2014 Annual Conference

Today, we are experiencing an epidemic of "infobesity" -- we are exposed to too much information, which impacts our ability to think critically, process facts and make effective decisions -- also known as "cognitive load."

Our audiences also experience cognitive load when we present them with too much information, unnecessary detail, content that is not tailored for them, or we "deck them" with endless slides covered with copious amounts of text. The definition of genius is making the complicated simple. In BRODY Professional Development’s “Scientifically Speaking” workshop, we will show you how to create clear, compelling presentations, even with the most technically dense and detailed information.

Amy Glass, Director of Training at BRODY, can help you create messages that resonate, tell stories which are memorable, make your visuals more visual (data visualization), and present complex information in a confident, engaging way. This program will enable you to successfully sell your ideas, projects, or products with ease -- while minimizing unnecessary details and tech speak. Come to the Scientifically Speaking workshop and learn the ART and SCIENCE of scientific speaking!

Learning Outcomes After attending this program, participants will be able to:

• Assess the technical acumen of an audience to develop optimally effective presentation strategies
• Present technical information in understandable terms with clarity and impact
• Educate and engage their audience through visual, vocal, and verbal signals
• Leverage powerful visuals to assist communication of complex ideas
• Achieve increased comprehension and retention

Non Personal Promotion Planning
Pharmaceutical companies need to fundamentally re-think how they interact with prescribers as the “Reach and Frequency” used for personal promotion is becoming increasingly limited as prescribers spend less and less time with medical reps. Some of the market trends reveal that reps are making fewer calls, reps are not getting detail time, or they are leaving without detailing. This is why non-personal promotion becomes more and more important.

By understanding the segment of physicians specially, those who are not open to personal promotion, and discovering their characteristics, effective non-personal promotion such as email campaigns or paid searches can be customized.

This presentation with discuss one approach that a team of consultants from Deloitte Consulting took to utilize internal and external sources of data to uncover physician characteristics and inform non-personal promotion planning.

Life Style Based Analytics
Life Style Based Analytics (LBA) models use non-traditonal data (3rd party marketing) to predict the health risks of a population where claims data is otherwise unknown or unavailable. This method is used by insurance companies to explore total claims as well as specific diseases and conditions.

In this tutorial, the Deloitte team will present the types of external information that can be used to profile the population based on their life style based data. Finally they will recommend approaches that this method can be used in Pharmaceutical and Medical Device marketing projects for designing more effective patient management programs as well as predicting the patient behavior which leads to better outcomes.

Accurate Prediction of High Cost Insurance Claims
In this part of the presentation the Deloitte team will discuss a case study from the Auto Insurance industry. The intent is to show that by using more data sources outside of what was used in conventional models, higher prediction lift and cost management can be achieved. There will be a follow-up discussion on how to apply such techniques to Pharmaceutical and Medical Device marketing projects.

Industry research shows that a small number of claims drive a high percentage of insurance claim costs. For example, in Healthcare, 5% of the population spends 50% of the total amount and in Auto Insurance, 10% of the claims account for more than 50% of claim payment. Early intervention on high risk claims can reduce the cost of high risk claims by 50%.

Advances in data mining technology allow models to handle more predictors, even when the predictors are strongly correlated. New techniques and large data samples guarantee higher performance of complex models without over-fitting.

The pharmaceutical business has always offered a highly dynamic and challenging environment within which to work and build a rewarding career. Although this aspect of our industry will not change, the stakes have never been higher. In today’s world, developing and delivering innovative therapies to patients in need requires companies to devote up to 15 years of concentrated effort and well over $1 billion dollars in capital investment. With the advent of healthcare reform, pipeline challenges and tighter margins, the need for robust, reliable and timely analytic support has never been greater.

Every senior leadership team in every pharmaceutical company across the country is looking to their analytics professionals to help them turn “information” into “knowledge.” Data – big or small – is not the answer. “Crunching numbers” or deploying complex CRM systems to gather more information is not the answer. The “answers” come from the individuals and teams who possess the skills and the drive to truly understand what the data means, and to turn that understanding into knowledge that informs decisions that benefit the business. In today’s talk, Greg Barrett will focus on what leadership is looking for when they ask their analytics teams to help turn information into knowledge.

