Roohi‐Azizi 2017.

Study characteristics
Methods	Randomised, double‐blind, placebo‐controlled, parallel group trial
Participants	Diagnosis: DSM‐IV‐TR major depressive disorder N: 54 Age: citicoline group M = 36.0 (SD = 8.11); placebo group M = 35.8 (SD = 11.10) Sex: citicoline group 64% female; placebo group 76% female Baseline depression severity: citicoline group HDRS score M = 24.76 (SD = 4.40); placebo group HDRS score M = 24.40 (SD = 4.24)
Interventions	Patients were randomised to receive either citicoline (100 mg) or placebo every 12 hours for 6 weeks. Both groups also received citalopram (20 mg/day first week; 40 mg/day subsequent 5 weeks).
Outcomes	HDRS Early improvement ((≥20% reduction in HRDS score within the first 2 weeks) Response to treatment (≥50% reduction in the HDRS score) Remission rate (HDRS score ≤ 7) Time needed to respond to treatment Adverse effects
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Randomization was conducted by using a computerized random number generator (blocks of 4, allocation ratio of 1:1)."
Allocation concealment (selection bias)	Low risk	Quote: "An independent party who was not involved elsewhere in the study was responsible for the generation of randomization codes. Concealment of allocation was performed using sequentially numbered, sealed, opaque, and stapled envelopes. Separate individuals were responsible for randomization and treatment allocation, as well as rating."
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Stated but not tested. Quote: "The participants, research investigators, raters, and the statistician were all blinded to treatment allocation".
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Stated but not tested. Quote: "The participants, research investigators, raters, and the statistician were all blinded to treatment allocation.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Diagram of patient numbers and reasons for withdrawal included in figure 1 (page 2). Patient withdrawals evenly spread across groups.
Selective reporting (reporting bias)	Low risk	Trial registration available online (IRCT201502191556N74). All outcomes reported.
Other bias	Low risk	None identified.