Then, Now, and Into the 21st Century
Published in: Injury, Volume 50, Issue 1, pages 2–3 (January 2019). doi: 10.1016/j.injury.2018.11.002
Posted on RAND.org on January 21, 2021
Accurately predicting the outcomes of injury is an important aspect of the clinical and administrative management of injured patients. But are we close to defining the perfect predictor? Is a perfect predictor even possible?
The urge to prognosticate the outcomes from injury at least as old as the history of medicine. Four thousand years ago the Egyptian physician Imhotep categorized injuries as survivable ("I can heal"), possibly survivable ("I will fight with"), and unsurvivable ("cannot be healed"). In the 20th century continuous measures of the probability of death were developed, but the approach remained unchanged: using past experience to predict outcomes for future patients. In this essay we review the recent history of injury outcome prediction and suggest how the future may play out. We will find that the measure we are currently using (ISS) has outlived its usefulness because that newer, better measures are available. Indeed, it is likely that modern outcome models are about as good as they can be. Further improvements will have to come from improved approaches to injury description and capture.