Brand teams make sizeable investments in acquiring data assets and commissioning resources to perform analytics. However, when a business question comes up, typically substantial time is spent in collecting, processing and integrating data. In most cases this leads to analysis not being ready in time or lacking the level of comprehension and analytic rigor needed to solve the problem. Wouldn’t it be great if the analysis was ready in a proactive way to support decision making at the time the question is asked? By taking a proactive approach to analytics, the focus can be on deriving key insights and supporting better decision making, instead of spending time setting up and conducting the analyses.

This presentation will highlight how brand teams can strengthen their overall analytic readiness while achieving greater efficiency and insightfulness. Case examples will demonstrate how the framework has helped the brand team answer several key questions spanning broad topics, level of complexity and urgency. We will conclude the presentation by offering some practical tips for management science professionals to facilitate the creation of agile systems to strengthen the brand teams they support.

Disruptive Innovation will be happening in healthcare in the next decade impacting access, quality and cost of healthcare. The innovation in products and services, care delivery models, consumer engagement programs will be driven by insights from big data. The future financial performance of the pharmaceutical companies depends on how pharmaceutical companies harness and leverage data in the patient outcome driven accountable care model. In this presentation the factors that are driving the growth of data in healthcare will be explored. Why this growth has created more the 200 companies in the last three years some with disruptive business models and what white spaces have created these companies? What does this mean to the future of pharma in the complex, uncertain and transformative healthcare system? Using design thinking approach and ongoing interviews with senior leadership the future of pharma, readiness of pharma and the challenges in creating and managing big data will be presented. A futuristic discussion on how the data can transform the current drug discovery to commercialization will be presented.

EMR adoption has increased dramatically across the US in recent years, in part fueled by the government’s Medicare and Medicaid EHR incentive programs to achieve meaningful use. EMR data contains incredibly rich patient information with key advantages over other secondary data sources and has the potential to provide pharmaceutical & biotechnology companies with a powerful data source from which they can glean key insights to understand their markets and businesses. However, manufacturers currently face several key challenges in utilizing EMR data sets, challenges that stem from hospitals and community practices implementing EMR systems from a wide variety of EMR vendors, resulting in capture limitations, inconsistent data fields, and imperfect data standardization.

This session will provide an overview of an ongoing collaborative initiative between ZS Associates and the PMSA that aims to assess the current EMR landscape, identify EMR data needs of manufacturers, evaluate current EMR data quality and usefulness (from a manufacturer standpoint), and ultimately provide recommendations to both manufacturers and EMR vendors regarding how each side can help shape the future of EMR data. Questions this research will aim to address include: To what extent have manufacturers utilized EMR data to-date, and what specific challenges have they faced along the way? What would manufacturers ideally like to see from EMR data moving forward? Who are the leading EMR vendors today? What best practices exist in collecting and providing EMR data to manufacturers, and where are those best practices being implemented? Where can improvements be made to EMR vendor systems and processes in order to increase the value of their data sets for manufacturers?

Key findings and insights from manufacturer and EMR vendor interviews to-date will be shared, with the hopes that some of you in the audience will be interested in participating in the rest of this exciting research moving forward!

Unstructured data – data found in Electronic Medical Records (EMRs) in the form of clinician notes – has traditionally been difficult to access much less quantify. Time-consuming chart reviews have been somewhat informational but labor intensive and based on small numbers. Advances in Natural Language Processing (NLP) – and the technology enabling it – have changed the rules of the game, offering deep insights into millions of patient records at faster speeds and adding never-before-seen color to the world of pharmaceutical marketing.

To succeed in a healthcare world increasingly driven by information, life sciences firms must use the entire palette of available information, including the deep color provided by physician notes. EMRs continue to evolve and become more sophisticated, but many critical components of patient care are still not found in structured data (i.e. pre-defined input fields of the EMR). The physician assessment itself – the very basis of medical care and prescribing decisions – is monochromatic without the notes. In the old black and white world, we see a diagnosis but not the chief patient complaint and complicating factors that prompted the office visit in the first place. In the old black and white world, we see a medication switch but not the physician rationale driving the medication change. In the old black and white world, it is far too easy to draw inaccurate conclusions without the light of notes.

