Better Decision-Making on Vaccines

The H1N1 influenza virus (Swine Flu) appeared after the 2009-10 flu vaccine was approved — causing  the vaccine for this specific strain of the flu to be rush-manufactured and delivered to consumers separately.

Soo-Haeng Cho, assistant professor of operations management at Carnegie Mellon's Tepper School of Business, believes this kind of situation could be avoided through an improved decision-making process.

"At least once a year the Vaccine and Related Biologic Products Advisory Committee meets to decide the composition of seasonal influenza vaccine for the United States," explained Cho. "Because the virus strains constantly mutate, and because manufacture of the vaccine, needed in October of that year, takes a long time, the committee must make their decision early in the year — before they have optimal data on the strains to be feared."

If the committee decides to retain the current vaccine composition instead of updating it to a new one — as they did in 2009 — Cho says there is lower uncertainty in production yields, but the current vaccine could be less effective if a new virus strain spreads later, as it did in the case of H1N1.

Cho derived a model to improve this decision-making, showing that the optimal timing of selecting a new strain is a few weeks later than that of retaining the current vaccine strain.

"The model suggests that the committee should delay its decision by three to four weeks to collect more information," said Cho. "This delay allows for better decision regarding the virus strains that are likely to threaten consumer health."

Cho's dynamic model analyzes the effect of cross-efficacy of vaccines, capacity expansion of incumbent manufacturers, entry of new manufacturers, and the severity and progress of virus activities. He also examines consumer behavior, as consumers place increasing value on obtaining a vaccination rather than suffering the risk of expensive illness or even death.

"By incorporating these key components, the model allows the committee to perform a what-if analysis under various scenarios. The value of this approach is potentially huge," Cho added. "This is an example of the Tepper School applying mathematical principles to a practical and important business — and social — problem."

Cho's award-winning paper, "The Optimal Composition of Influenza Vaccines Subject to Random Production Yields," will be published soon in Manufacturing & Service Operations Management, a leading journal in the field.

"This is an example of a dynamic decision-making situation prevalent in other industries," said Cho. "An optimized delay in making the choice can lead to a better social outcome."

Related Links: Cho's Homepage  |  Tepper School of Business

Homepage Story Archives