Abstract
Chronic exposure to physical, chemical, and biotic stressors can make animals more susceptible to infections. Such stress-induced susceptibility is widely expected to make disease-and epidemics-more likely and more severe. However, whether the impacts of stressors on individuals scale up to population-level outcomes is uncertain, both theoretically and empirically. We initiated ranavirus epidemics in replicate mesocosm populations of larval wood frogs (Lithobates sylvaticus) exposed to conditions known to impact their individual susceptibility to ranavirus infections: low and high salinity crossed factorially with ambient or elevated temperatures. Contrary to expectations, epidemics were no more likely or more severe in stressful conditions, although increased temperatures did speed their progression. We discuss several potential reasons why the effects of stressors did not scale up to epidemics, but our results suggest caution in assuming the individual-level effects of even well-studied stressors necessarily scale simply to population-level outcomes.