Industry showing the way for health system change

Biz guru helps pilot scale up fast

You might think fast-food purveyor McDonald’s has little to offer to TB patients in Peru with multidrug-resistant TB (MDR-TB). But when it came time to scale up a model for treating MDR-TB in resource-poor settings, the Harvard-affiliated nonprofit Partners in Health called on Donald Berwick, MD, the CEO and founder of the Boston-based Institute for Healthcare Improvement (IHI). When Berwick sees customers lining up for French fries and cheeseburgers, he doesn’t think about trans fats, but about queuing theory. That’s because Berwick (who also teaches pediatrics at Harvard University) is in the business of making systems — health care systems in particular — run better. For the past 12 years, he’s been adapting the same methods that helped Toyota revolutionize the auto industry 40 years ago and putting them to work in health care.

Those methods, pioneered by American industrial gurus such as Edwards Deming and Joseph Juran, discarded a system of quality assurance based on "gotcha"-style inspections and replaced it with worker empowerment, and analytical tools aimed at finding and weeding out a system’s built-in flaws.

Queuing theory, it turns out, is how McDonald’s figures out how many cash registers should be open at what time of day. Berwick used that concept to help some doctors’ offices serve their patients more efficiently — so that instead of fidgeting in the waiting room for an hour, the patients can now call any day of the week and book a same-day appointment. Even better, a doctor reportedly sees these lucky health care consumers as soon they walk in.

"It’s a matter of analyzing patient demand, and then matching capacity to fit that," says Jonathan Small, communications director at IHI. In the same way, Berwick analyzed how big hotel chains get room service to guests in a timely fashion each morning, and used the same techniques to streamline medication delivery at hospitals.

Change in small, fast packages

When Berwick teamed up with Partners in Health (PIH), the challenge wasn’t to make PIH’s model more efficient, but instead to scale it up quickly, by getting the best-practices package adopted all across Peru as fast as possible. "We think the same methods that worked for Toyota’s production line are going to help us scale up very quickly in a resource-poor setting," says Jim Yong Kim, MD, executive director and co-founder of PIH.

To do that, IHI will use what’s called the Break-through Theory. The idea grew out of Berwick’s frustration with more traditional methods of bringing about change, says Small. "We used to simply teach lots of courses, bringing people together and giving them these new concepts," he says. "But we found that when they got back to their workplace, they didn’t necessarily know how to get things done."

What worked much better, it turned out, was to assemble a group, teach everyone new approaches to a problem, and then send them all back home to try out a specific change or two. A few months later, the same group would reconvene to talk about how things had worked out, and to pick up a new assignment.

The result, Small says, was a revolution by increments — the accumulation of many small changes, which eventually equaled sweeping, systemwide alterations. "It’s basically a model for spreading new ideas very rapidly," he adds.

Even when proposed changes don’t work out, the strategy provides feedback loops that allow for corrections. That means it’s far more cost-effective than the usual way of bringing about systematic change, notes Small. "What’s really expensive are huge improvement models that are run from the top down," he points out. "What we’re doing amounts to thousands of little tests run at the front lines."

That’s the idea in Peru, where dozens of teams from health care centers are now meeting and working to spread the PIH model. There, the learning process is more a two-way street than usual, Small adds that since TB experts are having to show the business wonks a thing or two about life in the developing world. "We’re not accustomed to working in Spanish, for one thing," notes Small. "More to the point, we’re used to having substantial communication linkages among our clients, with teleconferencing and e-mail and so on. They’re having to show us how to operate effectively without all that."

Someday, Kim says hopes the Berwick model will be used to tackle other challenges, such as the AIDS pandemic. Again, the problem isn’t figuring out what to do — after all, rich countries have already devised effective treatments for their own AIDS patients — but how to deliver those treatments on a grand scale to the millions of poor people who urgently need them.