Targeted intervention in the early treatment of enduring positive and negative symptoms in first-episode psychosis is contingent on adequate case-identification procedures. Six methods of determining enduring negative symptoms in a sample of 238 individuals with first-episode psychosis who were assessed at three time points over a 12-month period are described. Results were examined at each of the seven combinations of follow-up points, revealing a highly fluid pattern. The percentage of the sample achieving caseness varied dramatically, depending on the method used. Compared with non-cases, the enduring negative symptom group had significantly lower scores on quality of life, poorer premorbid adjustment, and longer durations of prodrome and untreated psychosis. Notwithstanding the 'ebb and flow' of negative symptoms in first-episode psychosis, the results suggest that early identification of enduring negative symptoms is a feasible goal.