O'Connor 2007.

Methods	Allocation: random. Blinding: unblinded. Duration: 24 weeks.
Participants	Diagnosis: DSM‐IV delusional disorder, confirmed by an independent rater using the Structured Clinical Interview for DSM axis 1 disorders. N = 24, 17 completed. Age: mean ˜38 (completers). Sex: 8 women, 9 men (completers). History: referred by specialists in psychotic disorders, had stable symptoms for 2 months, most in receipt of psychiatric medication (not altered once trial began). It is not explicit if any hospitalised patients were included, but this can be inferred from the stability criterion that all patients were outpatients.
Interventions	1. Cognitive behavioural therapy + standard medication care*: Quote: "The CBT consisted of individualized weekly meetings [for 24 weeks] with 1 of 3 licensed psychologists specialized in CBT for DD. The CBT program was based on programs reported in the literature and followed the main stages of preparation, cognitive challenge, and reality testing." It was manualised for the therapist N = 12. 2. Attention 'placebo' + standard medication care*: Quote: "The APC consisted of individualized weekly meetings with 1 of 3 licensed psychologists specialized in CBT for DD. In the APC treatment, the therapist and patient discussed any immediate problems and recurrent themes in a nondirective and supportive manner, encapsulating the proper supportive psychotherapeutic approach to the paranoia patient of interested, attentive, relaxed, and unaffected attitude with an unfeigned air of detachment and suspended judgment, which has been shown to lead to some remission of symptoms.'' This was manualised for the therapist N = 12.
Outcomes	Social function ‐ self worth: Social Self‐Esteem Inventory. Leaving the study early. Unable to use: Mental state (BAI, BDI, MADS): data were skewed. Insight: authors did not fully report BABS.
Notes	* 15 of 17 completers were already taking medication; 11 of the completers were taking 1 of 3 types of antipsychotic medication (risperidone, olanzapine, quetiapine) and the other 6 were taking antidepressants or benzodiazepines.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Random allocation by consecutive referral at point of entry into the study per published article." We have established (via personal correspondence with the study's lead author) that random allocation occurred at the point of entry into the trial of consecutive patients through the use of a random number table (Fisher & Yates).
Allocation concealment (selection bias)	Unclear risk	Not specified in the study.
Blinding of participants and personnel (performance bias) All outcomes	High risk	The trial was unblinded.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Quote: "The main outcome measure was the MADS rating; this test was administered by an evaluator independent of the study." There is no mention that blinding was assured.
Incomplete outcome data (attrition bias) All outcomes	High risk	People who withdrew early were not accounted for in the final data, only completers. The study does acknowledge the attrition rate.
Selective reporting (reporting bias)	Unclear risk	There is some unevenness in selective reporting of positive findings, with less prominence given to less positive findings, although the study does acknowledge some weaknesses overall.
Other bias	Low risk	No other biases noted.

BABS: The Brown Assessment of Belief scale
 BAI: Beck Anxiety Inventory
 BDI: Beck Depression Inventory
 DSM‐IV: Diagnostic Statistical Manual IV
 MADS: Maudsley Assessment of Delusions Schedule