This presentation will explore how NLP data obtained during clinical encounters and recorded in unstructured data fields can lead to insights that drive the thinking, messaging and targeting of brand strategy – in a more efficient manner than chart reviews. We will highlight key content areas from one patient note to show how, when aggregating across an entire patient population, notes can provide the color to empower better decisions.

This study provides an overview of the publicly & commercially available anonymous patient-level data sources that could be used in forecasting. It discusses their ranges of applicability, projectability, and their applications in different forecasting situation.

Depending on the patient-level data source, different statistical and analytical techniques could be applied to analyze this data for the forecasting. Selection of the analytical techniques is also strongly dependent on the intrinsic properties of the data. Some analytical techniques could be used to analyze particular data source for forecasting, other techniques could be invalid and even dangerous. This review also discusses publicly available international databases, and compares & contrasts them with the US bases patient-level data sources.

E. Loren Buhle, Jr. PhD, is a Life Sciences Executive with over 25 years in Life Sciences in roles ranging from scientific & clinical research, business management, and information systems. He spent 10 years as a medical researcher and practicing oncologist at the University of Pennsylvania School of Medicine (and ASCO member), followed by several years as a Smithkline Beecham executive in research and development. In the late 1990s, he moved into consulting at (PwC)IBM, working with global pharma, biopharma and medical device clients. His expertise ranges from infrastructure through corporate strategy, with two approved drugs and one approved medical device. He is closely associated with the Watson work on oncology with the Memorial Sloan Kettering Cancer Institute. He is a graduate of The Johns Hopkins School of Medicine.

1994 was a simpler time, a time when a young graduate with some operations research training could dream of deploying the Simplex Method to solve the problems of the day, a time when only futurists imagined digital marketing analytics, a time when physician-level prescribing data was still novel, and a time when people were just beginning to automate spreadsheets with Excel 5.0’s VBA. It was also the year I joined the industry.

The last twenty years have ushered in evolutionary & revolutionary changes to our industry, both changes that are almost imperceptible as they occur and ‘step-function’ changes that restructure core pieces of our industry, the work we do and the way we work. Through too many projects to count and too many air miles to want to remember, I have, I believe, learned some hard-won lessons, as well as ideas on where we can all still improve. This talk includes twenty lessons roughly split into answering these five questions:

• How do you build a great team?
• What makes a project tick, click and rock?
• How can we speak analytics to non-analysts?
• What can each of us do to grow our careers?
• Where do we go from here?

Although the healthcare industry has changed dramatically in recent years, the primary methodologies utilized to target physicians have largely remained stagnant. One recent marketplace dynamic that in particular has implications for targeting prescribers is the Affordable Care Act’s Health Insurance Exchanges, which have begun and will continue to impact growth in patient populations state to state, patient makeup of physician practices, drug coverage state to state, and much more. However, traditional targeting methods have significant limitations in addressing these implications, limitations that demonstrate the importance of utilizing an innovative targeting methodology that does not simply take into account these marketplace dynamics, but helps pharmaceutical and biotech companies leverage and consequently benefit from them.

To that end, The Marketing Advantage, Inc. refined our Dynamic Targeting™ methodology, a powerful system that profiles prescribers using over 130 rich behavioral metrics that shed light on the prescriber’s loyalty to competitive products, our product’s formulary status in the Managed Care plan under which a competitive product was prescribed, and much more. These behavioral metrics are a superior indicator of a physician’s performance than are standard metrics and, accordingly, result in more focused and productive sales efforts.

The pharmaceutical industry lags behind other industry sectors in the sophistication of certain of its commercial environments. This is especially true in the critical function and process of Forecasting which, in life sciences, is often disjointed, inconsistent and ineffective, leading to the suboptimal generation and distribution of business-critical knowledge. This presentation will address the issue of best-practices for business analytics as it relates to forecasting in general and how this impacts overall strategic planning in the industry.

In an industry with exponentially more data available, it’s easy to fall into the pattern of using as many measures as possible to evaluate performance. This big data trend is easy to follow in the pharma industry as the electronic capture of more and more data sources provide more refined and reliable data at the territory and provider level, now including even patient data. In developing incentive compensation plans, a key goal is to develop an effective measurement of sales force performance while motivating the field through incentives.

