psychiatry from Paris Nord University.

Arnaud Troubat, PHARMD, MBA, MHEM • Dr. Arnaud Troubat is Principal RWE Solutions & HEOR at IMS Health. He has extensive consulting experience and special expertise in the development of registration dossiers and market access strategies across a large number of therapeutic areas.

• A pharmacist by training, Arnaud began his career at the French pharmaceutical industry association (LEEM), supporting members in understanding and interpreting the changing economical environment in France. He then spent a number of years in pharmaceutical affairs at ICI, leading regulatory work on registration submissions and reimbursement strategies, before subsequently moving into consulting. Most recently, he was Director at Carré-Castan Consultants, managing a team working for a wide range of pharmaceutical companies.

• Arnaud holds a Doctor of Pharmacy degree and an MBA from IAE Paris and a Master’s degree in Health Economics and Management from Paris-Dauphine University.

Dana Vigier, MD • Dr. Dana Vigier is Senior Principal RWE Solutions & HEOR at IMS Health, applying in-depth expertise and extensive experience in pharmaceutical pricing, reimbursement and market access to help clients meet the growing challenges of today’s increasingly complex product launch process.

• A medical doctor and INSEAD executive, Dana’s background spans 15 years in pharmaceuticals and includes roles in R&D, commercial, market access, strategy and government affairs at GlaxoSmithKline, Organon and 3M Pharma. She has worked on numerous pricing and reimbursement negotiations and designed and implemented national and international Phase II, III and IV studies across a wide range of therapy areas.

• Dana holds an MD from Bucharest Medical University, Romania and the Paris-Cochin Faculty, Paris, France.

Rolin Wade, RPH, MS • Ron Wade is Principal, RWE Solutions & HEOR at IMS Health, and a recognized expert in the applications and limitations of applying large retrospective datasets and late-phase datasets to health economics and outcomes research.

• Prior to joining IMS, Ron served as a Healthcare Executive and Principal Investigator with Cerner Research, directing retrospective database research using EMRs, administrative claims and other publically available databases. He was previously Research Director at HealthCore, where he led project teams in health outcomes research, economic modeling, safety/epidemiology and prospective observational research. He also has extensive experience generating evidence to support value messages to managed care, government payers and public health associations, gained in various leadership roles within the pharmaceutical industry.

• A widely published author with expertise across a broad range of therapy areas, Ron is an invited lecturer at colleges of pharmacy and he has served in leadership roles with the American College of Clinical Pharmacy and the Academy of Managed Care Pharmacy. He is a licensed pharmacist and holds a BS in Pharmacy and an MS in Pharmaceutical Sciences from the University of the Pacific, California.

Jovan Willford, MBA • Jovan Willford is Principal RWE Solutions at IMS Health, supporting growth strategy, offering development and commercialization of RWE solutions.

• Jovan’s background includes more than 10 years of strategic advisory experience across payers, providers, life science organizations and technology companies, including several cross-industry collaborations to advance quality and value of care delivery.

• Jovan holds an MBA from the Kellogg School of Management, Northwestern University, with majors in Management and Strategy, Managerial Economics and International Business, and an undergraduate degree from the University of Notre Dame with majors in Marketing and Philosophy.

AccessPoint - Issue 3 ACCESSPOINT • VOLUME 3 ISSUE 6

Page 73 PAGE 73

IMS REAL-WORLD EVIDENCE SOLUTIONS AND HEALTH ECONOMICS & OUTCOMES RESEARCH is based in 18 countries worldwide with regional headquarters in: EUROPE 210 Pentonville Road London N1 9JY United Kingdom Tel: +44 (0) 20 3075 4800

THE AMERICAS 1725 Duke Street Suite 510 Alexandria, VA 22314 USA Tel: +1 (703) 837 5150

LATIN AMERICA Insurgentes Sur # 2375 5th Floor Col. Tizapan México D.F. - C.P. 01090 México Tel: +52 (55) 50 62 52 00 and +1 917 542 5844

ASIA PACIFIC 8 Cross Street #21-01/02/03 PWC building Singapore 048424 Tel: +65 6412 7365

RWEinfo@imshealth.com www.imshealth.com/rwe

About IMS Health IMS Health is a leading worldwide provider of information, technology, and services dedicated to making healthcare perform better. With a global technology infrastructure and unique combination of real-world evidence, advanced analytics and proprietary software platforms, IMS Health connects knowledge across all aspects of healthcare to help clients improve patient outcomes and operate more efficiently. The company’s expert resources draw on data from nearly 100,000 suppliers, and on insights from 39 billion healthcare transactions processed annually, to serve more than 5,000 healthcare clients globally. Customers include pharmaceutical, medical device and consumer health manufacturers and distributors, providers, payers, government agencies, policymakers, researchers and the financial community.

Additional information is available at www.imshealth.com

©2013 IMS Health Incorporated and its affiliates. All rights reserved.

Trademarks are registered in the United States and in various other countries.

ACCESSPOINT0513