Two cross-sectional surveys were conducted at the University of Zurich Centre for Travel' Health during January and February 2009 and January 2010, respectively. Self-administered, anonymous questionnaires including 16 items were distributed to travellers waiting for pre-travel health advice. Participation was voluntary. Individuals above 17 years, understanding German or English, residing in Switzerland and planning to travel to a resource-limited destination were included. Questions included demographic data (gender, age, nationality, education, profession), travel-related characteristics (destination country, duration of stay, influenza risk perception, previous travel health advice, travel purpose, travel costs) and general attitudes and practices towards influenza vaccination (vaccination coverage, reasons to be vaccinated, reasons to refuse vaccination, motivations to consider vaccination with options for multiple answers except for the vaccination coverage). In 2010, an additional question targeting the pandemic influenza A/H1N1 vaccination coverage was included. The questionnaires were checked for completeness. A written letter of exempt was received by the Ethical Commission of the Canton of Zurich.
Statistical analyses were conducted by using SPSS® version 14.0 for Windows. Differences in proportions of demographics, travel-related data and attitudes and practices were compared using the Chi-square test. The significance level was set at p ≤ 0.05. For the multiple logistic regression analysis the surveys were analysed as well as pooled dataset and each survey, CTH-2009 and CTH-2010, separately. The seasonal influenza vaccination was used as outcome and all demographic, travel-related and attitude- and practices-related factors were evaluated as independent predictors. Odds Ratios (OR) were determined by stepwise backward elimination of variables with p > 0.150. For sensitivity analyses, each dataset of the CTH studies, 2009 and 2010, was analysed separately and additionally, predictors for pandemic influenza vaccination were determined by multiple logistic regression analyses.