The Health eHeart trial is being compared to the six-decade-old Framingham Heart Study, which led to important discoveries about the role of smoking in heart disease, the relationship between high blood pressure and stroke, and the effect of obesity on heart failure. However, unlike the Framingham trial, which collected data on patients only once every two years, the UCSF study will gather information from its subjects' mobile devices several times a day.
"We hope to be able to collect copious amounts of data on a large segment of the population so we can develop very robust and accurate models to predict the occurrence of heart disease in people who don't yet have heart disease, or slow the progression in people who already have heart disease," said cardiac electrophysiologist Jeffrey Olgin, MD, chief of the UCSF division of cardiology, in a news release.
[ Will data analysis revolutionize healthcare? Read 7 Big Data Solutions Try To Reshape Healthcare. ]
Participants, who may enroll directly or through their physicians, will receive free mobile apps that can measure their heart rate, blood pressure and pulse rate. The data will be transmitted via Bluetooth to their smartphones, which will forward it to the researchers' database.
Among the applications that will be used in the study, reports MobiHealth News, are Azumio's Instant Heart Rate app or Azumio's Stress Check app; AliveCor's Heart Monitor, which can record and transmit a user's EKG; iHealth's latest Bluetooth-enabled blood pressure cuff; and a heart-rate-monitor watch from Boston Scientific.
In addition, The Wall Street Journal reports, the researchers might use tools such as GPS-enabled phones that can record where a person is eating or shopping for food or whether they're in the hospital. A photo app might be used to estimate the portion size and calorie content of a meal.
Among the questions that the UCSF researchers would like to answer are what causes atrial fibrillation and which people are at high risk for heart attacks, based on daily variations in their blood pressure.
In addition, UCSF says, "a major goal of the Health eHeart Study is to make health care delivery more precise." This is also an important goal of other big-data research efforts aimed at enabling physicians to provide more personalized care.
Based on the UCSF news release, it appears that doctors of UCSF patients enrolled in the study will use the mobile data to help them make treatment decisions. But a UCSF spokesman could not answer a question about what would happen if the researchers receive data indicating that an enrolled patient might be suffering a heart attack or another type of cardiac event. Olgin was unavailable at press time.
"It's wonderful to see mainstream medicine paying this much attention to patient-derived data that doesn't come from a test in the office or the hospital," Joseph Kvedar, president of the Center for Connected Health, a branch of Partners Healthcare in Boston, said of the study in an interview with InformationWeek Healthcare. "That's an important step forward." However, he added, "if it's just a longitudinal epidemiological study that takes that epidemiology into the land of sensors, I'm not sure it's that interesting."
The study raises some other questions, he pointed out. First, it doesn't seem to be designed to engage patients at the level of evaluating and responding to their own vital signs. The structure of the study is "very top down and medical driven," he said.
Second, he noted, the announcement gives no details about how the researchers plan to cope with the huge stream of incoming data or what kinds of clinical algorithms they intend to use. And the mobile health apps reported in the media have not all had their data output clinically validated. AliveCor's app has been, he said, but the iHealth software is "a consumer app. There's an error rate there."
As large healthcare providers test the limits, many smaller groups question the value. Also in the new, all-digital Big Data Analytics issue of InformationWeek Healthcare: Ask these six questions about natural language processing before you buy. (Free with registration.)