Résumé : OBJECTIVE: This study examines risk recall and evolution of patients' anxiety after transmission of life-threatening risk information in an informed consent procedure for experimental HSCT. METHODS: Informed consent interviews were audio-recorded and transcribed. Patient risk recall was obtained through comparing information provided in the interview to information recalled by patients following the interview. The evolution of patients' anxiety was assessed through comparing patients' post- to patients' pre-interview anxiety using the STAI-State. Physicians' communication skills and risk framing were analyzed. RESULTS: Twenty patients were included. Patients recalled on average 4 risks (S.D.=1.6) out of 9 different risks transmitted (S.D.=2) which corresponds to a recall rate of 44% (S.D.=15.5). Patients' anxiety remained on average stable (Mean=0.4; S.D.=9.1). Linear regression analysis showed that risk recall was predicted positively by the number of risks transmitted (B=.30; P=.032) and by patients' problem-focused coping (B=.21; P=.008). The evolution of anxiety was predicted positively by the number of times benefits were transmitted (B=.83; P=.003) and negatively by the level of anxiety before the interview (B=-.50; P=.001). CONCLUSION: Results show the limits of patients' risk recall in the context of informed consent for a life-threatening procedure. PRACTICE IMPLICATIONS: This study highlights the necessity to develop strategies allowing tailoring of risk transmission to every patient's needs.