Methodological considerations for interrupted time series analysis in radiation epidemiology: an overview

Area of Science:

• Epidemiology
• Biostatistics
• Public Health

Background:

• Radiological events pose significant public health challenges.
• Evaluating health outcomes in affected populations requires robust analytical methods.
• Interrupted time series analysis (ITSA) is a valuable tool for assessing population health trends post-event.

Purpose of the Study:

• To discuss the strengths and limitations of ITSA for evaluating health outcomes after radiological events.
• To highlight methodological issues, including bias, confounding, data quality, and statistical aspects.
• To provide recommendations for enhancing the robustness of ITSA studies.

Main Methods:

• Utilizes aggregated time series data to analyze health indicators.
• Compares pre-event trend projections with post-event observed data.
• Assesses statistical significance of discrepancies between projected and observed trends.

Main Results:

• ITSA can detect statistically significant changes in health trends post-radiological events.
• The method is susceptible to bias and confounding, necessitating careful methodological considerations.
• Data quality and statistical modeling decisions significantly impact study robustness.

Conclusions:

• ITSA is a powerful method for public health surveillance after radiological incidents.
• Addressing potential biases and ensuring data quality are crucial for valid ITSA results.
• Recommendations are provided to improve the reliability and reduce spurious findings in ITSA studies.