Forecasting Hospitalization Trends in the Greek Elderly Population

Area of Science:

• Gerontology and Public Health
• Health Services Research
• Epidemiology

Background:

• The Greek elderly population is growing, necessitating accurate healthcare demand forecasting.
• Understanding hospitalization trends is crucial for effective public health planning and resource allocation.

Purpose of the Study:

• To forecast all-cause hospitalizations in the Greek elderly population up to 2032.
• To compare the performance of Autoregressive Integrated Moving Average (ARIMA) and Prophet models for this forecasting task.
• To identify age-specific forecasting accuracy for optimizing healthcare resource management.

Main Methods:

• Utilized historical hospitalization data from Greece between 2001 and 2019.
• Applied two distinct time-series forecasting models: ARIMA and Prophet.
• Evaluated model performance based on forecast accuracy across different elderly age cohorts.

Main Results:

• Both ARIMA and Prophet models were employed for forecasting, with ARIMA offering stable, conservative predictions and Prophet providing more flexible trend capture with higher uncertainty.
• Forecasting accuracy was found to differ significantly across age groups.
• The 80+ age cohort exhibited the highest forecasting precision, indicating more predictable healthcare utilization patterns in the oldest segment of the elderly population.

Conclusions:

• A multi-model approach is valuable for healthcare planning, especially when forecasting for aging populations.
• Accurate predictions of hospitalization trends in the elderly are essential for efficient healthcare resource allocation.
• The findings highlight the importance of considering age-specific patterns when forecasting healthcare needs in older adults.