OBJECTIVES: To explore the process of patients’ satisfaction with anticoagulant treatment using Structural Equation Modeling (SEM). METHODS: The Perception of AntiCoagulant Treatment Questionnaire (PACT-Q) includes 2 modules: a 7-item module assessing patient expectations of anticoagulant treatment and a 20-item module assessing treatment convenience and patient satisfaction. It was completed by deep venous thrombosis, atrial fibrillation and pulmonary embolism patients in 3 clinical trials assessing anticoagulant treatments. The first module was administered at baseline (BL) and the second after 3 (M3) and 6 months (M6). SEM was applied to pooled data from the three trials. SEM specifications were supported by the questionnaire conceptual model, satisfaction theory and the scoring rules of the instrument. The goodness of fit of the models was assessed using a set of commonly used fit indices including the Root Mean Square of Approximation (RMSEA). Association between latent variables was assessed using Standardized Path Coefficients (SPC). RESULTS: A total of 986 patients had fully completed PACT-Q at BL, M3 and M6. PACT-Q items allowed good measurement of Convenience (RMSEA=0.054) and Satisfaction (RMSEA=0.028) at M3. The 7 expectation items were kept independent. An overall model involving two expectation items (expectations of symptom relief, worries about making mistakes) and both Satisfaction and Convenience at M3 and M6 was estimated (RMSEA=0.032). In this final model, Convenience had a stronger impact on Satisfaction at M3 than at M6 (respective SPC: 0.57 vs 0.48). Convenience at M6 was very strongly related with Convenience at M3 (SPC: 0.80) while the relationship between Satisfaction at M3 and M6 was weaker (SPC: 0.52). CONCLUSIONS: The application of SEM to the PACT-Q data allowed the prominent role of patients’ expectations and perception of treatment convenience in the process of anticoagulant treatment satisfaction to be highlighted. Convenience was also shown to be more stable over time than satisfaction.