Perez Stable 2000.

Study characteristics
Methods	Single site study. Participants 'randomly assigned ... using a factorial design and a computer generated list of random numbers'. Propranolol and placebo groups matched for sex, race, annual household income, smoking status, alcohol intake, age, educations, body weight, BP and heart rate. Double blinded study. Duration: 1 year.
Participants	Geographic region: California, USA. Study setting: 'Potential participants were recruited from the community by mailings and by public service announcements and from patients seen at UCSF Medical Center'. Epoch: unclear. Age criteria: 18 to 59 years. Blood pressure entry criteria: average DBP 90 to 104 mmHg of three readings at two visits two weeks apart.
Interventions	Participants assigned to 'Lifestyle Focus Group' + placebo, 'Lifestyle Focus Group' + propranolol, placebo alone or propranolol alone. Initial dose 40 mg daily for three days then increased to 80 mg daily. 'Subsequent dosages were modified according to DBP at each of nine follow‐up visits'. Maximum study dose 400 mg propranolol daily in divided doses. Cognitive outcomes measured at baseline, three months and 12 months.
Outcomes	Analysed in this review: Stimulus Evaluation/Response Selection; Continuous Performance Task; Digit Symbol Substitution Test; California Verbal Learning Test Not analysed in this review: Center for Epidemiological Studies Depression Scale; Beck Depression Inventory; sexual function questionnaire
Notes	Exclusion criteria: insulin, bronchodilator or antidepressant treatment; antihypertensive treatment within the previous month; coronary artery disease; valvular heart disease; renal insufficiency; cerebrovascular disease
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: 'patients were randomly assigned ... using a factorial design and computer‐generated list of random numbers'
Allocation concealment (selection bias)	Low risk	Quote: 'propranolol and matching placebo were purchased from generic suppliers. Study medications were prepackaged...the maximum dose of study medications was five 80mg tablets of propranolol or five placebo tablets...'
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: 'propranolol and matching placebo were purchased from generic suppliers. Study medications were prepackaged...the maximum dose of study medications was five 80mg tablets of propranolol or five placebo tablets...study personnel in contact with participants were blinded to drug assignment.'
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: 'study personnel in contact with participants were blinded to drug assignment.'
Incomplete outcome data (attrition bias) All outcomes	High risk	Cognitive outcomes reported for 102/156 (propranolol group) and 101/156 (placebo group) at 12 months.
Other bias	High risk	Neither group had systolic hypertension at the outset, mean SBP 140 mmHg and 141 mmHg in the two groups. Quote: 'At 3 months of follow‐up, 72 participants were taking 80 mg of propranolol, 41 were taking 160 mg, and 21 were taking 240 to 400 mg per day. At 12 months, 63 participants were taking 80 mg of propranolol, 19 were taking 160 mg, and 35 were taking 240 to 400 mg per day.Twenty participants were not taking any study medications at the end of follow‐up, and 19 others were lost to follow‐up.' Dropouts not described for placebo group. Intention‐to‐treat analyses used.