The spread of pandemic influenza A (2009 H1N1 influenza) virus resulted in a global influenza pandemic in 2009. During the early stages of the pandemic, population surveillance was crucial. However, officials around the world realized that many of our surveillance and reporting systems were not prepared to respond in a coordinated, integrated way, which made informed public health decision-making very difficult. More accurate estimates of the total number of hospitalized 2009 H1N1 influenza cases were required to calculate population-based 2009 H1N1 influenza-associated mortality, morbidity and hospitalization rates. For instance, how many people were hospitalized with 2009 H1N1 influenza in Massachusetts? Of these, how many were admitted to the ICU and how many died? Compared to seasonal influenza, were some race/ethnic and age groups affected more than others, and what types of characteristics led to more severe manifestations of 2009 H1N1 influenza among these groups in Massachusetts? To address the above questions, I proposed a retrospective cohort study using data from the Hospital Discharge Database (HDD), which contains data for all inpatients discharged from 76 acute care hospitals in Massachusetts, as well as Census information to provide a measure of socioeconomic status (SES). My specific aims are as follows: 1. Develop methods to identify influenza cases precisely and describe characteristics of those hospitalized with ILI in MA between April 26-Sept 30, 2009; 2. Conduct analyses to identify race/ethnicity-related trends in reference to 2009 H1N1 influenza-related hospitalizations; 3. Conduct analyses to identify age-related trends in reference to 2009 H1N1 influenza-related hospitalizations. First, I established influenza case selection criteria using hospital discharge data. I addressed limitations in the published methods on defining cases of influenza using administrative databases, and evaluated ICD-9 codes that correspond with common and relatively serious respiratory infections and influenza using a ‘maximum’ and ‘minimum’ approach. Results confirmed that 2009 H1N1 influenza affected a younger population, and disproportionately affected racial minorities in Massachusetts. There were also higher rates of ICU admission compared to seasonal influenza. I then presented epidemiological data indicating race/ethnic disparity among 2009 H1N1 influenza cases in Massachusetts. I found that Hispanics had significantly lower odds of 2009 H1N1 influenza-related ICU stay. SES gradients calculated using five-digit zip code information did not account for these differences. Within race/ethnic strata, Hispanics Finally, I presented epidemiological data indicating differences among 2009 H1N1 influenza cases by age group in Massachusetts. I calculated measures of Diagnostic Cost Group (DxCG) comorbidity for the study population to provide a comorbidity measure at baseline. Main results indicate that although comorbidity scores were similar between the 2009 H1N1 influenza and seasonal influenza groups, 2009 H1N1 influenza caused more severe disease in younger age groups. This is the first study to report population-based statewide outcomes in all acute care centers in MA. In this dissertation I address challenges surrounding influenza surveillance to create case selection criteria within an administrative database. Using my case selection criteria, I then provide data related to fatality and severity of 2009 H1N1 influenza in Massachusetts in reference to sociodemographic variables such as racial/ethnicity and age groups, and provide evidence for patient-level interventions to those hardest hit by influenza. These findings provide valuable information about using large administrative databases to describe pandemic influenza cases and guide resource allocation to reduce disparities in relation to pandemic influenza preparedness.

Human influenza is a contagious respiratory disease resulting in substantial morbidity and mortality worldwide. With the recent cases of avian influenza infections in humans and the heightened concern for an influenza pandemic arising from these infections, it is essential to understand host responses that would confer protective immunity to influenza. The cell-mediated immune responses to influenza virus play an important role during influenza infection. To analyze the specificity and diversity of memory T-cell responses, we performed a genome-wide screening of T cell epitopes to influenza A virus in healthy adult donors. We identified a total of 83 peptides, 54 of them novel, to which specific T cells were detectable in interferon-(IFN-γ) enzyme-linked immunosorbent spot assays (ELISPOT) using peripheral blood mononuclear cells (PBMCs) from four healthy adult donors. We found that among 11 influenza viral proteins, hemagglutinin (HA) and matrix protein 1 (M1) had more T-cell epitopes than other viral proteins. The donors were not previously exposed to H5N1 subtype, but we detected H5 HA T cell responses in two of the four donors. To confirm that HA is a major target of T cell responses we also analyzed H1 and H3 HA-specific T-cell responses using PBMC of additional 30 adult donors. Fifteen out of thirty donors gave a positive response to H3 HA peptides, whereas five of thirty donors gave a positive response to H1 HA peptides. Because we detected T cell responses to the H5 HA peptides in donors without prior exposure to H5N1 subtype, we asked if cross-reactive T cells to H5 HA peptides can be attributed to a prior exposure to H2N2 subtype, the closest HA to the H5 based on their phylogeny. We compared younger donors who have no prior exposure to H2N2 subtype and older donors who were likely to be exposed to H2N2 subtype, and both groups responded H2N2 peptides at similar level, suggesting that memory T cells cross-reactive to H5 HA peptides can be generated by prior exposure to the H1N1 and H3N2 subtypes, and the exposure to H2N2 subtype is not necessary. We subsequently identified a CD4+ T cell epitope that lies in the fusion peptide of the HA. This epitope is well conserved in all 16 subtypes of HA of influenza A and the HA of the influenza B virus. A CD4+ T cell line specific to this epitope recognizes target cells infected with various influenza A viruses including seasonal H1N1 and H3N2, a reassortant H2N1, the 2009 pandemic H1N1, H5N1 and influenza B virus in cytotoxicity assays and intracellular cytokine staining assays. Individuals who have the HLA-DRB1*09 allele have ex vivo IFN-γ responses to this epitope peptide in ELISPOT. Although natural infection or standard vaccination may not induce strong T and B cell responses to this very conserved epitope in the fusion peptide, it may be possible to develop a vaccination strategy to induce these CD4+ T cells which are cross-reactive to both influenza A and B viruses.