The pharmaceutical industry is seemingly in agreement that electronic data capture (EDC) is a good idea. Even obligatory. But let’s back up: what is EDC? Is it software? If so, why can it cost hundreds of thousands of dollars to get every trial started? Is EDC more properly considered a service? Then why don’t some contract research organizations (CROs) warmly embrace it?

Such thoughts were occasioned by a visit to Akaza Research, in Cambridge, Mass., which you may want to think of as the Red Hat of the clinical trial realm.

Just as Red Hat helps with its own variant of the Linux operating system, Akaza supervises improvements to (and gives away) the OpenClinica software, which has its own community of users and developers. For sponsors that need extra help with OpenClinica, Akaza offers services and this library of case report forms.

The company’s CEO, Cal Collins, says business is brisk. The company (which has no venture capital backing) tripled its revenues in 2007 and will probably double them in 2008, he estimates. “We’re growing at a steady clip,” he says. “Our mission is to build a successful business by developing an open source ecosystem.”


At the moment, Akaza has 20 customers and a similar number of employees. There are 2,800 members of the OpenClinica community; the software has been downloaded approximately 10,000 times.

The open-source aspect of OpenClinica means, among other things, that the users are never tied to one particular company or commercial licensing fees. “There is no per-seat fee to use the software,” Collins notes. “We can get an organization up and running on a trial basis for a low cost.” Some users can use the software for more than one trial, which is unheard of in the commercial EDC world.

More Flexibility

Collins readily concedes that EDC is growing fast. “But there is still a vast number of studies that are still done on paper,” he notes. And some of his customers want more flexiblilty than traditional vendors can offer. Says Collins: “You don’t want a black box where you can only use it in a certain way or get your data out in a certain way.”

For some sponsors, of course, a commercial vendor is an indispensible partner in the clinical trial research process. But Akaza is proving that setting up other trials may not require a thousand man-hours or a building full of people in India. With one savvy IT person, some investigators can think about doing EDC themselves. Which is not to say the OpenClinica platform is a toy.

International Projects

Akaza is provisioning its EDC system for a number of gnarly projects, including a trial in Africa that has plans for a 1,000-patient trial in three countries. Then there are eight trials for the National Institute of Allergy and Infectious Diseases (NIAID), some of which will be used in FDA submissions.

In theory, one attraction of using open source software should be that (with a sufficiently large user community) good ideas can rise to the forefront more quickly than with commercial software. The fact that the Firefox web browser has four times as much market share as Apple’s web browser is evidence enough that great software doesn’t have to originate from a single company. Ordinary mortals (not just Linux nerds) can find open source software to be very decent stuff.

Pastel Hues

In a demo of the OpenClinica platform, we liked the clean look of the pages, right down to the calendar popup boxes to select dates. It’s easy to set up six recurring events once, not six individual times. There was a nice popup allowing the user to view, print, edit (and remove, with sufficient privileges) a patient record.

Akaza project manager Paul Galvin noted that many users love the ability to do a lot from a single page of patient data. People don’t have to click around quite as much to do what they need to do, he says. No arcane technical wizardry is required to get started with OpenClinica, Galvin notes. I watched him quickly set up an electronic CRF. “This is created with an Excel spreadsheet,” he said. “The user doesn’t need a lot of new knowledge. People who are not developers can do this.” It’s easy to export the data into SAS and other common formats, including CDISC ODM format, HTML, tab-delimited and SPSS.

Safety In Numbers

As Collins explains the ethos in the open source world, the merit of a particular improvement is collectively determined and affords the developer a slightly elevated status, which (as competition arises) spurs further improvements. “There is a hierarchy, based on what they’ve been able to contribute,” he says. “All of the achievements of open source have been attributed to meritocracy. That’s something we respect and embrace.”

The collective nature of innovation on the OpenClinica platform means that one’s person’s brainstorm can belong to the community. A case in point: a full Spanish version of OpenClinica and much of its international functionality was developed in Spain and donated.


Collins has more than a superficial interest in the convergence of electronic health record systems and EDC. “The holy grail is single source data capture,” he says. “We’re building partnerships and interfaces with other providers of that technology. That’s the only way that is going to be achieved.” Citing a few specific electronic health record systems, he says: “Open source is being embraced and has a lot of momentum.”

CRO interest in Akaza is just starting.“They can adopt the open source technology with minimal initial investment,” says Collins. “They can take ownership of the platform while still being supported by us. At least one biostatistical CRO, StatWorks, Inc., recently signed on with Akaza.

Perhaps the poster child for what Akaza can do is another NIH project, at the National Database for Autism Research (NDAR), which has adopted OpenClinica for electronic data capture and implemented a library of highly complex and lengthy autism assessments and instruments as CRFs in OpenClinica.

Some of the NDAR CRFs are as long as 50 pages. They incorporate automated scoring algorithms, skip logic, and edit checks. The CRFs cover karyotyping measure; the Vineland Adaptive Behavior Scales; IQ measures; the Autism Diagnostic Interview-Revised (ADI-R) and the Autism Diagnostic Observation Schedule (ADOS), for use according to their published manuals.