Validity of consumer-grade activity monitor to identify manual wheelchair propulsion in standardized activities of daily living

PLoS One. 2018 Apr 11;13(4):e0194864. doi: 10.1371/journal.pone.0194864. eCollection 2018.

Abstract

Background: Hypoactive lifestyle contributes to the development of secondary complications and lower quality of life in wheelchair users. There is a need for objective and user-friendly physical activity monitors for wheelchair-dependent individuals in order to increase physical activity through self-monitoring, goal setting, and feedback provision.

Objective: To determine the validity of Activ8 Activity Monitors to 1) distinguish two classes of activities: independent wheelchair propulsion from other non-propulsive wheelchair-related activities 2) distinguish five wheelchair-related classes of activities differing by the movement intensity level: sitting in a wheelchair (hands may be moving but wheelchair remains stationary), maneuvering, and normal, high speed or assisted wheelchair propulsion.

Methods: Sixteen able-bodied individuals performed sixteen various standardized 60s-activities of daily living. Each participant was equipped with a set of two Activ8 Professional Activity Monitors, one at the right forearm and one at the right wheel. Task classification by the Active8 Monitors was validated using video recordings. For the overall agreement, sensitivity and positive predictive value, outcomes above 90% are considered excellent, between 70 and 90% good, and below 70% unsatisfactory.

Results: Division in two classes resulted in overall agreement of 82.1%, sensitivity of 77.7% and positive predictive value of 78.2%. 84.5% of total duration of all tasks was classified identically by Activ8 and based on the video material. Division in five classes resulted in overall agreement of 56.6%, sensitivity of 52.8% and positive predictive value of 51.9%. 59.8% of total duration of all tasks was classified identically by Activ8 and based on the video material.

Conclusions: Activ8 system proved to be suitable for distinguishing between active wheelchair propulsion and other non-propulsive wheelchair-related activities. The ability of the current system and algorithms to distinguish five various wheelchair-related activities is unsatisfactory.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Activities of Daily Living*
  • Aged
  • Algorithms
  • Biomechanical Phenomena
  • Equipment Design*
  • Exercise
  • Female
  • Hand
  • Healthy Volunteers
  • Humans
  • Male
  • Middle Aged
  • Monitoring, Ambulatory / instrumentation*
  • Monitoring, Ambulatory / methods*
  • Movement
  • Persons with Disabilities
  • Predictive Value of Tests
  • Quality of Life
  • Reproducibility of Results
  • Sensitivity and Specificity
  • Software
  • Video Recording
  • Wheelchairs*

Grants and funding

Purchase of the Activ8 Professional Activity Monitors used in this study was supported by the Stichting Beatrixoord Noord-Nederland. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.