Computing Community Consortium Blog

The goal of the Computing Community Consortium (CCC) is to catalyze the computing research community to debate longer range, more audacious research challenges; to build consensus around research visions; to evolve the most promising visions toward clearly defined initiatives; and to work with the funding organizations to move challenges and visions toward funding initiatives. The purpose of this blog is to provide a more immediate, online mechanism for dissemination of visioning concepts and community discussion/debate about them.

“Remaking American Medicine”

September 28th, 2011 / in big science, research horizons / by Erwin Gianchandani

With Health Buddy, a patient's medical condition can be monitored on a continuous basis without requiring visits to a physician or hospital [image courtesy Robert Bosch/Healthcare/CACM].This month’s Communications of the ACM features a great piece about improving — and transforming — our nation’s healthcare system through the development of an information technology ecosystem:

…[Health] information technology need not be limited to doctor’s visits and lab tests. [A report by the President’s Council of Advisors on Science and Technology (PCAST) last December] envisions a more comprehensive, lifelong record that includes not only treatment history but also a genetic profile, psychological characteristics, behavior patterns, and exposures to risks that might be relevant to health. While such a record could benefit individual patients, it could provide even greater value when stripped of personally identifying information, combined with similar records, and subjected to data mining algorithms.


It would, for instance, create a sort of extended clinical trial for approved drugs, says [CCC Council member] Susan Graham, a computer science professor at the University of California, Berkeley, and a member of the report’s working group. Today’s drug trials stop with the approval of a medication, “yet while people are taking these drugs there’s an accumulation of experience about what the side effects are and what the potential benefits are,” Graham says. The healthcare group Kaiser Permanente has already demonstrated such a benefit; electronic records for its 8.6 million members helped identify the link between the painkiller Vioxx and an increased risk of heart attacks.


With an entire nation’s health records at their disposal, computers might also find early warnings of epidemics or identify which treatment approaches work best. Graham points out that only major diseases that affect millions of people tend to be studied. A huge database could provide valuable insights into less common disorders. “It’s only possible if all of the information on which that kind of insight is based is, number one, electronic, and number two, available,” she says.

Among other research challenges, the article highlights telemedicine…

…[Increased] monitoring could catch potential problems earlier, perhaps leading to more effective treatment or outright prevention of some conditions. It could also reduce costs…


The growth of the “Internet of Things,” in which now-discrete devices are networked, could provide both monitoring and feedback, suggests Isaac Kohane, professor of pediatrics and of health sciences and technology at Harvard Medical School and director of informatics at Boston’s Children’s Hospital. Your refrigerator, for instance, might offer suggestions to help you adhere to your diet, or the motion sensor in a gaming system could be used to guide physical therapy. “Pretty much everything we’re doing today could have a sensor,” Kohane says. “Your scale could have an IP address.”


There’s already a package of sensors that many people carry around with them every day: their smartphone. “People are walking around with devices that make it much easier to capture in-the-moment data,” says Deborah Estrin, director of the Center for Embedded Network Sensing at UCLA. Analyzing patterns of a smartphone’s GPS traces could reveal changes in a person’s behavior, perhaps signaling, for example, a bout of depression or an increased risk of suicide.

…and safeguarding patient privacy:

If all this is to work, strong privacy protections will be important. Latanya Sweeney, professor of computer science at Carnegie Mellon University [now at Harvard University], says data should be segmented and in the control of the patient. This way, a patient could share information about an HIV test only with her primary-care doctor while letting everybody know about her allergies. There also should be a way to track who sees patient data to help prevent abuse, Sweeney says. If a bank, for instance, is buying information about a customer’s cancer risk and using it to adjust their credit scores, a patient ought to know. Sweeney worries that a lack of privacy incentives in the health-care initiative will produce a backlash.

Perhaps most importantly, the CACM piece emphasizes the need for interdisciplinary approaches to health IT, as we’ve argued in this space before:

Computer scientists will have to work with doctors to figure out what is technically feasible and how IT can fit into the practice of medicine, says Graham. The capture of information in clinical settings has to fit into the workflow, so providers don’t find it burdensome. And they will have to guide the policy makers who will make the regulatory and financial decisions.


“It really needs to be interdisciplinary,” Graham says. “This is not just a computer science topic.”

Check out the entire article in the September 2011 CACM, available for free for one year thanks to our colleagues there.

And while we’re on the subject of health IT, it’s worth linking to another article — “Mining Data for Better Medicine” — that appeared in MIT’s Technology Review early last week:

The antidepressant Paxil was approved for sale in 1992, the cholesterol-lowering drug Pravachol in 1996. Company studies proved that each drug, on its own, works and is safe. But what about when they are taken together?


By mining tens of thousands of electronic patient records, researchers at Stanford University quickly discovered an unexpected answer: people who take both drugs have higher blood glucose levels. The effect was even greater in diabetics, for whom excess blood sugar is a health danger.


The research is an example of the increasing ease with which scientists now scour digitized medical results, like glucose tests and drug prescriptions, to find hidden patterns…


The spread of electronic patient records, with their computer-readable entries, is opening new possibilities for medical data mining. Instead of being limited to carefully planned studies on volunteers, scientists can increasingly carry out research virtually by sifting through troves of data collected from the unplanned experiments of real life, as preserved in medical records from scores of hospitals.


Such techniques are allowing researchers to ask questions never envisioned at the time of a drug’s approval, such as how a medicine might affect particular ethnicities. They are also being used to uncover evidence of economic problems, such as overbilling and unnecessary procedures.

Check it out here.

And be sure to review the CCC’s white paper describing research challenges in health IT broadly.

(Contributed by Erwin Gianchandani, CCC Director)

“Remaking American Medicine”

Comments are closed.