Realizing True Interoperability

We've all heard the promises. New data management systems will reduce the time required for clinical evaluation of drugs, resulting in faster, less expensive clinical trials. By eliminating the built-in inefficiencies in capturing and processing clinical trial data, it is said that pharmaceutical companies can slash some of the staggering costs associated with drug development today.

If these promises were true, we would already have seen a significant improvement in the clinical trial process. After all, the typical pharmaceutical company uses a dozen different software systems to automate its work. Unfortunately, these specialized software packages have, for the most part, failed to live up to their promise; it can still take up to nine years and $600 million dollars to bring a drug to market.

Part of the problem is that many of these systems are capable of handling only one discrete process in the complex course of a large-scale clinical trial. More significantly, the systems are not designed to integrate with other specialized clinical trial applications, despite the best efforts of industry groups promoting standards for data exchange.

Because these standalone applications were not designed and built to share information, researchers spend valuable time re-entering the same data at each step in the clinical trial process. Such manual data re-entry is a recipe for introducing errors into trial results, requiring still more time to edit and correct the data.

When the technologies don't work together, and there is no way to aggregate data across systems, then software becomes just another obstacle to completing a timely, affordable, and successful clinical trial.

More than standards

Industry groups have defined and published a variety of standards and definitions that enable data to be shared at different points in the development and distribution lifecycle of a drug. We are all familiar with the efforts of groups such as Health Level 7 (HL7) and the Clinical Data Interchange Standards Consortium (CDISC).

Adopting these standards is an important step toward sharing data across the industry, but the standards themselves are not enough. They provide a framework for consistent data but do not enable pharmaceutical companies or organizations participating in clinical trials to actually move data in an automated way from one phase, application, or platform to another.

In other words, adhering to standards can make data available in multiple contexts through a labor intensive process of data importing and exporting. But the next step, a critical one for data accuracy and process efficiency, is to develop seamless interoperability between systems so that data is automatically accessible and sharable across applications, platforms, and computer languages. To do that, the pharmaceutical industry requires systems that are based on shared technical standards and built with components that easily connect.

Joining forces

In spring of 2005, our two companies, Broadpeak and DataLabs, joined forces to demonstrate the power of interoperability for improving trial processes. We set out to illustrate how separate systems from separate vendors can provide seamless support for clinical trial data when each system is built using industry standard technologies and an open architecture for interoperability.

Broadpeak provides a site management package that helps pharmaceutical companies and contract research organizations (CROs) accelerate the recruitment, selection, and initiation of new sites for a clinical trial. Our portal software, Site Manager, integrates work at different locations and coordinates communication and collaboration among participants.

DataLabs provides a data management system that streamlines the entire process of data collection and analysis, while also promoting greater data integrity. DataLabs Clinical™ enables those conducting clinical trials to enter data directly into a Web-based template in real-time, or to efficiently transcribe data from paper records, all based on a common platform.

Although the value proposition of each product is compelling on its own, we recognized that what pharmaceutical companies really need is an interoperable solution that can integrate the entire process. After an introduction from our shared business partner, Microsoft, DataLabs and Broadpeak decided to work together to improve the overall process of managing data in a clinical trial.

The resulting Clinical Trial Management System supports electronic data capture (EDC), paper data entry, and comprehensive data management. Using the combined capabilities of our two applications, a clinical trials manager can progress from managing investigator credentialing, patient enrollment rates, and reimbursement to viewing specific case report forms (CRFs) and patient information on an investigator-by-investigator basis.

All the site, investigator, and logistical data are provided by Site Manager, while the clinical data comes from DataLabs' Clinical. But such distinctions are transparent to the user. The tight integration of the two software applications creates the connections automatically, behind the scenes.

Technology solutions

Because a standards-based open architecture is, by definition, nonproprietary, companies can choose technologies from many sources to achieve interoperable solutions. In the case of DataLabs and Broadpeak, the Microsoft stack of software provided assurances that our systems would integrate easily.

Both DataLabs Clinical and Broadpeak Site Manager were built using Microsoft^® .NET technologies, the Microsoft framework for connecting systems and exchanging data. Each system uses CDISC-compliant XML for data exchange and a robust set of Web services to enable communication between applications and across platforms.

