Big Data Poised To Change Pharma R&D, McKinsey Suggests

A McKinsey & Company white paper highlights the potential for big data gathered from across the health care spectrum to revolutionize pharma R&D.

As information technology systems continue to evolve and more data becomes available across all aspects of the health care environment, the availability of big data is poised to transform research and development in the pharmaceutical industry, the consultancy McKinsey & Company believes.

The idea of big data has caught on in some quarters of biopharma, with hopes it could offer new approaches, reveal and validate new treatments and also foster new types of collaboration (Also see "Orion Bionetworks MS Project Plumbs Big Data To Revitalize CNS Drug Development" - Pink Sheet, 19 Mar, 2013.). The promise is attracting new investors (Also see "Heavy Hitter In Digital Health Joins Aberdare" - Medtech Insight, 17 Apr, 2013.). And it is also is drawing in participants who first gained their expertise outside of pharma (Also see "Google's Doc: A Physician Leads Google Ventures Into Health Care" - Pink Sheet, 25 Mar, 2013.). But some observers wonder if big data is more than the latest buzz word and whether the tools and lessons used elsewhere can be applied easily and productively to the biopharma industry given.

Weighing in on the optimistic side, McKinsey & Company estimates big data could generate up to $100 billion in value annually across the U.S. health care system “by optimizing innovation, improving the efficiency of research and clinical trials, and building new tools for physicians, consumers, insurers and regulators to meet the promise of more individualized approaches.” The figure is contained in an April 2013-dated article “How Big Data Can Revolutionize Pharmaceutical R&D” posted on the company’s website and authored by McKinsey Principal Jamie Cattell and Associate Principals Sastry Chilurkuri and Michael Levy.

“Big data” is a term that generally references data sets that have grown so large that it becomes difficult for standard software to process information efficiently. Within the pharma industry, as health information technology expands, data is coming from a wide range of external sources beyond what manufacturers generate in-house as part of the development process, creating that big data environment. Sources such as electronic health records, online patient communities, claims and other data contribute to the big data that can be used to drive R&D decisions.

“The big-data opportunity is especially compelling in complex business environments experiencing an explosion in the types and volumes of available data,” the authors say. “In the health care and pharmaceutical industries, data growth is generated from several sources, including the R&D process itself, retailers, patients and caregivers. Effectively utilizing these data will help pharmaceutical companies better identify new potential drug candidates and develop them into effective, approved and reimbursed medicines more quickly.”

Expanding And Managing Data Collection

The consultancy suggests the impact may be far-ranging, identifying eight broad areas where pharmaceutical companies may be able to unlock the potential of big data and noting some issues pharma should consider to both expand the data it collects as well as manage that data to help improve efficiency.

The first two key areas are data integration and increased collaboration, both in-house and externally. “The ability to manage and integrate data generated at all stages of the value chain, from discovery to real-world use after regulatory approval, is a fundamental requirement to allow companies to derive maximum benefit from the technology trends,” the authors comment. “By breaking the silos that separate internal functions and enhancing collaboration with external partners, pharmaceutical companies can extend their knowledge and data networks, ” the consultancy adds, citing Eli Lilly & Co.’s Phenotypic Drug Discovery initiative as one example of a more open and broad-based approach to collaboration. (Also see "Lilly Hopes To Score Novel Compounds Early By Offering To Screen Discoveries" - Pink Sheet, 15 Jun, 2009.).

Third on the list is employing IT-enabled portfolio decision support for tasks such as deciding which compounds to progress through the company’ pipeline. The authors recommend the use of tools such as smart visual dashboards to allow for rapid and effective decision making. Fourth is a recommendation that pharmaceutical R&D continue to employ cutting-edge tools, such as sophisticated modeling techniques like next-generation human genome sequencers.

The fifth area is reaching into the consumer space by deploying sensors and other data-collecting devices. “Advances in instrumentation through miniaturized biosensors and the evolution in smartphones and their apps are resulting in increasingly sophisticated health-measurement devices,” the authors state. “Pharmaceutical companies can deploy smart devices to gather large quantities of real-world data not previously available to scientists.”

Sixth, authors recommend pharmaceutical companies look to raise clinical trial efficiency by, for example, employing new, smarter devices combined with an emphasis on fluid data exchange which could boost improvements in clinical trial design and efficiency. Examples of potential clinical trial efficiency gains include dynamic sample size estimation (or re-estimation) and other protocol changes to enable rapid response to emerging insights from the clinical data; adapting to differences in site patient recruitment to allow pharma to adjust to sites that are under-recruiting as well as expanding sites that are successfully recruiting; and the increased use of electronic health records to capture trial data (Also see "Low E-Health Record Adoption Presents Challenge For R&D Use, AstraZeneca Informatics Scientist Says" - Pink Sheet, 22 Apr, 2013.).

To improve drug safety and risk management, the authors note that data to add to safety monitoring is coming from a variety of non-traditional sources, such as patient inquiries through Internet search engines, online communities, social media and EHRs. “Bayesian analytical methods, which can identify adverse events from incoming data, could highlight rare or ambiguous safety signals with greater accuracy and speed.”

Finally, the authors recommend a sharpened focus on real-world evidence, noting that such outcomes are increasingly important as payers increasingly impose value-based pricing. Payers have been involved in big data approaches, as illustrated by Optum Labs, launched in January by UnitedHealth’s Optum Inc.’s research business unit and the Mayo Clinic to link clinical data with claims data to help find new ways to lower the cost of health care (Also see "UnitedHealth’s Optum, Mayo Clinic Link Claims And Clinical Data For Research" - Pink Sheet, 18 Jan, 2013.). Another example is Pfizer’s partnership with Humana (Also see "Pfizer Research Partnership With Humana Could Inform Drug Development Decisions" - Pink Sheet, 14 Oct, 2011.).