1978

Development of APACHE rating of illness severity: William Kraus and colleagues at George Washington University secure funding to develop first comprehensive severity of illness scoring system, the Acute Physiology and Chronic Health Evaluation Score (APACHE, followed by the more refined APACHE II). APACE was an early AI-like approach that inputted patient data points into a simple algorithm that generated a score linked to mortality prediction. It proved remarkably accurate, correctly predicting mortality rates up to 86% of the time. In the 1980s, APACHE II scores were also used to assess quality of care across hospitals by comparing a hospital’s predicted and actual mortality rates via a mortality ratio (actual mortality over predicted mortality), with a mortality ratio below 1.0 indicating a hospital outperformed expectation, e.g., of survival rate of ICU patients. Of course, APACHE data analysis lacked real-time clinical awareness. It also engendered a backlash from clinicians, who felt computational tools infringed on physician autonomy (Shahinpoor, 198-202).