Assessing Physical Activity in Inpatient Rehabilitation: validity, practicality, and sensitivity to change in the physical activity in inpatient rehabilitation assessment

Abstract

Objective: To validate a novel assessment of inpatient physical activity.

Design: Prospective cohort study for the evaluation of a novel questionnaire for physical activity in geriatric inpatients.

Setting: German geriatric inpatient rehabilitation unit.

Participants: Patients (N=96; 67 [72%] women; median age, 81y) with a variety of main underlying diagnoses, including musculoskeletal diseases, hip fracture, cardiovascular diseases, stroke, and others.

Interventions: Not applicable.

Main outcome measures: Ceiling and floor effects and administration time were measured. For criterion-related concurrent validity (convergent and discriminative), the Physical Activity in Inpatient Rehabilitation Assessment (PAIR) was administered in parallel to self-rated, proxy-rated, and performance-based measures of physical function at admission. Measurements were repeated at discharge and 4-month follow-up in the home environment, including a standard physical activity questionnaire to determine predictive validity. Spearman correlation coefficients were calculated to describe associations between parameters. Sensitivity to change was estimated using standardized response means (SRMs).

Results: Administration time of the PAIR ranged from less than 1 to 4 minutes. Ceiling effects occurred mainly at discharge (5%-14%), and floor effects (5%-11%), at admission. There were no missing values. Associations between convergent and predictive validity measures and functional measures (r=.43-.53, r=.49-.54, respectively) were clearly better when cognition was intact. Discriminative validity expressed as effect sizes ranged from .27 to 1.44. The SRM to describe sensitivity to change was .65 for the total score.

Conclusions: The PAIR is the first validated questionnaire to assess physical activity in geriatric inpatients. It is practical and its validity and sensitivity to change are similar to existing physical activity questionnaires for community-dwelling older persons.