A disease-specific measure of health-related quality of life in adults with chronic immune thrombocytopenic purpura: psychometric testing in an open-label clinical trial.

BACKGROUND: The Immune Thrombocytopenic Purpura Patient Assessment Questionnaire (ITP-PAQ) was developed to assess disease-specific quality of life (QoL) in adults with ITP. It is a 44-item questionnaire that includes scales for physical health (symptoms, fatigue/sleep, bother, and activity), emotional health (psychological and fear), overall QoL, social activity, women's reproductive health, and work. A previous study reported preliminary evidence of its reliability and validity.
OBJECTIVES: The present study was conducted to ascertain the responsiveness (ability to detect a clinically important treatment effect), reliability, and validity of the ITP-PAQ and to corroborate the earlier findings. The women's reproductive health scale was evaluated for psychometric evidence of the existence of separate menstrual symptoms and fertility subscales.
METHODS: The ITP-PAQ was evaluated in the context of an ongoing open-label extension study assessing the tolerability and durability of increases in the platelet count with AMG 531 (a thrombopoiesis peptibody that increases platelet production by targeting the thrombopoietin receptor) administered by subcutaneous injection once weekly in adult patients with ITP It was self-administered at baseline and at weeks 4, 12, and 24. The responsiveness of the questionnaire was evaluated by calculating and comparing the change scores of patients who showed clinical improvement-categorized as platelet responders (those with a platelet count > or =50 x 10(9) cells/L and a doubling of baseline values at week 24) and durable platelet responders (those with a platelet count > or =50 x 10(9) cells/L and a doubling of baseline values on > or =6 occasions during weeks 17-24)-with the change scores of patients wh did not show clinical improvement. The reliability (internal consistency and test-retest) and validity (convergent, discriminant, and known groups) of the questionnaire were also evaluated. Validity was examined in terms of correlations between the ITP-PAQ and the 36-item Short-Form Health Survey (SF-36), a generic measure of health-related QoL.
RESULTS: Thirty-four patients completed the ITP-PAQ. Most of the scales were found capable of detecting clinically important treatment effects, with the scales for symptoms, fatigue/sleep, bother, and activity being particularly responsive. All scales were found to have internal consistency reliability (Cronbach's alpha, 0.700-0.950), with the exceptions of the menstrual symptoms subscale (0.988 and 0.959 at weeks 12 and 24, respectively) and the work scale (0.691 at week 24). Test-retest reliability was acceptable (intraclass correlation coefficient, 0.725-0.867), with the exceptions of the scales for symptoms (0.677) and women's reproductive health (0.592) and the fertility subscale (0.171). Construct validity was supported by correlations between specific ITP-PAQ and SF-36 scales, with the exceptions of the menstrual symptoms and fertility subscales. Discriminant validity was reported for the symptoms, fatigue/sleep, bother, and activity scales. Durable platelet responders had significantly better scores than nonresponders on the symptoms (P = 0.022), bother (P = 0.008), psychological (P = 0.033), and overall QoL scales (P = 0.032). Compared with those who had undergone splenectomy, patients without splenectomy had significantly higher scores on the women's reproductive health scale (P = 0.03).
CONCLUSIONS: The results of this analysis indicate that the ITP-PAQ has acceptable responsiveness, reliability, and validity. Further study of the minimal clinically important difference in ITP-PAQ scale scores is needed.