Model Development & Scientific ValidationDevelopment of a Continuous Measure of Patient Acuity Across All Diseases, Conditions
The Rothman Index (RI) is at the forefront of predictive analytics in healthcare. The RI is the only general measure of patient condition, and is integral to an array of effective predictive analytics in healthcare. It leverages the investment hospitals have made in electronic health records (EHRs), utilizing existing, real-time data to quantify and visualize patient acuity and risk.
The RI is based on a heuristic model that uses a range of physiological measures, including labs, vitals, and most importantly nursing assessments. The model transforms each input into a common representation of univariate risk, allowing heterogeneous data to be summed, solving the data fusion problem. The result is a continuous measure of patient condition, integrated into the EHR, computed on a real-time basis across all conditions, diseases and care settings.
The RI correlates well with many measures of risk, including:
24-hr Mortality (AUC = 0.93)
Unplanned Transfers to the ICU
Code Blue Events
Length of Stay
Rothman Index has been presented in over 45 oral and poster presentations
The Rothman Index continues to play a leading role in digital health — from analytics to patient care.
- In hospital: Analytics and machine learning can be applied to longitudinal view of RI graph patterns, providing a new window into understanding of disease progression, treatment effectiveness, and population health.
- In home: statistically validated patient acuity score and visualization software can reach beyond the acute care setting and into the home for improved monitoring and actionable clinical alerts.