The surveillance population of interest was all active component service members who served at some point during the period of September 1, 2010 through April 30, 2011. Data were obtained from the Defense Medical Surveillance System (DMSS), a large relational database maintained at the Armed Forces Health Surveillance Center (AFHSC) which contains longitudinal data including demographic characteristics, occupations, immunizations and medical encounters for US military personnel 
. Additionally, data on laboratory-confirmed cases and test-negative controls were obtained from a Department of Defense influenza reference laboratory at the United States Air Force School of Aerospace Medicine (USAFSAM) 
and represented cases detected at US military treatment facilities in the United States and internationally.
Cases were defined as active component service members with a laboratory-confirmed influenza-associated illness, detected by reverse transcriptase polymerase chain reaction (RT-PCR) or viral culture methods. Cases were identified through one of the service-specific notifiable disease reporting systems 
or by collaborating investigators at USAFSAM 
. Since cases provided by USAFSAM were identified by a nasal wash or nasopharyngeal (NP) swab specimen submitted to the laboratory, they were additionally required to have an inpatient or outpatient medical encounter occurring within seven days of the specimen collection date in order to be considered a case. An individual was eligible to be a case only once during the study period; if an individual had more than one case-defining diagnosis, then only the first episode was included. Nasal wash/NP samples were mostly taken within 72 hours of symptom onset as per DoD influenza surveillance recommendations. For cases reported via service-specific notifiable disease reporting systems, respiratory illnesses that met a clinical case definition (sudden onset of fever >102.2°F, respiratory symptoms and either myalgia or headache) and were laboratory-confirmed as influenza were evaluated.
Two groups were selected for comparison, 1) a test-negative control group which consisted of individuals suffering from non-influenza respiratory illnesses, and 2) a healthy control group, consisting of individuals who had a medical encounter for a non-respiratory illness within three days of the case’s medical encounter. Healthy controls were required to have a musculoskeletal (International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) 700–739, 810–848, or V54) or mental health encounter (ICD-9-CM 700–739, 810–848, or V54) with no documented respiratory problems (ICD-9-CM 001–139, 320–326, 380–382, 460–519, 780.6, 780.7, 786, or 787.0 ) during the medical visit. In order to control for baseline immunity, gender and geographic influenza distribution, healthy controls were additionally matched to cases on age, sex, and location. Individuals in both control groups were excluded if they had an inpatient, outpatient, or reportable medical event encounter at any point during the study period with a diagnosis of influenza (ICD-9-CM 487 or 488). A maximum of four controls were matched to each case.
Immunization data from DMSS were used to determine whether cases and controls received any influenza vaccination during the period of September 1, 2010 through April 30, 2011. Subjects who received an influenza vaccine at least 14 days prior to the date of their qualifying medical encounter were considered immunized; all others (those immunized less than 14 days prior to or after the medical encounter, or those not vaccinated with any of the 2010–11 seasonal influenza vaccines) were considered non-immunized for the purposes of this evaluation. Additionally, immunization data from DMSS for the period of 2004–05 through 2009–10 seasons was also obtained in order to control for prior immunization history, an important potential confounder in our highly-immunized military 
. Non- immunization of personnel may be due to their non-availability due to deployment, medical contraindications (such as history of hives or other significant allergic reactions to previous influenza vaccination) or simply due to non-availability of vaccine at the unit, clinic or hospital medical level.
Crude odds ratios (OR) were calculated for comparison of cases to controls by multiple factors including sex, age group (<25 years, 25 to 29 years, 30 to 39 years, 40 years and over), race-ethnicity (White, Black, Hispanic, Asian/Pacific Islander, American/Alaskan Indian, Other/unknown), Branch of service (Army, Air Force, Coast Guard, Navy, Marine Corps), hospitalization status, location at diagnosis (US versus non-US), month of diagnosis, and any prior influenza vaccinations since 2004.
Adjusted odds ratios (AOR) for vaccination status were calculated using logistic and conditional logistic regression (depending on whether the cases and controls were matched). The test-negative control analysis adjusted for sex, age group, number of prior vaccinations and month of diagnosis. The healthy control analysis adjusted for sex, age group, and number of prior vaccinations. VE was defined as (1–AOR)*100. Vaccine type (TIV versus LAIV) and influenza subtype (A/H3, A/H1 and B) specific analyses were also conducted. The subtype analysis was restricted to USAFSAM cases and healthy control subjects due to availability of data. All analyses were performed using SAS 9.1.3 (SAS Institute, Cary, North Carolina, USA).
This study was reviewed by the US Air Force Research Laboratory (AFRL) Institutional Review Board and was determined not to constitute human use research according to the 32 CRD 219.102 (d) (AFRL IRB number: FWR20110097N). The opinions and assertions contained herein are solely those of the authors and do not reflect the official policy or position of the US Department of Defense (DoD) or of its subordinate services (Army or Air Force) medical authorities.