Big Data Can Bring Patients to Water But It Can’t Make Them Think; Aetna can tell its members if they’re likely to develop cardiovascular disease. It does this by tracking data from lab results, pharmacy data and claims data of its 18 million members

March 20, 2013, 7:30 PM ET

Big Data Can Bring Patients to Water But It Can’t Make Them Think

Michael Hickins

As it prepares to vie for new business from some of the 30 million additional people entering health exchanges through the Affordable Care Act next year Aetna Inc. is looking to analytics as a means of lowering the cost of some coverage. According to Michael Palmer, head of innovation for the Hartford, Conn.-based insurance company, Aetna is using a new analytic platform to predict which ailments its members are likely to contract over the coming year in order to lower the odds that they will develop cardiovascular disease, one of the more expensive and endemic diseases it has to cover.

This information could help improve health outcomes for patients, dramatically lowering health care costs for themselves, their employers and Aetna itself, says Mr. Palmer. “Better outcomes also lead to better costs. It’s a virtuous cycle,” he told CIO Journal Wednesday after a presentation at the Structure: Data conference in New York. But Mr. Palmer also noted that it’s difficult to get people to act on the information they’re given, even if it’s for their own good.

For example, Aetna can tell its members if they’re likely to develop cardiovascular disease. It does this by tracking data from lab results, pharmacy data and claims data of its 18 million members, looking for data showing that a given individual suffers from three of any of five factors – high cholesterol, high blood pressure, low HDL (so-called good cholesterol), high triglyceride levels, and abdominal girth – all of which are indicative of metabolic syndrome. “We found we can predict at the individual level the probability of their getting metabolic syndrome in the coming year,” Mr. Palmer said.

Metabolic syndrome, he said, is a combination of disorders that, when occurring together, increase the risk of cardiovascular disease and diabetes. According to Mr. Palmer, having three of the five factors present makes an individual five times more likely to contract metabolic syndrome than other people.

Aetna can also, for patients presenting two of those factors, predict which third factor those people are likely to develop next over the coming year. According to Mr. Palmer, Aetna spends $6.5 billion per year on cardiovascular disease on behalf of its customers, and some of Aetna’s largest customers spend more than $1 billion on employee health plans. It costs companies 1.6 times more to cover people with cardiovascular disease than the average employee population. Aetna and its employer customers thus have a huge incentive to reduce the incidence of cardiovascular disease in the workforce. “You get a more productive and healthier workforce, but there are also direct economic savings,” he said. He said Aetna is trying to help its customers by telling them “what are the right interventions you can do with your workforce to allow them to improve their health and reduce metabolic syndrome,” he said.

Using “Big Data, you can push these capabilities out to the individual level, give them access to the data so they can manage their own health,” he said.

The company has to be careful to respect the privacy of people whose data is being mined. But Mr. Palmer notes that members can’t be denied coverage due to the analytics Aetna performs because the ACA prohibits insurers from denying coverage to individuals because they have preexisting conditions.

A greater difficulty is getting people to act on the information they’re given. Mr. Palmer said some employers using Aetna plans have tried charging employees more for their coverage if they refuse to take certain preventative tests; more common incentives, such as lower costs, have not proven particularly effective, he said. Aetna tried increasing the adherence rate – the rate with which people take their drugs as prescribed – by eliminating co-pays for medications prescribed to people who had just had a heart attack. According to Mr. Palmer, adherence went from the low 40s percentage of adherence before the new policy was tested, to the mid-40s after the test. “Getting people to do the right things for themselves… is still a challenge,” he said.

Aetna uses analytic technology from GNS Healthcare for its predictive work on metabolic syndrome, and works with another half-dozen startups for other Big Data projects. He said Aetna works with larger vendors as well, but says the smaller companies innovate more quickly and “drive things at speed.”