Epidemiology in anaesthesia: a method for predicting hospital mortality

Area of Science:

• Anesthesiology
• Medical Informatics
• Health Services Research

Background:

• Anesthetic records provide a valuable data source for understanding surgical outcomes.
• Identifying predictors of in-hospital mortality is crucial for patient care and resource allocation.

Observation:

• Analysis of 108,878 anesthetic records from 13,043 operations identified nine surgical groups with high in-hospital mortality.
• A logistic regression model was developed to predict the probability of death.

Findings:

• Key variables influencing mortality included age, sex, operation type (elective/emergency), and five intercurrent diseases.
• Age emerged as the most significant predictor in six out of nine high-mortality surgical groups.
• The developed model effectively calculated individual patient risk of in-hospital death.

Implications:

• This predictive model can aid clinicians in assessing patient risk and informing treatment decisions.
• Further refinement could lead to broader clinical application in predicting surgical outcomes.
• The findings highlight the importance of age as a critical factor in surgical mortality risk stratification.