Lynette Ferrara

An aging population, the end of the "blockbuster drug era," and mounting drug discovery and development costs are pressuring organizations to dramatically improve clinical trial efficiency. Innovation in trial design and management to address the growing pressures requires a new approach to information management—one that is collaborative and allows information to be aggregated, accessible, and reusable.   

Next generation information management also needs comprehensive data mining and analysis tools to support broader and deeper understanding of the impact of the trial drug on patient populations. Collectively, integrated data management and the associated new processes, business models, and technology architecture are part of what industry is calling enterprise health intelligence. 

Historically, pharmaceutical organizations have excelled at collecting and analyzing vast and complex data required for clinical trials, while complying with the detailed regulations at many levels. The result, unfortunately, is robust data management at the study level—creating silos of data, most in different data formats, and using different coding schema. To move beyond siloed clinical trials data and study focused analysis, an organization needs to start at the top to implement an approach to enterprise-wide management of internal and external information. 

Real-world health care data in the form of administrative claims data, lab data, electronic medical records, and the longitudinal electronic health record (EHR) have important roles to play. In de-identified forms, these data can be incorporated into standard practice to support all stages of research, development, and postmarket support. Protected health information accessed through models involving patient consent will play an important role too in areas such as personalized medicine and outcomes management. 

When all these changes and technologies are put into place, information becomes a strategic asset that takes time, cost, and risk out of clinical trials. Following are a few examples of how health intelligence solutions can improve clinical trials:  

 Large-scale patient population databases enable study designers to test clinical trial protocol inclusion/exclusion criteria to optimize protocol design. 

Research sponsors can easily identify investigators from sites known to have eligible patients, which speeds patient recruitment. 

Existing business relationships with organizations having HIPAA and IRB compliant business processes for identifying and recruiting patients reduces clinical trial time and costs. 

Broader clinical insights can be derived from studies that include noninterventional study control groups' data.    

The health intelligence platform and information management program structure depicted in Figure 1 shows the major components that help organizations come to consensus on technology, business, and data decisions related to building their specific business solution. Health intelligence program success depends on a top-down approach to clinical information planning, monitoring, and coordination, starting with overall program stewardship, and including data management services and core business and support services—all needed to create custom health intelligence business solutions. 

 Stewardship defines the group's clinical information management strategies, creates a plan for building the technology solution and the first set of high-priority business solutions, and establishes the processes and governance for ongoing support and development. This process not only produces an action plan, it also quantitatively declares expected gains and builds executive awareness and sponsorship needed to sustain the development of this new competency. 

 Data services use a transformational process that not only maps data from disparate sources into a consistent data model using standardized code sets, but progressively increases their quality and utility for reuse with sophisticated tools and industry recognized vocabularies. Robust management and security function, also part of data services, will monitor and audit changes applied to source data, and ensure that data confidentiality and privacy policies are enforced. 

 Data across studies are now available to researchers who may use a variety of specialized statistical tools to query, mine, and analyze the data. These can include both third-party products such as SAS and SPSS, and custom-developed applications. 

 Infrastructure services support application management, data management, end-user presentations, business process integration, and overall systems management. 

 Health intelligence business solutions combine core business and support services to create custom applications that support specific initiatives such as clinical research optimization, patient safety, health outcomes and economics, health management, and market intelligence.

When implementing a health intelligence program, organizations typically focus on one business solution. Part of the initial project includes establishing the overall program governance and building the technology architecture that will be the platform for this and future health intelligence business solutions. The benefits from the initial project alone can be substantial. 

For example, one large pharma firm focused on combining research data from multiple clinical trials. The health intelligence business solution that included a complex subject selection algorithm used this combined data to identify more and better-qualified candidates, resulting in shorter clinical trials. They projected an average of $18 million reduction in clinical trial operational costs by using their health intelligence solution to locate recruits in geographic areas known to have patients meeting the trial criteria. Even more significant, the time saved recruiting patients and the resultant accelerated speed to market is projected to generate an estimated $120 million in peak market sales. 

