OBJECTIVES: Markov models characterize disease progression as specific health states based on clinical measures. Heart failure (HF) has a substantial impact on costs and patients’ quality-of-life. This study aimed to estimate the cost-effectiveness analysis and budget impact (BI) using real-world data (RWD) to define cost and health states in a local setting from the Brazilian private payer perspective (BPPP).

METHODS: Data derived from clinical trials (CONFIRM-HF) were used to develop a model for predicting HF hospitalization rates and NYHA class distribution over 52 weeks. RWD from BPPP was used to estimate risk of hospitalization and specific costs. An algorithm was used to cluster all patients (n=4,246,930) and the following prognostic characteristics: age, hospitalization rates, hospital readmission, Cardio-intensive Care Unit rates, New York Heart Association (NYHA) class, outpatient medication rate, and inpatient rates. CEA and BI were carried out considering a time horizon of 52 week and five years, respectively. Data was used as for target population to be treated, medical direct costs and drug costs.

RESULTS: Cluster Analysis was carried out on 10,657 patients with HF. The class cluster was NYHA I (n=2,458), NYHA II (n=5,085), NYHA III (n=945), NYHA IV (n=2,169). The weekly inpatient cost was NYHA I (BRL 72,51), NYHA II (BRL 219,94), NYHA III (BRL 337,57), NYHA IV (BRL 530,24). The weekly outpatient cost was NYHA I (BRL 81.01), NYHA II (BRL 84.20), NYHA III (BRL302.88), NYHA IV (BRL269.36) The Incremental Cost Effectiveness Ratio (ICER) was dominant BRL 2.675,29/QALY. The BI of introducing FCM shows a potential cost-saving of BRL 36.380.871 over five years.

CONCLUSIONS: The algorithm used for clustering was effective and showcased a substantive savings of BRL 36.380.871 over 5 years. Ferric Carboximaltose proved to be a cost- effective option with an ICER below the Brazilian threshold of BRL 30-40K/QALY.