Venous thromboembolism (VTE) is a common cause of death and morbidity in stroke patients. There are few data concerning the effectiveness of intermittent pneumatic compression (IPC) in treating patients with stroke. To establish whether or not the application of IPC to the legs of immobile stroke patients reduced their risk of deep vein thrombosis (DVT). Clots in Legs Or sTockings after Stroke (CLOTS) 3 was a multicentre, parallel-group, randomised controlled trial which allocated patients via a central randomisation system to IPC or no IPC. A technician blinded to treatment allocation performed compression duplex ultrasound (CDU) of both legs at 7-10 days and 25-30 days after enrolment. We followed up patients for 6 months to determine survival and later symptomatic VTE. Patients were analysed according to their treatment allocation. We enrolled 2876 patients in 94 UK hospitals between 8 December 2008 and 6 September 2012. patients admitted to hospital within 3 days of acute stroke and who were immobile on the day of admission (day 0) to day 3. age < 16 years; subarachnoid haemorrhage; and contra-indications to IPC including dermatitis, leg ulcers, severe oedema, severe peripheral vascular disease and congestive cardiac failure. Participants were allocated to routine care or routine care plus IPC for 30 days, or until earlier discharge or walking independently. The primary outcome was DVT in popliteal or femoral veins, detected on a screening CDU, or any symptomatic DVT in the proximal veins, confirmed by imaging, within 30 days of randomisation. The secondary outcomes included death, any DVTs, symptomatic DVTs, pulmonary emboli, skin breaks on the legs, falls with injury or fractures and duration of IPC use occurring within 30 days of randomisation and survival, symptomatic VTE, disability (as measured by the Oxford Handicap Scale), quality of life (as measured by the European Quality of Life-5 Dimensions 3 Level questionnaire) and length of initial hospital stay measured 6 months after randomisation. We allocated 1438 patients to IPC and 1438 to no IPC. The primary outcome occurred in 122 (8.5%) of 1438 patients allocated to IPC and 174 (12.1%) of 1438 patients allocated to no IPC, giving an absolute reduction in risk of 3.6% [95% confidence interval (CI) 1.4% to 5.8%] and a relative risk reduction of 0.69 (95% CI 0.55 to 0.86). After excluding 323 patients who died prior to any primary outcome and 41 who had no screening CDU, the primary outcome occurred in 122 of 1267 IPC participants compared with 174 of 1245 no-IPC participants, giving an adjusted odds ratio of 0.65 (95% CI 0.51 to 0.84; p = 0.001). Secondary outcomes in IPC compared with no-IPC participants were death in the treatment period in 156 (10.8%) versus 189 (13.1%) (p = 0.058); skin breaks in 44 (3.1%) versus 20 (1.4%) (p = 0.002); and falls with injury in 33 (2.3%) versus 24 (1.7%) (p = 0.221). Among patients treated with IPC, there was a statistically significant improvement in survival to 6 months (hazard ratio 0.86, 95% CI 0.73 to 0.99; p = 0.042), but no improvement in disability. The direct cost of preventing a DVT was £1282 per event (95% CI £785 to £3077). IPC is an effective and inexpensive method of reducing the risk of DVT and improving survival in immobile stroke patients. Further research should test whether or not IPC improves survival in other groups of high-risk hospitalised medical patients. In addition, research into methods to improve adherence to IPC might increase the benefits of IPC in stroke patients. Current Controlled Trials ISRCTN93529999. The start-up phase of the trial (December 2008-March 2010) was funded by the Chief Scientist Office of the Scottish Government (reference number CZH/4/417). The main phase of the trial was funded by the National Institute for Health Research Health Technology Assessment programme (reference number 08/14/03). Covidien Ltd (Mansfield, MA, USA) lent its Kendall SCD™ Express sequential compression system controllers to the 105 centres involved in the trial and donated supplies of its sleeves. It also provided logistical help in keeping our centres supplied with sleeves and training materials relevant to the use of their devices. Recruitment and follow-up were supported by the National Institute for Health Research-funded UK Stroke Research Network and by the Scottish Stroke Research Network, which was supported by NHS Research Scotland.