Several North American bumblebee species have recently undergone dramatic declines. The use of managed, pathogen-carrying bumblebees for pollination of greenhouse crops began shortly before these declines, and wild bumblebees near greenhouses now have high pathogen loads. This has led to speculation that pathogen spillover from commercial bumblebees caused declines of these species. We test this hypothesis using a large dataset of bumblebee occurrence records and agricultural census data. We find support for the pathogen spillover hypothesis for two species but no evidence that pathogen spillover caused the near disappearance of the previously widespread Bombus affinis. Furthermore, we show that pesticide use and habitat loss are unlikely to be major causes of decline for any of the Bombus species examined. Collectively, our analyses underscore that there remains an urgent need to identify causes of pollinator population losses.