By Todd Skrinar, principal in the Advisory Life Sciences practice and Thaddeus Wolfram, manager in the Advisory Life Sciences practice, Ernst & Young LLP
Near the end of 2013, many in the life sciences industry were looking for clear evidence that the FDA was willing to work with industry to get more needed drugs to patients. Eyes were focused on the “scorecard” of new drugs approved, which for the first eight months of 2013 reached 18.
While this number was down from 22 during the same period in 2012, it still outpaced what was a very sluggish approval pace through much of the 2000s. The FDA’s most recent trend seems encouraging.
The bigger challenge is the sustainability of the R&D process itself. Clinical trial costs, in particular, are driving intolerably high R&D expenditures. In the midst of slow sales growth, it is no surprise these costs remain a focus of the industry’s continued belt-tightening.
The industry is demonstrating an interesting range of strategies to make R&D more efficient, including dispersing risk through open innovation, collaboration, and partnerships, as well as diversifying by targeting personalized medicine and orphan and niche disease markets. But a key question remains: How does one reduce clinical trial costs while still meeting the rising demands of regulators and payers for more data that demonstrates that the drug is a significant improvement over current standards of care?
Effective use of Big Data is increasingly seen as the path forward. It offers opportunities for cutting clinical trial costs while providing the type of robust data required for both approval and reimbursement.
The Value Of Big Data
Big Data for R&D is less about velocity and more about variety, viability, and sometimes volume. The key analytics capability for this data is the ability to visualize relationships and patterns. By combining real-world outcomes data with clinical data and through the mining of genetic data and a broader understanding of regional and population data, analytically savvy organizations can begin to recognize research failures faster, design more efficient clinical trials, and speed the discovery and approval of new medicines while lowering costs along the way.
Whether it is “-omics” data, patientrelevant social media, payer claims, or patient electronic health records, a limitless amount of patient data is now available and is enabling companies to achieve impactful R&D goals including:
Given the rapid changes in technology and the ambiguous regulatory environment surrounding the use of such data, many companies are struggling to meet compliance, privacy, data quality, and other challenges. However, there are several steps companies can take today to keep pace with, or even leapfrog over, their competitors in harnessing the power of Big Data to improve their R&D efforts.
STEP 1: Establish a clear analytics strategy. The first step in incorporating Big Data into your R&D operations and decision making is to define an analytics strategy and operating model that includes a center of excellence. The center of excellence provides a sustainable core to drive the ongoing execution of the analytics strategy within the day-to-day running of the business, and fosters a collaborative environment that generates the necessary tools for both creators and consumers of analytics to extract the greatest value from Big Data. The R&D analytics strategy will be driven by the needs of the business, not technology. A strategy motivated by available or interesting tech will too often cloud decision making, not clarify it.
STEP 2: Identify the most relevant sources of Big Data. Given the large and disparate amount of data now available to life sciences companies, it is easy for an organization to quickly become overwhelmed. The defined R&D analytics strategy (mentioned in Step 1) will, by nature, provide initial guidance for why any data source is valuable or not. The data that offers value can be filtered further based on the potential impact it holds for the business. The process then advances from targeting the right Big Data opportunities to an assessment of key factors, including accessibility of the data, security requirements surrounding the data, and the effort it will take to make the data usable.
STEP 3: Master large-scale data management. The capability to appropriately access, pool, and maintain large volumes of data from varied sources will, of course, be critical to success. While there is a wide range of tools, technologies, and platforms available for delivering this capability, the appropriate choices depend on the sources of Big Data and the analytics targets identified in Step 2. Assessing the current foundational IT and analytical state will present a clear picture of the steps needed to reach the appropriate level of large-scale data management.
STEP 4: Pursue meaningful collaborations. The structure and demands of today’s healthcare ecosystem mean no one organization can go it alone. Data access is one of the many activities that at times requires cooperation between two or more parties. For example, collecting patient information is certainly not something that a biopharma company can just go out and do. However, what that company can do is gain access to the right information from electronic health record data by partnering with the institutions that are able to collect and maintain it. Establishing data partnerships with other life sciences companies as well as academic institutions, providers, and payers is key to gaining access to the widest range of Big Data possible. Companies pursuing these partnerships also need to think “win/win” when determining their positions on intellectual property, risk, and resource commitments. Success will depend on selecting like-minded business partners and using trusted third parties to support the datamanagement challenges.
STEP 5: Optimize your analytics organization for performance, value, and continuous learning. Improving the performance of R&D requires a constant search for new insights by combining and analyzing nontraditional data sources. Complacency around Big Data will eventually lead to missed insights, overlooked efficiencies, and an inadequate analytics function. To guard against this, establish a continuous feedback loop to understand the results of analytics and apply them to future analytics efforts. This process requires skills, structure, and management behavior that all drive a culture of continuous learning and improvement.
STEP 6: Derive and define your value. Successfully utilizing Big Data is ultimately about deriving value from the data in a manner that drives effective decision making and that enables a company to demonstrate the value of its product to patients and to the healthcare system overall. After providing the right information to drive better decision making, the biggest challenge remains: articulating both the quantitative and qualitative benefits of R&D analytics efforts and the downstream R&D efforts to the appropriate stakeholders, including internal clinical development and operations teams, as well as payers and patients.
In communicating the analytics benefits, it is important to have predefined short-term and long-term metrics for assessing the impacts. By doing so, the results are aligned to specific targets and can easily be used to inform the selected audience of the value that R&D has created for them.
Companies that master these steps will build a more sustainable approach to R&D and develop competitive advantages in the life sciences space. They will be better poised to deliver products and solutions that meet the specific needs of patients — individually, stratified, and at a population level. And they will be able to do this with lower R&D costs, with a streamlined route to market, and with a clear knowledge along the way of exactly which patients can benefit from the therapies they deliver.