Can data aggregation cure healthcare?

How do we translate digital health data into meaningful results?
By Samir Damani
09:45 AM

There’s growing buzz around Apple’s HealthKit, billed as the latest and greatest in data aggregation. Rather than living in separate databases, information about blood pressure, sleep, fitness, nutrition and lab results will be funneled into HealthKit, providing a single health dashboard from which users, and hopefully clinicians, can make better decisions.

While solid information is sparse, it appears Apple is discussing this technology with Mayo Clinic, Kaiser and other healthcare organizations. In one scenario, data from HealthKit feeds into a patient’s electronic health record. Aberrant readings, such as dangerously high blood pressure, are flagged. Physicians are then encouraged to contact patients and discuss solutions. The hope is that real-time information will improve care.

First, let me applaud these efforts. Aggregating health data is a good first step towards proactively improving health. But the larger question is: How will this data advance society’s broader healthcare goals?

Let’s think through the high blood pressure scenario. A patient’s reading is digitally transferred to the health record and flagged. The physician contacts the patient with this information, but what would he or she say? There would be little time to discuss what it means, why it’s dangerous and what the patient should do next. Few physicians have the time to provide such comprehensive coaching.

But the real systemic flaw is even expecting a phone call. In 2012, there were 90 active primary care physicians for every 100,000 patients. Now, consider the millions of patients who need this level of care. How many physicians can promptly follow-up on non-emergency indicators?

Under the present system, case management is difficult to scale past a small percentage of high-risk patients. How do we translate digital health data into meaningful results?

A two-part solution
Aggregating health data is a great idea, but we need to do more. We are faced with an expensive and dauntingly complex healthcare crisis. Millions of people are dying from preventable chronic diseases. However, rather than finding better ways to prevent these conditions, we wait for disease to set in. We’re chasing the horse instead of locking the barn.

In that context, does data aggregation represent progress or simply the illusion of progress? The verdict is still out; however, as with all data, it’s not the information gathering that’s important, it’s what we do with it. We can outfit millions of people with apps and devices that count steps, calories, sleep, workouts, heart rates, ad infinitum. We can integrate information into one appealing dashboard. But can we give patients actionable information that will improve health outcomes? And can we build a scalable system that builds in accountability?

For that we need more than an app. We need better infrastructure.

Helping patients
Tracking health information is an evolutionary step when we need a revolution. America is buckling under the cost of healthcare. More than 115 million people have at least one chronic disease and these conditions account for 84 percent of healthcare spending.

This is both bad news and a tremendous opportunity. The World Health Organization estimates that 80 percent of heart disease, stroke and type 2 diabetes cases could be prevented if we simply tackled major risk factors. In addition, 40 percent of cancers could be prevented with the same approach.

However, we cannot generate these changes by grafting new technologies onto a broken system. We need to make prevention the centerpiece of a redesigned healthcare infrastructure. We need a system of physician extenders: nutritionists, fitness professionals, health coaches and nurses to interact frequently with patients.

By continuously engaging with patients, these professionals can interpret aggregated data (from HealthKit or any other source) and provide the personalized feedback so critical to motivate new behaviors and improve health. In addition, a “smart” clinical interface would sort the data to present the most relevant information—in other words, separating signal from noise.

We know this approach can work because it has been proven, over and over. The University of Pittsburgh’s SMART Trial measured activity in three groups. One used a paper nutrition/activity diary; one received a personal digital assistant (PDA) and tracking software; and one received daily feedback from trained coaches to complement their PDAs. Not unexpectedly, the third group had better results—adhering to their goals, increasing their activity and losing weight.

Another study, conducted by Partners Center for Connected Health, compared participants who were given a pedometer with those who received both a pedometer and coaching. Again, the second group increased their activity significantly compared to those who received no coaching.

What this tells us is the human interaction is central to any successful long-term intervention.

Bucking tradition
Healthcare must be personalized, preventive, and most importantly continuous. Our current episodic approach, in which patients get occasional feedback from physicians, fails everyone. We need to use what we know about sustaining behavior change and build that into new healthcare delivery models.

Quite simply, we need a re-engineering of our current care model, rather than a slow iterative approach to innovation. That means creating a network of physician extenders to help patients increase awareness of their own health issues. But we also need to increase accountability through highly personalized coaching and a prescriptive care plan. 

Technology will be an important component of both increasing awareness and accountability. If leveraged properly, apps, wearable technologies, genomic testing, wireless medical devices and other emerging technologies can help millions improve their health.

But information is worthless without accurate interpretation and, even more importantly, an action plan. At present, we simply don’t have the infrastructure to provide these for patients. Given this confluence of impressive technologies, we have a once-in-a-generation opportunity to make radical improvements in our nation’s health. We just need to remind ourselves, gadgets and data aggregation can only do so much. For real change, we need people.