OBJECTIVES: Patients with HF can suffer from poor HRQoL particularly as their disease progresses. We wanted to determine whether HRQoL and mortality differences existed between patients following an intervention. We defined an event as either death or a drop from baseline of at least 10 points in the PCS or MCS summary scales of the SF-36 that was sustained on at least 2 consecutive intervals with no recovery prior to study termination. METHODS: This was an RCT (n=134) with HRQoL collection from baseline every 3 months to 1 year. All data collection and outcomes assessment were done by study personnel who were blinded to patient treatment allocation. Complete SF-36 data were available on 114 patients. We did a Cox regression time-to-event survival analysis adjusting for Arm. RESULTS: According to our definition of events 14.9% of patients had a PCS event and 16.7% of patients had an MCS event. There was no significant effect of Arm. CONCLUSIONS: A 10-point drop in the PCS or MCS can be considered to be a minimal clinically important difference in patients with a sustained drop indicating no response to the intervention. The analytic techniques we used expand the interpretability of HRQoL changes and incorporate clinical outcomes.