By Mark D. Uehling
July 15, 2003 | Has there been one redeeming idea to come out of the hypnotic and violent world of children's video games? There is now: the notion that it might be possible to give kids a medically useful obsession with a handheld game tracking their own blood sugar.
Managing diabetes is not easy at any age. But children have an especially difficult time, with scientific literature showing them to be less attentive to their condition than adults are. Common problems include failures to check their glucose levels, inject insulin, and follow an appropriate meal plan.
Enter DiaBetNet, dreamt up by Vikram Sheel Kumar, a medical student in a joint program at Harvard and MIT. He persuaded a diverse set of parties to get their technology and wireless networks to operate as one. The list includes a maker of handheld computers (Handspring), a maker of handheld glucose meters (Roche Diagnostics), a phone company (Verizon Wireless), a modem company (Novatel Wireless), and a major research center (Boston's Joslin Diabetes Clinic).
Not Playing With Sugar
The design process had to reflect a basic obligation of diabetes that no technology can change: pricking one's finger for a drop of blood to know if food or insulin is needed. "Patients are expected to check their insulin levels three to five times to day," Kumar says. "Patients don't want to do this. It's painful." At an exercise at the MIT Media Lab, Kumar and professor Alex (Sandy) Pentland tried to explore a subtler medical-IT issue.
Why do kids find some interactive medical games so boring? Kumar's conclusion: "There wasn't a community aspect and an interactive aspect." His inspiration? "The most effective behavioral change-assistance tool we have is Alcoholics Anonymous," he notes. "You get the patients together, and it works. They feel better; their outcomes are better."
Applying those insights to diabetes, and exploiting the competitive instincts in children, Kumar came up with a game in which children could guess not only their own insulin levels but also those of classmates or friends. After a day of monitoring the disease, a graph on the Handspring summarizes the patient's day, carbohydrate intake, and prior glucose data. With that data in hand, the patient guesses the blood sugar. More accurate answers earn higher scores.
Then Kumar began working the phone to put the pieces of his network together. The starting point was the glucose meter, made by Roche Diagnostics, and its Accu-Chek Pocket Compass software loaded onto the donated Handsprings. (It can run on other PDAs too.) But writing code for glucose monitors is not something one can do during a lunch hour. "Roche gave us access to its database specs," Kumar says. "We didn't reinvent software that had to be FDA approved. We relied on stuff that was out there."
Watching Your Meter
At Roche, in Indianapolis, Schuyler Buck is manager of IT business development. For him, lending a hand with Kumar's project was part of a larger effort to make the technology as easy to use as possible. "The software becomes important for those people who are looking to very closely manage their diabetes," Buck says. "One of our goals is to try to fit into the everyday lives of people with diabetes."
A wireless solution, obviously, could slightly simplify the hassle (not to mention cabling and docking stations) that comes with diabetes. Roche's product line -- unlike those of competitors -- includes both tethered and wireless glucose meters. "[Wireless] is certainly one of those things that could throw the scale one way or the other," Buck says.
Next came the Joslin Diabetes Center, where Lori Laffel, chief of the pediatric and adolescent unit, says she was impressed with Kumar's brainchild. "The group that got the game checked their sugars slightly more than the group that just got the [PDA] technology. To us, that was very encouraging. Kids and families are excited to try something new. It brings along this hope."
Lafell says one critical question was whether patients got their data into a secure server at MIT. In a previous PDA trial, she says, data transfer was problematic. This time was different. "Ninety percent of the patients were able to do that successfully," she says. "That in itself is remarkable in a pilot study." The one family that had trouble transmitting was in a mountainous hamlet in Vermont.
Still, she notes the pilot study, with 40 children aged 7 to 18, did not offer a clear determination of whether the technology really made a difference to their condition: "We did not take it to the next step, which is to see if using the technology actually makes a difference in health, in levels of blood sugar control." Lafell is trying to arrange that next step now, with help from Roche or others.
Even so, it's clear she thinks the technology has potential. "It's the young adult population that is at risk of disengagement from healthcare," Lafell says. "They are technologically savvy, and this may be a way to maintain their engagement."