2017 Winter Symposium

Just yesterday, PLD was a new data asset primarily geared towards understanding patient journey, identifying sources of business, and measuring patient compliance and persistence. Identifying KOLs and mapping out referral patterns and spheres of influence were regarded as cutting edge as it required a mental shift from the patient as the object of the study to the patient as a mere thread that connects more important entities, namely, physicians. Patient-level claims data assets were so big compared with physician-level data assets they displaced (at least by one order of magnitude) that one had to be insane to suggest that this behemoth of data asset needed to be combined with yet other data assets to maximize insights that can be gleaned.

How things have changed in the past 5 years! Today, patient-level data sources are routinely combined with all kinds of other data sources: other claims data assets, EMR's, registries, lab results, pharmacy and medical formulary data, patient demographics (including credit score, ethnicity, education level, and lifestyle segments), enhanced physician profile, and the list goes on. Why did that happen? For two reasons mainly. First, to achieve a larger footprint. Marrying two databases of different footprints obviously results in a database of a larger footprint. Second, to answer clinically more pointed questions. Case in point: What is the market share by line of therapy of a drug taken by a particular subpopulation of metastatic cancer patients. Claims data does not indicate line of therapy, nor does it indicate that the cancer is metastatic.

We also see patient-level data assets deployed in ways we have not seen before. Normally, patient-level data provides data for analysis and analysis provides insights to act upon. These two tasks are distinct and take place sequentially. Lately, EMR's have been used not only to provide data for analysis, but also to identify physicians in real time and even take a shot at changing their prescription behavior. EMRs can now be programmed to fire pop-ups to physicians as soon as their patients fulfill pre-defined characteristics. The pop-up takes the physician to a web site that invites the physician to opt in, and if the physician consents, the physician grants the manufacturer the permission to target the physician. Another trend also worth noting is the site alert which owes its success to its real-time dimension. Replenishment signals coming from medicine cabinets located in physician group practices are meant to tell the GPO when to ship which products to which practices. These signals have been repurposed and are now dispatched to reps so they would know which physicians to call on the following day.

How did we get there? There are essentially two forces at play. The first one is a tapestry of trends that have guided the marketplace. They include the rise of the EMR, explosion of the digital space along with the big data phenomenon, release by CMS of several databases in keeping with the Freedom of Information Act, monetization of data by any outfit that holds some kind of data. The second force has to do with the implications of the shift in emphasis from primary care drugs to specialty and orphan drugs. Sadly enough, data sharing has bucked the trend of openness. Manufacturers of specialty products in particular have gone to great lengths to ensure that the competition does not have access to their data. Interestingly, relentless data blocking has led to very creative ways to track the market.

In this session we'll go over the two forces that gave birth to what we refer to as PLD 2.0, the tangle of creative combinations of databases to plug holes and answer clinically ever more complex questions. We'll pause and reflect on what PLD 2.0 may have in store for us, and conclude by taking a stab at what we believe constitutes a good data strategy moving forward.

We all know what the challenges are in the industry:

• Increasing study complexity -- involving niche patient populations and tighter I/E criteria, sophisticated endpoints, and adaptive designs
• Fewer investigators taking part in research, and patient recruitment becoming even harder
• Protocol amendments -- which delay timelines and add costs
• Sites -- that are initiated but fail to recruit a patient, or under-enroll
• New and expanding data sources – which are difficult to identify and incorporate to improve development
• Different evidence requirements for regulators and payers

By merging the “big data” access, technology and analytics with clinical trial capabilities, this allows us to replace the assumptions that are inherent in clinical development today - with much more credible pulls of real-life data. So we can replace those “pretty good guesses” (that we ourselves have relied on too) with real-world evidence.

