Roy‐Byrne 2010.

Methods	Study design: Randomised controlled trial
Participants	Setting: Primary care Diagnosis: Meeting DSM‐IV criteria for one or more of panic disorder, generalised anxiety disorder, social anxiety disorder, or post‐traumatic stress disorder based on the Mini International Neuropsychiatric Interview and scoring at least 8 (moderate anxiety symptoms on a scale ranging from 0–20) on the Overall Anxiety Severity and Impairment Scale (OASIS) Inclusion criteria: 18–75 years Exclusion criteria: Persons unlikely to benefit from the Coordinated Anxiety Learning and Management (i.e. unstable medical conditions, marked cognitive impairment, active suicidal intent or plan, psychosis, bipolar I disorder, substance abuse of dependence except for alcohol and marijuana abuse), receiving ongoing CBT or medication from a psychiatrist, unable to speak English or Spanish Age: Mean 43.5 (SD 13.4) years Gender: 71% female Ethnicity: 57% white Country: United States Sample size (randomised): Total participants 1004, intervention 503, control 501
Interventions	Intervention: Stepped Co‐ordinated Anxiety Learning and Management (CALM) Contains the four elements of collaborative care: 1) a multi‐professional approach to patient care: Primary care provider (PCP), social workers, nurses, psychologists (CM), psychologist/psychiatrist (MH specialist) 2) a structured management plan: Patient choice of CBT, medication or both during 10‐12 weeks. The computerised CBT programme included 5 generic modules (education, self‐monitoring, hierarchy development, breathing training, and relapse prevention) and 3 modules (cognitive restructuring and exposure to internal and external stimuli) tailored to each anxiety disorder and included psycho‐educational materials and instructions for skills practice and exposure. CMs entered data which then created a personalised workbook and homework assignments. The medication algorithm emphasised selective serotonin reuptake inhibitors (SSRIs) or serotonin and norepinephrine reuptake inhibitors (SNRIs), dose optimisation, adverse effect monitoring, followed by second and third step combinations of 2 ADs or an AD and benzodiazepine for non‐response. CMs provided adherence monitoring, counselling to avoid alcohol and optimise sleep hygiene and behavioural activity. Non‐responders could receive more of the same (stepping up) or the alternative modality (stepping over) for up to 3 more steps of treatment. After treatment completion, patients were entered into continued care and received monthly follow‐up telephone calls to reinforce CBT skills, medication adherence, or both. If symptoms re‐emerged within the first 9 months patients were referred back a step 3) scheduled patient follow‐ups: computerisedCBT=6 to 8 weekly sessions in 3 months, maintenance =monthly follow up calls. 4) enhanced inter‐professional communication: CMs relayed medication suggestions from MH specialist to the PCP. CMs interacted regularly with the PCP both face‐to‐face and via written communication. MH specialist provided PCPs with a medication algorithm and as needed consultation by telephone or email. CMs had weekly supervision with MH specialist plus cross‐site monthly conference supervision calls Control: Treatment as usual
Outcomes	Anxiety (BSI‐12): 6, 12, 18 months Medication use: 6, 12, 18 months Quality of Life (mental and physical health): 6, 12, 18 months Satisfaction: 6, 12, 18 months
Notes	CBT: cognitive behaviour therapy; CM: case manager; DSM‐IV: Diagnostic and Statistical Manual fourth edition; MH: mental health; PCP: primary care provider
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer generated in blocks
Allocation concealment (selection bias)	Low risk	Block size was masked to all clinical site study members
Incomplete outcome data (attrition bias) All outcomes	Low risk	Short‐term loss to follow‐up based on primary anxiety outcome (BSI‐12 response) was: overall 128/1004 (13%), 57/503 (11%) intervention and 71/501 (14%) control. Reasons for loss to follow‐up provided, with similar reasons across groups. Used intention‐to‐treat analysis
Selective reporting (reporting bias)	Low risk	Protocol available and all prespecified outcomes reported
Other bias	Unclear risk	Insufficient information available to assess
Implementation Integrity	Low risk	Attempts were made to assess implementation integrity (e.g. direct observation or rating of tapes)
Blinding of participants and personnel (performance bias) All outcomes	High risk	Participants and personnel could not be blinded, outcome likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Assessor was not aware of treatment allocation