Problem definition: The U.S. influenza (flu) vaccine supply chain is decentralized and experiences frequent supply and demand mismatches caused by two key factors: (1) the vaccine production process (yield) is highly uncertain; and (2) individuals are self-interested and do not completely take into account positive and negative externalities that they impose on others. To improve matching of supply and demand, we counteract these factors by developing an ex ante budget-neutral incentive program. Academic/practical relevance: We establish the sources of inefficiency in the flu vaccine supply chain. To eliminate the inefficiency, we develop a two-sided incentive program that policymakers can implement to finance vaccines under an ex ante balanced budget. Methodology: We model the flu vaccine supply chain as a decentralized system consisting of self-interested individuals on the demand side, and a profit-maximizing manufacturer with uncertain yield on the supply side. We use backward induction to characterize the subgame-perfect equilibrium of the sequential game that models the interactions between individuals and the manufacturer. Results: We develop a two-sided incentive program that proposes âvaccination incentivesâ to be given to individuals on the demand side, and âa menu of transfer paymentsâ between the social planner and manufacturer on the supply side. When the realized vaccine supply is high (or low), our incentive program provides positive (negative) vaccination incentives for individuals to stimulate (or curb) the demand and eliminate positive (or negative) externalities by making vaccination more affordable (or costly). When social benefits from vaccination are significantly high, our incentive program uses a menu of transfer payments to penalize (or subsidize) the manufacturer for low (or high) yield realizations so that it produces the socially optimal quantity. We show that our incentive program can attain the social optimum, maintain an ex ante balanced budget (i.e., budget-neutral in expectation), and distribute the maximum social welfare between individuals and the manufacturer arbitrarily. Managerial implications: Vaccination incentives to individuals can ensure their access to the vaccine, but they are not enough to entice the manufacturer to ensure vaccine availability. A menu of contracts contingent on realized yield provides necessary incentives to the manufacturer and assures the availability.
Graeme Laver, one of Australiaâs greatest influenza research scientists, died after collapsing en route to an influenza meeting in Portugal in September 2008. Graeme was born in Victoria, educated at Ivanhoe Grammar School and graduated from the University of Melbourne in biochemistry. He received his PhD in biochemistry at the University of London before returning to the John Curtin School of Medical Research at the Australian National University (ANU) in 1958. After the Burnet School at the Walter and Eliza Hall Institute in Melbourne changed its focus from influenza to immunology, the influenza research program moved with Frank Fennerâs group to ANU, where it was anchored by Stephan Fazekas de St. Groth. There the pioneering influenza work of Sir MacFarlane Burnet continued, with Graeme Laver as biochemist and Stephan Fazekas as the mathematical modeler of virus neutralization. I (Rob Webster) joined the group in 1959 to carry on the virus neutralization studies. To elucidate the subunit structure of influenza viruses, Graeme gently dissociated the lipid bilayer of the virus with the mild detergent sodium deoxycholate and established the basis for the first Australian subunit influenza vaccine. Before that accomplishment, only intact inactivated influenza vaccines had been used in Australia and they were considered undesirably reactogenic. Studies first in rabbits and then in children established the proof of principle for the vaccine. The ANU paid Laver and myself 10 shillings each for the patent rights. Australian subunit influenza vaccines have been further improved, but the fundamental principle established by Laver remains the same: that the proteins must be kept in their native conformation for the vaccine to be effective. Further studies with Robin Valentine of the National Institute of Medical Research at Mill Hill, London, established the morphology of the hemagglutinin (HA) and neuraminidase (NA) subunits by electron microscopy, and later work with Nick Wrigley produced the first electron microscopy images of antigen-antibody complexes. After the emergence of the Asian influenza pandemic in 1957 the influenza community, led by Martin Kaplan of the World Health Organization (WHO), began searching for the origins of influenza pandemics. During a fishing trip to the Southeast coast of Australia, near Batemanâs Bay, New South Wales, Laver and I found the beaches littered with dead mutton birds (Puffinus pacificus). Knowing that A/Tern/South Africa/61 (H5N3) influenza virus had been isolated from seabirds, we speculated that the mutton birds might have been killed by an influenza virus. In 1970, serological studies of mutton birds on the Great Barrier Reef of Australia found the first evidence of human N2 neuraminidaseâinhibiting antibodies, and in 1973 influenza viruses were isolated for the first time from wild migratory seabirds. These findings provided the first links in the chain of evidence that subsequently showed that the migratory aquatic birds of the world are the natural reservoirs of all influenza A viruses. Studies on the structure of influenza virus neuraminidase began with Graemeâs keynote paper in 1978 on crystallization of the heads of the human N2 neuraminidase molecule (Virology 1978; 96: 78â87). These and subsequent studies culminated in the resolution of the three-dimensional structure of the neuraminidase, with Peter Colman and Jose Varghese. With this information, Mark von Itzstein was able to design the first structure-based anti-influenza drug, zanamivir (Relenza). Further development using NA crystals provided by Graeme led to the development of the orally available anti-influenza drug oseltamivir (Tamiflu). The WHO Global Influenza Surveillance Network was established in 1952 to keep up with antigenic drift in influenza viruses and to recommend changes in vaccine strains to stay abreast of antigenic variation. However, there was no fundamental understanding of the mechanisms involved until Graeme Laver began peptide mapping of the hemagglutinin (HA) molecule and provided evidence of multiple antigenic determinants on HA spikes (Virology 1974; 59: 230â244). The advent of monoclonal antibodies by Georges Kohler and Caesar Milstein and their use to select influenza virus escape mutants, together with sequence analysis of the HA molecules (Virology 1979; 98: 226â237), established that a single amino acid change in the HA would allow the influenza virus to escape neutralization. Antigenic mapping of the HA and NA molecules soon followed and, ultimately, establishment of the antibody binding domains and visualization of the epitopes on the NA subunits (Nature 1987; 326: 358â363). Thus, the mechanism of antigenic drift in influenza and the epitopes involved were elucidated thanks to the seminal work of the Laver laboratory. Graeme Laver had a sense of adventure in all aspects of his life, from family to friends to research, and was always a free spirit who was willing to challenge bureaucracy when they âgot it wrongâ. He was an avid skier, mountaineer, gardener and axeman. In later life, his scientific passion was the production of perfect crystals of NA through studies in zero gravity on the Russian MIR space station and crystallization chambers set up in every laboratory he visited. Some of the most beautiful crystals were the N9 neuraminidase of an influenza virus isolated from a noddy tern on the Great Barrier Reef. Photographs of these crystals appeared on the covers of many scientific journals, and Graeme and photographer Julie Macklin were awarded the Nikon âSmall Worldâ prize in 1987. In addition to influenza, Graeme made contributions to our knowledge about adenoviruses, sialidases and paramyxoviruses. His multiple scientific achievements were recognized by his election to the Royal Society of London in 1987 and by co-receipt of the Australia Prize with Peter Coleman, Mark von Itzstein and Paul Janssen in 1996. In its efforts to prepare for an influenza pandemic and to identify the best options for controlling seasonal influenza, the global community owes a very special debt to Graeme Laver. The current strategies for control of both pandemic and seasonal influenza, the stockpiles of anti-influenza drugs and prepandemic vaccines are products of his lifeâs work. His final mission was to have anti-influenza drugs made available over the counter so that every household could have an immediate supply in the face of a pandemic. He argued that this option would have no effect on the emergence of resistance and could save millions of lives in the event of infrastructure failure. New Zealand has taken the lead in following his visionary advice.