Travel destinations were grouped as follows: Africa, Asia, Europe, Latin America/Caribbean, and the USA. A “Multiple/other” category was used for cases that traveled to more than one destination group or that traveled to other parts of the world (eg, Australia). INK 128 manufacturer The data variables used for the analysis included age; gender; onset date; disease; symptoms

(abdominal pain, abdominal bloating, chills, dehydration, diarrhea, bloody diarrhea, greasy diarrhea, chronic diarrhea, dizziness, fatigue, fever, headache, loss of appetite, weight loss, muscle soreness, nausea, vomiting, and weakness, depending on the disease); hospitalization; recovery date; travel destination; resort accommodation; mode of transportation; and departure and return dates. Where data were available, disease duration was computed as recovery

minus onset dates, and travel duration as return minus departure dates. The TRC were broken down by illness and then further described by hospitalization, symptoms, disease duration, age, gender, travel duration, travel destination, mode of transportation, and resort accommodation. When available, onset, travel departure, and return dates were plotted by month. For each of these three dates, a Poisson regression model of the monthly count was used to test the differences LDK378 between both years and months at the same time. The year was defined as a successive 12-month period starting from June and ending in May the following year; eg, 2005 to 2006 encompassed June 2005 to May 2006 Pazopanib datasheet inclusively.

Multiple correspondence analysis (MCA) was used to probe the existence of travelers’ subgroups within the data and, if these existed, to explore any association between these subgroups and illness. MCA is a descriptive statistical technique designed to explore and visualize the relationships between three or more categorical variables (see Appendix 1 for more details).23 MCA was conducted on age (<5, 5–14, 15–24, 25–39, 40–59, and 60+ y), gender, travel duration (<8, 8–28, and 29+ d), travel destination, and accommodation in resort. In addition, disease was the supplementary variable used to assess any relationship between the subgroups identified and the illnesses. Finally, TRC and DC were overall described by, and tested for differences in, gender, age, illness distribution, and hospitalization. For every disease with at least 30 TRC, TRC and DC were compared for gender, age, hospitalization, disease duration, the various symptoms relevant to the disease, and the delay of reporting (difference between onset and report dates). Campylobacter species and Salmonella serotypes were also compared. The Kruskal–Wallis test was used for continuous data and the bilateral Fisher exact test or the Chi-square test for categorical data. The p-value threshold was set at 0.01 for all statistical tests.