Measuring Mobility in Low Functioning Hospital Patients: An AM-PAC Replenishment Project.

Abstract Objective Expand an existing validated measure of basic mobility (AM-PAC) for patients at the lowest levels of function. Design Item replenishment for existing item response theory (IRT) derived measure, via: 1) idea generation and creation of potential new items, 2) item calibration and field-testing, and 3) longitudinal pilot test. Setting: Two tertiary acute care hospitals Participants Consecutive inpatients (n=502) > 18 years old, with AM-PAC Inpatient Mobility Short Form (IMSF) raw score ≤15; for the longitudinal pilot test, eight inpatients were evaluated. Results Fifteen new AM-PAC items were developed, two of which improved mobility measurement at the lower levels of functioning. Specifically, with the 2 new items, the floor effect of the AM-PAC IMSF was reduced by 19%, statistical power and measurement breadth were greater, and in longitudinal pilot testing there was greater measurement sensitivity. Conclusions Adding 2 new items to the AM-PAC IMSF lowered the floor, increased statistical power, measurement breadth, and sensitivity.