VTE included: nonfatal deep vein thrombosis (DVT) or nonfatal pulmonary embolism (PE). All index events, DVT and/or PE were adjudicated/confirmed by a central independent adjudication committee blinded to treatment. DVT assessed by compression ultrasound and/or venography; PE assessed by spiral computed tomography scanning, pulmonary angiography, and/or ventilation/perfusion lung scan. Event rate (proportion of participants with event) calculated as n/N (n=number of events; N=number of participants). Intended treatment period: longer of the dosing period plus 2 days (completed treatment) or 355 days (discontinued early). Composite endpoint included events at any time from randomization until end of the intended treatment period, regardless whether drug treatment was received. For missing endpoint data, participants were imputed as having had a primary efficacy outcome event.

Intent to treat: all randomized participants with consent. Analyzed per randomized treatment assigned. (n) number of events=32, 34, 96 in apixaban 2.5 mg, 5 mg, and placebo arms, respectively; number of events imputed=13, 20, 19, respectively. Confidence interval (CI) for event rate was calculated based on the Wald asymptotic confidence limits.

VTE includes nonfatal DVT or nonfatal PE. All index events, DVT and/or PE were adjudicated/confirmed by a central independent adjudication committee blinded to treatment. DVT assessed by compression ultrasound and/or venography; PE assessed by spiral computed tomography scanning, pulmonary angiography, and/or ventilation/perfusion lung scan. Event rate (proportion of participants with event) calculated as n/N (n=number of events; N=number of participants). Intended treatment period: longer of the dosing period plus 2 days (completed treatment) or 355 days (discontinued early). Composite endpoint included events at any time from randomization until end of the intended treatment period, regardless whether drug treatment was received. For missing endpoint data, participants were imputed as having had a primary efficacy outcome event.

ITT: all randomized participants with consent. Analyzed per randomized treatment assigned. (n) number of events=27, 34, 92 in apixaban 2.5 mg, 5 mg, placebo arms, respectively. Number of imputed events=13, 20, 19 in apixaban 2.5 mg, 5 mg, placebo arms, respectively. CI for single event rate was calculated based on Wald asymptotic confidence limits.

VTE included: nonfatal DVT or nonfatal PE. Event rate (proportion of participants with event) calculated as n/N (n=number of events; N=number of participants). Intended treatment period: longer of the dosing period plus 2 days (completed treatment) or 355 days (discontinued early). Composite endpoint included events at any time from randomization until end of the intended treatment period, regardless whether drug treatment was received. No imputation was done for these endpoints; participants who had an event during the intended treatment period were counted. Confidence interval (CI) for single event rate was calculated based on the Wald asymptotic confidence limits.

Intent to treat: all randomized participants with valid consent. Analyzed per randomized treatment assigned.(n)number of events = 19, 14, 77 in apixaban 2.5 mg, 5 mg, and placebo arms, respectively. All events were counted; no events were imputed.

VTE includes nonfatal DVT or nonfatal PE. All index events, DVT and/or PE were adjudicated/confirmed by a central independent adjudication committee blinded to treatment. DVT assessed by compression ultrasound and/or venography; PE assessed by spiral computed tomography scanning, pulmonary angiography, and/or ventilation/perfusion lung scan. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). Composite endpoint included events that occurred any time from randomization until end of the intended treatment period, regardless of whether the participants were receiving drug treatment. Intended treatment period was defined as the longer of the dosing period plus 2 days or 355 days. If there were missing endpoint data, participants were imputed as having had an efficacy outcome event.

Intent to treat: all randomized participants with consent.(n)number of events=27, 34, 95 in apixaban 2.5 mg, 5 mg, placebo arms, respectively. The (n)number of imputed events were = 13, 20, 19 in apixaban 2.5 mg, 5 mg, placebo arms, respectively. CI for single event rate was calculated based on the Wald asymptotic confidence limits

DVT was adjudicated/confirmed by a central independent adjudication committee blinded to treatment and assessed by compression ultrasound and/or venography. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). Intended treatment period: longer of the dosing period plus 2 days (completed treatment) or 355 days (discontinued early). Endpoint included events at any time from randomization until end of the intended treatment period, regardless whether drug treatment was received. Participants with missing endpoint information were classified as having had the efficacy event (imputation). Confidence interval (CI) for single event rate was calculated based on the Wald asymptotic confidence limits.

Intent to treat: all randomized participants with valid consent. Analyzed per randomized treatment assigned. (n) number of events = 19, 28, 72 in apixaban 2.5 mg, 5 mg, placebo arms, respectively.

