Development and Validation of PT-PENCIL: The Physical Therapy Frequency Clinical Decision Support Tool to Increase Hospital Discharge to Home.

The study used historical treatment frequency, not assigned treatment, so it cannot show that daily physical therapy caused discharge to home. Patients in neurology, oncology, or orthopedics and patients with intensive care unit stays longer than 48 hours were excluded, so the model may not apply to them. Patients discharged to long-term care facilities were excluded, and some prehospital living settings were unclear, so some home discharges may have been misclassified. Daily physical therapy was not a major predictor in the external validation sample. The model was not tested prospectively as a clinical decision support tool. The model used only information available in the electronic health record, so factors such as staffing, fatigue, or patient personality were not captured. Calibration slightly underestimated the probability of discharge to home for patients with a low probability.

Funding was a pilot grant from the Learning Health System Rehabilitation Research Network, supported by the National Institutes of Health under award number 1P2CHD101895-01. The funder had no role in the design, conduct, or reporting of the study. The authors reported no conflicts of interest. Joshua Johnson is a member of the PTJ Editorial Board.

Do not read the development result that daily therapy was associated with 2.15 times higher odds of going home as proof that daily physical therapy causes discharge to home. The data were retrospective, treatment frequency was not assigned, and daily therapy was not a major predictor in the external validation sample.