Ascertainment of aortic valve disease using administrative claims

Abstract

Abstract Background Administrative claims may be useful for characterizing patients with aortic stenosis (AS) and aortic regurgitation (AR) and estimating disease prevalence. However, the accuracy of diagnostic codes for aortic valve disease has not been well studied. Purpose To evaluate the validity of International Classification of Diseases, 10th Revision (ICD-10) codes for identification of AS and AR. Methods Using a large, transthoracic echocardiographic (TTE) report dataset linked to Medicare Fee-for-service (FFS) claims, 2017–2018, the performance of candidate claims to ascertain AS/AR status using standard TTE definitions was evaluated. Recursive partitioning with 10-fold cross validation was used to build the optimal prediction tree for AS/AR status using all ICD-10 codes as candidate predictors. The optimal performing claims algorithm was tested against patient outcomes in a separate 100% sample of Medicare FFS inpatient and outpatient claims, 2017–2019. Results Of those included in the derivation dataset (N=5497, mean age 74.4±11.0 years, 49.7% female), any AS or AR was present in 24% and 38.8%, respectively. The code I35.0 was optimal for identification of any AS with a sensitivity and specificity for any AS of 53.1% and 94.8%, respectively (Table 1). Amongst those with an I35.0 code, 40.3% had severe AS. Claims were unable to distinguish disease severity (i.e. severe vs. non-severe) or subtypes (e.g. bicuspid or rheumatic AS), and were insensitive and nonspecific for AR of any severity. Among all Medicare beneficiaries who received an TTE (N=3,783,249), those with an I35.0 code, compared to those without, had a higher risk of all-cause mortality (HR 1.65, 95% CI 1.63–1.67), heart failure hospitalization (HR 2.17, 95% CI 2.11–2.24), and aortic valve replacement (HR 32.35, 95% CI 31.46–33.27) (Table 2). Conclusions Amongst those receiving TTE, the ICD-10 code I35.0 in any position was optimal for identification of AS and identified a population at significant greater risk of all-cause mortality, heart failure hospitalization, and receipt of aortic valve replacement. Though 40.3% of those with I35.0 had severe AS, claims were unable to distinguish disease severity of subtype. Claims may be feasibly used to identify those with AS who may be at risk for adverse valve-related cardiovascular events and require future treatment. Funding Acknowledgement Type of funding sources: Public Institution(s). Main funding source(s): National, Heart, Lung, and Blood Institute