To effectively assess and reward performance, sales force leadership is adding more variables within performance measurement models, but are these complex models effective in measuring performance? Does the inclusion of more measurements provide a more accurate and fair assessment of performance? Do certain incentive compensation plan designs lend themselves to more data? Is there a challenge in field force confidence in emerging metrics, such as patient data-based metrics which cause the inclusion of these metrics to be additive rather than a replacement of previously-used metrics? If so, how can this be managed? What do representatives in the field think about these more complicated models? Does the increased complexity provide comfort or confusion?

Big Data Analytics is the hottest new trend in the pharmaceutical industry. The presentation addresses how the advent of Big Data will affect analysis approaches and generation of insights. The emergence of EMR data and expansion of Claims data presents a unique set of opportunities to generate actionable business intelligence. In particular, the historical focus in data analytics has been on addressing the “whats” whereas the new data sources open doors to gain a richer understanding of the “whys”. This is a subtle but critical paradigm shift with game changing long term implications for what data we use, how we use it, and the role of data analytics teams.

This presentation will cover innovative Big Data analytics, powerful visualizations, and scope for addressing business questions that were previously relegated to the realm of primary market research. The presentation will also address the caveats associated with these nascent data assets and the need for a thorough understanding of the data infrastructure.

The linkage between Incentives and Call Planning continues to be critical in the US Pharmaceutical industry. Overall, as an industry, many companies continue to strive on aggressive Call Plan Compliance (“CPC”) attainments, and CPC metrics continue to form a key part of field force effectiveness discussions. These companies are either mandated by global headquarter offices to strive and achieve almost ~ 80% CPC, or are influenced by history that pegs 80% CPC as an industry benchmark. However, field behavior is more driven by IC attainments, and in many cases there are disconnects observed between Call Plan Compliance and IC attainments. Reps would like to discover that higher CPC would lead to higher attainments, but this is not always the case. A rep may have high CPC but may only achieve 94% attainment while another rep might have a lower CPC with 110% attainment. Inherent disconnects between call plan models and IC models is often the root cause of this issue, and leads to distrust in call plans provided by the Home Office.

In this presentation, we discuss:

• Fractures in the process that causes this break in relationship to emerge
• What can we do in IC designs to account for a better linkage to call planning process. This would also include discussion on modeling approaches need to be aligned to achieve better linkage
• Finally, we will close with a discussion on cycle times and IC administration impact that our suggested improvements may cause

Launching a successful new pharmaceutical product is more challenging than ever. With declining approvals, payer pressure, fragmentation and competition, new product uptake and sales are down. It’s critical to focus on high potential assets, and maximize those opportunities. Through trend analysis of the Nielsen BASES new product database of 2000+ product profile evaluations spanning 75 therapeutic categories, and close examination of more than 50 product launches or approval submissions, Nielsen examines the impact of key drivers such as physician uptake, and market development on success. We then show how to use market research results to identify strong launch candidates and optimize commercialization.

Does a strong physician value proposition predict in-market success? What role does differentiation play? How do trends vary by category and disease severity? What is the role of a fragmented category vs. open territory? What about 1st to market vs. successors? Analyzed physician metrics include prescribing interest, adoption urgency, prescribing frequency, product differentiation, benefits vs. need, fulfilment of unmet needs and side effect burden.

Nielsen’s analysis will culminate with a category-specific Success Index framework that helps manufacturers set expectations, manage pipeline development, and focus efforts on assets that are most likely to pay out.

Over time, the pharmaceutical industry has considered various measures outside of territory level sales for motivating field force performance. In competitive environments, motivating collaboration and teamwork may prove challenging. To that end, a component of team performance has been included in many incentive compensation plans. Do these team components work? Do they effectively motivate collaboration? Most importantly, do they have an impact on an organization’s bottom line?

Using both subjective and objective data, including sales data and field surveys, we will provide the pros and cons of using plans that contain team components. We will examine various types of team components and discuss the benefits and hazards of each. Using plan results, we will compare the growth in plans with and without team measures for similar sales forces. We will utilize benchmark data from several sales forces for the analysis and illustrate key takeaways by focusing specifically on a case study of one organization’s utilization of team performance metrics.

