Risk factors for and a prediction nomogram of physical frailty in older patients hospitalized with acute calculous cholecystitis

Abstract

Frailty increases the risk of complications and delays recovery in older patients with acute calculous cholecystitis (ACC). Early identification is crucial to improving outcomes. Subjects were 386 older inpatients with ACC at two hospitals who were randomly divided into a training set (n = 270) and validation set (n = 116). Patients were categorized into frail and non-frail groups. Binary logistic regression identified significant predictors, and a nomogram was developed. The incidence of frailty was 27% (n = 73). Smoking, a sleep disorder, impaired ADL, and malnutrition were independent predictors for frailty (p < 0.05). The nomogram showed good discrimination (AUC = 0.752), with a sensitivity of 82.6% and a specificity of 67.4%. Calibration was acceptable (Hosmer-Lemeshow χ² = 4.407, p = 0.732), and decision curve analysis demonstrated clinical utility. The developed nomogram reliably predicts the risk of frailty in older patients with ACC and may facilitate targeted early interventions in clinical practice.

Keywords: acute calculous cholecystitis; elderly; influencing factors; physical frailty; prediction model.

Figure 1.

Nomogram to predict the risk of physical frailty in older patients with acute calculous cholecystitis. This nomogram includes smoking, exercise, multimorbidity, sleep disorder, polypharmacy, depression, sleep disorders, ADL, and malnutrition. The horizontal scale labeled "Points" reflects the impact of each variable. A line was drawn up to the points axis for each variable. The total score was calculated by summing all of the variables. Then, the probability of non-frailty and frailty was determined by drawing a line down from the total points axis to the horizontal axis "Risk of non-frailty" and "Risk of frailty" below.

Figure 2.

The receiver operator characteristic (ROC) curve for the model to predict the risk of frailty. (A) Training set; (B) Validation set.

Figure 3.

Calibration curve for the frailty risk model in training and validation sets of older patients with acute calculous cholecystitis. (A) Calibration curves for the training group show that the apparent curve closely matches the ideal curve, indicating predictive probability. (B) Calibration curves for the test group also show that the apparent curve aligns well with the ideal curve, confirming the model's strong predictive performance. Note: The X-axis represents the predicted possible risk of physical frailty in patients with ACC. The Y-axis represents the actual diagnosed ACC. The dashed line represents the original performance, and the solid dashed line represents the performance during internal validation by Bootstrapping (B =1,000 repetitions).

Figure 4.

Decision curve analysis (DCA) of the nomogram to predict frailty risk in training and validation sets.