PE was adjudicated/confirmed by a central independent adjudication committee blinded to treatment and was assessed by spiral computed tomography scanning, pulmonary angiography, and/or ventilation/perfusion lung scan. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). Intended treatment period was defined as the longer of the dosing period plus 2 days or 355 days. Endpoint included events at any time from randomization until end of the intended treatment period, regardless whether drug treatment was received. Participants with missing endpoint information were classified as having had the efficacy event (imputation). CI for single event rate was calculated based on the Wald asymptotic confidence limits.

Intent to treat: all randomized participants with valid consent. Analyzed per randomized treatment assigned. (n) number of events = 23, 25, 37 in apixaban 2.5 mg, 5 mg, placebo arms, respectively.

VTE-related death defined as: PE (based on objective diagnostic testing, autopsy), unexplained death (and VTE cannot be ruled out), sudden death (and VTE cannot be ruled out). DVT and/or PE were adjudicated/confirmed by a central independent adjudication committee blinded to treatment: DVT assessed by compression ultrasound and/or venography; PE assessed by spiral computed tomography scanning, pulmonary angiography, and/or ventilation/perfusion lung scan. New/recurrent VTE, death, venous/arterial thromboembolic events, bleeding, thrombocytopenia, acute myocardial infarction and stroke were also adjudicated. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). Intended treatment period was defined as the longer of the dosing period plus 2 days or 355 days. Participants with missing endpoint information were classified as having had the efficacy event (imputation).

Intent to treat: all randomized participants with valid consent. Analyzed per randomized treatment assigned. (n) number of events = 17, 24, 26 in apixaban 2.5 mg, 5 mg, placebo arms, respectively. CI for single event rate was calculated based on the Wald asymptotic confidence limits.

CV-related death was defined as myocardial infarction, stroke, or other specified cardiovascular event and were adjudicated/confirmed by a central independent adjudication committee blinded to treatment. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). Intended treatment period was defined as the longer of the dosing period plus 2 days or 355 days. Endpoint included events at any time from randomization until end of the intended treatment period, regardless whether drug treatment was received. Participants with missing endpoint information were classified as having had the efficacy event (imputation). CI for single event rate was calculated based on the Wald asymptotic confidence limits.

Intent to treat: all randomized participants with valid consent. Analyzed per randomized treatment assigned. number of events (n)= 17, 24, 29 in the apixaban 2.5 mg, 5 mg, placebo arms, respectively.

DVT and/or PE were adjudicated/confirmed by a central independent adjudication committee blinded to treatment: DVT assessed by compression ultrasound and/or venography; PE assessed by spiral computed tomography scanning, pulmonary angiography, and/or ventilation/perfusion lung scan. New/recurrent VTE, death, venous/arterial thromboembolic events, bleeding, thrombocytopenia, acute myocardial infarction and stroke were also adjudicated. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). Intended treatment period was defined as the longer of the dosing period plus 2 days or 355 days. Participants with missing endpoint information were classified as having had the efficacy event (imputation). CI for single event rate was calculated based on the Wald asymptotic confidence limits.

Intent to treat: all randomized participants with valid consent. (n) number of events = 22, 25, 33 in apixaban 2.5 mg, apixaban 5 mg, and placebo arms, respectively.

All index events, DVT and/or PE were adjudicated/confirmed by a central independent adjudication committee blinded to treatment. DVT assessed by compression ultrasound and/or venography; PE assessed by spiral computed tomography scanning, pulmonary angiography, and/or ventilation/perfusion lung scan. Intended treatment period was defined as the longer of the dosing period plus 2 days or 355 days. Endpoint included events at any time from randomization until end of the intended treatment period, regardless whether drug treatment was received. First event category was the first primary event for each participant and each participant was counted once. CV-related death was presented excluding VTE-related death. In participants with event category, each participant was counted once in each event category but could have been counted in multiple categories. No imputation was done for these endpoints; participants who had an event during the intended treatment period were counted.

Intent to treat: all randomized participants with valid consent. Analyzed per randomized treatment assigned. In first event (first primary event) each participant counted once. In event category, each participant was counted only once in each event category but could have been counted in multiple categories.

Major bleeding was adjudicated/confirmed by a central independent adjudication committee blinded to treatment and was defined as acute clinically overt bleeding: associated with a fall in hemoglobin of 2 grams per deciliter (g/dL) or more, or leading to a transfusion of 2 or more units of packed red blood cells or 1000 milliliters (mL) or more of whole blood, or in a critical site: intracranial, intraspinal, intraocular, pericardial, intra-articular, intramuscular with compartment syndrome, retroperitoneal, or another critical organ; or is fatal. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). Confidence interval (CI) for event rate was calculated based on the Wald asymptotic confidence limits. Treated population includes randomized participants who received at least one dose of study drug.