Current patient assistance programs are designed to attract the maximum amount of patients and enable the maximum amount of utilization by those patients. The vast majority of those program designs, however, neglect to observe true consumer behavior over the course of that program’s implementation and rarely revisit program design over time, leading to overspending and a decline in potential profitability, even though revenue may remain the same.

By examining patient behavior over time, especially price sensitivity to out of pocket prescription costs and persistence of therapy, it is possible to design an assistance program with maximum efficiency which attracts and retains profitable patients while simultaneously being as cost conscious as possible, thus generating the maximum ROI.

CRM is not just another buzz word in the established arsenal of pharmaceutical communication tactics. It is a broad-based strategy grounded in proven impact across industries, and the next critical stage in the evolution of HCP and Patient engagement.

Prescription drugs and OTC medication are two traditional channels that align with the evolution of CRM. On the surface, a natural extension of this approach should be vaccines. While on the surface this appears intuitive, there are significant hurdles associated with CRM and vaccine communication. Despite this, if you rise the challenge, think about the opportunity non-traditionally and address the data gaps, the rewards will be worth the effort.

Join us as we take a look at a combination of common - sense and non-conventional approaches to building out vaccine targeting, segmentation and CRM strategies that work.

Every pharma organization aspires to identify and reach influential customers. However, the concept of “influence” has been difficult to articulate and challenging to quantify, since treatment decisions involve complex interactions between patients, physicians, and institutional stakeholders such as payers and provider networks. Using techniques derived from social network analysis, which deals with the complexities of mapping and measuring stakeholder influence, we have created a novel framework for combining multiple pharma-specific data sources and applying ranking algorithms to measure different types of influence at the customer level.

Specialist-driven markets such as oncology and rare diseases are especially appropriate for this type of analysis since treatment decisions in these markets involve a significant amount of peer-to-peer interactions, local expertise, and patient referrals. The insights gained from this analysis can be used to reshape traditional approaches to developing product strategy and engaging with customers.

This session will cover:

• Three notions of customer influence that are relevant for pharmaceutical sales and marketing: authority, credibility, and the prescribing environment
• Techniques for mapping, measuring, and visualizing influence within specialty markets
• Recent project examples that illustrate the social network approach and its applications to specialty markets.

Call Planning has been well integrated into the planning and execution of the pharmaceutical sales forces in the last several years. Call Planning is a great tactical step to tradeoff brand, sales and operational goals and help implement the best strategy for the organization as a whole. Several companies have integrated measuring call plan adherence and tasking their reps to follow the call plan. There also have been studies that try to correlate call plan adherence to rep performance. On the flip side, we know that the data is not perfect on several dimensions at a micro level - Data vendor captures a percentage of the actual data (& rest is projected), doctors treating & the doctor (pad) writing the Rx could be different, reps report what is asked of them (self-fulfilling prophecy). Given both sides of the coin, some fundamental questions still remain. How valuable are call plans to the field reps? How would the reps perform if we did not provide any call plans (compared to those who have a call plan)? Do we facilitate the reps to build their own call plans? What type of products and reps may need call plans? What if we did not measure any adherence (would we get more accurate call activity data)? We believe that Pharma companies, based on their internal cultures, could set up controlled experiments to answer these questions. Based on a real experiment, this talk will explore the approaches to answering these questions:

• We will present the operational challenges in setting up such experiments?
• We will present the results of the experiment at a couple of levels?
• How did the reps go about their activity?
• How did their activity measure up against the optimal?
• What were the results?

Physician valuation measures employed by pharmaceutical companies focus on past and current prescription volumes, and the subsequent promotional strategies are aimed at maximizing returns in the short term rather than building long term relationships. As the landscape of the industry undergoes constant changes, the focus is shifting towards creating targeting strategies with a future outlook.

In this discussion, we focus on customer life time value, a concept common to industries like banking and telecommunications. We will illustrate how the use of statistical models can be utilized to predict the net present value of physicians’ future prescribing. Physician life time value could create deep customer insight into the strategies and personalized interactions.

