Calibration of accelerometer output for adults.

Area of Science:

• Exercise physiology
• Biomedical engineering
• Physical activity research

Background:

• Activity monitors are widely used to measure physical activity intensity.
• Calibration studies aim to translate monitor output (activity counts) into meaningful intensity levels.
• Existing calibration methods exhibit significant methodological diversity.

Purpose of the Study:

• To review and critique existing calibration study methodologies for adult physical activity monitors.
• To identify sources of variation in calibration prediction equations and cut points.
• To propose strategies for enhancing the comparability of future calibration research.

Main Methods:

• Review of published adult monitor calibration studies.
• Analysis of common criterion measures (e.g., oxygen consumption).
• Critique of statistical approaches (e.g., linear regression) for developing calibration equations.

Main Results:

• Methodological heterogeneity in calibration studies leads to substantial variation in prediction equations and cut points.
• This heterogeneity limits the comparability of physical activity intensity data across different research groups, populations, and monitor types.
• Despite robust monitor technology, calibration processes splinter data into less comparable summary measures.

Conclusions:

• There is a critical need for standardized and flexible research methods in physical activity monitor calibration.
• Improved calibration strategies will enhance the interpretability and comparability of accelerometer data.
• Future research should focus on methods that increase consistency in translating activity counts to physical activity intensity.