Army of Women allies with CaBIG for online longitudinal study

This poster encouraged British women to join the Women’s Land Army by becoming farm workers during World War II; today, women are flocking to a different kind of army in the form of the Army of Women.  Image courtesy Blitz and Peaces website.

What happens when you combine a celebrity doctor, cancer researchers, social media, and experts in distributed computing? An Army of Women over 340 000 strong and growing, standing ready to participate in innovative research projects such as the Health of Women study.

“The percentage of people who participate in clinical studies is very low,” said Naz Sykes, executive director of the Dr. Susan Love Research Foundation. “Our goal is to educate the public about the importance of research, the importance of taking part in research ... throughout their lifecycle.”

That’s why the DSLRF joined forces with the Avon Foundation for Women to create the Army of Women website. Collaborating with the Beckman Research Institute, City of Hope, and the National Cancer Institute’s Cancer Biomedical Informatics Grid to create the Health of Women study was simply the logical next step.

“No one has ever done an online longitudinal study,” explained Sykes. “The ultimate goal is that we’re going to drill down and answer the questions women have always wanted to know. Even the stupid questions like does hair dye cause cancer, or do underwire bras cause breast cancer. We know that there’s very little chance that they do, but through the Health of Women study we’ll be able to confirm that.”

Launching a questionnaire that could handle thousands of responses in the space of a few weeks is no easy task.

“One of the challenges we were concerned about is how could we would we deal with what could potentially be a tsunami of requests to use the infrastructure,” said Kenneth Buetow, NCI associate director for bioinformatics and information technology. “Part of the solution was to use cloud technology as a way that we could have expanding capacity to respond to user requests for the infrastructure in a very dynamic way.”

To test the infrastructure they had created before launch, they pitted cloud against cloud, with one cloud making requests while another received and executed them.

The first module in the HoW study was launched in December 2009 via an email sent to the members of the Avon/Love Army of Women. Within a few weeks, 30 000 women (and a few good men) had filled out the questionnaire, making the study about 10 to 20 times the size of most large studies.

As with any major effort, everything did not go smoothly.

“We had some issues with the website, navigating through the website, navigating through the questionnaire,” Sykes said.

Although the intention is that the study will eventually take results continuously, they made the decision to put the study on hiatus and do a post-mortem.

Health of Women 2.0

After tweaking the interface to make it as user-friendly as possible, HoW is set to relaunch in 2011. But their work won’t end there.

“The power of a standing cohort like this is that it can be used to answer all sorts of questions,” Buetow said. “There’s plans to continue to expand that Health of Women study with additional modules.”

At the moment CaBIG is working with their HoW partners to build a tool that will allow researchers to create their own modules.

Of course, not everyone will be able to create a module. As with the Army of Women, researchers who wish to create their own module will have to show that their study is funded and peer-reviewed. Modules will have to gain approval from the HoW scientific advisory committee, and also pass an ethics review by their Institutional Review Board. Finally, to pay the bills of the HoW and Army of Women infrastructures, they’ll have to pay a fee.

Participants will receive an email inviting them to sign on to fill out each module as the modules are released, while new participants will be asked to fill them out in order, so that no module has to repeat questions contained in prior modules. Each module should take no more than 20 minutes to complete. Researchers, in turn, will be able to log in to see snapshots of the data.

According to Sykes, they hope to have the “rent-a-module” system in place and running by the end of summer 2011.

—Miriam Boon, iSGTW