More specifically:

• Understand the interconnected factors that influence customer lifetime value.
• Improve the profitability of every customer interaction.
• Boost profits and reduce costs by targeting customers with the highest lifetime value potential.

In an era of patent expirations and shrinking sales forces, pharma companies have begun to move away from a “one size fits all” to a differentially resourced model, which enables customer centric deployment models at the local (sub-regional) level. Differential or Flexible resourcing replaces the idea of national sales teams with sales forces that are flexible, dynamic, and focused on local markets. This primarily involves establishing territory footprints that stays relatively stable over time, in order to maintain representatives’ customer relationships. Territories carry a multitude of products, and the product bag is customized to meet local needs.

While differential resourcing provides immense flexibility and efficiency to selling teams, motivating the sales force to successfully operate in this new model often requires adjustments to the sales compensation program. Some of the most common incentive compensation (IC) related questions and challenges are as follows:

• How should IC plans be designed to address varying product focus by region?
• How can we ensure fairness across territories with different product focus?
• What are the implications on Annual recognition programs?
• What are some IC operations/administration challenges arising due to constantly changing footprints and product focus?

In our experience, organizations need to develop capabilities in a number of sales operations areas in addition to IC design and administration to properly support differential resourcing. In many cases legacy processes and systems are the biggest barrier to implementation.

We would like to present a disciplined and analytical approach for adapting incentive plans to a flexible deployment model. We will describe how companies have addressed the most critical dilemmas and complications associated with IC design and administration, namely:

• Sales crediting for field sales reps and managers – Should reps be paid on total geography performance versus a sub-set of customers? What are considerations that drive this decision?
• Differential IC product weighting – How can product emphasis vary by territory to reflect the local differences in promotional effort?
• Designing fair IC plans – What are techniques companies can balance earning potential across territories and to address ongoing changes?
• Annual recognition programs – How can annual programs like President’s Club be adapted to address different product mixes, and ensure that winners are not always ones that carry high-performing products or territories with favorable selling environments?
• Incentives operations – How can incentive operations teams and systems adapt to readily accommodate variable product emphasis and local mid-period changes, while ensuring efficient and error-free operations?

Commercial effectiveness is increasingly more dynamic and complex. Physician focus is shifting towards overall patient health management. Payers are exerting more control over physician behavior to mitigate costs. At the same time, more generic products are entering into market and reducing the revenue potential of branded therapies. It is critical to quantify Payer, provider and patient level dynamics to optimize promotional effectiveness.

SHS has constructed High Value Target (HVT) solutions that help us understand these dynamics better at the customer level. SHS leverages its Integrated Dataverse to derive these solutions. Solution aims at better understanding the customer behavior, identifying the right set of customers, targeting them the right way and at the right time. Solution looks at all the transaction level forces that are in play starting with patient visits a physicians, physician diagnoses the condition, evaluates the options looking at cost, efficacy and safety, physician makes a decisions, payer processes the claims, and patient receives the treatment. Bryan Galatro has 17 years of experience in the healthcare and pharmaceutical industries, with a particular focus on Sales Force Effectiveness, marketing science, contract strategy and business development. He specializes in rapid and continuous innovation.

Reckitt Benckiser Pharmaceuticals has identified a growing trend in the payer space – a payer type that significantly discounts prescriptions for the patient, but is sponsored by an independent third party. Further, many insured patients have benefit designs that require significantly high co-pays 27 above and beyond reasonableness. This creates an environment where typically insured patients behave similar to cash paying patients. This analysis seeks to identify, group and quantify those payers whose membership is at risk due to the above criteria.

This presentation covers a new algorithm we developed to profile users of a branded Statin that over the years have grown sicker and smaller in number as patients that have their LDL well under control have been lured away by the low co-pay of generics. The algorithm has three distinctive features. First, it uses only a small number of the most relevant attributes out of a large pool of potential attributes to profile patient segments. Second, it highlights attributes which when turned on or off are accompanied with a dramatic change in drug usage. Third, it transcends the LDL problem that motivated its creation and is applicable to a large class of profiling problems.