The next phase in the development of the health intelligence program is typically to extend the initial business solution by adding data from other internal and external sources or to build new business solutions. 

In today's challenging climate for drug discovery and development, developing collaborative partnerships can provide a more effective model for clinical research when supported by a health intelligence program. Together, the combined data and other resources from health care organizations, pharmaceutical companies, payers, and other research organizations can further accelerate patient recruitment and optimize clinical trial design and data capture, leading to faster approvals and controlled launches. 

Some of the leading health care collaborative efforts include work at Partners Healthcare, Cleveland Clinic, Mayo Clinic, and Regenstrief. There are also a growing number of pharma-related initiatives for clinical trials. For example, M2Gen Total Cancer Care, a partnership between H. Lee Moffitt Cancer Center and Merck & Co., is building a database that combines a patient's phenotype data with genotype data. Researchers expect that analyzing patients' responses to specific treatments will lead to more individualized care aimed at providing rapid improvement with fewer side effects. To date, tissue samples from nearly 6000 patients have been collected and thousands more have signed consents.

For collaborative partnerships, coming to consensus on data standardization, governance, technology platform architecture, and business solution design is by far more challenging than for a single organization. Fortunately collaborative leaders have found that using health intelligence framework allows them to address these challenges in a structured manner, focusing first on solving their business, data sharing, and governance challenges, rather than technology issues. 

Specifically, the collaborative uses the health intelligence program framework and processes to determine how to acquire, manage, and analyze data, and how it makes decisions on new data sources, new research methods, and new business solutions. Health intelligence supports the initial build as well as additions, creating a long-lasting program management platform for the collaborative. 

Health intelligence lays the foundation for personalized medicine by enabling the capture and integration of both phenotype and genotype information that can be used through standardized, automated, and compliant processes to drive the entire R&D cycle. Health intelligence business solutions can be used to integrate clinical trial data into the patient care workflow. While promising, these solutions will not reach their maturation for years. 

Two major technology-related challenges today are the lack of data standards and the limited adoption of EHRs in physician practices. But changes are coming that will help address both these obstacles. 

Publication of the Clinical Data Acquisition Standard Harmonization (CDASH) by CDISC—the Clinical Data Interchange Standards Consortium—is an important step toward standardizing clinical research data capture.

  The recently enacted Health Information Technology for Economic and Clinical Health (HITECH) Act will set EHR data and interoperability standards starting this year. HITECH has the potential to significantly improve EHR adoption by providing financial incentives for implementing applications that support data exchanges.

Many factors are creating the need for and breaking down the barriers to making the necessary changes to improve clinical trials design and conduct. Escalating drug development costs, public demand for safe and effective medications, the growth of collaborative partnerships, and the recent push by the federal government to solve the data interoperability and technology adoption issues are providing a favorable environment for clinical trial innovations supported by health intelligence. 

1. D. Foltz and L. Ferrara, "Thriving on Disruption," CSC White Paper, 2007. 

2. C. Gentry, "Have Merck, Moffitt Found Cure?" 

3. M2Gen, Moffitt Cancer Center's new research initiative and collaboration with Merck, & Co., Moffitt Cancer Center Web site, 

4. Clinical Data Acquisition Standards Harmonization (CDASH) v 1.0, CDISC, October 1, 2008 (

5. The Health Information Technology for Economic and Clinical Health Act, 

http://library.ahima.org/xpedio/groups/public/documents/ahima/bok1_029033.hcsp?dDocName=bok1_029033

 is a partner in CSC's Global Healthcare Sector. 

 is the director of health informatics in CSC's Global Healthcare Sector. 

* is a research principal in Emerging Practices, the applied research arm for CSC's Global Healthcare Sector, 266 Second Avenue, Waltham, MA 02451 email: 

*To whom all correspondence should be addressed. 

Articles

Uses for next generation information management, called health intelligence, in research.

Lynette Ferrara

Articles

Information as a Strategic Asset