Zoonotic diseases are likely to be affected by climate variations but to reveal such dependence long time series extending over several decades are required. Human cases of Hantavirus infection in Northern Sweden shows regular peaks that occurs every fourth year and the virus is associated with the bank vole (Clethrionomys glareolus). From a perspective of climate change it is important to analyse variations in bank vole population size and human disease incidence and how these interactions is affected by climate forcing. Here I show that a time lag of two years exist between NAO index and response in the vole population and human disease incidence. In some way triggers the NAO index two years previously a build up and peak in vole density and with corresponding peak in human Hantavirus disease. Likewise, high NAO index two years previously is associated with low population density of voles. The change in number of human disease cases is positively correlated to density of bank voles and both bank voles and disease peaks at a regularity of about every fourth year. Further, the time series revealed a match between NAO index, vole population and human disease cases at a longer periodicity of 8.5 years that has not been anticipated.These contrasting climate effects makes it hard to predict how a future warmer climate will affect disease dynamics in this particular system because of complex relationship between long and short term effects.