|
|
|
By BIO-IT World
The promise of electronic data capture (EDC) is simple: fewer steps to a clean set of clinical data. Electronically recorded data, in theory, are ready to be analyzed long before data collected on paper. Some pharmaceutical companies have announced they are saving tens of millions of dollars annually using EDC. At other companies, that payoff remains a mirage, thanks to complex clinical workflows, massive regulatory requirements to warehouse paper no matter what electronic solutions are used — and a decade of unsatisfactory solutions from vendors.
The Great Paper Chase
A physician, nurse, or clerk (1) uses a paper worksheet, known as a "source" document, to record information about a patient. This information is transcribed onto a formal document called a case report form (CRF). Some trials have thousands of patients, and each patient may require 100 CRFs recording various types of medical information.
Paper chronically piles up (2) at sites conducting trials. It gets lost. It gets misfiled. It gets carted to the basement of the hospital. As deadlines loom, and independent field monitors are about to arrive at the site to inspect the work, the CRFs may be processed in binges of late-night filing and double-checking.
A practice called double data entry (3) is the norm. It could make executives at Wal-Mart blanch. Each CRF is scanned or typed into a database twice by two people. Software can spot some discrepancies. But if the system receives two different values for, say, a patient's blood pressure, a third person gets involved. In many cases, the nurse (1) who originally recorded the information gets another piece of paper, a query (6), asking her to verify the questioned number.
Weeks or even months pass between when a medical fact is recorded and a question is asked about that fact. This strains the memory of every person who saw the patient. As such "queries" (6) about the data are slowly resolved, using still more paper, the data in the clinical database (4) are slowly cleaned and prepared for rigorous statistical analysis.
In the worst-case scenario, data or the patients underlying the data may be excluded if queries cannot be resolved. There are long periods of uncertainty, during which data monitors and sponsors of trials (5) have no idea about the quality of the data. As costs rise or new scientific information comes to light, it is difficult to adjust the process.
EDC Keeps Data Flowing, and Trees Growing
An electronic process may use any number of devices (1) to record clinical data, but the most common information-collection technology remains paper. PDAs, tablet PCs, phones, and other devices can upload the data directly to a database (2) on a regular basis, avoiding the need to fax documents or ship them overnight.
With paper source forms, nurses or clerks enter data themselves, often via the Internet. Software can catch basic mistakes. When the data comes directly from devices, the tools can impose a gentle coercion that keeps bad data from entering the system. Such systems can be programmed to pose a question to the nurse if, for example, a patient in a trial suddenly begins taking blood pressure medicine.
It may take longer to set up an electronic trial, but once it gets going the savings of time and money can be significant. Queries (4) are now handled electronically, with missing pieces of data flagged for each patient. Monitoring the progress of a site is much easier.
Electronic trials generally have fewer queries, and queries that do arise can be resolved more continuously, as the trial unfolds. This gives the data monitor (3) and sponsor of the trial a more continuous perspective on the progress of various sites.
What Regulators, Staff, and Sponsors Worry About in the Middle of the Clinical Night...
 Data Quality. At the FDA, the primary concern is the integrity of the data, whether in paper or electronic form. For now, comparing data on the same drug in different trials — or different drugs for the same disease — is difficult because it requires looking at actual paper documents or scanned versions of those documents on CD-ROM. Tracking adverse reactions to drugs could be radically improved if all the clinical data were digitally available at the FDA.
 Tools Profusion. Clinical trial administrators (CRAs) and coordinators (CRCs) may initially feel that paper is easier — and that EDC shifts work onto their already tired shoulders. That may change over time, as CRAs and CRCs become familiar with EDC, but the profusion of solutions from the vendor community means that they must often learn different application interfaces to do the same work.
 Persnickety AMCs. Not every EDC solution connects well to highly customized back-end data-processing applications and storage systems in large drug companies. And some EDC solutions may antagonize vital participants in the process — academic medical centers (AMCs) gathering the data under contract. The best AMCs can pick and choose which trials they participate in. That makes it difficult for large drug companies to force a particular application or approach.
COMPILED BY MARK UEHLING; ILLUSTRATION BY JOHN BLECK
Back to Data Rapture?
|
“Storage for Science – Methods for Managing Large and Rapidly Growing Data Stores in Life Science Research Environments” sponsored by Isilon
Large and rapidly growing stores of file-based and other data are a hallmark of life science research and bioinformatics. Determining how best to manage those data stores has become a significant challenge for Researchers and IT Pros alike.
This paper is intended to:
- Provide guidance on the many storage requirements common to Life Science research;
- Explain the evolution of modern storage architectures;
- Summarize the major data storage architectures currently in use.
Additionally, it will present the Isilon IQ clustered storage product as a strong and flexible solution to those needs. Download now
| |  Next-Generation Technologies Revolutionizing Oncology and Diagnostics
underwritten by Definiens
This “Briefing On” collection of Bio-IT World features, commentaries and analysis, presents some of the latest thinking on high-throughput technologies that are being applied to the fields of research and drug discovery, with particular emphasis on oncology, diagnostics and imaging technologies. Download now at no charge compliments of the underwriting sponsor, Definiens. Download This Free Paper
| |
.png "metaminer image(1)")
MetaMiner™ Cystic Fibrosis Report, Sponsored by GeneGo
This paper discusses the MetaMiner™ (CF) data analysis platform for a broad range of CF researchers designed to: 1. Easily assemble important biological and chemical experimental data available today in cystic fibrosis research. 2. Visualize key mechanisms leading to the disease through pathway maps and network models 3. Provide the CF community a “one stop shop” tool for uploading and analyzing experimental data in a disease-centered interface. Download now
|
Storage for Science
Methods for Managing Large and Rapidly Growing Data Stores in Life Science Research Environments
Sponsored by Isilon
Large and rapidly growing stores of file-based and other data are a hallmark of life science research and bioinformatics environments. Determining how best to manage those data stores has become a significant challenge for the Researchers and IT Professionals that support them.
This webcast is intended to:
More Podcasts
Isilon Systems ~ Senior Marketing Communications Manager
Isilon Systems is the worldwide leader in clustered storage systems and software for digital content and unstructured data. We seek an experienced marketing communications professional/writer expert in creating and delivering effective and persuasive business communications. The ideal candidate can think at the strategic and conceptual level and act, simultaneously, as a highly-effective and productive individual contributor. The position is based in Seattle, WA. For additional information click here: Lilly Singapore Center for Drug Discovery (LSCDD) - Associate Director of Informatics
Lead and mentor a strong team for the Bioinformatics group at the Integrative Computational Sciences (ICS) department at LSCDD towards the development of novel algorithms, data analysis methods and software tools for drug discovery. Work closely with the Software Engineering group at ICS, and collaborate with the Discovery IT organization in Europe and USA. For additional information, or to apply visit: LSCDD
|
|
|