Time trend and seasonality of community-acquired bacteremia in a Danish county as assessed from hospital registers.

When investigating infectious diseases caused by bacteria or fungi in the blood an important distinction has to be made between on one hand community-acquired microbes and on the other hand hospital acquired microbes. Bacteremia is the preferred term when bacteria (or fungi) are detected in the blood stream, and in this study we are concerned with the seasonality and long-term trend ofcommunity-acquired bacteremia, which for the time being is analyzedusing Poisson regression models. The seasonality is described using atrigonometric function whereas the trend is modelled using a thirddegree polynomial. The two main objects of analysis are the totalnumber of blood samples taken each week and the number of positiveblood samples encountered each week.Community-acquired bacteremia is an important health issue and this study, which is built on a unique database, is the first step to uncover determinants of disease patterns.In cooperation with Henrik C. Schønheyder, Department of ClinicalMicrobiology, Aalborg Hospital, Aarhus University Hospital.