Project 456786

Immune thrombocytopenia (ITP) is a common blood disorder caused by low blood platelets and a high risk for bleeding. Doctors often get the diagnosis wrong because other conditions can mimic ITP. This can be dangerous for patients because it results in long delays in getting the appropriate care, and patients are exposed to unnecessary treatments that can be harmful and costly. Mistakes in making the diagnosis of ITP is a top concern for patients. Our group developed the Predict-ITP Tool for doctors to use when faced a patient with low blood platelets. We determined specific patient-related features that were associated with the reference standard for ITP, which we defined as a platelet count rise after ITP-specific treatments. Using statistical models based on a group of 629 patients who had been referred to a hematologist with suspected ITP, we identified the following variables for the Predict-ITP Tool: 1) extent of platelet count fluctuations over time; 2) lowest platelet count value; 3) highest mean platelet volume; and 4) history of a major bleed. We completed the internal validation of the Predict-ITP Tool, which showed very promising results. The objective of this study is to conduct an external validation of the Predict-ITP Tool across a broad range of patients from academic and community hematology clinics. We require 765 patients to complete this external validation study and so we are collaborating with hematologists from 10 clinics across Canada. We will design a web-based application so that the tool is readily available to help doctors make an accurate diagnosis. Our team includes ITP researchers, hematologists, biostatisticians, methodologists and a patient representative. Our research will improve patient care, avoid unnecessary treatments, and save on healthcare costs.