Physical activity self-reports: past or future?
Matteo C. Sattler, Barbara E. Ainsworth, Lars Bo Andersen, Charlie Foster, María Hagströmer, Johannes Jaunig, Paul Kelly, Harold W. Kohl, Charles E. Matthews, Pekka Oja, Stéphanie A. Prince, Mireille N. M. van Poppel
Abstract
The measurement of physical activity (PA) is fundamental to health-related research, practice and policy. For decades, self-report measures have provided unique insights into the role of PA for human health and society. In fact, studies, in which participants reported their behaviours—or the behaviours of others—using diaries, logs, questionnaires and recalls, have historically provided the evidence that underpins global PA guidelines.1 Self-reports have been used extensively in various settings, including population surveillance, observational and intervention studies and routine assessment as part of healthcare. The field of PA measurement is rapidly evolving. We have a wealth of measurement instruments and achieved remarkable advancements in the use of device-based information such as raw accelerometry, novel algorithms for pattern recognition and worldwide initiatives for data harmonisation.2 3 The technological evolution has changed the practice of PA self-reports as well, and led to electronic surveys and ecological momentary assessments (EMAs) for the measurement of PA in natural environments and in ‘real time’. Despite significant improvements, an established standard for the measurement of PA does not exist due to the complexity of the behaviour.4 PA is multifaceted and encompasses different domains (eg, leisure, occupation, transport, household), dimensions (eg, frequency, duration, intensity, …