Predictive Capacity of COVID-19 Test Positivity Rate

Area of Science:

• Epidemiology
• Public Health
• Biostatistics

Background:

• COVID-19 spread is often silent, with data lacking for asymptomatic to mild cases.
• Existing hospitalization data is reliable but doesn't capture the full scope of infection.
• Accurate forecasting tools are crucial for healthcare planning and resource allocation.

Purpose of the Study:

• To investigate the correlation between a new Test Positivity Rate (TPR) formulation and hospitalization data.
• To develop a reliable forecasting model for COVID-19 related hospital admissions.
• To provide decision-makers with a tool for predicting future healthcare needs.

Main Methods:

• Utilized a novel Test Positivity Rate (TPR) formulation.
• Applied the Seasonal Auto Regressive Moving Average (SARIMA) statistical model.
• Employed stochastic processes theory for rigorous analysis and forecasting.

Main Results:

• A strong lagged correlation was found between the standardized TPR index and hospitalized patient numbers.
• The SARIMA model provided reliable forecasts of hospital and intensive care unit admissions approximately 12 days in advance.
• The standardized TPR index proved to be a simple yet accurate metric for monitoring epidemic spread.

Conclusions:

• The standardized TPR index is a valuable tool for monitoring COVID-19 epidemic growth.
• This metric enables accurate, daily-based forecasting of hospital and intensive care unit demand.
• The proposed approach offers an optimal balance of simplicity and accuracy for predicting healthcare system strain.