Robinson 2000a.

Study characteristics
Methods	Study design: parallel design Number of arms: 2 Experimental arm: fluoxetine (SSRI) Control arm: matched placebo
Participants	Geographical location: USA and Argentina Setting: mixed Stroke criteria: all subtypes Method of stroke diagnosis: via clinical signs and CT (100%) Time since stroke: on average 8 weeks (treatment) and 5 weeks (control) prior to randomisation Inclusion criteria: 1) stroke within 6 months of recruitment; 2) 18 to 85 years of age Exclusion criteria: 1) other significant medical illness; 2) severe comprehension deficit; 3) prior history of head injury; 4) prior history of other brain disease (with the exception of stroke); 5) participants on antidepressants (other than fluoxetine) were allowed to stop their antidepressants for a 2‐week washout period Total number randomised in this study: 25 Number randomised to treatment group: 17 (88% men, mean age 66 years, SD 13) Number randomised to control group: 8* (75% men, mean age 67 years, SD 9) Total number included in final analysis: unclear Number included in treatment group for final analysis: unclear Number included in control group for final analysis: unclear
Interventions	Treatment: fluoxetine (SSRI),10 mg daily (3 weeks), 20 mg daily (3 weeks), 30 mg daily (3 weeks), 40 mg daily (3 weeks) Control: matched placebo Treatment duration: treatment continued for 12 weeks Follow‐up: none
Outcomes	Primary outcomes Depression measured using the HDRS Secondary outcomes Cognitive function measured using the MMSE Activities of daily living measured using the Johns Hopkins Functioning Inventory Adverse events Death
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "All patients were randomly assigned to either active or placebo medication." pp. 352 Comments: method of sequence generation not reported.
Allocation concealment (selection bias)	Unclear risk	Comments: method of allocation concealment not reported.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "To maintain the blind, doses were decreased for equal numbers of placebo patients. The active and placebo pills were identical." pp. 353
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Comments: blinding of outcome assessment not reported. However, the study authors state that this trial is 'double‐blind'
Incomplete outcome data (attrition bias) All outcomes	High risk	Comments: Table 1 presents ITT analysis but per protocol analyses reported. The number excluded from analysis varies
Selective reporting (reporting bias)	Unclear risk	Comments: no trial protocol available to compare with the publication
Other bias	Low risk	Comments: there was no statistically significant difference between the treatment and control group in baseline demographic characteristics