Fluphenazine depot 1982.

Study characteristics
Methods	Randomisation: participants were matched for age, sex, duration of illness, and severity of symptoms in the preceding episode and then assigned based on a randomised schedule. Allocation: procedure not described. Blinding: double‐blind, evaluating psychiatrist and participants were unaware of the contents of their injections. It seems that the treating psychiatrist was aware of the treatment. Duration: 12 months. Design: parallel. Location: single‐centre. Setting: outpatient.
Participants	Diagnosis: schizophrenia (ICD‐9 and Present State Examination), with two or more episodes and several first rank symptoms in previous episode, free of psychopathology for at least 12 months, on fluphenazine decanoate for at least 2 years. N = 70. Gender: n.i.. Age: n.i.. History: duration stable‐ at least 12 months free of psychopathology, duration ill‐ n.i., number of previous hospitalisations‐ n.i., but at least two previous episodes, age at onset‐ n.i., severity of illness‐ BPRS < 10 in all participants, baseline antipsychotic dose‐ n.i..
Interventions	1. Drug: fluphenazine decanoate. Fixed dose of 50 mg IM four/eight weekly. N = 35. 2. Placebo: vitamin B complex IM. Duration of taper: 0 days. N = 35. Rescue medication: nitrazepam for sleep and benzhexol for extrapyramidal side‐effects; additional antipsychotic drugs were not allowed.
Outcomes	Examined Relapse (re‐emergence of definite schizophrenic psychopathology necessitating hospital admission or other major treatment change). Leaving the study early. Death. Adverse effects: tardive dyskinesia (Aquired Involuntary Movements Scale). Unable to use/Not included Mental state: BPRS (no mean, no SD/no predefined outcome of interest). Adverse effects: extrapyramidal symptoms ‐ use of antiparkinson medication (combined with nitrazepam), use of additional nitrazepam for sleep (combined with use of antiparkinson medication).
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Participants were matched for age, sex, duration of illness, and severity of symptoms in the preceding episode and then assigned based on a randomised schedule.
Allocation concealment (selection bias)	Unclear risk	Procedure not described.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Double‐blind, evaluating psychiatrist and participants were unaware of the contents of their injections. It seems that treating psychiatrist was aware of the treatment.
Blinding (performance bias and detection bias) Subjective outcomes	Unclear risk	Double‐blind, evaluating psychiatrist and participants were unaware of the contents of their injections. It seems that treating psychiatrist was aware of the treatment.
Blinding (performance bias and detection bias) Objective outcomes	Low risk	Double‐blind, evaluating psychiatrist and participants were unaware of the contents of their injections. It seems that treating psychiatrist was aware of the treatment.
Incomplete outcome data (attrition bias) All outcomes	High risk	Overall dropout rate drug 40% versus placebo 66%, most due to relapse. This poses a risk for bias for other outcomes. Completer analysis.
Selective reporting (reporting bias)	Low risk	No evidence for selective reporting.
Other bias	Low risk	No clear evidence for other sources of bias.