The study is a retrospective analysis of longitudinal medical and pharmacy claims data from Symphony Integrated Dataverse (IDV) database. The IDV is an open claims administrative health claims database that contains prescription, medical, and hospital claims across the US for all payment types, including commercial plans, Medicare Part D, cash, assistance programs, and Medicaid. The IDV contains over 10 billion deidentified prescriptions claims linked to over 280 million unique patients with an average of 5 years of prescription drug history. These prescription drug claims are linked to hospital and physician practices claims with medical procedure (i.e., current procedural terminology [CPT] and diagnosis codes [International Classification of Disease Tenth Revision (ICD-10)] for nearly 180 million patients. The full database includes claims from over 65,000 pharmacies, 1,500 hospitals, 800 outpatient facilities, and 80,000 physician practices across the US, capturing approximately 75% of the total prescriptions dispensed in the US. The IDV database was selected due to its longitudinal nature and comprehensive coverage of claims required to meet the study objectives. The distribution of Symphony Health patients across census regions is very similar to that of the US population.

Patients of interest were identified from January 1st, 2017 to April 30th, 2021, with a 12-month index period. The first relevant ICD-10 claim of HCM diagnosis was considered as the index diagnosis date. The date of first treatment with a beta-blocker, calcium channel blocker, disopyramide, or a procedure of interest (e.g., alcohol septal ablation, septal myectomy, pacemaker) after diagnosis date was considered as the index treatment date. Patient demographics and clinical characteristics were captured at baseline and costs and HRU were captured over the follow-up period. The data used in this study were de-identified in compliance with the Health Insurance Portability and Accountability Act.

The patient population for this study consisted of prevalent patients with oHCM identified from the IDV (Figure 1). Adult (Age ≥ 18 years at index treatment date) patients with oHCM who met the following criteria were included in this study (1) ≥two claims of oHCM (ICD-10 diagnosis code: I42.1) at least 30 days apart, or (2) one diagnosis of HCM (ICD-10 diagnosis code: I42.2) along with either a diagnosis of oHCM (ICD-10 diagnosis code: I42.1) at least 30 days apart or a septal reduction therapy procedure any time after HCM diagnosis. Patients were required to have 12 months of activity prior to index treatment date, 3 months of activity post index treatment date, and be symptomatic. Patients were categorized as symptomatic if they had a diagnosis for fatigue, chest pain, syncope, dyspnea, heart failure, palpitations, insertion of a pacemaker, or septal reduction therapy, in the 3 months before or after their index diagnosis date. We excluded patients with a diagnosis of Fabry disease, amyloidosis, and patients with HCM treatment in the 12 months prior to index treatment date (i.e., patients were treatment-naïve at index treatment date). Patients were followed from their index treatment date until the end of index treatment due to either (1) discontinuation, (2) treatment switch, (3) treatment augmentation, or (4) end of activity in the database.

Patient demographics and comorbidities, included the Charlson Comorbidity Index, were calculated 12 months prior to index date (excluding index date). All-cause and HCM-related HRU and costs were evaluated. Healthcare resource utilization and costs were defined for inpatient admissions, outpatient visits, ER visits, urgent care visits, other visits, and pharmacy costs (HCM-related prescriptions and non-HCM related prescriptions). HCM-related outcome variables included in this analysis were predefined based on the AHA/ACC treatment guidelines and confirmed by clinical experts who care for patients with HCM (13). Regarding the financial data available in the IDV, final adjudicated costs were only available for all-cause related prescriptions. For all other cost categories including HCM-related prescriptions, the amount billed by the payer (charged amount) is reported.

Descriptive statistics were presented as means and standard deviation (SD), median and interquartile range (IQR) for continuous variables, and frequencies and proportions for categorical, and dichotomous variables. Baseline characteristics were compared with Chi-squared test for categorical variables or Kruskal–Wallis test for continuous variables. Systematic differences between sex were evaluated by Kruskal–Wallis for continuous variables and Chi-square tests for categorical variables. Generalized linear models with a gamma distribution and log link clustered on the patient was used to estimate per-person-per-year (PPPY) costs and generalized estimating equations with a negative binomial distribution clustered on the patient was used to estimation PPPY visits. The mean difference in PPPY costs/visits, 95% confidence intervals (CIs), and p-values were estimated for the patient cohort and by sex. These estimates were evaluated for each healthcare setting (inpatient, outpatient, ER, urgent care, other medical visits, and prescriptions).

