EDC brings clinical development into focus

By Chris Connor, Senior Research Analyst, Clinical Development at Health Industry Insights, an IDC Company

Key datapoints, statistically distilled from large or highly variable sets of data, are commonly employed to guide our decisions every day. The allure of statistics is inescapable. The need to partake of its intoxicating ability to offer simplicity amidst chaos derives from our inability to handle and process calculations on large sets of data. It turns out that the human brain’s capacity to discern potable results, obtained from a clean source of data, versus those results pumped from a polluted data source, exponentially decreases as the size of the data set increases.

This propensity to become overwhelmed by lots of data has created a general uneasiness in our use of statistics. In clinical development we compensate by attempting to scrutinise every dimension of the analytical process. And no functional area tends to receive more attention than database lock and the subsequent bioanalytical review, which occurs during the final stages of clinical development process.

But no manner of vigilance, validation or vetting can resolve issues that result from the use of inherently poor source data. Simply stated, if you put garbage in, the best you can hope to get out is a set of impressive, yet hollow metrics and some very colourful curves that still statistically stink.

In clinical development, statistical power is the keystone that holds firm conclusions on the viability of a clinical endpoint. The ability to demonstrate statistical power is a core competency of the industry. To develop statistical power, it’s important to isolate the signal from any noise. Or stated more simply, to compare apples to apples and oranges to oranges. The more variability in the data, the larger the data set must be. The larger the data set, the less we’re able to detect defects in data quality. If we were to allow the irony in this vicious cycle to continue, clinical development would be crushed under its own weight.

The only sustainable approach to increasing statistical power, while maintaining our ability to actually conduct clinical trials, is to improve the quality of data at its source. Electronic data capture (EDC) provides both the tools and the process infrastructure necessary to achieve the needed data quality as well as process scalability.

As an industry built on the rigorous execution of redundancy, coupled with the duration of the average clinical trial, the potential efficiency improvements and cost savings of EDC are enormous. Health Industry Insight estimates that half of all new clinical trials will be initiated using EDC by the end of 2007. Armed with the clinical data provided by EDC, and more importantly the metadata associated with that clinical data, sponsors can begin to manage their trials in something approaching real time. But to continue to the adoption EDC, sponsors must take the following steps:

Rationalise EDC investments

The success of EDC in penetrating the market beyond simple CRF data capture demonstrates the desire of life science companies to select a single vendor that can address both sides of the same coin: EDC providing the clinical data management while CTMS managing the operational and administrative aspects of the trial. Sponsors should develop a short list of strategic EDC vendors and then focus on adopting enterprise-wide solutions versus project based technology acquisitions. After selecting the short list of vendors, sponsors should carefully consider the benefits of hosted, subscription, and SaaS solutions as an effective strategy to divest from the cost and responsibility of maintaining 21 CFR Part 11 compliant solutions.

Support process transformation

Some companies suggest that pilot studies are the best method for adopting EDC. However, pilot studies have a “go / no go” as their primary endpoint. To paraphrase something that Ted Chin at GlaxoSmithKlein said to me recently, there is no “no-go” decision when it comes to successfully deploying EDC. In order to achieve the necessary data quality and scalability results, sponsors must be fully invested in not only the technology but, more importantly, the cultural and processes changes that are supported by the technology. Everyone in the company must be on board for the benefits of EDC to be realized.

Develop an eClinical services architecture

EDC certainly plays a critical role in the eClinical infrastructure. However, it is only a component of a well-architected eClinical ecosystem. Every company navigates their own route, but the journey is significantly improved if the sponsor develops a services oriented architecture (SOA) roadmap. In support of this, sponsors should evaluate their eClinical technology vendors on their support of data interchange standards and their ability to integrate data from other systems.

For their part EDC vendors, as well as other eClinical technology vendors, should look to increase their investments in developing a scalable partnership strategy. In the end, extending the reach of their technology through the use of data interchange standards such as CDISC and HL7 will benefit the vendor as well as the overall market.