The algorithm takes a geometric approach. Each patient is represented as a string of 0’s and 1’s where each bit of the string corresponds to a specific attribute. Each string is in turn viewed as coordinates of a hypercube, where properties such as number of patients and drug usage are assigned to each vertex. The algorithm then proceeds with a local exploration of the hypercube, analyzing one pair of vertices at a time, looking to recognize one of two situations. When conditions that correspond to the first situation are fulfilled, vertices are merged, which amounts to discarding dimensions and elimination of irrelevant attributes. When conditions that correspond to the second situation are fulfilled, the vertices are left intact but instead the edges (attributes) that connect the pair of vertices under scrutiny are earmarked as triggers of dramatic change in drug usage.

The algorithm discussed here has distinctive features and offers specific insights that traditional techniques such as regression, factor analysis, and clustering do not. In our humble opinion, this algorithm arguably deserves a place in the market researcher’s toolbox.

At the root cause of issues facing the Pharmaceutical industry today is the tension between access and incentive – influenced by healthcare reform, generic drug market and changing sphere of global market. While healthcare reform seeks to reduce the cost of pharmaceutical drugs and extend coverage, it is also expected that the share of biologics, among all pharmaceutical drugs, will grow to 50 per cent by 2015 (IMS Institute of Healthcare Informatics, 2011). At the same time, contracting increasingly continues to be a differentiator in the managed markets space for pharmaceutical companies as they look to increase profitability. Therefore, at first glance, internal commercial functions within pharmaceutical companies addressing these changes in external environment – healthcare reform, increasing share of specialty drugs and growing importance of contracting – seem disparate in operations. But the fundamental need across these functions for effective and innovative strategies, policies, processes, data and technology support is stronger than ever. Given this situation, there is a need for a pioneering approach to effectively integrate specialty markets, commercial managed markets, government policies and contracting under a common assessment and solution framework, especially with respect to data and reporting. This paper/presentation introduces such a pioneering approach that uses just the foundational operational elements of data and reporting in order to accomplish the complex integration between specialty markets, commercial managed markets, government policies and contracting.

The market definition for a new product is one of the most important decisions to make ahead of product launch. Success, when a market is defined wide, requires an improved product profile, lower prices or superior market powers. When superiority studies are risky to attempt, when the market is so established that the lion’s share is owned by non-drug therapies and a significant percentage of the remainder is owned by generics, market powers are the only remaining leverage. If these too are sadly limited, then the only strategy that makes winning sense is to smartly array these resources against a narrower target market. The question then becomes how narrowly to define the market so that deployable market powers are both effective and economical. Our case is that of a product characterized as me-too from an efficacy perspective, superior from a convenience stand point, and that was launched in a very wide market. Novelty and convenience conspired for a while to give it a small market share in the relatively small market of drug therapies. But a ceiling seems to have been hit now, and the time to re-evaluate initial assumptions has come. One of these assumptions is precisely that of a market defined as the totality of available drug therapies, 29 branded and generics. As is commonly practiced, prescribers of the market were segmented along the market volume dimension. Promotional resources were assigned with that segmentation in mind. Because the product was new, no response modeling was possible and many of the resulting high prescribers received intense personal promotion despite the fact that most or even all of their prescribing was for generics. The nature of this product precludes the existence of switch data to examine its sources of business; the percentage of Rx’s sourced from generics vs. the percentage sourced from other brands. Despite this limitation, we postulate a state space based approach to derive the switching behavior of prescribers and are able to deduce that only a very small percentage of Rx’s are effectively a switch from what would have been a generic. Our conclusion is that the sources of business for our product have historically depended little on generics. And therefore, the market definition can safely ignore them in order to allow a potentially more promising deployment of limited personal promotional resources.

This presentation provides both an understanding of the transition in Life Sciences of Primary Care anchored portfolios to a hybrid of primary care and specialty products and the challenges and transformations that are associated with shift. The focus is on both embracing, demystifying and showcasing a path forward to both support a crossfunctionally effective and strategic infrastructure of the “new” normal.

The pharmaceutical landscape will continue to evolve at a rapid rate with changing payer environment, increased Payer controls and emerging Integrated Delivery Networks (IDNs) influencing treatment choice. The variation of these stakeholders and how they interact and influence each other by local geography has led to a new U.S pharmaceutical landscape of individual regional markets.

