Organisms across a wide range of taxa use migration as a strategy to avoid, reduce or recover from parasitic infection. Previous work has identified three different processes by which migration can help reduce infection risk and/or costs: migratory escape from infection, migratory culling of infected individuals, and migratory recovery from infection. However, most theoretical modelling of host migration in response to infection assumes that individuals have a single strategy during both infected and susceptible states, meaning an individual’s state (susceptible or infected) is irrelevant to its decision to migrate. Here, we construct a model with two independent strategies of migration for an individual based on whether they are infected or susceptible, to study when the decision to migrate is favourable. We show that the best strategy for individuals of a given infection state is to either always migrate or never migrate. When infected and susceptible individuals differ in their migration strategy, this leads to partial migration at the population level (only some individuals migrate). Whereas previous theory showed that partial migration can be optimal in the face of infection,  our work further parses out the contributors to partial or complete migration by determining which individuals in the population are involved in migration for different sets of conditions.