Objective: To assess the health-related quality of life (HRQOL) of immune thrombocytopenic purpura (ITP) patients.
Research design and methods: This was a cross-sectional, descriptive study comparing ITP patients' HRQOL to age and gender matched controls. ITP patients from the Platelet Disorder Support Association were recruited until 1000 surveys had been completed. Controls were randomly sampled from the Harris Interactive Online Panel. ITP patients and controls completed a one-time web-based survey, including a comprehensive HRQOL assessment. ITP patients completed the SF-36, the EQ-5D, and the ITP-Patient Assessment Questionnaire (ITP-PAQ). Controls completed the SF-36 and EQ-5D only. ITP patients' SF-36 and EQ-5D scores were compared to controls in unadjusted and adjusted analyses. Associations between splenectomy status, duration of illness, and platelet count with ITP patients' HRQOL scores were also examined.
Results: This analysis included 1002 ITP patients and 1031 controls. ITP patients scored worse on seven of eight SF-36 domains and the Physical and Mental Summary scores (all p < 0.05) and on the EQ-5D visual analog scale (65.5 vs. 82.3; p = 0.002). ITP patients who had undergone splenectomy had similar SF-36 and EQ-5D scores to non-splenectomy patients but scored significantly worse on 5 of 10 ITP-PAQ scales: Bother, Psychological, Fear, Social Activity, and Work (all p < 0.05). ITP patients diagnosed within the past 5 years had worse Bother and Overall Quality of Life scores than less recently diagnosed patients but were similar on other ITP-PAQ scales. Lower platelet count was consistently associated with worse ITP-PAQ scores and had weaker associations with SF-36 and EQ-5D scores.
Conclusions: ITP was associated with consistent and statistically significant deficits on generic HRQOL measures. The ITP-PAQ demonstrated differences based on disease severity and treatments. The self-selection bias in the two samples limits the generalizability of the results to all patients with ITP. Further research is needed in more generalizable samples.