Improving the Delivery of Function-Directed Care During Acute Hospitalizations: Methods to Develop and Validate the Functional Assessment in Acute Care Multidimensional Computerized Adaptive Test (FAMCAT).

This paper reports instrument development and selected psychometric checks, not a test of whether FAMCAT changes therapy, discharge or patient function. Validation designs for discharge location, readmission, convergent validity, mode and proxy effects are described, but their results are not reported. Participant counts differ across the text and tables, for example 2341 calibration patients recruited, 2024 in a table, and 2154 validation patients versus 2050 in a table. Therapist assessments occurred for only 15% and 12% of participants, so cut scores relied mainly on nurse ratings. The item bank was adapted from post-acute care measures, and items about gait aids and wheelchairs were removed because they were less relevant to hospital settings. The final three-factor model was chosen even though a four-factor model fit better, because the cognitive factor was not explored further. Response times reduced uncertainty in patient estimates but were not included in the final algorithm. Eligibility required English fluency and excluded patients needing ventilatory support or certain medications, so it may not apply to more impaired acute patients. Patients admitted to neurology or readmitted to neurosurgery were not recruited because therapy was routine for them.

Do not conclude that FAMCAT improves function-directed care or predicts discharge and readmission. This paper reports development and some measurement checks; it does not report the clinical validation results.