In addition, Broadpeak built its secure online portal using Microsoft Windows^® SharePoint^® Services. The portal, which is customized for clinical trial processes, provides a platform for managing data exchange with DataLabs' Clinical as well as with other openly architected systems.

The combination of Web services and the SharePoint framework enabled our two organizations to integrate solutions quite easily. In a matter of 30 hours, developers had created a Web part out of the DataLabs software to plug into the collaboration workspace provided by Broadpeak. The technical people described it as a simple and straightforward mapping exercise.

First, they exported all the CDISC XML data points from the DataLabs software using its existing Web services. Then they imported that information into the Broadpeak system using the same XML and Web services. The technologies used to build the products made the marriage between them easy.

Key similarities

The success of the partnership between Broadpeak and Data-Labs is based on more than a common architecture. At its heart is a mutual desire to work toward the larger business needs of our customers, and a shared willingness to use open, accessible technology standards to deliver shared data.

For the pharmaceutical industry, the advantages of such a partnership are clear. With an interoperable solution, companies can expedite data collection and aggregation and get a consolidated view of clinical trial status and results. By using systems that can pull data together from different levels, phases, and locations, trial decision-makers can access critical knowledge at a pace not previously possible.

The strategic use of technology facilitates this kind of process improvement. DataLabs and Broadpeak were able to partner successfully because we use the same Microsoft technology stack and we believe in interoperability of systems. Because our technology is based on an open architecture, it is highly adaptive. With proprietary software, technology can be a barrier to change. But with a standards-based architecture supporting interoperability, technology becomes an enabler of process change.

While the partnership between Broadpeak and Data-Labs worked quite well—so well, in fact, that DataLabs completed an acquisition of Broadpeak in November 2005—it has not been one effortless path forward for our companies. Merging technologies was relatively simple; educating customers about the value of interoperable systems has been far more difficult.

Although pharmaceutical companies have great interest in digitizing their communications, their options up to this point have been limited to standalone solutions or, in some cases, hosted service offerings. That being the case, they often don't immediately perceive the value of an integrated approach.

Addressing this problem will require greater cooperation among software vendors as well as a greater commitment to open standards. What the industry needs is a larger vision for how open standards can drive interoperability, and, by extension, make significant improvements across the clinical trial process. Ultimately, that's where the merger of Broadpeak and DataLabs provides its greatest value. It demonstrates that system interoperability is not only possible but also practical, and that the business advantages of standards-based interoperable systems include lower costs and faster time to market for the pharmaceutical companies that deploy them.

Eye on the future

Automating and integrating the clinical trial process can yield major competitive advantages. But the interoperability that companies, including Broadpeak, DataLabs, and Microsoft, envision for the pharmaceutical industry go far beyond joining data across clinical trials. For example, a portal that can integrate clinical trial data can also provide other data that an investigator might want to see, if that data is accessible in standards-based systems. In fact, the whole communication process between the sponsors, CROs, Institutional Review Boards (IRBs), investigators, and others involved in the trial process could be automated.

Other systems could be integrated as well. The clinical trial stage, as complex as it is, is only one piece of the whole process of developing, approving, manufacturing, marketing, and delivering new products. Key pieces of data are reused—and re-entered into different applications—throughout the various stages. But a truly interoperable architecture could allow data to be entered only once and used appropriately at predetermined points.

To move from vision to reality, however, pharmaceutical companies must deploy technology solutions that are based on an open architecture. Slowly, we are seeing the development of an ecosystem of companies that believe in open architecture and use Microsoft's .NET framework and Web services to build their applications.

If the industry demands this kind of open, shared architecture from all vendors, no longer settling for proprietary stand-alone niche applications, then software companies will follow. Vendors that want to offer their customers the best services at the lowest cost will build to open standards and integrate their offerings with strategic partners to provide a higher level of integrated services.

James Langford is President of DataLabs, Inc., University Research Park, 101 Academy, Suite 250, Irvine, CA 92617, (888) 773-0099, email: jlangford@datalabs.comwww.datalabs.com.

Zikria Syed is Vice President, Product Management ' Collaborative Solutions, DataLabs, Inc., 1180 Welsh Road, Suite 100, North Wales, PA 19454, (215) 631-7030, email: zsyed@datalabs.comwww.datalabs.com.