Amphetamines and cannabis are the most commonly used illicit substances in Australia and are associated with a range of severe mental health problems, including psychosis. While a number of studies have previously examined the symptom profiles of amphetamine users and cannabis users with psychosis, it remains unclear whether these patients experience differences in the severity and clinical course of their symptoms and behaviour compared to non-using patients with psychosis. This thesis examined the demographics, psychiatric and family history, premorbid adjustment, clinical symptoms, and disturbed behaviours of 98 inpatients admitted to hospital with an early psychosis. The TimeLine Follow Back method was used to determine substance use in the 30 days prior to admission, allowing participants to be categorised according to the following drug classes: amphetamines, cannabis, amphetamines + cannabis, none. Participants were assessed at admission and then weekly until discharge (or for a maximum of eight weeks) with the Brief Psychiatric Rating Scale (positive symptoms, negative symptoms, mania symptoms, depression-anxiety symptoms) and the Disturbed Behaviour Rating Scale. Multi-level modelling (MLM) was used to determine statistically significant differences between each of the substance-using groups and their respective non-using groups on the severity of their symptoms and behaviour at admission and over the course of their hospitalisation. Importantly, MLM also allowed for these groups to be compared on the rate of symptom and behaviour change during the first eight weeks of hospitalisation, providing evidence of any differences in clinical course.