Many pharmaceutical companies are seeking to minimize costs while simultaneously increasing their 30 attention on changing and emerging stakeholders, regional based go-to-market strategies and structure for differential resourcing and tactics. Most approaches to sales force optimization and contracting effectiveness are done somewhat in isolation, and rarely while assessing the interaction of payer controls and the impact of groups like IDNs. A consistent set of metrics or framework available to measure the total national impact of stakeholders and the variable influence Payers or IDNs exert in local areas is lacking.

An approach utilizing metrics derived from integrating data across Payers, IDNs and local geographies to better understand their influence and impact on brand performance is critical for pharmaceutical companies to adapt to the changing Payer and IDN environment. This poster outlines new measures and a framework to guide enhanced strategic decisions at the local level, keep pace with the changing landscape and also provide the basis to model resource allocation consistently across Payers, Provider Networks, and physicians.

Advances in analytic methodology, network science and “big data” have allowed the healthcare industry to accelerate its use of new techniques to help improve performance in core pharma promotion activities. This presentation highlights, with case examples, the blending of network science with multiple secondary data sources that provide relevant, actionable insight. The use of secondary data instead of survey-based data provides a broader, more cost effective and objective perspective. Coupled with Network science techniques, targeting for multiple promotion channels can be optimized.

A more effective and operative use of Big Data has become a primary focus of the marketing and analytics groups within pharmaceutical companies. Identifying and, more importantly, accessing the appropriate patient data that drives improved outcomes and more efficient marketing practices, however, is a complex issue that marketing and analytics teams must navigate effectively and carefully (i.e. HIPAA compliant) in the new world of health care. Lab data represents a key component in physician diagnosis and treatment of patients in each patient’s journey. This presentation will discuss the implications of lab data, as well as the corresponding analytics of this patient-level data--on advanced patient analysis and its role in the development of pharmaceutical marketing programs designed to accurately identify prospective prescribers. It will also describe how the application of analytics to clinical algorithms based on patient-level lab data can be used to more precisely identify treaters by specific condition and lead to more effective promotional activities. For example, greater access and understanding of the lab data opens up new opportunities for improving sales force planning and optimizing the rep/physician interaction. An 32 analysis of the data can validate current physician targets, as well as identify new accounts with high rates of specific types of patients for whom the physician is considering treatment options. Case studies will be used to demonstrate the power of lab data within the diabetes population and how the utilization of this patient-level data can drive more effective results at the point of care. The robust and readily-usable database of patient-level data in this case study is composed of 50 million patients representing 200,000 physician practices.

Branded pharmaceutical products face lossof- exclusivity and generic competition in the marketplace. Manufacturers must understand the impact of generic entry on a mature drug category in order to better inform targeting and promotional efforts for loss of exclusivity. Adheris performed a two-part analysis to elucidate key drivers of brand-loyal patients and characteristics of those who switched to generic. Data was obtained from the Adheris computerized pharmacy prescription database of more than 22,000 retail pharmacy outlets.

An observational analysis was performed across three analog products to determine commonalities at the patient, prescription, and prescriber level in regards to brand-loyalty and switching activity. Characteristics associated with brand-loyalty include patient experience on therapy, decreased co-pay amounts, and patients with an increased percentage of branded medications.

After the observational analysis, we sought to create two predictive models for each analog using logistic regression. The first model identified characteristics predictive of brand loyalty postgeneric entry, while the second model identified characteristics predictive of switching to generic. Five variables (days supply, number of refills prescribed on index prescription, patient copay amount, prescriber specialty, and median household income) were forced into both predictive models for each analog. Other variables were selected in a stepwise fashion, based on their 33 statistical significance at the level of α=0.05. The overall fit and predictive ability of the final logistic regression models were assessed using the Hosmer-Lemeshow goodness-of-fit test and c-statistic, respectively. Variables found to be significant predictors of brand loyalty include an increased day’s supply of medication, lower patient co-pay amounts, and an increased number of refills prescribed. Variables found to be significant predictors of switching activity include higher patient co-pay amounts and higher patient overall medication cost burden. These results can be used to formulate actionable recommendations to improve patient retention leading up to and following product loss-of-exclusivity.