Venous thrombembolism (VTE) is a frequent event in the developed world and is associated with serious long-term complications, including post-thrombotic syndrome, recurrent VTE, poor health-related quality of life (HRQL) and death.  VTE can manifest as deep venous leg thrombosis or lung embolism.  OBJECTIVES:   To examine factors associated with HRQL in a randomized, double-blind phase III trial comparing dabigatran etexilate to dose-adjusted warfarin in acute VTE (RE-COVER). METHODS:   Following parenteral treatment of acute VTE, patients were randomized to oral dabigatran etexilate or warfarin for six months.  Patients completed the EQ-5D questionnaire at baseline, three and six months. EQ-5D index scores (UK weights) were regressed on treatment, time since index VTE, age, gender, race, ethnicity, smoking status, body weight, and various clinical characteristics/conditions. Multivariate Censored Least Absolute Deviations (CLAD), Tobit and Ordinary Least Squares (OLS) regression methods were compared.  EQ-5D questionnaire responses were also examined to identify the most affected domains. RESULTS: There were 1245 patients on warfarin and 1264 on dabigatran with valid EQ-5D scores at baseline; 1149 and 1150, respectively, at trial end.  After controlling for covariates, the following factors were statistically significant (p<0.05; CLAD) and exhibited the largest magnitude changes in EQ-5D index scores (from largest to smallest): 6 months post VTE (+0.21), 3 months post VTE (+0.192), recurrent DVT (-0.17), underweight (-0.09), female (-0.08), morbidly obese (-0.07), recurrent pulmonary embolism (-0.06), heart failure (-0.05), age >65 (-0.04) and clinically relevant bleeding (-0.03). There were no statistically significant differences between treatment groups.  All regression methods yielded comparable results. Domains most affected by VTE were mobility, usual activities and pain/discomfort. CONCLUSIONS:   Results of the RE-COVER trial demonstrate that HRQL after VTE is largely dependent on the time from event rather than choice of treatment.  Further, significant differences in HRQL were associated with certain clinical and patient characteristics.