This article originally appeared at Healthcare Dive on Feb. 12, 2015
The University of Pennsylvania Health System is among those piloting the Apple HealthKit platform to utilize health data uploaded from patients’ iPhones. So far, the project is raising more questions than it’s answering on what to do with all the new data.
The project is currently in its initial stages.
“We have pilot users who have demonstrated that they’re able to get information from their devices, in to their phones, and from their phone into their EMR,” says Chief Medical Information Officer William Hanson, III, MD.
Next, the project will engage with medical practices that focus on heart failure and diabetes. Hanson calls this the logical first step because it confines the scope of the pilot to a limited number of variables, and relies only on monitors that are well understood, unlike non-medical fitness tracking devices, etc.
“We’re looking for practices that have a discrete group of patients that have a one-to-one relationship with a monitor and something that’s medically important—not so much something fitness oriented,” Hanson says.
In about six months, following the initial rollout and analysis, he expects to have a better sense of whether the pilot is to be expanded and the likeliest next groups to incorporate.
Questioning data out of context
The first issue Hanson sees with the concept of utilizing HealthKit data for medical purposes is that the information is being acquired outside the typical medical environment, and is coming from devices that those who are receiving the data don’t own or maintain.
“The second issue,” Hanson says, “is we lack the contextual information we typically have with patients.”
For example, if a physician sees a patient with an elevated heart rate or blood pressure in the office, they can see the patient in the context in which it’s being measured. Hanson notes that as an anesthesiologist he frequently sees pre-op patients with an elevated blood pressure because they’re anxious about surgery.
“If I get a high blood pressure reading from a device on a patient in their home environment and I don’t have any other context—that’s not necessarily an issue, but it’s a new way of grappling with information,” he says.
Managing the data
The next question is what to do with this potential overload of information. Healthcare providers are telling Hanson they don’t want to have “a whole new source of alerts that they don’t know what do about.”
To answer this, the program is looking at the question of whether the data needs to be handled by an individual, or by a processing algorithm that could spot patients in trouble. They’re also considering how to set boundaries on what kind of data should prompt alerts. “These are things that are not packaged with the Apple product,” Hanson notes.
“Our EMR is Epic, and we just have a place to put the information in the medical record—we haven’t yet figured out how to deal with this new source of information and what to be alerted about.”
The problems and the potential
Hanson says his main concern is that if this paradigm of patients measuring health data out in their daily activities—and then wishing to put it into their medical record—becomes the norm, physicians have to figure out how to use that new onslaught if information. “But, in a sense that’s also the promise,” he says.
Hanson thinks having better visibility on what patients are doing in their day-to-day lives will provide the opportunity for healthcare practitioners to intervene before patients get to the point that they’re really sick, and prevent hospitalizations or other short and long-term consequences.
For example, Hanson says, “If you can recognize that a patient’s weight is going up because they’re retaining fluid and call them up to intervene two days before they would have otherwise have to be admitted to the ER, that’s a beneficial outcome.”
“There’s the Yin and the Yang of this thing,” Hanson says.