Treated population includes randomized participants who received at least one dose of study drug. (n) number of events = 2, 1, 4 in apixaban 2.5 mg, and 5 mg, and placebo arms, respectively.

Major bleeding and clinically relevant non-major bleeding were adjudicated/confirmed by a central independent adjudication committee blinded to treatment. Major bleeding was defined as acute clinically overt bleeding: associated with a fall in hemoglobin of 2 g/dL or more, or leading to a transfusion of 2 or more units of packed red blood cells or 1000 mL or more of whole blood, or in a critical site: intracranial, intraspinal, intraocular, pericardial, intra-articular, intramuscular with compartment syndrome, retroperitoneal, or another critical organ or is fatal. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). CI for single event rate was calculated based on the Wald asymptotic confidence limits. Treated population includes randomized participants who received at least one dose of study drug.

Treated participants were those who received at least 1 dose of study drug. (n)number of events = 27, 35, 22 in apixaban 2.5 mg, 5 mg, and placebo arms, respectively.

Non-major clinically relevant bleeding was adjudicated/confirmed by a central independent adjudication committee blinded to treatment and defined as: acute clinically overt bleeding compromising hemodynamics; leading to hospitalization; traumatic subcutaneous hematoma; intramuscular hematoma; epistaxis that lasted for more than 5 minutes, was repetitive or led to an intervention; spontaneous gingival bleeding (or lasting more than 5 minutes); spontaneous hematuria (macroscopic or lasted more than 24 hours after instrumentation of the urogenital tract); macroscopic gastrointestinal hemorrhage (including at least 1 episode of melena or hematemesis (if clinically apparent with positive results on a fecal occult-blood test); rectal blood loss. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). CI for single event rate was calculated based on the Wald asymptotic confidence limits.

Treated population includes randomized participants who received at least one dose of study drug. (n)number of events = 25, 34, 19 in apixaban 2.5 mg, 5 mg, and placebo arms, respectively.

All bleeding events were reviewed by the central independent adjudication committee blinded to treatment and classified as major bleeding, clinically relevant non-major bleeding, minor bleeding or no bleeding. If event was not major or clinically relevant non-major, it was judged to be minor. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). Confidence interval (CI) for single event rate was calculated based on the Wald asymptotic confidence limits. Treated population includes randomized participants who received at least one dose of study drug.

Treated population includes randomized participants who received at least one dose of study drug. (n)number of events = 75, 98, 58 in apixaban 2.5 mg, 5 mg, and placebo arms, respectively.

All bleeding events were adjudicated/confirmed by a central independent adjudication committee blinded to treatment. Total bleeding was defined as any major, clinically relevant non-major, or minor bleeding. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). CI for single event rate was calculated based on the Wald asymptotic confidence limits. Treated population includes randomized participants who received at least one dose of study drug.

Treated population includes randomized participants who received at least one dose of study drug. (n) number of events = 94, 121, 74 in apixaban 2.5 mg, 5 mg, and placebo arms, respectively.

VTE related death defined as PE (based on objective diagnostic testing, autopsy), unexplained death (and VTE cannot be ruled out), sudden death (and VTE cannot be ruled out). DVT and/or PE were adjudicated/confirmed by a central independent adjudication committee blinded to treatment: DVT assessed by compression ultrasound and/or venography; PE assessed by spiral computed tomography scanning, pulmonary angiography, and/or ventilation/perfusion lung scan. New/recurrent VTE and death, were also adjudicated. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). Intended treatment period was defined as the longer of the dosing period plus 2 days or 355 days. Endpoint included events at any time from randomization until end of the intended treatment period, regardless whether drug treatment was received. No imputation was done for these endpoints.

Intent to treat: all randomized participants with valid consent. Participants analyzed per randomized treatment assigned. (n) number of events = 14, 14, 73 in apixaban 2.5 mg, 5 mg, and placebo arms, respectively. CI for single event rate was calculated based on the Wald asymptotic confidence limits.

CV-related death was defined as myocardial infarction, stroke, or other specified cardiovascular event. Index events of DVT and/or PE, along with myocardial infarction and stroke were adjudicated/confirmed by a central independent adjudication committee blinded to treatment: DVT assessed by compression ultrasound and/or venography; PE assessed by spiral computed tomography scanning, pulmonary angiography, and/or ventilation/perfusion lung scan. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). Intended treatment period was defined as the longer of the dosing period plus 2 days or 355 days. Composite endpoint included events at any time from randomization until end of the intended treatment period, regardless whether drug treatment was received. No imputation was done for these endpoints; participants who had an event during the intended treatment period were counted.

