Objective The aim of this study was to explore the use of first (FGAs) and second generation antipsychotics (SGAs) in older outpatients with schizophrenia and schizoaffective disorder. Factors associated with schizophrenic relapses were also studied. Methods The study sample consisting of 8792 patients aged 64 years or more was collected from Finnish nationwide registers. The register data on the use of FGAs and SGAs were followed up between 1998 and 2003. Factors associated with psychiatric hospitalization in 1999 indicating relapse were studied using logistic regression analysis. Results The use of SGAs increased from 2.8% to 12.4%, and the use of FGAs decreased from 57.5% to 39.4%. The use of a combination of SGAs and FGAs increased from 4.0% to 8.5%. The proportion of those who did not buy any antipsychotics varied between 35.8% and 39.7%. The number of patients hospitalized on psychiatric wards within a year (1999; relapsed) was 8.8%. Factors independently associated with relapse were use of combined FGAs and SGAs [odds ratio (OR) 1.70, p = 0.001] and use of antidepressants (OR 1.27, p = 0.019). Diagnosis of cardiovascular disease was negatively associated with risk of schizophrenic relapse (OR 0.84, p = 0.040). Conclusion The use of SGAs increased while the use of FGAs decreased in older outpatients with schizophrenia. Almost 40% of the study sample did not use any antipsychotic medication. The 1‐year relapse rate was 8.8%. Several factors, such as combined use of FGAs and SGAs, or antidepressants, were associated with schizophrenic relapse, whereas cardiovascular disease showed a negative association with the relapse. Copyright © 2013 John Wiley & Sons, Ltd.