Provider type could not be determined for 25% of shipments,

the information on state and local decisions and processes was not always complete, and databases could have errors. Finally, the number of dependent variable observations is fairly small (51), and many factors may potentially be associated with H1N1 coverage. The distribution and administration of the H1N1 vaccine was a test of the health emergency response systems, and it is an opportunity to identify specific approaches that may result in higher vaccine uptake in a future event of this nature. Several of the findings warrant further consideration. The findings suggest that continued efforts to increase uptake of influenza vaccination may result in increased uptake in an emergency response. The negative association between see more order lags and coverage is an important aspect of the supply chain and distribution. It is possible Icotinib that time lags are a function of the system design or processes, which would suggest monitoring and/or designing the system for fast response within the states in an emergency is needed. There can be many decisions made at the state level that can affect lead-time

including ordering frequency, number of delivery locations, on which days orders were placed, use of third parties, etc. Further study would be useful in this area. Our results on type of location to which vaccine was directed may provide some guidance on increasing coverage, e.g., in a campaign with limited resources and time pressures, sending to general access or public locations may be beneficial. As more adult and specialty providers, including pharmacies, take on the role as vaccinators, this strategy may change. This, too, remains an area where additional analysis is useful, such as collecting information on shipments by type of provider, examining the small number of states where registry information records the location of vaccine administration, or additional analysis on where vaccination occurred for different target groups. C. Davila-Payan collected

data, performed statistical analysis, and aided in drafting the manuscript. J. Swann designed the study, advised on methodology and logistical factors, Rutecarpine and drafted the manuscript. P. Wortley advised on public health and vaccination programs, assisted in acquisition of data, aided in interpretation of results, and editing the manuscript. All authors approved the final manuscript. C. Davila-Payan was partially supported by the ORISE Fellows program during the research. J. Swann was partially supported as the Harold R. and Mary Anne Nash professor, by the Zalesky Family, and by Andrea Laliberte in gifts to the Georgia Institute of Technology, and was partially supported by the Centers for Disease Control and Prevention (CDC) in an Intergovernmental Personnel Act agreement between the CDC and Georgia Tech. The ORISE Fellows program and the donors to Georgia Tech had no role in this research.