Department of Human Biology, Maastricht University Medical Centre, Maastricht, The Netherlands
ActiGraph Digital Endpoint Accelerator Research GrantCall for Letter of Intent Learn more
Reliable assessment of physical activity in disease: an update on activity monitors
- Published on Sept. 2014
Purpose of Review: Physical activity is the main determinant of variability of individual energy expenditure. Reliable assessment of the activity level of an individual provides information on energy requirement and vital health. Here, evidence is presented on the validity of methods to assess physical activity as applied in health and disease.
Recent Findings: Improvement of technology has resulted in a growing number of physical activity monitors to evaluate the activity level of an individual. Outcome measures vary from raw data on body acceleration and posture to activity levels derived from proprietary algorithms based on body acceleration combined with sensor information on additional activity-induced physiological responses. Data interpretation is limited by evidence-based studies on the value of measured parameters for the assessment of physical activity.
Summary: The optimal tool for reliable assessment of physical activity is an accelerometer providing a valid measure of body movement. Overall validity is derived from studies using doubly labelled water-assessed activity-induced energy as a reference. Thus, out of the large range of activity monitors, three came out with a better validity. The ultimate tool is a single unobtrusive device allowing valid and long-term monitoring of activity with regard to type and intensity.
Conclusion: Accelerometry is the most promising method to objectively assess physical activity behaviour as a reflection of metabolic health or determinant of metabolic risk. The validity of accelerometer-based activity monitors is judged with doubly labelled water-assessed activity energy expenditure or PAI as a reference, where the optimal observation time is 1 or 2weeks. The outcome of an activity monitor should not be a function of proprietary equations where agreement between subsequent versions can be large. Most monitors explained less than 50% of the variation in activity energy expenditure or PAI. Only three out of the 11 tested, Actigraph GT3X, Dynaport, and TracmorD, came out better. The final choice of a valid monitor depends on further specifications including memory and battery capacity, additional options for data processing including movement quality, and options for user feedback and coaching.