Algorithms for Estimating Suppressed Disease Data

This technical report presents an analysis of algorithms designed to estimate suppressed disease rate values caused by data suppression practices, particularly in the context of public health data. The report outlines the impact of data suppression on the accurate representation of disease loads, especially in rural areas, where small population sizes lead to greater biases. It details three distinct algorithms that utilize demographic adjustments and area illness risk to estimate suppressed data values, highlighting the trade-offs between implementation complexity and accuracy. The algorithms are applied to synthetic heart disease mortality data at the county level, demonstrating how integrating regional risk can improve estimation accuracy. The report emphasizes the importance of understanding these methodologies to mitigate biases introduced by data suppression, which is crucial for maintaining the integrity of health data while complying with privacy regulations such as HIPAA. Overall, the work aims to contribute to the field by providing techniques that enhance the reliability of disease rate estimations in public health research.