Aim: To identify predictors for developing a higher suicidal tendency during treatment of first-episode psychosis. Methods: In a prospective follow-up study, we examined clinical factors collected at treatment initiation as predictors for developing a higher suicidal tendency among patients in the first year of treatment of psychosis. Patients were grouped and ranked according to their highest suicidal tendency in the year before treatment: not suicidal, suicidal thoughts, suicidal plans or suicide attempt(s). Predictors for becoming more suicidal in the first year of treatment were examined on group level in multivariate logistic regression analyses. We assessed patients' suicidal tendency and clinical factors using validated interviews and rating scales. Analysis included 386 patients. Results: Among patients with some suicidal tendency a one-point increase score on hallucinations significantly predicted developing a higher suicidal tendency, whereas a one-point increase score on delusions was preventive of this. Feeling hopeless was highly associated with suicide attempt in those with earlier suicide attempt. Conclusion: The risk of suicide attempt did not differ between patient groups with suicidal thoughts, plans or suicide attempt. In first year of treatment of psychosis, hallucinations increased the risk for becoming more suicidal, whereas delusions reduced this risk in already suicidal patients.