2023 Volume 3 Issue 1 Pages 14-20
Objective: External validation of the model is an important step toward clinical implementation. This study aimed to externally validate the prediction model for gait independence at discharge in older patients who had suffered a hip fracture.
Methods: In this retrospective cohort study, we collected data from 163 older patients with hip fractures who were admitted to convalescent rehabilitation wards. Discrimination and calibration were examined in a validation cohort of a model that predicts gait independence at discharge based on cognitive function and balance at admission.
Results: Compared with the development cohort, the area under the curve was comparable in discrimination, with higher sensitivity and lower specificity. Using calibration plots, the predicted and observed probabilities were consistent for patients with a high probability of gait independence. However, for patients with a moderate to low probability of gait independence, the prediction was an underprediction, with a higher observation probability than the prediction probability.
Conclusion: External validation of the prediction model for gait independence at discharge after hip fractures resulted in acceptable performance, despite some calibration problems. These results suggest that the predictive model can be used for older patients with hip fractures who are in convalescent rehabilitation wards.
