# Comprehensive Geriatric Assessment as a predictor model for post-operative delirium in older adults undergoing major non-cardiac elective surgery

## TL;DR

This study shows that age, cognitive status, and nutrition can predict delirium after surgery in older adults.

## Contribution

Develops a prediction model for post-operative delirium using CGA factors in older surgical patients.

## Key findings

- Post-operative delirium incidence was 6.8% in older adults undergoing non-cardiac surgery.
- Age, cognitive status, and nutritional status were significant predictors of delirium.
- The prediction model showed good calibration and moderate performance with an AUC of 0.750.

## Abstract

The needs for surgery in older adult populations is increasing every year. Postoperative delirium is one the most common complications and will impact many adverse outcomes. Comprehensive Geriatric Assessment (CGA) and perioperative risk stratification of older adults are needed as an initial prevention strategy as well as an efficient and applicable prognosis predictor model.

This study aims to determine the incidence of post-operative delirium and develop a prediction model for delirium in older adults after major non-cardiac elective surgery based on predictor factors.

This research is a retrospective cohort study using secondary data from medical records of older adult inpatients who underwent major elective non-cardiac surgery at Cipto Mangunkusumo Hospital between January 2020 and March 2023. Data analysis using SPSS 20.0 and STATA 10. Development of a prediction model for post-operative delirium complications using the Hosmer- Lemeshow test and Area Under the Curve of the Receiver Operating Characteristic (AUC ROC).

Total of 370 subjects that met the criteria were analyzed. The incidence of post-operative delirium was 6.8%. The predictor factors analyzed were age (HR = 3.43; 95% CI 1.544–7.635), cognitive status (HR = 2.74; 95% CI 1.156–6.492), and nutritional status (HR = 3.35; 95% CI 1.459–7,679). The postoperative delirium complication prediction model had good calibration (p > 0.05) and moderate score performance for predicting the incidence of delirium in older adults [AUC 0.750 (p < 0.001; 95% CI 0.640–0.860)].

Age, cognitive status, and nutritional status are strong predictors of postoperative delirium in older adults undergoing major non-cardiac elective surgery. The postoperative delirium prediction model has good calibration and moderate score performance.