We also examined measures of income inequalities [22], and segregation and disparities [23]. We extracted the geographical area, number of counties, and federal government expenditure per capita from the Census. We estimated the total number of healthcare practitioners [24], the number of active physicians [25] per thousand population (PTP), and the percentage of the population who have not visited a doctor in the last year because

of cost [2]. We determined whether states were characterized by state control, local control, or by inference, mixed control, from the 2008 National Profile of Local Health Departments [26]. To capture health-seeking behaviors and use of preventive services, we obtained state-specific influenza vaccination rates for previous seasons [7], the percent of women who had a Pap smear in the past 3 years [2], and population PFI-2 ic50 percentages associated with various health conditions [27]. We obtained information on the emergency funding provided to states for the H1N1 pandemic Androgen Receptor Antagonist order from CDC reports including amounts spent or unobligated for assessment, planning and response [28] and [29]. Reports from the Outpatient Influenza-like Illness Network (ILINet) [5] obtained from the CDC, provided weekly values for the proportion of outpatient visits for influenza-like illness (ILI) at participating

providers, by state, from which we calculated several measures including the percentage of weeks with % ILI above 2.3, after week 30. We extracted information on state processes and decisions from a survey [30] of immunization

program managers conducted by the no University of Michigan to provide CDC with situational awareness during the H1N1 campaign on allocation of vaccine, expansion date beyond priority groups, whether a state focused on school vaccination or not, and vaccine distribution methods. We obtained information on the amount of vaccine allocated to each state over time, the maximum number of provider sites to which each state could have vaccine shipped through the centralized distribution system (“ship-to” sites) [8], and self-reported data from states on doses distributed to or administered in public settings [31]. Information on the date, address, and number of doses shipped to each location, from the beginning of the campaign through December 9, 2009 (which covers the major shortage period) was obtained from the centralized distribution shipping records [4]. We calculated measures such as the number of unique sites to which vaccine was shipped (ship-to sites), the average number of shipments per site, the variation in doses PTP across counties within a state, and the lead-time from allocation to shipment (i.e.