Bouchard 1987.

Methods	Six‐week multicentre, controlled, double‐blind trial.
Participants	Patients who suffered from a depression which required drug treatment and which was of a severity corresponding to a total score of at least 15 on the Montgomery and Asberg Depression Rating Scale (MADRS) after a wash‐out period of 3‐7 days. The depression was classified as endogenous, doubtfully endogenous or non‐endogenous, using the Newcastle rating scale and the DSM‐III, as belonging to one of the following groups: major depressive episode with melancholia, major depressive episode without melancholia, atypical depression. Exclusion criteria: pregnancy or absence of use of an effective contraceptive methods, severe somatic disease (particularly severe cardiac, renal or hepatic disease), organic brain syndrome, schizophrenia or paranoid psychosis, epilepsy, abuse of alcohol or narcotics, treatment with MAO‐inhibitors within the last 3 weeks preceding entry into the trial, previous unsuccessful treatment with one of the test drugs, patient's refusal to participate in the trial.
Interventions	Citalopram: 48 participants Maprotiline: 48 participants Citalopram dose range: 40‐60 mg/day Maprotiline dose range: 75‐150 mg/day Among psychotropic drugs, only benzodiazepines were allowed.
Outcomes	Primary outcome: mean score on MADRS or Clinical Global Impression (CGI).
Notes	This study was funded Lundbeck (citalopram manufacturer). One death for suicide in the citalopram group.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "the patients were allocated at random in blocks of four to double‐blind treatment with either citalopram or maprotiline once daily for a period of 6 weeks". Probably done.
Allocation concealment (selection bias)	Unclear risk	No data provided
Blinding (performance bias and detection bias) All outcomes	Low risk	Quote: "Using the double‐ dummy half patients received active citalopram tablets and placebo maprotiline tablets and the other half was given placebo citalopram tablets and active maprotiline tablets". Probably done.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	No clear data provided
Selective reporting (reporting bias)	High risk	CGI‐S score at baseline are missing. Remission rate are reported only at endpoint (week 1‐4 are missing).
Other bias	Unclear risk	Sponsorship bias cannot be ruled out.