Sit to Stand Transitions as a Digital Biomarker

A sit to stand transition is the act of rising from a seated position to standing. It is worth measuring because it is the most mechanically demanding movement most adults perform in ordinary life, requiring lower limb strength, postural control and the coordination to combine them. Clinicians have measured it for decades with the Five Times Sit to Stand test, which simply times five repetitions, and it forms one of the three components of the Short Physical Performance Battery. As a digital biomarker the same construct is captured continuously: sensors count how many transitions a person makes across a day and how long each one takes. That shifts the measure from a single supervised effort to a record of how the movement is actually performed at home, where fatigue, furniture and confidence all apply. Both the count and the duration carry information, and they can move in opposite directions.

Transition detection uses accelerometer and gyroscope data, most reliably from a sensor at the lower back or thigh where the trunk rotation and vertical acceleration of rising are unambiguous. The algorithm identifies the characteristic forward trunk lean, vertical acceleration peak and settling that make up a transition, then derives its duration and, in some pipelines, the power generated. Thigh-worn inclinometry has an advantage here because it knows posture directly, so a transition is a state change rather than a pattern to be inferred. Wrist-worn devices perform poorly for this measure because the wrist is often supporting the movement on an armrest or simply hanging still. Validation compares detected transitions against video observation or instrumented chairs and force plates in the laboratory.

Sit to stand measures are used wherever lower limb function is the clinical question: sarcopenia and frailty research, stroke and orthopaedic rehabilitation, Parkinson's disease, and recovery after hospitalisation. The clinic test they derive from has well established reference values, and a Five Times Sit to Stand time beyond roughly twelve seconds is commonly treated as a marker of functional decline and elevated fall risk in older adults, which gives the digital measure an anchor to interpret against. In the public DiMe endpoint library, posture transition endpoints are registered in trials including as primary outcomes. The measure is also directly relevant to WeGuide's own first biomarker use case, a mobility recovery study validating wearable-derived movement against the Short Physical Performance Battery.

Detection accuracy depends heavily on sensor placement, and wrist-worn consumer devices, which dominate long studies, are the weakest option. Transitions performed slowly or with arm support, which are exactly the ones that signal impairment, are the hardest to detect and are systematically under-counted, so the measure can look better than the participant is. Daily transition counts also reflect the environment as much as capacity: someone with a desk job and a single chair produces a very different count from someone moving between rooms. Duration is generally the more informative half of the measure, but it requires the higher-quality sensing that consumer devices do not provide.

• Bohannon RW. Reference values for the five-repetition sit-to-stand test. Percept Mot Skills. 2006. pubmed.ncbi.nlm.nih.gov
• Guralnik JM, et al. A short physical performance battery assessing lower extremity function. J Gerontol. 1994. pubmed.ncbi.nlm.nih.gov
• Regterschot GRH, et al. Accuracy and concurrent validity of a sensor-based analysis of sit-to-stand movements. Gait Posture. 2016. pubmed.ncbi.nlm.nih.gov
• Regterschot GRH, et al. Sensitivity of sensor-based sit-to-stand peak power to the effects of training. Gait Posture. 2014. pubmed.ncbi.nlm.nih.gov