Irvine, CA, July 26th, 2012 - - According to the World Health Organization 25 percent of women and 8 percent of men worldwide have felt the excruciating pain of a migraine headache. Symptoms, which vary from person to person, are often under-diagnosed or blamed on inflamed sinuses.
A SURF-IT summer research project is collecting and analyzing data in an effort to develop technologies to help sufferers manage their migraines. Lead researcher Yunan Chen, assistant informatics professor, and her graduate student, Sun Young Park, discussed their efforts at this week’s SURF-IT lunchtime seminar event.
“Migraine has not really been explored in the human-computer interaction and informatics fields,” said Park, who opened the presentation. Most studies are medically driven, focusing on diagnosis, treatments and headache triggers. She and Chen, however, are interested in developing tools to help sufferers cope with their affliction, balance work and home life around the condition, and find social and emotional support.
First, though, they have to understand what their target audience really needs.
“The purpose of this study is to examine people’s migraine-management experience in order to understand common and effective practices,” Park said. “This knowledge will serve as a guide to design technologies that will help migraine sufferers find their triggers and better manage their illness in daily life.”
Methods employed by human-computer interaction studies are somewhat different than those used in other scientific studies, Chen told the audience. “We are trying to understand users as a way to design these technologies.”
She and her team began the process by scouring online migraine forums in order to learn more about the sufferers and what information was important to them.
Then they interviewed 12 patients with symptoms ranging from mild to severe. They are currently interviewing some doctors as well. “The disease is so individualized that sometimes it’s difficult for doctors to get the information they are looking for,” Chen said.
The team also searched for already-available apps but found a dearth of well-designed products. “There is a gap between the technology you want to design and the technology people really need.”
Chen said her team relies heavily on what she calls “semi- structured” interviews, which are based on several pre-determined questions but are flexible enough to accommodate other information. They attempt to contextualize the material by conducting the interviews in the subjects’ homes, taking photos and returning to the lab with artifacts like calendars or diaries that document the patients’ struggles.
Interviews are usually done by pairs of researchers; one takes notes while the other makes an audio recording. Those recordings are transcribed word for word, and often yield additional insight, including emotions and frustrations.
Then there’s the matter of qualitative analysis. “This takes a really long time to think about,” Chen said. “How do we relate all those pieces to one another? That’s a challenge.”
Her preferred method is known as affinity diagraming. Researchers isolate information gleaned from their interviews into small chunks. They write these tidbits on post-it notes, which they group by categories on a large empty surface like a wall.
“This is a good way for us to see things from the bottom up,” Chen said. These groupings are followed by additional groupings of higher-level categories like social support or disease management – themes that Chen says don’t take shape until more specific categories are assembled and analyzed.
Researchers often conduct multiple rounds with these affinity diagrams, disassembling and reassembling the post-it notes in different formations. “There’s a lot of things you can do by reshuffling those affinity notes and talking about them from a different perspective,” said Chen, adding that this method of data analysis is “a really fun and engaging methodology.”
After the assembling and analyzing of data are complete, the researchers plan to develop a prototype of their technological intervention. “We are trying to identify the needs and decide what we can do from there,” Chen summarized. “Then we will try it out with migraine patients and see if it works for them.”