Approval for this study was obtained from Medical IRB-2 at UCLA. Written consent was obtained from all participants.
Participants were recruited for participation via convenience sampling on the UCLA campus. Anyone who fulfilled the inclusion criteria (at least 18 years of age, able to give informed consent in English and affiliated with the UCLA community) was offered enrollment in the study. Potential participants were recruited from large gatherings of UCLA students, faculty and staff in order to maximize participation in the study. Following the process of informed consent, all participants completed a comprehensive questionnaire containing questions about their basic demographic information including their date of birth, gender, race, location of residence (on/off campus), and affiliation with UCLA (undergraduate, graduate, faculty, staff or other). The questionnaire also inquired about their vaccination status (both seasonal and H1N1 vaccine) and history of flu-like symptoms including chills, cough, diarrhea, fever, headache, muscle aches, nausea, runny nose, sore throat, stuffy nose and vomiting. They were asked whether or not they sought medical care, whether or not they had been diagnosed with H1N1 and whether or not they had been hospitalized due to H1N1 infection. They were also asked whether or not they had known exposures to individuals who had been diagnosed with H1N1 infection. In order to attempt to minimize recall bias, for each question participants were specifically prompted to think about the time frame from April 2009 until the current date (May 2010). All subjects were recruited in May of 2010.
A 10 ml venous blood sample was then collected from all willing participants. Participants were free at any point to decline to give detailed information and/or biological specimens and they were free to end their participation in the study at anytime. In total, 300 participants were enrolled in the study.
Blood specimens were centrifuged and separated into plasma and buffy-coat aliquots prior to being frozen and stored in a - 20°C freezer at UCLA until they were ready for laboratory analysis at the University of Florida. Antibody responses were detected by use of a routine standard Centers for Disease Control and Prevention protocol hemagglutination antibody inhibition (HAI) analysis to detect antibodies to influenza A(H1N1) 2009 (A/Mexico/4108/09(pandemic H1N1)). The human H1N1 strain was grown in MDCK cells. Serum samples were pretreated with receptor-destroying enzymes (1 part serum to 3 parts enzyme) from Vibrio cholerae overnight, and then were hemadsorbed with guinea pig blood. Serial two-fold dilutions of serum were tested beginning with a 1:10 dilution and a final dilution of 1:640. Suitable control samples were included in all assays. HI titer results were reported as the reciprocal of the highest dilution of serum that inhibited virus-induced hemagglutination of a 0.5% (from guinea pigs) solution of erythrocytes [14, 15]. In order to be consistent with other published papers in the field, sera were considered positive at an HI titre of 1: 40 or greater since titres greater than 1: 40 are correlated with a reduction of 50% of the risk of contracting an influenza infection [16–23]. While the authors acknowledge this is a highly sensitive threshold, it may come at the compromise of specificity.
Analyses were conducted using the SAS software (SAS, Inc., V.9.1). A cross-sectional analysis was performed using the questionnaire data obtained from the participants. Descriptive variables were examined and then we cross-tabulated variables of interest to explore any apparent associations. When analyzing symptomatic characteristics, we looked at which participants reported the aggregate of symptoms which met the May 2010 CDC definition on influenza-like illness (fever AND cough and/or sore throat). Because of small sample size, more complex statistical analyses were not performed.