Rate of Bleeding-Rated Episodes (BREs) in Elderly Patients with Primary Immune Thrombocytopenia (ITP): A Population-Based Retrospective Cohort Study Using Medicare 20% Sample Data

Background: ITP is a rare disorder characterized by low platelet counts and a tendency toward increased bleeding and bruising. Several studies have shown that ITP incidence increases with age. Elderly ITP patients may have a higher risk of bleeding and their disease management can be challenging due to increased burden of other health problems. However, little information is available regarding rates of BREs in elderly ITP patients. As the population ages, understanding the frequency of bleeding events and/or use of ITP therapy to treat or prevent bleeding in routine clinical practice could provide insights into the real-world burden of this disease in elderly patients.

Methods: Using Medicare 20% sample data 2007-2012, we identified elderly Medicare fee-for-service (FFS) enrollees  newly diagnosed with primary ITP (ICD-9-CM diagnosis code 287.31) at age ≥67 years (to ensure ≥2 years of Medicare FFS enrollment before their first claim with an ITP code to determine study eligibility), between January 1, 2009, and September 30, 2012. BREs were defined as ≥1 bleeding event (of any severity) and/or use of ITP therapies commonly considered for rescue or emergency therapy (platelet transfusion, intravenous immunoglobulin [IVIg], anti-D, or IV steroids). Claims with relevant codes with dates of service separated by ≤3 days were considered a single BRE. The rate of BREs (per person-year) was calculated for all BREs and by event type incorporated into BREs for the ITP cohort overall and stratified on ITP phase (newly diagnosed: 0 to <3 months; persistent: 3-12 months; chronic: >12 months). ITP phase was defined based on ITP disease duration, which was calculated as the time between initial ITP diagnosis and last ITP claim. Patients were followed from ITP diagnosis until the last ITP claim (end of ITP duration), death, disenrollment from FFS coverage, or December 31, 2012, whichever came first.

Results: We identified 3007 elderly patients with primary ITP (mean [SD] age: 79.6 [7.5] years; 55% female) in the database of 12.8 million enrollees. The median (IQR) ITP duration was 6 (1, 17) months; 37%, 30%, and 34% of patients had ITP duration <3, 3-12, and >12 months, respectively. During 2687 person years follow-up, 1806 patients (60%) experienced at least one BRE for a total of 6305 BREs; 42% were bleeding events only, 50% were therapy only, and 8% were both. BRE rates (95% CI) per person year were 2.35 (2.29-2.41) for any BREs and 0.99 (0.95-1.03), 1.18 (1.14-1.22), and 0.18 (0.16-0.20) for BREs with bleeding only, therapy only, and both, respectively. The most common bleeding types were gastrointestinal hemorrhage, hematuria, epistaxis, and ecchymoses. Intracranial hemorrhage was reported in 107 patients (3.6%). Newly diagnosed patients had higher BRE rates (Table).

	Newly diagnosed ITP (n=3007)		Persistent ITP (n=1907)		Chronic ITP (n=1010)
BREs	Count	Rate per person year (95% CI)	Count	Rate per person year (95% CI)	Count	Rate per person year (95% CI)
All	2447	4.46 (4.28,4.63)	2196	2.05 (1.96,2.14)	1662	1.56 (1.49,1.64)
With bleeding only	1087	1.98 (1.86,2.10)	835	0.78 (0.73,0.83)	733	0.69 (0.64,0.74)
With therapy only	1022	1.86 (1.75,1.98)	1262	1.18 (1.11,1.24)	884	0.83 (0.78,0.88)
With both bleeding and therapy	338	0.62 (0.55,0.68)	99	0.09 (0.07,0.11)	45	0.04 (0.03,0.05)

Conclusion: We provided current real-world estimates of BRE rates in elderly patients with ITP. About 60% of ITP patients experienced at least one BRE, and half of all BREs were defined by therapy use alone. This demonstrates the importance of examining both bleeding and use of rescue or emergency ITP therapy in the assessment of disease burden in elderly patients with ITP.