For patients who are diagnosed with venous thromboembolism (VTE) provoked by transient risk factors, clinical guidelines typically recommend 3 months of oral anticoagulation, with longer treatment considered for unprovoked VTE. Describing treatment patterns of VTE in a real world population may identify remedial gaps in patient care.


We aimed to characterize oral anticoagulant treatment with vitamin K antagonists (VKA) following incident VTE and assess persistence of VKA therapy in patients with provoked vs. unprovoked VTE in a real world setting.


We used the linked administrative healthcare databases of the province of Québec, Canada, including the hospitalization, universal healthcare services, and out-patient prescription databases. We identified all beneficiaries with an incident DVT or PE between 2000 and 2009, which we classified as definite or probable VTE using a priori determined diagnostic algorithms based on ICD-9-CM or ICD-10-CA diagnosis codes. We formed two patient cohorts, one with definite and the other including definite or probable first-time VTE, that were followed until death or end of study (December 31, 2009). Anticoagulant out-patient prescription patterns were analyzed for both patient cohorts.


From 245,452 Québec residents between 2000 and 2009 with at least 1 VTE diagnosis in RAMQ or MED-ÉCHO, we formed the definite VTE cohort including 40,776 definite cases and the any VTE cohort consisting of 54,803 definite or probable cases. From the 40,776 patients with a first definite VTE, there were 24,860 patients with DVT alone (61%) and 15,916 with PE with or without DVT (39%). Furthermore, there were 78% of patients over the age of 60 and 58.3% of patients were women. In all, 8,998 (22.1%) patients had an unprovoked VTE event while 19,010 (46.6%) patients had a provoked non-cancer event. Similar findings were found in the any VTE cohort.

Among definite VTE cohort patients with a provoked non-cancer VTE, 68.6% of patients received anticoagulation after the VTE event. Most were dispensed VKA (64.9%) and 23.9% received a prescription for low molecular weight heparin (LMWH). Among patients with an unprovoked VTE, 86.3% of patients were prescribed anticoagulation (84.5% used VKA and 39.9% used LMWH). Overall, a greater number of patients received anticoagulation following PE (85.2%) than DVT alone (66.2%). Results were similar in the any VTE cohort.

Persistence with VKA therapy among patients with provoked VTE was 86.9% at 90 days, 59.5% at 180 days and 19.9% at 365 days. Treatment persistence for patients with unprovoked non-cancer VTE was 88.8%, 66.8% and 22.9% for 90, 180 and 365 days, respectively. Similar findings were found in the any VTE cohort.


Our study provides useful information on VTE management in a real world population. Treatment persistence was similar for patients with provoked and unprovoked VTE. VKA therapy duration after provoked VTE was longer than the recommended 3 months, whereas treatment was shorter than suggested in patients with unprovoked VTE. Further investigation is needed to determine reasons for non-adherence to clinical guidelines.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.