The objective is to systematically review previous literature on the diagnostic stability of schizophrenia, particularly to investigate prospective and retrospective consistency. We carried out a systematic literature search in PubMed and other minor sources from 1980 to July 2017. Specifically, prospective and retrospective consistency were examined. Thirty-nine studies were included, 5 focused on schizophrenia, 23 on psychotic episodes and 11 on psychiatric disorders in general. Samples sizes range from 60 to 10 058 subjects (total N = 39 965). The majority of studies (n = 26, 66.67%) were performed in Europe and North America and they had a prospective design (n = 27, 69.23%), with a median follow-up of 3 years. Prospective and retrospective consistency means were 84.29% and 67.15% respectively. Diagnostic change was also frequently measured (n = 12, mean 31.28%). The factors more commonly associated with diagnostic stability were: male sex, older age at the study inception, older age at onset, late stages of illness, family history of mental illness, poorer functioning and longer length of stay. Schizophrenia was found to have high diagnostic stability over time, although research on this topic is mainly focused in first psychotic episodes. More standardized methods are needed to further research diagnostic stability of schizophrenia over time and its determinants.