We appreciate the comments of Dr. Miller and colleagues, and also the efforts of the British Thoracic Society. Our recommendations were developed using the following key principles: a transparent link to the strength of the relevant evidence, evaluation of the balance between risks and benefits, and explicit identification of the underlying values and preferences.1Our recommendation that patients with a first episode of idiopathic pulmonary embolism receive treatment for 6 to 12 months was supported by the aggregate evidence available at the time for patients with idiopathic venous thromboembolism.2–5 This evidence supported three conclusions: (1) stopping treatment at 3 months resulted in a high incidence of recurrent thromboembolism, (2) an extended duration of anticoagulant treatment was effective for preventing recurrent thromboembolism while patients continued therapy, and (3) the optimal duration of anticoagulant therapy remained uncertain. The study by Professor Agnelli and colleagues4in patients with pulmonary embolism did not include sufficient patients with idiopathic pulmonary embolism to definitively conclude that 1 year of treatment was not more effective than 3 months, since the 95% confidence interval for the relative risk of recurrent thromboembolism in this subgroup ranged from 0.45 to 2.16. A similar conclusion applied to the study by Pinede and colleagues,5 in which the 95% confidence interval for the relative risk of recurrent venous thromboembolism for 6 months vs 3 months of treatment ranged from 0.47 to 1.87 in the subgroup with idiopathic venous thromboembolism. The aggregate evidence available at the time, and particularly the study by Kearon et al,2 suggested strongly that 3 months was an insufficient duration of treatment for idiopathic venous thromboembolism. We therefore recommended treatment for at least 6 to 12 months (grade 1A), providing the clinician some flexibility to tailor the duration of treatment to the patient’s specific clinical situation. We also made a separate recommendation to consider patients with idiopathic venous thromboembolism for indefinite anticoagulant treatment (grade 2A).6 Finally, we also included an explicit statement of the values and preferences underlying our recommendation, namely, “This recommendation ascribes a relatively high value to preventing recurrent thromboembolic events and a relatively low value on bleeding and cost.”6 Therefore, by including this explicit statement, we acknowledged that our recommendation was weighted toward preventing recurrent thromboembolism, and that clinicians may select a shorter duration of treatment, such as 3 months, for those patients who place a relatively higher value on avoiding bleeding than on preventing recurrent thromboembolism.