Professors use industrial lean methods to re-engineer clinic

What began as an MBA class project at Purdue University’s Krannert School of Management ended up with participants applying factory management techniques that hold promise for improving medical clinic productivity and quality.

Purdue University photo/Mark Simons
Krannert professors Suresh Chand, from left, and Herbert Moskowitz with a screen display of their research.

Herbert Moskowitz, the school’s Lewis B. Cullman Distinguished Professor of Manufacturing Management, said the student-faculty team’s approach to improving health-care clinic operations is, of necessity, “multi-pronged.”

“We started with the idea of the need to provide better service to more patients,” Moskowitz said. “We soon realized that we could employ manufacturing principles emanating from factory physics and established industrial quality techniques to cut patient waits and better utilize medical personnel.”

What makes a factory efficient is getting the right parts to the right places at the right time. In these days of just-in-time inventory and “lean” methods, factory managers use high-tech scheduling to manage global supply chains.

Operations management professor Suresh Chand said established industrial management processes provide tools to re-engineer a medical clinic so it could deliver more and better care while increasing productivity.

“We decided to apply factory physics at a high level,” Chand said. “Our goals were to increase the efficiency of the clinic in terms of patient time in the office, utilization of clinic resources, quality of care and, ultimately, revenue for the physician.”

Physician revenue has been conspicuous in its absence from discussions about controlling health-care costs, Chand said. In fact, many organizational health-care arrangements limit physicians’ revenue.

Moskowitz said that’s a mistake in terms of making a clinic work more efficiently and, ultimately, in maximizing the quality of patient care.

“If we organize a clinic so a patient spends less time there, then we are maximizing the utilization of the doctors,” Moskowitz said. “That means physicians are being more productive and giving better service while they also are generating more revenue.”

Chand explained that better care also can cost less when the essential efficiencies are realized.

“What we’ve come to understand is that increased revenue and improved quality go hand-in-hand,” he said. “The simple fact is that if a patient’s clinic experience is fast and pleasant, they will avail themselves of more care earlier – a prescription for better health.”

In fall 2004, Moskowitz, Chand and the MBA class started working with the Indiana University Medical Group’s physicians at Grassy Creek, one of its 18 outpatient clinics in Indianapolis serving low-income patients. Moskowitz describes the initial assignment as “collecting, recording and tracking data to evaluate, create designs and optimize systems using process control.”

The Purdue professors and students worked closely with the clinic’s nine physicians, led by Dr. Deanna Willis, an Indiana University assistant professor of family medicine. The team mapped clinic patient and physician flow and identified bottlenecks and flow improvement factors resulting in the clinic’s being able to serve 37 percent more patients on a given day. Those recommendations included:

• Scheduling existing patients to arrive five minutes prior to appointments and new patients 10 minutes early to complete necessary paperwork.
• Centralizing incoming phone calls and not assigning that task to registration assistants.
• Using a single “pooled queue,” or one line, in place of the three separate physician-dedicated lines the clinic was using.
• No “batching” of patient records, which means handling records one at a time in the order patients arrive.
• Including some same-day scheduling in addition to the traditional appointment model.

While the team used an animated simulation model, Moskowitz stressed that the success of the project stemmed from the cooperative relationship between the medical staff and the researchers.

“One of the keys to our success was establishing rapport with the medical personnel so we could work with them to solve problems,” he said. “The physicians and staff were wonderful.”

The team is planning to continue its analysis in future projects with Grassy Creek.

“In our initial work, we didn’t consider clinic demand or appointment scheduling,” Moskowitz said. “There are other variables – finite number of patient slots, congested days, slow days, patient preferences – that we need to come to terms with.”

Chand said that while the project’s first phase was to reduce the time patients spend in the clinic, a new goal is to reduce the time between when patients call for appointments and when they are seen. The logic is two-fold: Sick patients need prompt treatment; and when they go untreated, the consequences tend not to be good.

The effects on the clinic’s business are not good either because patients often fail to show up for appointments made too far in the future. This results in a loss of clinic and physician revenue and wastes slots in which other patients could have been served.

Chand said business models could supply solutions.

“We might want to employ a revenue management model – that is, charge different prices for services booked at different times – along the lines of airlines or hotels,” Chand said. “After all, we’re after the same thing. We want to use all the seats, beds and clinic slots, and there are established ways to manage these variables.”

Moskowitz said, “We’re starting at the end of the chain and working backward to show how to make the clinic model work efficiently to provide both better care and a financial incentive for medical professionals to buy into a different way of doing things. The next step is to evaluate the broader implications of our work for health-care delivery.”

John Norris, a Krannert School doctoral candidate in quantitative methods, said the team is considering using RFID (radio frequency identification) tags in the clinic to more precisely track patient and doctor flow. The tags are the same devices big retailers have started to use to track products from the assembly line all the way to the cash register and out the store door.

“We’re still considering the best way to utilize the technology,” Norris said. “But we need the seamless collection of data that RFID technology can provide so we can build the ‘industrial’ tools, such as statistical process control charts and process capabilities.”

Moskowitz also will use the study as part of a course he is teaching this fall that focuses on factory physics and continuous business process improvement and their applications in health-care engineering and management.

The research was supported by the Regenstrief Center for Healthcare Engineering at Purdue’s Discovery Park. The Regenstrief Center, funded by a grant from the Regenstrief Foundation Inc. in Indianapolis, applies the principles of engineering, management, science and information technology to improve the delivery of health care to consumers

Related Web sites:
Krannert School of Management: http://www.krannert.purdue.edu/
Regenstrief Center for Healthcare Engineering: http://web.e-enterprise.purdue.edu/wps/portal/rche