+ research support expressed emotions increase risk of relapse. Hogarty et al 1986 found that when relatives were trained to reduce EE, relapses decreased. Leff & Vaughn 1985 found that reducing EE reduced hospitalisation rates. however, these studies do not prove causation as correlation does not equal causation. It could be argued that those with higher EE are less able to cope with the illness so are more likely to need hospitalisation. Also, the training given to relatives was quite intensive and expensive so cannot be applied easily to all families. Therefore, generalisability is low.