OBJECTIVES: The study objective was to estimate health status changes using the EuroQol (EQ-5D) in immune thrombocytopenia purpura (ITP) patients treated with romiplostim in two, 24-week randomized placebo controlled phase 3 trials. METHODS: The EQ-5D was scored from 0 to 1 for both the summary score (Index) and the visual analog scale (VAS) and was given at baseline and every four weeks thereafter.  Patients received either romiplostim (titrated to maintain platelets between 50 to 200 x 109/L) or placebo at a 2:1 ratio.  One trial was conducted in patients (n=62) with ITP without splenectomy and the other was among patients (n=63) refractory to splenectomy.  We pooled data across studies and evaluated the EQ-5D change from baseline to week 24.  We compared romiplostim with placebo, as well as responders with non-responders.  Responders had mean platelet counts ≥50 x 109/L for 6 of the last 8 weeks on the study (regardless of treatment group).  Effects were estimated using linear regression (final score as the dependent variable adjusting for baseline score).  Responder analyses incorporated age and gender as covariates.  Missing follow up data were imputed using the last available value, unless a subject died, in which case zero was imputed.  Patients with insufficient data were excluded (n=11). RESULTS: At baseline the median age was 52 years (range 21 to 88), 65% were females, 82% were white, and the mean platelet count was 16.5 x 109/L.  The differences in mean change from baseline for romiplostim compared to placebo was 0.06 for the Index (p=0.017) and 0.05 for VAS (p=0.040).  Adjusted mean changes were similar for responders versus non-responders (0.06 and 0.10 for the Index and VAS, p<0.05). CONCLUSIONS:  Romiplostim use was associated with significant improvement in health status.  Furthermore, these results will be useful for assessing the value of romplostim in economic models.