Estimating the corrected prevalence of cardio-renal-metabolic conditions in elderly patients using an actuarial methodology

Abstract Background Despite evidence of the utility of incretin therapy across multiple indications, the prevalence of individual and overlapping cardio-renal-metabolic (CRM) conditions among elderly patients remains poorly characterized, limiting a comprehensive estimate of the treatable population. A newer method from actuarial science provides an alternative estimate of disease prevalence in this population. Methods We used data from the 100% sample of United States (US) Medicare fee-for-service beneficiaries age 65 years and older continuously enrolled during 2018–2023 (n=14,939,076) to estimate the prevalence of older adults with coronary artery disease (CAD), heart failure (HF), chronic kidney disease (CKD), diabetes (DM), and clinical obesity (CO) based on associated administrative data codes. Applying the actuarial methodology described by Stocking et al. (2023), we identified two groups: observed individuals—beneficiaries first diagnosed in an index year (2019), and latent individuals—those first diagnosed after the index year (2020–2023). The Stocking et al. correction assumes that latent patients had disease in 2019 but had not been identified. In a separate sample of beneficiaries continuously enrolled in 2023, we calculated prevalence for the five conditions individually and combined, both as observed and as corrected with the Stocking et al. methodology for identification of undiagnosed latent individuals. Results Of the 22,121,159 beneficiaries in the separate 2023 sample constructed, 18.2% were diagnosed with CAD, 9.7% with HF, 16.2% with CKD, 23.0% with DM, and 13.7% with CO, while 46.3% were diagnosed with any of the five CRM conditions. Stocking corrections for underdiagnosis (or ratio of corrected to observed prevalence) ranged from 1.37 for DM to 2.32 for HF. After applying the Stocking corrections, prevalence was 34.7% for CAD, 28.0% for HF, 37.5% for CKD, 31.7% for DM, 27.8% for CO, and 70.0% for any of the five conditions. Conclusion Applying an actuarial methodology to administrative claims data on older adults suggests a 50% higher corrected prevalence of having at least one of five CRM indications, indicating the population of older adults suitable for incretin therapy may be larger than considered previously.