Description

Two methods are presented for predicting METs of manual wheelchair users- a general prediction equation and a set of four activity-specific regression equations. Mean absolute error in kcal/min across activities was 0.3 to 1.3 for the general model and 0.2 to 0.9 for the activity-specific models. Intraclass correlations were 0.72 and 0.86 for the general and activity-specific models, respectively.

Development/Validation

Sample: 45 (37 males) adult manual wheelchair users with spinal cord injury, 18-60 years of age

Setting: Laboratory

Activities: Arm ergometer, overground and treadmill manual propulsion, stationary

Criterion: Cosmed K4b2 (VO₂)

Accelerometer(s): Sensewear armband on right tricep

Validation approach: 6-fold cross-validation, holdout of 9 participants

Phase Designation

(What's this?)

This model is in Phase 1.

Instructions

The general prediction model and activity-specific models are below. A description of the input variables for these models is also provided.

Source Information

Hiremath, S. V., Ding, D., Farringdon, J., & Cooper, R. A. (2012). Predicting energy expenditure of manual wheelchair users with spinal cord injury using a multisensor-based activity monitor. Archives of Physical Medicine and Rehabilitation, 93(11), 1937-1943. http://dx.doi.org/10.1016/j.apmr.2012.05.004 Link to Paper

Corresponding author: Dan Ding, dad5@pitt.edu

Contact

Kimberly Clevenger at accelerometerrepository@gmail.com