Almeida 2006.

Study characteristics
Methods	Study design: parallel design Number of arms: 2 Treatment arm: sertraline (SSRI) Control arm: matched placebo
Participants	Geographical location: Australia Setting: inpatient Stroke criteria: acute ischaemic or haemorrhagic stroke Method of stroke diagnosis: via clinical signs (ICD‐10) and CT (100% imaged, 10/111 CT scan did not show acute ischaemia) Time since stroke: stroke on average < 2 weeks prior to randomisation Inclusion criteria: not reported Exclusion criteria: 1) severe communication difficulties; 2) unstable medical condition; 3) severe cognitive impairment and depression; 4) taking antidepressants within 4 weeks of stroke; 5) contraindication to sertraline Depression criteria: HADS score > 8 Total number randomised in this trial: 111 Number randomised to treatment group: 55 (67% men, mean age 68 years, SD 13) Number randomised to control group: 56 (62% men, mean age 67 years, SD 13) Total number included in the final analysis: 99 Number included in treatment group for final analysis: 48 Number included in control group for final analysis: 51
Interventions	Treatment: sertraline (SSRI), 50 mg, daily (night) Control: matched placebo Treatment duration: treatment continued for 24 weeks Follow‐up: 28 weeks post treatment end (52 weeks from baseline)
Outcomes	Primary outcomes Depression measured using the HDRS Secondary outcomes Cognitive function measured using the MMSE Disability measured using the mRS
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Subjects were allocated to 24‐week treatment with placebo or sertraline (fixed daily dose of 50mg at night) according to a computer‐generated random list of numbers ..." pp. 1105
Allocation concealment (selection bias)	Low risk	Quote: "Placebo and sertraline were delivered in capsules that had the same size, shape, colour, smell, and weight" pp. 1105
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "Both the research team and participants were unaware of treatment allocation …" pp. 1105
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Quote: "We designed this randomized, double‐blind, placebo control trial ..." pp. 1105
Incomplete outcome data (attrition bias) All outcomes	High risk	Quote: "... last observation carried forward (LOCF) analyses to investigate the primary endpoint of interest for this study ..." pp. 1106 Comments: per protocol analysis presented. 11/111 were not included in the analysis
Selective reporting (reporting bias)	Unclear risk	Comments: no trial protocol available to compare with the publication
Other bias	Unclear risk	Comments: more participants in treatment group with previous heart attack and stroke, also higher levels of hypertension