Marshall 2007.

Study characteristics
Methods	NEW TRIAL ADDED TO UPDATED REVIEW Design: multi‐centre trial, randomised, placebo‐controlled, parallel arm, flexible‐dose, double‐blind and maintenance study Duration of intervention: 10 weeks acute, 12 weeks maintenance Placebo run‐in: 1 week of single‐blind placebo Post‐treatment: Not specified
Participants	Setting: New York Sample size: 52 randomised to paroxetine and placebo Mean age: 39.8 years (SD: 11.2) Sex: 17 men and 35 women, no combat‐related trauma, 62.5% (30/48) MDD Diagnostic measure: DSM‐IV, SCID Inclusion criteria: Quote: “Subjects were adults between 18 and 65 years of age, who met DSM‐IV criteria for chronic PTSD (duration at least 3 months) according to the Structured Clinical Interview for DSM‐IV and who provided written informed consent after the study procedures were explained. Comorbid affective, anxiety, and personality disorders were allowed provided that PTSD was the primary diagnosis in need of clinical intervention. Maintenance phase. We entered subjects rated as much improved or very much improved, on the Clinical Global Impression Improvement scale (CGI‐I score of 2 or 1, respectively) at week 10 into a 12‐week maintenance phase with continuation of double‐blind treatment and independent evaluations every 4 weeks, to examine whether clinical gains were maintained and/or continued to accrue" Exclusion criteria: Quote: “Concomitant psychoactive medications and concomitant trauma‐focused psychotherapy were not allowed. Patients were excluded if they had previously been treated with paroxetine 20 mg or more for at least 4 weeks; if they manifested psychotic symptoms or serious suicidal ideation; if they met criteria for substance abuse disorders within the past 3 months or lifetime criteria for schizophrenia, schizoaffective disorder, or bipolar disorder; if they had been successfully treated in the past with an antidepressant for major depressive disorder; or if they had unstable medical illness or significant abnormalities on laboratory or electrocardiogram (ECG) examinations. Prior to entering the trial, patients were medication‐free for 2 weeks if previously taking any psychoactive medication, or for 4 weeks if previously taking fluoxetine or a monoamine oxidase inhibitor" Comorbidity: SCID: social phobia, MDD, panic disorder, personality disorder Dropout rates: 12/52 (8/25 in the paroxetine and 4/27 in the placebo group)
Interventions	Pharmacological intervention: Quote: “Non responders to placebo were randomized to paroxetine–placebo initiated at 10 mg daily. If patients remained symptomatic, dosage was increased to 20 mg for weeks 2 and 3, 40mg daily for weeks 4–6, and 60 mg daily for weeks 7–10 as tolerated"
Outcomes	Primary outcomes: CGI‐I, CAPS Secondary outcomes: CAPS‐2, IIP, DES, HAM‐A, HAM‐D‐24, and items 22 (social functioning) and 23 (occupational or important role functioning) of the CAPS‐2 Time points: Quote: "Primary outcome was assessed using x2 analysis of the CGI‐I scale at weeks 0, 5, and 10. The treating physician assessed adverse effects at each visit using an adverse effects checklist" Data estimation: ITT and observed cases (i.e. mixed effects model)
Notes	Dates of trial: Not stated Industry‐funded: Yes. Quote: "National Institutes of Mental Health; Contract grant number: MH‐01412; Contract grant sponsor: Glaxo SmithKline (investigator‐initiated grant)" Medication provided by industry: No information Any of the authors work for industry: No information
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	It was reported that the patients were randomised however no mention is made of the method of randomisation Quote: "Those not rated as significantly improved were then randomly assigned to placebo (N527) or paroxetine (N525) for 10 weeks, with a flexible dosage design (maximum 60mg by week 7)"
Allocation concealment (selection bias)	Unclear risk	There was insufficient information provided to determine if study medication allocation was concealed
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	It is not clear whether the participants were blinded, only the personnel Quote: "Treating physicians and IE staff were kept blind to randomized assignment throughout the course of the study"
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "Outcome was assessed by independent evaluators blind to treatment status or side‐effect profile. Maintenance phase. We entered subjects rated as much improved or very much improved, on the Clinical Global Impression ‐ Improvement scale (CGI‐I score of 2 or 1, respectively) at week 10 into a 12‐week maintenance phase with continuation of double‐blind treatment and independent evaluations every 4 weeks, to examine whether clinical gains were maintained and/or continued to accrue"
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	More participants withdrew from the paroxetine group (8/25, 32%) compared to the placebo group (4/27; 15%). Reasons for withdrawal were not provided. ITT and observed cases were used for the estimation of data. Quote: "Patient demographics (age, gender, ethnicity, language status, education, and income) did not differ significantly between the paroxetine and placebo groups"
Selective reporting (reporting bias)	Unclear risk	The study protocol was not available for this study
Other bias	Unclear risk	Quote: "Contract grant sponsor: Glaxo SmithKline (investigator‐initiated grant)"