"I don't like that standing water out there." Dr. William McGuire, the former CEO of UnitedHealth Group, is peering into the summer heat from the new Guthrie Theater in Minneapolis, which he helped pay for. He is staring at a puddle on the lawn of the adjacent park. Which he also helped pay for. He punches buttons on his cell phone. "Luther?" McGuire outlines the puddle problem, then listens. Covering the mouthpiece, he points across a sidewalk to a sad strip of parched yellow grass. "The city owns that grass," McGuire says with a sly smile. He gestures to the lush green carpet where a mother and toddler are now ambling. "That's our grass."
McGuire has experience getting exceptional results from less-than-fertile ground. Under his 15-year stewardship, he transformed a struggling Minnesota HMO into the second-largest health insurer in the country, a $70 billion profit generator that returned nearly 30% annually to investors. He introduced card-swipe technology for benefits information, eliminated gatekeeper referrals, and pushed for treatments based on results—all controversial steps. "He was always focused on disruption," says Sheryl Skolnick, a health-care analyst for CRT Capital.
But these days, he's best known as the latest highly paid CEO to leave his job in the wake of allegations involving the timing of stock options. (His unexercised options were valued at $1.6 billion when the issue surfaced.) He's facing government investigations, shareholder lawsuits, and a battle to retain his hefty retirement package. It is the elephant in the room that we have agreed not to discuss, due to the pending legal issues.
Instead, we are here to talk about the other elephant in the room, the messy health-care system. We first met at an event at this theater last fall, just days before he was slated to leave UnitedHealth. Even then, with an unwanted transition looming, he offered to school me on his great passion—the role data can play in medicine. I'm here to take him up on it.
Why is data so crucial to improving the health-care system?
Because it's the key to measuring the quality that translates into better care. Traditional insurance didn't collect care data; it collected payment histories not specific to each person. We began accumulating data and found that we could see where the disparities were. What percentage of patients were given hemoglobin tests? What medications? We developed feedback mechanisms that we shared with doctors; some people called them report cards. But really we were looking for performance relative to standards that exist. What's clinical evidence? What do we really know?
How did doctors respond?
There was resistance to making the information public. We saw it as an opportunity to help care providers. We know that if you have certain protocols in place, you can reduce infection. Dramatically. We thought we should judge based on who gets the job done. By looking at the data, it's easy to see that not every place is the best at everything. Not every business deserves to stay in business.
And did that data drive better care?
We saw some behavior change, for example, with the centers for excellence around certain transplants. We found a handful of places that really got great results on pediatric cardiac cases. Data can create evidence-based treatments that help make sure people get the right care at the right time.
You insist that doctors should lead health-care change.
They should absolutely lead the way, and they're not now. I happen to think that, on a comparative basis, doctors are underpaid. Doctors gravitate to profit centers when there are low reimbursement rates. You see that kind of turf battle in cardiology and radiology—who reads specialty diagnostics for the heart? We have to change reimbursement. If we differentiated between physicians in terms of quality, they'd take a stronger leadership role as champions of quality.
What about prescription drugs?
Right now, there's no way to track the side effects across a patient population. Say 20 million people are taking a drug, and some have a [negative] event. A claim is generated. You could create a system that says, I want to look for all people with the following event on this drug. Think about Vioxx—that was done. We've got this drug, and all of a sudden, there's four times the cardiac toxicity and nobody realized it. Let the manufacturer and the FDA have access to the data.
You've talked about creating a set of "essential" benefits. But "essential" might sound like "not enough."
Sometimes less medicine is more. Newer isn't always better. I take three medications, all generic—ones for blood pressure and cholesterol, and baby aspirin. Not expensive, but effective.
Basic should also include prevention. Recent data show that 40% of people had not been checked for or were not being treated for hypertension. Ultimately a proportion of those people will have a vascular event. Lifestyle education around diabetes, obesity, and smoking is key.
So everyone should have access to affordable, essential health care. Someone has to pay for that.
Employers should cover some; state and federal governments provide for those who are unable to pay, at some percentage related to the poverty line. Then the market can come forth and offer incremental benefits. We can purchase as individuals, and employers can add on to attract people on a competitive basis.
Why don't we have electronic medical records?
What data should we be collecting? A researcher may want some things; a doctor may need others. There have been no agreed-upon standards. And data collection is tough in a fragmented health-care system. I would set some standards in the use of administration technology. Say, in five years, all claims would be received and paid electronically. We have the technology.
Is universal care possible?
It has to come from the President. Somebody has to stand up and say we expect accessible health care for all people. Not just insurance for all people—that doesn't get you health care if you can't afford the insurance, or there aren't enough facilities and practitioners, or you don't have a way to get to them. Insurance isn't enough.
A version of this article appeared in the October 2007 issue of Fast Company magazine.