Changes in extreme precipitation are expected to be one of the most important impacts of climate change in cities. Urban floods are mainly caused by short duration extreme events. Hence, robust information on changes in extreme precipitation at high-temporal resolution is required for the design of climate change adaptation measures. However, the quantification of these changes is challenging and subject to numerous uncertainties.This study assesses the changes and uncertainties in extreme precipitation at hourly scale over Denmark. It explores three statistical downscaling approaches: a delta change method for extreme events, a weather generator combined with a disaggregation method and a climate analogue method. All three methods rely on different assumptions and use different outputs from the regional climate models (RCMs). The results of the three methods point towards an increase in extreme precipitation but the magnitude of the change varies depending on the RCM used and the spatial location. In general, a similar mean change is obtained for the three methods. This adds confidence in the results as each method uses different information from the RCMs. The results of this study highlight the need of using a range of statistical downscaling methods as well as RCMs to assess changes in extreme precipitation.