An adjusted analysis of all-cause and HCM related HRU and costs was conducted. This included generalized linear models to estimate adjusted healthcare costs and generalized estimating equations to estimate adjusted healthcare visits. Covariates in the adjusted models were selected based on differences seen in a univariate analysis of baseline variables between males and females (Table 1). These included the baseline patient values of age, region, insurance type, hypertension, atrial fibrillation and flutter, heart failure, ventricular fibrillation, ventricular tachycardia, chronic pulmonary disease, obesity, valvular disease, stress cardiomyopathy, and coronary artery disease. Missing or unavailable data were not included in the analyses.

Among 9,490 patients with symptomatic oHCM, 5,309 (55.9%) were female and male patients had longer length of follow-up time (Table 1). Female patients were older (64 ± 13 vs. 59 ± 14; p < 0.0001) compared to males, respectively. Regardless of sex, the majority of patients in this cohort were 65 years of age or older (N = 4,726, 99.6%). Female patients also had a higher Charlson Comorbidity Index (1.9 vs. 1.7; p < 0.0001) compared to males, respectively. Baseline patient clinical characteristics for this cohort of patients with symptomatic oHCM are summarized in Table 2.

After adjusting for patient characteristics, female patients had significantly greater number of HCM-related hospitalizations (0.24 vs. 0.20 PPPY, p = 0.0014), LOS (5.08 vs. 4.30 PPPY; p = 0.0235), number of outpatient visits (4.98 vs. 4.59 PPPY; p = 0.0387), and number of distinct drugs (0.59 vs. 0.55 PPPY; p = 0.0010), compared with males, respectively (Table 3). Additionally, female patients had significantly greater number of all-cause hospitalizations (0.96 vs. 0.82 PPPY, p = 0.0002) and number of distinct drugs (7.34 vs. 6.31 PPPY; p < 0.0001), after adjusting for patient characteristics (Table 4). Unadjusted HCM-related and all-cause HRU are presented in Supplementary Table S1 and Supplementary Table S2, respectively.

Among symptomatic oHCM patients, total all-cause costs were greater compared to HCM-related costs, respectively ($51,835 vs. $23,048).     In adjusted models, only HCM-related pharmacy costs were significant, with female patients having slightly higher costs compared to males ($70 vs. $61 PPPY; p = 0.0465; Table 3). There were no significant differences in all-cause cost of care between male and female patients with oHCM (Table 4). Unadjusted HCM-related and all-cause costs are presented in Supplementary Table S1, and no significant differences in any all-cause costs categories between male and female patients with symptomatic oHCM (Supplementary Table S2).

This study is subject to several limitations, which are common across claims database analyses. The diagnoses, comorbidities, HRU and costs of patients with HCM were identified based on ICD-10-CM diagnosis code. The presence of a diagnosis code on a medical claim does not necessarily indicate a positive presence of disease because the medical record may have been incorrectly coded or included as a rule-out criterion rather than the actual disease. Also, diagnosis codes only signify the presence of the disease and do not detail the characteristics or the nature of the disease as you would find in electronic medical record data. This limitation was overcome by requiring eligible patients to have at least two claims with diagnosis codes for HCM. It was also ensured that generic codes, such as codes for “other cardiomyopathy” and “unspecified cardiomyopathy,” which could be used for HCM, were not included in the identification of HCM patients for the study.

Diagnosis codes were used to identify patients with symptomatic oHCM, and a combination of symptoms, comorbidities, and procedures was used to identify symptomatic oHCM in this study. Since the claims database does not have a record of all symptoms of a patient, and some of the symptoms could be attributed to comorbidities, there is the possibility of difference in actual proportion of patients with symptomatic oHCM and the estimates in this study. Additionally, due to restrictions of the database regarding HIPAA policies, we are unable to report separate costs for surgical procedures including septal reduction therapy. Furthermore, without patient level medical record data, the question remains whether these differences in HRU (including adjusted analyses) are driven primarily by HCM or the additive comorbidity burden. Lastly, the definition of costs in this analysis was charges, meaning the amount billed by the payer. This may not be reflective of what a patients pays for their cost of care, but what the charged amount is by the payer in the U.S. healthcare system.