Institute for Disease and Disaster Preparedness Plans for Emergencies

Several years ago, Daniel Wattson '00, an ORIE alumnus working at Weill Cornell Medical College in Manhattan, pointed out to his boss, Dr. Nathaniel Hupert, that some of the problems Hupert was working on sounded a lot like those addressed by operations research. One consequence of this observation was the establishment of a collaboration between ORIE and Weill Cornell called the Institute for Disease and Disaster Preparedness (IDDP). The Institute is headed jointly by Dr. Hupert and ORIE's Jack Muckstadt, the Acheson-Laibe Professor of Business Management and Leadership Studies.

In the past year, the co-Directors have assumed new responsibilities in the field and the Institute has changed its location. Muckstadt is now one of 10 members of the Board of Scientific Counselors of the Coordinating Office for Terrorism Preparedness and Emergency Response (COTPER) in the U.S. Centers for Disease Control and Prevention (CDC). Hupert is the inaugural Director of the new Preparedness Modeling Unit in the CDC. And IDDP, originally housed at ORIE's offices in Wall Street, OR Manhattan, has moved uptown to the offices of the Weill Cornell Department of Public Health at 402 E. 67th Street, a block from the main Weill Cornell campus.  

With all of the changes, Muckstadt and Hupert remain immersed in the research issues, many of them in supply chain logistics, associated with handling calamities of all sorts, from pandemic flu to hurricanes to bioterror attacks to nuclear bombings. Uncertainty and variability, always a hallmark of both supply chain systems and epidemiology, is a major component in preparing for such calamities, impacting the prepositioning of pharmacuetical supplies in what is now called the Strategic National Stockpile, the establishment of prophylaxis and treatment locations and coverage areas, and the development of procedures for handling large numbers of impacted individuals.  

According to Hupert, dealing with these problems require both discrete event simulation modeling and optimization techniques. Moreover information of all sorts, from capacities to diagnoses, must flow smoothly through systems that are not currently interconnected and do not necessarily code data in a standard way. Models can have an impact in establishing policy, in supporting crucial planning decisions, and in managing an ongoing emergency. 

Muckstadt points out that responding to a pandemic requires insuring a steady stream of medical supplies while limiting contagion. "Many things that would be required to deal with sick people, simple things like masks, are not made in the U.S. - they're made in China - and some of the plans call for dramatically reducing the amount of material that comes in from other countries, trying to quarantine yourself to some extent," Muckstadt told Beth Saulnier, author of a recent article in the Weill Cornell Medicine magazine. "But if you need to do that, you're liable not to get some needed supplies. So part of what we're trying to do is to stimulate thinking about what is really required, and then help address how you would go about doing it. We've been developing mathematical and computer-based models, representations of these problems, to try to understand where the constraints are going to fall, what capacities are going to be limited, what people are going to have to think about so they can insure that these bottlenecks can be dealt with."  

Muckstadt has developed a relationship with the Cayuga Medical Center in Ithaca to help develop concrete ideas about emergency preparedness in hospitals. This has led to a series of Master of Engineering projects with the hospital. The latest team, co-advised by Muckstadt and Professor Peter Jackson, used computer simulation to assess the potential impact of a pandemic influenza outbreak on the hospital's operation. Muckstadt noted that such an epidemic would by definition be a worldwide event, and "it will cause great demand for resources - some medical, some facilities related," he said. "Our county hospital would be under stress, and it has limited resources to deal with respiratory problems." Cornell has a disaster plan that calls for evacuation of more than two-thirds of the student budy, leaving thousands still on campus, not to mention treamtent for employees and their families. "The question then is, how do you do that? And how do you coordinate that across a county? What resources are going to be required? So on a national scale, these questions arise," Muckstadt told Saulnier. 

Since September of 2000, Hupert has collaborated with several agencies in New York City, providing critical simulation expertise in development of emergency response protocols for bioterrorism. As a physician, he learned how to do discrete event simulation and applied it to "forecasting how clinics in New York, designed to rapidly give out thousands of courses of antibiotics to New Yorkers every hour, how they might function, and most important, whether the staffing would be adequate so that it wouldn't just cause large lines around the city, causing havoc," Hupert told Harry Greenspun, Executive Vice President and Chief Medical Officer for Perot Systems in an audio podcast interview. Hupert's work on the National Strategic Stockpile led to interactions with the CDC, and in 2006 the Assistant Secretary of Health and Human Services suggested to the Director of the CDC, Julie Gerberding, that a larger investment be made in preparedness modeling," according to Hupert. 

"For one of the first times in recent memory," Hupert said, there has been "an effort to bring quantitative assessment techniques to very complex questions for which there was a lot of ground knowledge by the experts -for example, the folks who were engaged in the smallpox eradication campaign - and to speculate on how things would go in the current day."  

Hupert credits ORIE Master of Engineering students projects, working with Muckstadt and others in ORIE over two years, for the insight that "it's possible to actually change the readiness posture and at the same time, save the hospital money simply by ordering certain items differently from our suppliers." The students worked with the material management folks at New York Presbyterian Hospital "just to understand more about two things - one is how we order what we order for the hospital and the other is how that impacts our readiness. What we found was that we actually have a very interesting 'just in time' ordering system for material and that ordering system puts us in a very definable readiness posture for the eventuality of being cut off from our supplies." The students made recommendations that both saved money for the hospital and increased the amount of certain goods that we have on hand for the event of some emergency where we might get cut off from our suppliers upstate. This was a project where we hit two birds with one stone." 

In the Greenspun interview, Hupert also addressed the issue of information for use in models. With standardization of information definitions beginning to take place, "we're now at the stage where we can start to create simulation models and create optimization models of health care systems in near real time without the need for someone to sit down at a computer screen, call up each floor in the hospital, have the Chief Resident and the Chief Nurse walk around, room to room, with a pencil and paper, and laboriously enter information that by the end of the data entry process will be inaccurate because it took so long for you to enter the data. There are ways to automatically, seamlessly use information," Hupert told Greenspun, "and these are techniques that are used basically in every other realm of web-based commerce and information control and industrial applications but they're lacking to a large degree in health care and they're lacking almost to a full degree in preparedness modeling. So we're trying to jump into the information age in these ways."  

After graduation Wattson, the ORIE alum who helped bring Hupert and Muckstadt together, had joined First Manhattan Consulting, the financial industry consulting firm founded by James McCormick '69, M.Eng. '70, where he did technology strategy and risk management modeling. After deciding to make a career switch to medicine, he joined Hupert's group at Weill Cornell and then went to Washington University Medical School in St. Louis. He recently began his residency with the Harvard Radiation Oncology program at the Harvard Medical School and Massachusetts General Hospital. 

Hupert credits Wattson with bringing "the theories of operations research methodology, such as queuing theory, to the models we had been making of patient flow through clinics. Instead of having algebraic models that ran in spreadsheets and gave estimates of the number of staff you might need, we were able to create much more vivid representations of a clinic, such that you were able to watch over time as queues developed and were resolved by staffing. You could see efficiency changes as various aspects of patient flow were changed," Hupert said.