Digitalization of the Modified Physical Performance Test

Abstract Physical medicine and rehabilitation clinicians frequently measure the physical functions of their patients with neuromuscular motor dysfunction, such as those recovering from stroke, surgery, or trauma. The Modified Physical Performance Test (MPPT) is a widely accepted measure of physical function. However, the MPPT requires significant time investment from the clinician overseeing the test. Our objective was to create a digital MPPT (dMPPT) based on wearable technology which can be performed without extensive clinician oversight and validate the dMPPT using a clinician directed analogue MPPT (aMPPT). Wearable sensors were developed to capture movements performed in the MPPT. A time score and balance score were calculated for each task. A preliminary clinical trial with ten patients in the inpatient section of a rehabilitation unit was conducted. Patients performed both the aMPPT version without the sensors and the dMPPT with the sensors. We compared scores of the aMPPT with the dMPPT. The Pearson correlation test demonstrated a high (r = 0.99) and significant (p < 0.001) correlation between the aMPPT and dMPPT. Bland–Altman analysis determined that data from all ten participants were within the range defined by the limits of agreement, indicating a high level of practical agreement between the aMPPT and dMPPT. The dMPPT results were significantly correlated and demonstrated a high level of practical agreement with the aMPPT clinician-directed method. Future larger sample clinical trials to validate the dMPPT for clinical use are recommended. With this validation, the dMPPT would allow for more efficient monitoring of patients' physical performance.