Moderate to Vigorous Physical Activity as a Digital Biomarker
MVPA is the time a person spends moving at moderate or vigorous intensity. It is the activity measure public health guidelines are written in, and accelerometers estimate it directly.
Evidence maturity
Graded with the V3 framework: whether the sensor measures accurately, whether the algorithm has been validated against a reference standard, and whether the measure has been shown to matter clinically.
Technically mature and clinically meaningful, but analytically fragile: intensity classification depends entirely on the cut-points and wear location chosen, so MVPA values are rarely comparable across studies with different processing pipelines.
What is Moderate to Vigorous Physical Activity
Moderate to vigorous physical activity, usually shortened to MVPA, is the time a person spends moving at an intensity of roughly three metabolic equivalents or more. Brisk walking sits at the lower boundary, running well above it. MVPA matters because it is the currency of public health guidance: the World Health Organization recommends adults accumulate 150 to 300 minutes of moderate activity, or 75 to 150 minutes of vigorous activity, each week. Where step count answers how much someone moved, MVPA answers how hard. The two often diverge, and a participant can accumulate a respectable step total almost entirely below the moderate threshold. As a digital biomarker, MVPA converts a guideline written for populations into a measurable daily quantity for an individual participant, which is why it appears as a primary endpoint in trials testing whether an intervention actually changes how hard people move.
How it is measured
Accelerometers estimate intensity from the magnitude of body acceleration. Older pipelines convert raw acceleration into proprietary counts per minute and apply published cut-points, such as the Freedson or Troiano thresholds, to classify each epoch as sedentary, light, moderate or vigorous. Newer pipelines work from raw acceleration in gravitational units, most commonly the Euclidean norm minus one, which is comparable across device brands in a way counts are not. Hip and thigh placements track ambulatory intensity most faithfully; wrist placement is far more practical for long wear but needs its own thresholds. The reference standard is indirect calorimetry, which measures oxygen consumption directly. Note that the definition itself changed: guidelines before 2018 required activity to be accumulated in bouts of at least ten minutes, and current guidance counts all minutes, so older and newer studies are not directly comparable.
Clinical use
MVPA appears across trials where the question is whether an intervention changes activity intensity rather than volume. It is used in musculoskeletal and rheumatological studies, in heart failure and cardiac rehabilitation, in multiple sclerosis, and in metabolic trials where the intended mechanism runs through exercise. In the public DiMe endpoint library, MVPA is registered as a primary endpoint in several trials, more than most activity measures relative to how often it is used at all. Behaviour change and rehabilitation programmes commonly define success as a shift of participants across a guideline threshold rather than as a continuous minute change, because crossing the threshold carries an interpretation that a raw minute count does not. Studies typically pair MVPA with a self-reported activity questionnaire so that measured intensity can be read against perceived exertion.
Regulatory status
No standalone regulatory qualification to date. MVPA endpoints are written into registered trial protocols, including as primary outcomes, and the measure is anchored to World Health Organization physical activity guidelines rather than to a regulatory qualification.
Limitations
Cut-point dependence is the central problem. The same raw data classified with different thresholds can produce MVPA estimates that differ by a factor of two or more, so values are not comparable across studies unless the processing pipeline matches. Wrist-worn devices further complicate this, since arm movement is a poor proxy for whole-body intensity. Activities that load the body without moving it much, such as cycling, carrying loads or walking uphill, are systematically under-classified. Any study reporting MVPA should state its device, wear location, epoch length and cut-points, and any comparison that does not match on all four should be treated with caution.
References
- Bull FC, et al. World Health Organization 2020 guidelines on physical activity and sedentary behaviour. Br J Sports Med. 2020. pubmed.ncbi.nlm.nih.gov
- Troiano RP, et al. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc. 2008. pubmed.ncbi.nlm.nih.gov
- Migueles JH, et al. Accelerometer data collection and processing criteria to assess physical activity and other outcomes: a systematic review and practical considerations. Sports Med. 2017. pubmed.ncbi.nlm.nih.gov
- Freedson PS, Melanson E, Sirard J. Calibration of the Computer Science and Applications accelerometer. Med Sci Sports Exerc. 1998. pubmed.ncbi.nlm.nih.gov
Objective counterpart of self-reported activity questionnaires such as the SQUASH and the Physical Activity Scale for the Elderly, which ask participants to recall intensity that accelerometers measure directly.
Collect MVPA data with WeGuide, the all in one patient engagement platform
Capture accelerometer derived activity intensity alongside patient reported outcomes in a single study workflow.
