Researchers Develop New Statistical Model to Help Predict Chronic ITP at Diagnosis
When a child is diagnosed with immune thrombocytopenia (ITP), one of the hardest questions for families and clinicians is whether the disorder will resolve within a year or progress into a chronic condition requiring years of monitoring and ongoing care.
Now, a new study is helping bring much-needed clarity. Published in Blood (American Society of Hematology), this retrospective analysis of 611 children with ITP introduces a statistical risk model designed to estimate the likelihood of chronic ITP at the time of diagnosis.
Read the full study:
The model showed strong predictive performance and clinical utility. Led by first authors Mark Zobeck, MD, MPH and Kirsty Hillier, MD, alongside senior authors Taylor Olmsted Kim, MD, and Amanda B. Grimes, MD, the study examined a range of diagnostic-time factors. The team found that age, sex, antibody and immune cell levels, platelet count, direct antiglobulin test (DAT) positivity, and the presence of relevant underlying conditions each contributed meaningfully to chronic ITP risk.
Immune thrombocytopenia (ITP) occurs when the immune system mistakenly destroys platelets — important blood components that help form clots, stop bleeding, and start healing. This can cause easy bruising, frequent nosebleeds, or, in rare cases, serious internal bleeding. The model developed in this study offers a framework to support early risk stratification and guide more individualized treatment planning for children who may be at higher risk of chronic disease.
This work reflects a collaboration among Baylor College of Medicine, NYU Langone Health, the NYU Grossman School of Medicine, and many other institutions. Additional support came from organizations such as the Sala Elbaum Pediatric Research Scholars Program (NYU Langone) and the National Cancer Institute. See the publication for the full list of supporters.