In order to detect levels of pre-existing cross-reactive antibodies in different age groups and to measure age specific infection rates of the influenza A 2009 pandemic in Germany, we conducted a seroprevalence study based on samples from an ongoing representative nationwide interview and examination survey for adults that had started 6 months prior to the first registered case of influenza A 2009 in Germany. For the subsequent regression analysis,DMAT only three birth cohorts were used in order to increase the power of the analysis. As the observed titre values range over several orders of magnitude we logtransformed the response for variance stabilization. A special problem of the data is that a standard linear regression model for log does not apply, because a large proportion of the measurements are below the detection limit of 1:10, and hence are left-censored. Instead, we used a Tobit regression model for the analysis of log with a value just below log as left censoring limit. The effects of sampling period, age, sex, and vaccination on log could now be investigated. Model selection was performed using a manual stepwise forward selection procedure based on p-values from twosided likelihood ratio tests. Altogether,ARM390 a p-value below 0.05 was considered to be statistically significant. All statistical analyses were performed using the statistical software STATA version 11 and R version 2.12.0. By analysing samples from a representative nationwide health survey collected in the year preceding the start of the pandemic, we show that the level of pre-existing antibodies at titre $40 crossreacting with the pandemic influenza 2009 virus ranged between 2.3, and 12.5%, depending on age group. The highest proportions of cross-reactive antibodies before the pandemic were observed among 18–29 year olds. Our findings of higher titres of cross-reactive antibodies among young adults is in contrast to other studies showing higher levels of pre-pandemic cross-reactive antibodies among elderly. However, published findings vary markedly among different studies.