Safren 2005.

Methods	Randomised controlled trial
Participants	Country: USA Setting: ambulatory Age: adults (18‐65 years old) Sample size: 31 Sex: 17 women, 14 men Inclusion criteria: have had a principal diagnosis of Attention‐Deficit Hyperactivity Disorder with external validation of childhood onset and clinical severity of at least a moderate level (Clinical Global Impression; CGI of 4 or above); have been able to give informed consent and comply with study procedures; and have been stabilised on medications for ADHD or related symptoms. Stabilisation on medications was defined as no more than 10% change in medication dose over a 2‐month period with clinical evidence of improvement compared to the patients' unmedicated status. Exclusion criteria: moderate to severe major depression; clinically significant panic disorder; organic mental disorders, psychotic spectrum disorders, bipolar disorders, active substance abuse or dependence (past three months), pervasive developmental disorder; active suicidal ideation; history of cognitive‐behavioral therapy (CBT); estimated or documented verbal intelligence quotient (IQ) of less than 90
Interventions	Intervention: CBT (12‐15 weekly sessions) + continued psychopharmacology (n = 16) Control: continued psychopharmacology alone (n = 15) Dosage, timing of dosage and administration of pharmacotherapy were not specified.
Outcomes	Primary outcome ADHD symptoms ‐ Current ADHD Symptoms Scale (self‐report and clinician rating versions) Secondary outcomes Functioning ‐ Clinical Global Impression Scale‐NIMH (CGI) Depression ‐ Hamilton Depression Scale (HAM‐D) Anxiety ‐ Hamilton Anxiety Scale (HAM‐A)
Notes	We contacted authors to get the information about random sequence generation and allocation concealment that we included in this table (Safren 2014 [pers comm]) Study start date: September 2001 Study end date: August 2003 Funding source: this study was supported by grant NIMH 60940 (Steven A Safren, PhD). Declarations of interest: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Comment: the study authors created a randomisation table in blocks of 2, stratified by severity (CGI scale) and sex. After the person was assessed and the team agreed that they met the inclusion/criteria, they were randomised based on the table.
Allocation concealment (selection bias)	Low risk	Comment: the interventionist was blinded to randomisation until the team had met and it was deemed that the person met criteria.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Comment: it is not possible to blind personnel in a psychosocial intervention.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "The baseline and outcome assessments consisted of a clinician‐administered interview by an evaluator who was blind to treatment condition, and a battery of self‐report measures".
Incomplete outcome data (attrition bias) All outcomes	Low risk	Comment: no dropouts
Selective reporting (reporting bias)	Low risk	Comment: the study protocol is not available, but it is clear that the published reports include all expected outcomes, including those that were prespecified.
Other bias	Low risk	Comment: there was no evidence of other bias.
Conflict of interest	Low risk	Comment: no evidence of conflicts of interest.