Beyth 2000.

Study characteristics
Methods	Multicentre, randomised controlled trial.
Participants	325 hospitalised patients aged 65+ years (mean age 75 years) commencing warfarin therapy of at least 10 days duration. The study was based in several university hospitals (Cleveland, Ohio, USA). Exclusions included: warfarin therapy within previous 6 months; admission from nursing home; too ill to give informed consent.
Interventions	Self‐monitoring vs usual care The intervention group (n = 163) used home self‐testing using Coumatrak Protime Test System® to self‐monitor prothrombin time. 1‐hour education session, patients phoned results to coach who made recommendations on dosage adjustment. The conventional management group (n = 162) received medical care including management, dosing and medical information managed by primary care physician as per usual care. Randomisation was stratified according to baseline risk for major bleeding.
Outcomes	Primary outcome: first major bleeding event during the 6‐month intervention period. Secondary outcomes: death; recurrent venous thromboembolism at 6 months; major bleeding after 6 months; percentage time INR within target range.
Trial identification
Study duration	6 months
Oral anticoagulant used	Warfarin
Notes	One‐to‐one training, lasting 30‐60 minutes. Participants instructed to check prothrombin 3 times in the first week after hospital discharge and weekly in the first month, and monthly thereafter depending on the results. 100% up at 6 months.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Randomisation stated to have been done but no method reported.
Allocation concealment (selection bias)	Low risk	Clearly adequate concealment.
Intention to treat analysis	Low risk	ITT analysis was performed.
Reporting of losses of follow‐up	Unclear risk	< 20% losses to follow‐up.
Blinding	Low risk	Blinded data collectors.