Shawver 1959.

Methods	Allocation: randomly assigned, pairs matched on the morbidity score and ward location (randomised block design). Blinding: double, identical capsule. Duration: 6 months.
Participants	Diagnosis: schizophrenia. N=120 (80 to relevant interventions). Age: < 50 yrs. Sex: unknown. History: in hospital mean 8 yrs, on chlorpromazine >6 months. Excluded: leucotomy or organic conditions, ECT for the past year. Setting: hospital.
Interventions	1. Chlorpromazine cessation: placebo. N=40. 2. Chlorpromazine continuation: dose 200 mg/day. N= 40. 3. Reserpine continuation: dose 2 mg/day. N=40.
Outcomes	Relapse. Global state: MSRPP (morbidity score). Unable to use ‐ Behaviour change: AAMI (no usable data).
Notes	A team of two including the ward psychiatric made a joint interview to increase the reliability of the rating. AAMI consisted of five separated measures: Level of activity, level of anxiety, mental disorganization, interpersonal relationships and abnormality of activity. No data about whether relapses occured during the short term or medium term. All considered medium term.
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Unclear risk	B ‐ Unclear