Shalev 1993.

Methods	Randomisation: “randomly assigned”. Blinding: “The treating psychiatrists were blind to the sequence” of drug administration. No further details. Duration: minimum 4 weeks (first phase of the trial up to the point of first cross‐over). Design: cross‐over study also investigating levomepromazine. Location: not indicated. Setting: inpatients.
Participants	Diagnosis: schizophrenia (DSM‐III). N = 39. Gender (including participants of the levomepromazine study group, N = 60): 35M, 25F. Age (N = 60): mean 33 years. History: duration stable: not indicated, duration of illness: not indicated, number of previous hospitalisations: mean 4.2, age at onset: not indicated, severity of illness: not indicated, baseline antipsychotic dose: not indicated.
Interventions	1. Haloperidol: fixed/flexible dose: not indicated, dose range: not indicated, mean dose: 29.3 mg/day. N = 18 (study completers). 2. Perphenazine: fixed/flexible dose: not indicated, dose range: not indicated, mean dose: 35.8 mg/day. N = 21 (study completers).   “Therapists were instructed…to reach average daily doses (defined as….32 mg/day of perphenazine and 20 mg/day of haloperidol) within one week and remain within 50% of that dose for another 3 weeks.”   “Neuroleptics were administered orally. In cases of severe agitation, intramuscular administration was allowed for no more than 2 days…..Rapid increase in doses during the first days (rapid neurolpetization) was strictly avoided.”   Rescue medication: “An anti‐parkinsonian drug (trihexyphenidyl, up to 10 mg per day) was used according to the patient’s condition.”
Outcomes	Examined: Clinically important response to treatment: Assessment of therapeutic success (“a decrease in the patients BPRS score of at least 30% [Psychometric criterion] and improvement of the patient’s clinical state to the point that allows the patient’s return to the community [clinical criterion]"). Unable to use: Mental state general: Brief Psychiatric Rating Scale (BPRS) (no raw data were provided for each study group separately).
Notes	“The minimal duration of the illness required for inclusion in this study was….6 months.” “Newly hospitalised acutely exacerbated schizophrenics” were included in this study. Three antipsychotics (haloperidol, perphenazine and levomepromazine) “were administered one after the other, for 4 weeks each, in randomly determined order.”
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	“randomly assigned”. No further details.
Allocation concealment (selection bias)	Unclear risk	No further details.
Blinding (performance bias and detection bias) (performance bias)	Low risk	“The treating psychiatrists were blind to the sequence in which the 3 drugs were to be given to the patient.”
Blinding (performance bias and detection bias) (detection bias)	Unclear risk	No further details.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	There were only data regarding the whole study sample (treated with haloperidol, perphenazine and levomepromazine) available: 15 of 75 participants (20%) left the trial early before the termination of the investigated first cross‐over‐phase. Thus the overall attrition can be considered as being moderate. Completers‐only analyses were used.
Selective reporting (reporting bias)	High risk	The outcome data were not fully addressed (in terms of the BPRS no raw data were provided for each study group separately). The adverse effects were not reported. Data regarding the number of participants receiving rescue medication with antiparkinson drugs were missing.
Other bias	High risk	Cross‐over study design. “The primary goal of this study was the evaluation of drug responsiveness in the natural clinical environment and not a comparison between drugs.” Trial duration was depended on the degree of response.