The demand for geriatric rehabilitation will drastically increase over the next years. It will be increasingly important to demonstrate the efficacy and effectiveness of geriatric rehabilitation. One component is the use of objective and valid assessment procedures. These should be understandable to patients, relevant for goal attainment, and able to document change. A number of currently used physical capacity measures have floor effects. The use of body-fixed sensor technology for monitoring physical activity is a possible supplement for the assessment during geriatric rehabilitation to overcome floor effects and directly monitor improvement of mobility as a component of geriatric rehabilitation in many patients. The observational study with a pre-post design examined 65 consecutive geriatric hip fracture inpatients. Measurements were performed on admission and 2 weeks later. The capacity measures included gait speed, chair rise time, a balance test, 2-Minute-Walk test and the Timed-Up-and-Go test. Physical activity was measured over 9 h using body-fixed sensor technology and expressed as cumulated walking and walking plus standing (time on feet). Body-fixed sensors allowed direct measurement of physical activity in all patients available for testing. Cumulated walking and standing (time on feet) increased from a median 83.6 to 102.6 min. Cumulated walking increased from a median 7.0 to 16.3 min. The comparison with the physical capacity measures demonstrated a modest to fair correlation (r(s) = 0.455 and 0.653). This indicates that physical capacity measures are not the same construct as physical activity. Body-fixed sensor-based assessment of physical activity was feasible even in geriatric patients with severe mobility problems and decreased the number of patients with missing data both on admission and 2 weeks later. Body-fixed sensor data documented change in activity level.