Innovative Algorithm for Incidence of Leukemia and Lymphoma in the U.S. Military Health Care System.

Abstract

Introduction: There is limited research on cancer incidence in pediatric and adolescent/young adult patients using health care claims data and no standard algorithm for such a query. This study aimed to establish an algorithm to estimate incidence rates for multiple cancer types from 2013 to 2017 within the Military Health System and compare them to rates reported from the National Cancer Institute Surveillance, Epidemiology, and End Results (SEER) program.

Methods: The Military Health System Data Repository was queried by International Classification of Diseases 9 and 10 codes using look-back-periods to identify incident cases of leukemia and lymphoma diagnoses. Age-specific incidence rates within the Military Health System were compared to SEER incidence rates using standardized incidence ratios (SIRs).

Results: Military Health System incidences derived from our algorithm were similar to SEER incidences for all evaluated cancers in primary analyses. For pediatric dependents: Acute lymphoblastic leukemia (ALL) SIR 1.01 (95% CI, 0.91-1.12), acute myeloid leukemia (AML) 1.10 (95% CI, 0.86-1.36), Hodgkin's lymphoma 0.93 (95% CI, 0.73-1.16), and non-Hodgkin's lymphoma (NHL) 1.07 (95% CI, 0.88-1.28). For adult dependents: ALL SIR 1.09 (95% CI, 0.79-1.43), AML 1.19 (95% CI, 0.86-1.59), Hodgkin's lymphoma 1.19 (95% CI, 1.00-1.40), and NHL 0.85 (95% CI, 0.66-1.06). For active duty patients: ALL SIR 1.38 (95% CI, 0.96-1.87), AML 1.35 (95% CI, 0.94-1.84), Hodgkin's lymphoma 1.23 (95% CI, 1.01-1.48), and NHL 0.88 (95% CI, 0.67-1.12).

Discussion: Pediatric incidence rates were similar when compared to SEER rates for leukemia and lymphoma type, suggesting that this algorithm may be used to capture the burden in the military pediatric population and serve as a research tool to evaluate cancer epidemiology and patient outcomes.