"We hear a lot about bioterrorism and pandemics," says Sheldon H. Jacobson, a professor of computer science and pediatrics at University of Illinois at Urbana-Champaign,"but the fact of the matter is, the threat to routine immunization is one of the greatest threats we face.   If we had problems with our vaccine supply chain, it would have the potential to cause more deaths than any of those other issues."

Key in 'herd immunity', the smallest percentage of a population that must be immunized against a disease so that unvaccinated individuals are also protected, for society is the pediatric vaccine stockpile, he says.   95% of the population vaccinated is a good buffer, for example, while 100% is unrealistic.   With an average of 11,000 children in need of immunization on any given day, pediatric vaccine policy shouldn't simply be limited to increasing the levels of the stockpile itself but in optimizing it.   The CDC's current policy of maintaining a six-month rotating vaccine stockpile isn't the most optimal solution for achieving herd immunity, Jacobson says.   "When you have finite economic resources, you have to make choices. But we also want to create a buffer that will create the widest possible public health safety net."

Jacobson considered the relative mortality and morbidity of diseases to determine what levels stockpiles should be set at to minimize the risk of shortage and maximize coverage opportunities while minimizing costs and created a multi-attribute model using utility theory to solve for appropriate levels.   Jacobson said that when actual vaccine shortages have occurred over the last 10 years, the duration of the shortages were between 16 and 18 months.

"Even though we're preparing for six months out, we never see six months," he said. "The shortest time period in the last 10 years has been a 7-month period. That's a byproduct of one-size-fits-all policy for stockpiles. Although it's easy and simple to do, it's not the most efficient policy. We need to re-engineer the objectives of our pediatric vaccine stockpiles and establish more flexible policies for maximizing their utility."

Jacobson's research proposes a stochastic model to determine the stockpile levels that minimize the risk of a vaccine shortage during a supply interruption while also maintaining a given coverage rate.

"For some vaccines, you need very little stockpiles; for others, you need much larger ones," he said.  Pertussis, or whooping cough, as it's more commonly known, can be a "very deadly, virulent disease if it spreads, while mumps, on the other hand, rarely causes death.   When we're talking about vaccines, equal is not effective. The recent pertussis outbreaks in California and Ohio highlight the needs for differentiated stockpile levels, meaning we have to look at the characteristics of the diseases in terms of achieving herd immunity as well as how deadly the disease is."

Jacobson said that although routine immunization is the most effective public health strategy to prevent the occurrence and spread of infectious diseases, there's always going to be a certain small percentage of the population who will not be immunized, because of religious beliefs or allergies to the vaccine.   Ultimately, vaccination is a critical public health issue that can't be run by the vagaries of emotion.

"If we allow our emotions to guide our policies, we'll pay the price somewhere down the line. The public health system saves rather than costs our nation money. And any way that we can reduce mortality and morbidity through immunization would be beneficial to the health-care system both in cost and value and to the nation as a whole."

Citation: Ruben A. Proano, Sheldon H. Jacobson, Janet A. Jokela, 'A multi-attribute approach for setting pediatric vaccine stockpile levels', Journal of Industrial and Management Optimization (JIMO) Pages: 709 - 727, Volume: 6 , Issue: 4 , November 2010 doi:10.3934/jimo.2010.6.709