POS1428 IDENTIFICATION OF A WARM AUTOIMMUNE HEMOLYTIC ANEMIA (wAIHA) POPULATION USING PREDICTIVE ANALYTICS OF A KNOWN CLINICALLY PROFILED COHORT

Autor: K. R. Mccrae, T. Gooljarsingh, G. K. Jones, M. L. Tjoa
Rok vydání: 2022
Předmět:
Zdroj: Annals of the Rheumatic Diseases. 81:1057.1-1057
ISSN: 1468-2060
0003-4967
Popis: BackgroundThe disease burden of warm autoimmune hemolytic anemia (wAIHA), a rare disorder characterized by the destruction of red blood cells (RBCs) by pathogenic IgG autoantibodies, has not been well studied. At the time this claims database analysis was carried out a diagnostic code specific for wAIHA did not exist, which presented challenges in identifying patients with wAIHA.ObjectivesDespite the absence of a diagnostic code specific for wAIHA we aimed to 1) identify a wAIHA cohort using a collection of diagnostic codes, 2) bisect severe versus non-severe wAIHA patients, 3) compare the frequency of comorbidities, anemia symptoms, treatments, diagnostic tests, and healthcare provider visits in these two groups, and 4) use a predictive model to validate clinical variables and prevalence estimates.MethodsA de-identified, longitudinal, patient-level claims database of >300 million US patients was used for this study. Patients with wAIHA were identified based on a diagnosis code of “autoimmune hemolytic anemia” (AIHA), chronic use of steroids (≥30 days) in the last 36 months, and chronic use of non-steroidal immunosuppressants. Patients were classified as severe if claims related to transfusion, high frequency of blood testing, high frequency interactions with a hematologist, and/or ≥2 ER visits per year were observed in the 36-month period. Codes for anemia symptoms, comorbidities, treatments, and diagnostic tests were grouped and analyzed within the most recent 12 months for each patient. Prevalence was estimated using Artificial Intelligence/Machine Learning (AI/ML) lookalike modeling, using known patients with wAIHA as the positive training class.ResultsA cohort of 1,548 patients with wAIHA was identified (n= 631 were classified as severe while n= 917 as non-severe). Median patient age was >65 years, and patients were evenly distributed by gender. The rate of disease-relevant claims was disproportionately higher in the severe cohort versus the non-severe cohort. Over the 12-month study period, there was a 61% higher rate for anemia symptomatology codes and a 570% higher rate for wAIHA specific testing and monitoring codes in the severe cohort. Primary hypertension, hyperlipidemia, gastro-esophageal reflux, and evidence of chemotherapy use were also present in wAIHA patients. All these conditions were observed more frequently in severe patients with the exception of lupus. Almost 44% of wAIHA claims for the full cohort were associated with Hospital/Emergency care - 48% for the severe group. AI/ML modeling predicted patients using claims variables for hemolytic anemia, other blood count abnormalities, and medical procedure claims commonly used for the diagnosis and management of wAIHA. The predicted population supports reported US prevalence estimates of 30,000-49,000 patients.ConclusionWe developed and validated a method for defining wAIHA patients using de-identified claims data based on AIHA ICD-10 codes and relevant treatments. We observed that while disease manifestations are generally the same in the severe and non-severe wAIHA cohorts, there is an increased rate of occurrence in the severe cohort during the same 12-month period. This increase was also associated with higher utilization of healthcare resources. The comorbidity of lupus was more commonly associated with non-severe wAIHA patients. This may indicate that patients with a known diagnosis (in this case lupus) who are more closely monitored are less likely to reach the level of severity that would categorize them as patients with severe wAIHA.Disclosure of InterestsKeith R McCrae Consultant of: Momenta and Novartis, Tricia Gooljarsingh Employee of: Momenta and Janssen (previously), Graham K Jones Consultant of: Momenta, Employee of: IPM AI, May Lee Tjoa Employee of: Janssen
Databáze: OpenAIRE
Externí odkaz: