A research team used the Prosthetic Limb Users Survey of Mobility (PLUS-M) to develop and test a prediction model of mobility outcomes for people with lower-limb amputations. After testing, the researchers concluded that their “neighbors-based prediction approach” allowed for accurate estimates of PLUS-M T-score trajectories.
Eight hundred thirty-one patient records (1,860 PLUS-M observations) were used to develop and test the neighbors-based prediction model, using previous patient data to predict the six-month PLUS-M T-score trajectory for a new patient, based on matching characteristics. The prediction model was developed in a training data set (n = 552 patients) and tested in an out-of-sample data set of 279 patients with later visit dates. Prediction performance was assessed using bias, coverage, and precision. Prediction calibration was also assessed.
The average prediction bias for the model was 0.01 standard deviations, average coverage was 0.498 (ideal proportion within the 50 percent prediction interval = 0.5), and the prediction interval was 8.4 PLUS-M T-score points (40 percent improvement over population-level estimates). In the study describing the testing, the authors wrote that the research predictions were well calibrated, with the median predicted scores falling within the standard error of the median of observed scores, across all deciles of the data.
The study, “Development of a physical mobility prediction model to guide prosthetic rehabilitation,” was published in Prosthetics and Orthotics International.
Â