Intent to treat: all randomized participants with valid consent. Participants analyzed per randomized treatment assigned. (n) number of events = 14, 14, 76 in apixaban 2.5 mg, 5 mg, and placebo arms, respectively. CI for single event rate was calculated based on the Wald asymptotic confidence limits.

DVT was adjudicated/confirmed by a central independent adjudication committee blinded to treatment: DVT assessed by compression ultrasound and/or venography. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). Intended treatment period was defined as the longer of the dosing period plus 2 days or 355 days. Endpoint included events at any time from randomization until end of the intended treatment period, regardless whether drug treatment was received. No imputation was done for this endpoint; participants who had an event during the intended treatment period were counted.

Intent to treat: all randomized participants with valid consent. Participants analyzed per randomized treatment assigned. (n) number of events = 6, 8, 53 in apixaban 2.5 mg, 5 mg, and placebo arms, respectively. CI for single event rate was calculated based on the Wald asymptotic confidence limits.

PE was adjudicated/confirmed by a central independent adjudication committee blinded to treatment: PE assessed by spiral computed tomography scanning, pulmonary angiography, and/or ventilation/perfusion lung scan. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). Intended treatment period was defined as the longer of the dosing period plus 2 days or 355 days. Endpoint included events at any time from randomization until end of the intended treatment period, regardless whether drug treatment was received. No imputation was done for this endpoint; participants who had an event during the intended treatment period were counted.

Intent to treat: all randomized participants with valid consent. Participants analyzed per randomized treatment assigned. (n) number of events = 8, 4, 15 in apixaban 2.5 mg, 5 mg, and placebo arms, respectively. CI for single event rate was calculated based on the Wald asymptotic confidence limits.

VTE related death defined as PE (based on objective diagnostic testing, autopsy), unexplained death (and VTE cannot be ruled out), sudden death (and VTE cannot be ruled out). DVT and/or PE were adjudicated/confirmed by a central independent adjudication committee blinded to treatment: DVT assessed by compression ultrasound and/or venography; PE assessed by spiral computed tomography scanning, pulmonary angiography, and/or ventilation/perfusion lung scan. New/recurrent VTE, and death, were also adjudicated. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). Intended treatment period was defined as the longer of the dosing period plus 2 days or 355 days. Endpoint included events at any time from randomization until end of the intended treatment period, regardless whether drug treatment was received. No imputation was done for this endpoint.

Intent to treat: all randomized participants with valid consent. Participants analyzed per randomized treatment assigned. (n) number of events = 2, 3, 7 in apixaban 2.5 mg, 5 mg, and placebo arms, respectively. CI for single event rate was calculated based on the Wald asymptotic confidence limits.

CV-related death was defined as myocardial infarction, stroke, or other specified cardiovascular event and these were adjudicated/confirmed by a central independent adjudication committee blinded to treatment. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). Intended treatment period was defined as the longer of the dosing period plus 2 days or 355 days. Endpoint included events at any time from randomization until end of the intended treatment period, regardless whether drug treatment was received. No imputation was done for this endpoint; participants who had an event during the intended treatment period were counted.

Intent to treat: all randomized participants with valid consent. Participants analyzed per randomized treatment assigned. (n) number of events = 2, 3, 10 in apixaban 2.5 mg, 5 mg, and placebo arms, respectively. CI for single event rate was calculated based on the Wald asymptotic confidence limits.

DVT and/or PE were adjudicated/confirmed by a central independent adjudication committee blinded to treatment: DVT assessed by compression ultrasound and/or venography; PE assessed by spiral computed tomography scanning, pulmonary angiography, and/or ventilation/perfusion lung scan. New/recurrent VTE, death, venous/arterial thromboembolic events, bleeding, thrombocytopenia, acute myocardial infarction and stroke were also adjudicated. Event rate is proportion of participants with event; calculated as n/N (n=number of events; N=number of participants). Intended treatment period was defined as the longer of the dosing period plus 2 days or 355 days. Endpoint included events at any time from randomization until end of the intended treatment period, regardless whether drug treatment was received. No imputation was done for this endpoint; participants who had an event during the intended treatment period were counted.

Intent to treat: all randomized participants with valid consent. Participants analyzed per randomized treatment assigned. (n) number of events = 7, 4, 14 in apixaban 2.5 mg, 5 mg, and placebo arms, respectively. CI for single event rate was calculated based on the Wald asymptotic confidence limits.