We will demonstrate through the use of case studies how to:

• Increase predictability – by creating study plans that are based on evidence, in order to mitigate the operational risks.
• Shorten timelines – through using data to help us select the right sites and recruit the right patients, so we can deliver best-in-class delivery timelines
• Maximize asset value – through helping you make decisions based on real insights, to generate evidence that is relevant to stakeholders, earlier in the process

We will leverage four main cases studies to demonstrate how the inclusion of real-world data leads to better clinical trials. Those case studies will include:

• Multiple sclerosis trial feasibility in Europe
• US chronic heart disease trial recruitment
• US inflammatory bowel trial recruitment
• Diabetes global study feasibility and recruitment

Public data (incidence, utilization, census, open payment) integrated with claims data can provide compelling insights to understand how patients are treated. We leveraged big data technologies to procure, clean up, stitch together and analyze multiple data sources to identify local areas where patient care was not being delivered in an optimum manner.

Perhaps no industry is drowning in data more than life sciences. This is especially true when considering Real World Evidence (RWE) data, which includes everything from physician utilization patterns to patient treatment options to drug effectiveness. Market estimates suggest big pharma spends more than $20 million annually on RWE data, but the industry is still struggling with how this impacts pharmacologic treatment on patients and the healthcare system and how a multitude of sources can be unified under a common model to drive normalized insights for real-world decision-making.

This session will focus on:

• How the increase in data sources and the demand to leverage them for business benefit requires a common data model;
• How the translations of data in accurate and actionable insights requires comparative analytics and the right blend of business and technical acumen;
• How the derivation of analytics and dissemination of information requires a technology solution that can harness the power of RWE and broaden its use across the clinical-commercial continuum of an organization

The pharmaceutical industry is rapidly shifting towards specialty products in terms of promotion. With these specialty products, academic centers and IDNs become a greater portion of both sales and influence into overall distribution models. Gaining further insights into ever changing market dynamics, including disease specific utilization on treatment pathways in oncology, has become crucial in planning and implementing sales and marketing strategies.

This presentation will provide a real world data application that demonstrates how use of data aggregation coupled with other information can provide powerful insights for marketing and sales implementation.

To highlight the value of expanded data and insights for academic and IDN entities, a business case study will be presented. The case study will focus on scenarios demonstrating how data aggregation provides additional insight and action ability into challenged outlets of care from the historic perspective. This will evidence expanded insight into prescribing behavior for complex oncology markets and multi-indication products for insight and treatment sequencing can be used to effectively (and more accurately) rank physicians based on where in the treatment spectrum they prescribe a particular oncology drug. This information is paramount in tumor types with increased competition and complexities where understanding potential and performance at very detailed levels becomes increasingly important.

The expansion of real-world healthcare data coupled with innovation in advanced machine learning presents unique opportunities for using predictive analytics to gain deeper market insights, identify next best customers and design proactive solutions.

In general, predictive analytics is helpful in:

• gaining a better understanding of the key drivers of a certain dynamic (ex. key predictors of non-adherence, disease, other medical outcome, etc.)
• predicting the impact of an event, and taking action prior to the occurrence of the event

We will define predictive analytics, then showcase how it can be used with real-world data to find undiagnosed patients, predict non-adherence, and provide key treatment insights for clinical decision support tools.

We will be sharing examples from several therapeutic areas such as neurology, ophthalmology, and rare diseases:

• Increasing Accuracy and Speed of Diagnosis: Do delays in diagnosis result in your brand’s under-penetration of the market?
• Disease Progression and Focused HCP Trigger Alerts: Are you interested in targeting the right HCP at the right time?
• Treatment Response Profiling: Will identifying patient segments that best respond to treatment help you differentiate your brand among payers and providers?
• Predictions for Non-Adherence: Would you like to understand the drivers of adherence and design interventions that help patients stay on treatment?

The ubiquitous volume/share framework can be profitably augmented with Anonymous Patient Level Data (APLD). In our research, we identify the relevant patient characteristics that drive choice and we associate these patient level variables with physicians. Thus we are able to isolate specific physician preferences in treatment net of the variation in the types of patients that they treat. It’s only by taking a close look at patient level data that we can uncover such HCP preferences in patient care.

There are currently a number of sources of APLD and they have various strengths and weaknesses. We open our presentation with a review of three major datasets; we then note the relative advantages and disadvantages of each in the diabetes therapeutic area. While all three of the datasets provide the research with a wide array of patient characteristics on a massive volume of patients, each presents its distinct challenges in data integrity checking, temporal aggregation, coverage/projection and data staging. Once the patient typologies have been distilled, the data has been rolled up to the HCP level, the key drivers identified and has been temporally aggregated, then we can use a variety of dashboarding tools to present information to decision-makers.

