Banerji 1989.

Methods	Randomised controlled trial, 4 weeks
Participants	104 patients in general practice suffering from neurotic or reactive depression (for at least 2 weeks) entered the study, 3 were lost to follow‐up, 104 patients were included. An RDS score of at least 6 with a CAS score of no more than the RDS was required to be eligible for entry Patients who were withdrawn from the study before week 2 were not included in the analysis of data for therapeutic efficacy, but were evaluated for side effects. 21 patients failed to complete 2 weeks treatment ‐> 80 participants (40 each group) were evaluated for treatment Aged 18 to 70 years
Interventions	Alprazolam versus amitriptyline Maximum dose: alprazolam 3 mg and amitriptyline 150 mg Average daily dose alprazolam 1.8 mg and for amitriptyline 63.8 mg
Outcomes	Primary outcome: Mean HDRS after 4 weeks: alprazolam 9.2 (initial 21.9) and amitriptyline 5.9 (initial 21.1) 50% reduction of baseline HDRS score: alprazolam 26/40 (72%), 32/40 amitriptyline (84%) Secondary outcome: All‐cause withdrawals: alprazolam 10/51 (20%), amitriptyline 13/53 (25%) Withdrawals due to side effects: alprazolam 6/51 (12%), amitriptyline 11/53 (21)% Physicians' evaluation of no side effects: alprazolam 36%, amitriptyline 54% Side effects: insomnia alprazolam 22%, amitriptyline 6%, headache alprazolam 20%, amitriptyline 4%, dry mouth 30%, amitriptyline 65%
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"Randomised design". No information provided
Allocation concealment (selection bias)	Unclear risk	No information provided
Blinding (performance bias and detection bias) Patients	Low risk	"The drugs were presented in identical capsules." The drugs looked the same, but there appears to have been a difference in side effect profile.
Blinding (performance bias and detection bias) Physicians	Unclear risk	"The drugs were presented in identical capsules." No further information provided; unclear how physicians administered the drugs, while the tablets looked the same.
Blinding (performance bias and detection bias) Investigators	Unclear risk	"The drugs were presented in identical capsules"
Incomplete outcome data (attrition bias) HDRS	High risk	20% of the included patients were withdrawn from the study and were not evaluated for treatment
Incomplete outcome data (attrition bias) Withdrawals/side effects	Unclear risk	“...21 patients failed to complete two weeks treatment with the investigational drug (seven protocol violations and 14 withdrawals)” From table: 7 protocol violations: 5 alprazolam and 2 amitriptyline, 13 withdrawals due to side effects: 4 alprazolam and 9 amitriptyline and 1 withdrawal due to feeling improved: alprazolam
Selective reporting (reporting bias)	Unclear risk	No information provided
Other bias	Unclear risk	Low therapeutic dosage of amitriptyline in comparison to that of alprazolam
Placebo washout	Low risk	No information provided