The activPAL as a measure of physical activity and sedentary behaviour in people with Parkinson's Disease: a validity study

Abstract

Purpose To establish the criterion validity of the activPAL activity monitor in measuring sedentary behaviour (sitting/lying) and physical activity (standing/walking) in people with Parkinson's Disease, using direct observation as the criterion measure. Design A cross-sectional observational study, evaluating the criterion validity of the activPAL in a laboratory setting. Participants 12 individuals with a Parkinson's Disease diagnosis and Hoehn and Yahr Disease Severity Classification of ≤ 3 who could walk five minutes independently or with an aid. Methods Participants wore one activPAL monitor and completed a test protocol involving 3 modules: body-postures, activities of daily living and indoor/outdoor walking. Each trial was video-recorded where the camera was synchronised with the activPAL. Researchers classified the time spent in different postures and compared it to the activPAL output. Results Using the Bland and Altman method, the activPAL appeared to over-estimate sedentary behaviour compared to direct observation. Mean difference for module 1 was -35.6s (limits of agreement -75.3s to 4.1s) and for module 2 -16.5s (limits of agreement -29.4s to -3.6s). When measuring physical activity, in module 1, the activPAL slightly underestimated compared to direct observation. Mean difference of 2.3s (limits of agreement -21.1 to 25.7s). Modules 2 and 3 had better agreement. Mean differences of -0.2s and -0.1s with limits of agreement of -16.9s to 16.5s and -1.2s to 1.01s, respectively. Conclusion The activPAL may be a valid measure of physical activity and sedentary behaviour in people with Parkinson's Disease however, further research is needed to build-on and strengthen these findings.