Аннотация
A BSTRACT Background: Cardiovascular diseases (CVDs) remain the leading cause of mortality worldwide, representing a major public health challenge. Understanding temporal trends and forecasting future mortality are essential for effective planning and prevention strategies. Objective: This study aimed to analyze temporal trends in cardiovascular mortality and to forecast future mortality using time-series modeling based on publicly available data. Materials and Methods: Annual cardiovascular mortality data for the period 2000–2023 were obtained from the Centers for Disease Control and Prevention WONDER database. The dataset was structured as a univariate time series. An AutoRegressive Integrated Moving Average (ARIMA) model was applied to evaluate temporal patterns and generate forecasts. The data were divided into training (2000–2018) and test (2019–2023) sets for model validation. Model performance was assessed using root mean square error, mean absolute error, and mean absolute percentage error. Results: Cardiovascular mortality demonstrated a declining trend from 2000 to 2015, followed by a gradual increase and a sharp rise in 2020–2021. The ARIMA model showed acceptable predictive performance under stable conditions but underestimated mortality during periods of abrupt change. Forecasting results indicated a relatively stable trend for 2024–2028, with mortality levels remaining elevated compared to pre-2020 values. Conclusion: Time-series modeling using ARIMA provides a useful framework for analyzing and forecasting cardiovascular mortality trends. However, its predictive performance may be limited during periods of sudden disruption. Continuous monitoring and advanced modeling approaches are needed to improve forecasting accuracy and support public health decision-making.