IPPPSH 1985.

Methods	Multicentre study Randomisation: random allocation of participants was achieved by providing to the investigating centres participant numbers randomised into balanced blocks each having 6 numbers. Sealed envelopes containing the treatment code were provided to each investigator. Loss to follow‐up: 0.6% Duration of follow‐up: 3 to 5 years (mean 4 years)
Participants	Geographic region: UK (36.4%), Canada (12.0%), the Netherlands (3.6%), Israel (20.9%), Italy (11.7%), Federal Republic of Germany (15.4%) Number of participants: 6357 (50.2% men) Age range: 40 to 64 years (mean age: 52.2 years) Entry BP criteria: diastolic BP of 100 mmHg to 125 mmHg (Korotkoff Phase V) measured in seated position using standard mercury sphygmomanometer; mean SBP at entry 173 mmHg (SD 18.4) Race: Exclusion criteria: past or present history of angina pectoris or MI; heart failure; relevant cardiac valvular disease; atrio‐ventricular blocks grades II and III or sick sinus syndrome; bradycardia (< 50 beats per minute); intermittent claudication; previous cerebrovascular accident; insulin‐dependent diabetes; pregnancy; obstructive airways disease or history of bronchial asthma; renal, hepatic, gastrointestinal or any other severe disease Comorbid conditions: current smokers (29.1%)
Interventions	Beta‐blocker group: Step 1: oxprenolol slow release 160 mg/day Control group: Step 1: film‐coated placebo of identical appearance Additional treatment for both groups: Step 2: diuretic or sympatholytic or vasodilator Step 3: diuretic + sympatholytic, or diuretic + vasodilator, or sympatholytic + vasodilator Step 4: diuretic + sympatholytic + vasodilator During study, 30% of participants remained on beta‐blocker only while 15% remained placebo only. Total diuretic use was 67% in the beta‐blocker group and 82% in the placebo group.
Outcomes	CHD mortality: fatal MI, sudden death CHD morbidity ‐ non‐fatal MI Cerebrovascular mortality ‐ fatal stroke Cerebrovascular morbidity ‐ non‐fatal stroke Cardiovascular mortality Cardiovascular morbidity Total mortality Adverse effects
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Block randomisation used so assumed to be computer‐generated.
Allocation concealment (selection bias)	Low risk	Random allocation of participants was achieved by providing to the investigating centres participant numbers randomised into balanced blocks each having 6 numbers. Sealed envelopes containing the treatment code were provided to each investigator.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Participants and personnel blinded
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Blind outcome assessment
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Not indicated whether reasons for missing outcome data were similar across treatment groups.
Selective reporting (reporting bias)	Low risk	All outcomes reported as stated in protocol.
Other bias	Unclear risk	Other antihypertensive drugs added to randomly allocated treatment to control BP. The observed effects may equally have resulted from the different additional drugs.