Prospective public-health surveillance

Health-related applications use prospective scan-based methods to detect increased rates of mortality or morbidity and to detect bioterrorism or active clusters of disease, which raises the question of how such methods should be evaluated [1].

During the COVID-19 pandemic, decision makers needed to monitor several characteristics simultaneously, such as transmission, infection, recovery and mortality rates, in order to implement and evaluate prevention and treatment strategies [2, 3].

Prospective scan-based methods are reviewed and related to the moving average chart used in industrial applications, and issues in evaluating spatiotemporal scan methods are discussed [1].

A Susceptible, Exposed, Infected, Recovered, Death (SEIRD) model is estimated regularly with an augmented particle Markov chain Monte Carlo scheme, and its posterior samples are monitored with a multivariate exponentially weighted moving average (MEWMA) chart [2].

MEWMA and multivariate CUSUM (MCUSUM) charts are applied to the parameters of a Susceptible, Exposed, Infected, Recovered, Death and Vaccination (SEIRDV) model, with an augmented particle MCMC scheme that enables real-time monitoring [3].

A clarified definition of the recurrence interval for surveillance methods [1], and a framework that links epidemic-model estimation with multivariate statistical process monitoring of the model parameters [2, 3].

The recurrence interval does not reflect some important aspects of the statistical performance of scan-based and other surveillance methods [1].

The approach was illustrated on COVID-19 data for the State of Qatar [2]; MEWMA and MCUSUM charts were effective for monitoring SEIRDV model parameters, helping to assess interventions and track the impact of emerging variants, and the estimation scheme offered improved accuracy and robustness compared with traditional approaches [3].

Detection of increased mortality or morbidity, bioterrorism and active disease clusters [1], and monitoring of the COVID-19 pandemic in Qatar to support public-health decision-making [2, 3].