Safety through Standards
July 2008
Paula Brown Stafford, executive vice president of global data management at Quintiles, discusses how universal data standards can help ensure better patient safety across all phases of clinical research.
Photography by
The sheer volume of data collected for any given clinical study can be overwhelming. From basic ECG and lab results to patient reported outcomes and adverse event data, the constant flow of incongruous clinical trial information is vital to the success and integrity of any clinical study. Nevertheless, clinical research data needs to be better harnessed to ensure a safer experience for trial participants. By embracing data collection and reporting standards, trial sponsors and contract research organizations can improve both the quality of safety information, as well as the speed with which the data are collected, integrated and analyzed.
A major problem with today’s data collection practices is that all of the critical information is housed separately, requiring an intensive, manual reconciliation at a study’s end. For that reason, most pharmaceutical companies have a data management group as well as a safety group. For example, information related to expected adverse events such as dizziness is recorded in the standard Electronic Case Report Form. But should the dizziness in turn lead to convulsions or other unexpected events, a separate, serious adverse event protocol is followed. The problem is that these entries reside in different databases making safety trends, both major and minor, difficult to spot.
Data, Data Everywhere
Most trial sponsors have a variety of systems in place that allow them to look at data on an ongoing basis. But in most cases one set of people looks at the adverse event data, another set looks at the lab data and yet another looks at the serious adverse event data. Many companies have processes and tools to look across all of these data sets within one patient, but they are not able to systematically and comprehensively observe across all of these sources of data, and across all of their patients in real time.
Furthermore, most companies aren’t reconciling all of these data until the end of a study, which hinders the possibility of conducting a valuable trending analysis to identify potential safety concerns. When I joined the industry in 1985, you’d receive all of these data at once, but it would be on paper and worse yet, it wouldn’t be available until four to six months after the trial was completed. Eventually, we’d enter it all in and have all the data at once, but it was far too late to identify and act upon trends that might indicate a potential safety issue.
In an ideal scenario, all of the collected data would be available in one system, within 24 hours of the patient’s visit. This would enable a physician to look at the data to identify possible safety concerns as they materialize—even in a global study with thousands of patients and hundreds of sites. The tools that we have today just are not set up to provide this level of oversight. One solution is to create a data repository. There are data warehouses that simply store data, but a relational data repository can link together all the various pieces of information in one place.
Simply building a repository and integrating the data isn’t enough, however, the process also must include real-time access to reconciled data. Only then can you get the full picture of one patient as easily as across all patients in a study. But a proprietary data repository will be limited if only developed for one company’s use. For a system to truly have an effect on patient safety, data must be collected, shared and reconciled in a standardized format.
Prep Time
At the leading edge of the data standards bandwagon is the Clinical Data Interchange Standards Consortium (CDISC). Along with Health Level Seven (HL7), a standards organization for health care companies, CDISC is working on a project called BRIDG (the Biomedical Research Integrated Domain Group model), which is essentially linking CDISC to HL7. The project teams primarily evaluate the set of standards that we need in the pharmaceutical industry and compare those to the standards that the health care companies need in order to work with insurance providers. BRIDG looks across those two domains to make sure that the standards for the health care industry are not contrary to what is needed by the pharmaceutical industry for clinical research.
Most importantly, BRIDG is about having real-time access to the data and enables you to create trending reports. Trial sponsors can do a little of that now within their case report forms or even within their lab data, but you need to have one place across a pharmaceutical/biotech product to look at all of the data. Generally, companies run a few studies in tandem, and once a drug candidate gets into Phase II or III, they’re running several at the same time. For that reason, they need the ability to query the repository in real-time, not just for a particular study, but for all studies of a particular product.
Group Effort
The only way we’re ever going to get to the point that we can integrate the process, and therefore integrate the data, is to implement standards to help us get there. If everybody’s using a different way of collecting clinical trial data, we’re never going to be able to integrate it in real time, or it will be cost prohibitive to do so. We’ve got to move faster and smarter as an industry.
It is imperative that slow adopters increase their efforts to help move the process along. We need to make sure to get these standards implemented within companies and between companies so we can integrate the process and the data, making it possible to have real-time patient safety.
Send us Your Comments
Web Exclusives
- An Unbalanced Burden
-
A Q&A with Dr. Kathleen Squires, director of the infectious disease department at Jefferson Medical College in Philadelphia, illuminates the gender differences in the diagnosis and treatment of HIV.
- Turning the Corner
-
After decades of neglect, leading to a rapid increase in incidence, tuberculosis started gaining attention from industrialized nations over the past 10 years. With this new focus, TB’s incidence has started to decrease—but barely.