Carey 2012.

Study characteristics
Methods	NEW TRIAL ADDED TO UPDATED REVIEW Design: multi‐centre trial, randomised, placebo‐controlled, parallel, flexible dose, double‐blind Duration of intervention: 8 weeks Placebo run‐in: 1 week of single‐blind placebo run‐in Post‐treatment: Not specified
Participants	Setting: Primary care clinics, South Africa. Sample size: 28 randomised to olanzapine and placebo Mean age: 40.75 years Sex: 11 men and 17 women Diagnostic measure: DSM‐IV and MINI Inclusion criteria: Quote: “Men and women aged 18 years and over with DSM‐IV, non‐combat, chronic PTSD (PTSD symptoms of at least 3 months) were eligible to participate in the study if they (i) had a minimum score of 50 on the Clinician Administered PTSD Scale (CAPS), (ii) were willing to provide written informed consent to their participation, and (iii) were free of disallowed psychotropic medication for a washout period of 5 days for all except fluoxetine (5 weeks) prior to randomization. Trauma types reflect the profile of trauma in South Africa (i.e. domestic violence and criminal violence". Exclusion criteria: Quote: “Exclusion criteria included current major depressive disorder, a Montgomery Äsberg Depression Rating Scale (MADRS) score ≥20 at baseline, and the presence of a significant suicide risk according to the clinical judgement of the investigator. Additional exclusion criteria included a substance use disorder within 6 months of randomization, a positive urine drug screen for illicit substances, a history of severe personality disorder (based on clinician judgement), a lifetime history of schizophrenia or other psychotic disorder, pregnant or breastfeeding women, women of child‐bearing potential not willing to use contraception, an unstable medical condition, and unresolved clinically significant laboratory or electrocardiogram findings. Previous failure to respond to or intolerance of a second generation antipsychotic (SGA), failure of two or more trials of an SSRI or an SNRI given in adequate doses for an adequate duration, and initiation or change in psychotherapy within 8 weeks of screening receipt of electroconvulsive therapy in the 3 months before screening, participation in a clinical trial in the 6 months before screening, and finally an improvement of 2 or more points on the Clinical Global Impressions (CGI) severity score from screening to randomization were also criteria for exclusion from participation”. Comorbidity: MADRS (but participants with comorbid severe depression were excluded), DTS, SDS Dropouts: 10/34 (insufficient information to determine dropout rates for the 2 groups separately)
Interventions	Pharmacological intervention: Quote: “Following randomization, participants received 5mg olanzapine or matching placebo for 1week. This was followed by 7.5mg for 1week and then 10 mg for 2 weeks. If the response was satisfactory (CG I≤2) and treatment was well tolerated at the end of week four, participants were maintained at that dose. However, if response was not satisfactory (CGI ≥3) but adequately tolerated, olanzapine/placebo was increased in 2.5mg increments two weekly to a maximum dose of 15 mg. In cases of poor tolerability following a dose increase, a single dose reduction of 2.5 mg was permitted".
Outcomes	Outcomes: CAPS, CGI, MADRS, DTS, SDS (not clear which outcomes are secondary or primary) Time points: Quote: “Severity of PTSD symptoms was assessed at baseline, week 4, and week 8 using the CAPS. Participants were assessed by clinicians at baseline, then weekly for the first 4weeks and every 2 weeks thereafter, with these symptom measures, and with open‐ended questions regarding tolerability". Data estimation: ITT and LOCF
Notes	Dates of trial: Not stated. Industry‐funded: Yes. Quote: "Funding for this study was received from Eli Lilly" Medication provided by industry: Yes. Quote: "Matching olanzapine/placebo was supplied by Eli Lilly and then packaged for individual participants by an independent contract pharmacist". Any of the authors work for industry: No
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Randomization codes were kept only by the contracted pharmacist. Following successful screening, eligible participants were randomly assigned to receive either olanzapine or placebo using a computer generated 10‐number block randomization schedule".
Allocation concealment (selection bias)	Low risk	Quote: "Matching olanzapine/placebo was supplied by Eli Lilly and then packaged for individual participants by an independent contract pharmacist".
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Information on the blinding of participants was provided; nothing on whether personnel were blinded, however. Quote: "There was no instance of unblinding until after study closure and after data entry had been completed".
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	The study was described as "double‐blind", although no information was provided on blinding of outcome assessors
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	There was insufficient information to determine dropout rates for the 2 groups separately. Reasons for treatment withdrawal were provided. No information was provided on sample characteristics at endpoint. Data estimation was based on ITT and LOCF values. Quote: "Of the 10 early withdrawals, four were lost to follow‐up and one participant was withdrawn because of poor/erratic compliance at week 2. Of the remaining five participants, four were withdrawn because of lack of treatment efficacy or withdrawal of consent, and one was withdrawn at week 1 because of severe sedation (olanzapine group). No participants were withdrawn because of a serious adverse event. By using an intent‐to‐treat sample, defined as all participants with at least one post‐baseline CAPS assessment and a last observation carried forward (LOCF) approach, efficacy for olanzapine versus placebo was tested using repeated measures analysis of variance (RMANOVA)".
Selective reporting (reporting bias)	Unclear risk	The study protocol was not available for this study
Other bias	Unclear risk	Funding for study provided by industry. Quote: "Funding for this study was received from Eli Lilly"