Evaluating clinically meaningful change on the ITP-PAQ: preliminary estimates of minimal important differences

OBJECTIVE: Interpretation of data from health-related quality of life (HRQoL) questionnaires can be enhanced with the availability of minimally important difference (MID) estimates. This information will aid clinicians in interpreting HRQoL differences within patients over time and between treatment groups. The Immune Thrombocytopenic Purpura (ITP)-Patient Assessment Questionnaire (PAQ) is the only comprehensive HRQoL questionnaire available for adults with ITP. RESEARCH DESIGN AND METHODS: Forty centers from within the US and Europe enrolled ITP patients into one of two multicenter, randomized, placebo-controlled, double-blind, 6-month, phase III clinical trials of romiplostim. Patients enrolled in these studies self-administered the ITP-PAQ and two items assessing global change (anchors) at baseline and weeks 4, 12, and 24. Using data from the ITP-PAQ and these two anchors, an anchor-based estimate was computed and combined with the standard error of measurement and standard deviation to compute a distribution-based estimate in order to provide an MID range for each of the 11 scales of the ITP-PAQ. RESULTS: A total of 125 patients participated in these clinical trials and provided data for use in these analyses. Combining results from anchor- and distribution-based approaches, MID values were computed for 9 of the 11 scales. MIDs ranged from 8 to 12 points for Symptoms, Bother, Psychological, Overall QOL, Social Activity, Menstrual Symptoms, and Fertility, while the range was 10 to 15 points for the Fatigue and Activity scales of the ITP-PAQ. These estimates, while slightly higher than other published MID estimates, were consistent with moderate effect sizes. CONCLUSIONS: These MID estimates will serve as a useful tool to researchers and clinicians using the ITP-PAQ, providing guidance for interpretation of baseline scores as well as changes in ITP-PAQ scores over time. Additional work should be done to finalize these initial estimates using more appropriate anchors that correlate more highly with the ITP-PAQ scales.