Take a deep dive into the essentials of Real Word Data in this informative session. Attendees will take part in hands-on exercises using patient level data. Exercises will focus on how to summarize and visualize longitudinal patient data using a ‘patient journey’ framework, and will have the group exploring data and defining business rules to generate insights about selected patient populations.

We will focus on the most frequently encountered sources, including electronic health records, medical and Rx claims, and diagnostic testing data from labs. The workshop will address practical challenges such as setting up an environment for data access considering the daunting size of many RWE sources. We will discuss the ‘workbench’ required for RWE data sources and appropriate technologies for storing, blending, analysis and visualization.

As RWE data sources are a significant cost, we will discuss how to make the most of this investment. This includes tips for demonstrating value quickly, removing barriers to data access and integrating the data into practical business applications. We’ll also share a powerful approach to problem solving that combines ‘deductive’ hypothesis-driven analysis with ‘inductive’ analysis that lets the data speak. Case examples will be presented from a practitioner’s perspective, covering:

• Defining objectives
• Specification of required data
• Developing an analysis plan
• Creating outputs

When done properly, RWE becomes integrated into a broad range of analyses across the organization. We’ll leave you with a capability building blueprint that provides a template for developing your end-state vision and roadmap to get there.

For many pharmaceutical companies today, communicating to healthcare practitioners has become more challenging. Sales force numbers have been on a declining trend while rep access to physicians has been getting more and more limited. As a result, the need for an effective integrated multi-channel communications program has never been more urgent.

While many pharmaceutical companies have taken note of this reality and are communicating across multiple channels to get their message in front of physicians, just 20% use cross-channel data and attribution to evaluate all marketing touchpoints.1 Employing analytic strategies that evaluate multi-channel campaigns within the individual channel silos will not tell the full story and may lead to inaccurate insights on the value of each channel and therefore, result in a less than optimal multi-channel campaign.

In this presentation, we will share an analytic framework for data integration and analysis that will help pharma companies move from the traditional HCP targeting approach based on prescribing volume to a next generation: more preference-driven, individual-centric way of communicating to HCPs. We will show how through a deep analysis of engagement patterns by content and channel and how these patterns influence prescribing, we can leverage insights from the analysis to create promotional strategies that deliver and incent high-value, emotionally-resonant brand engagements.

The presentation will include a case study and we will take the audience through the framework starting from an overview of the methodology to the data used and finally, to the analysis itself along with the results and strategic recommendations for implementation. It will include lessons learned as well as data/analytic-driven innovative ideas to embed these new segmentation schemas into machine learning/artificial intelligence tools to implement optimal personalized multi-channel engagements.

Take a deep dive into the essentials of Real Word Data in this informative session. Attendees will take part in hands-on exercises using patient level data. Exercises will focus on how to summarize and visualize longitudinal patient data using a ‘patient journey’ framework, and will have the group exploring data and defining business rules to generate insights about selected patient populations.

We will focus on the most frequently encountered sources, including electronic health records, medical and Rx claims, and diagnostic testing data from labs. The workshop will address practical challenges such as setting up an environment for data access considering the daunting size of many RWE sources. We will discuss the ‘workbench’ required for RWE data sources and appropriate technologies for storing, blending, analysis and visualization.

As RWE data sources are a significant cost, we will discuss how to make the most of this investment. This includes tips for demonstrating value quickly, removing barriers to data access and integrating the data into practical business applications. We’ll also share a powerful approach to problem solving that combines ‘deductive’ hypothesis-driven analysis with ‘inductive’ analysis that lets the data speak. Case examples will be presented from a practitioner’s perspective, covering:

• Defining objectives
• Specification of required data
• Developing an analysis plan
• Creating outputs

When done properly, RWE becomes integrated into a broad range of analyses across the organization. We’ll leave you with a capability building blueprint that provides a template for developing your end-state vision and roadmap to get there.