Mayang 1990.

Methods	Allocation: randomised (no further details). Blind: not blind. Duration: 4 weeks. Setting: hospital, Michigan, USA.
Participants	Diagnosis: schizophrenia. N=18. Age: 20‐55 years. Sex: all female. History: hospitalised, recent history of physical or verbal aggression, frequently secluded or restrained because of aggressive behaviour, deficient in adaptive problem solving. Exclusion: not stated.
Interventions	1. Problem solving: weekly 45 min individual sessions with trained psychology graduate; sessions involved use of videos showing i. problem identification, ii. problem articulation, iii. formulation of solutions, iv. identification of best solution, v. evaluation of solution and translate into action, additional help from trainer in form of feedback and role play situations. N=6. 2. Interaction group: weekly 45 min individual sessions with trainer but receive no problem solving training. N=6. 3. Routine care: no additional sessions with trainer. N=6.
Outcomes	Mental state: problem solving ability improved/not improved (staff rated). Behaviour: improved/not improved (staff rated). Social interaction: improved/not improved (staff rated). Leaving the study early. Unable to use ‐ Mental state: problem solving ability score (scoring system used not peer reviewed). Behaviour: number of hours in seclusion (no SD).
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Unclear risk	B ‐ Unclear