Potential Biases in Estimating Absolute and Relative Case-Fatality Risks during Outbreaks.

Area of Science:

• Epidemiology
• Infectious Disease Dynamics
• Biostatistics

Background:

• Accurate estimation of case-fatality risk (CFR) is vital for emerging and re-emerging infectious diseases.
• Existing surveillance data, often collected for other purposes, presents challenges for precise CFR calculation.
• Understanding biases in CFR estimation is critical for effective public health response.

Purpose of the Study:

• To identify and describe key biases affecting overall CFR estimation.
• To review proposed and implemented solutions for mitigating these biases in past epidemics.
• To explore biases in estimating the causal impact of interventions on survival and discuss mitigation strategies.

Main Methods:

• Description of two primary biases in overall CFR estimation: preferential ascertainment of severe cases and reporting delays.
• Review of historical epidemic data and proposed solutions for bias reduction.
• Identification of additional biases (confounding, survivorship, selection) in intervention impact analysis using observational data.

Main Results:

• Preferential ascertainment of severe cases and reporting delays can significantly skew CFR estimates.
• Observational data for intervention impact analysis is prone to confounding, survivorship, and selection biases.
• Systematic cohort ascertainment, such as through contact tracing, offers a method to reduce bias.

Conclusions:

• Careful consideration of biases is essential for accurate CFR estimation and causal inference regarding interventions.
• Strategies to mitigate bias include improved data collection and the use of systematically defined cohorts.
• Non-causal interpretation of risk factors for death requires understanding potential biases in observational studies.