Tang, KT; Spence, WD; Simpson, D; Richardson, A; Maxwell, D; Stansfield, BW
1st International Conference on Ambulatory Monitoring of Physical Activity and Movement
INTRODUCTION The population of people with cerebral palsy (CP) has varying levels of mobility impairments [1] and one of the primary aims of clinical interventions in CP is to maintain and improve walking ability. Therefore knowledge of activity levels will provide critical information about the success of the clinical intervention. There is limited information pertaining to the mobility of people with CP in their free-living environments [2] and typically no routine monitoring of activity levels is performed. The activity monitor (activPAL™, PAL Technologies Ltd, Glasgow, UK) is a small, 50x35x7mm and lightweight (20g) self contained unit designed to monitor free-living activity. The aim of this study was to determine daily physical activity levels of children with CP. An integral part of this study was the evaluation of the activity monitor in a laboratory-based environment in children with CP. METHODS 10 subjects with the condition of CP (3 girls and 7 boys; age 5-17 years; 7 diplegia and 3 hemiplegia) were recruited. A laboratory based evaluation study was carried out before the subjects wore the device in a freeliving environment. The activPAL was attached to the right thigh of each participant during a gait analysis session. They were asked to walk, stand and sit for different periods of time, whilst being video recorded. Subjects then wore the device during waking hours (except for water based activities) for 7 consecutive days during their normal daily life. RESULTS For the laboratory based evaluation study, it was shown that the activity monitor could detect sitting/lying, standing, walking and the number of steps with average accuracies of 96.4%, 94.2%, 92.1%, 97.3% respectively. Fig 1 shows the range of activity levels for the 10 subjects for the 7 days periods, indicating the average stepping time, quiet standing time and number of steps per day. DISCUSSION The evaluation study verified good agreement between video and the activPAL data. However, it was noticed that for specific children who had greater mobility impairment such as severe crouch gait, the accuracy decreased as misclassification occurred between standing and walking. The evaluation study was laboratory based, involving short intervals of each activity. This may have increased the proportion of time spent in transition between activities compared with daily life. We therefore believe that the values achieved in the evaluation will be under estimates of accuracy of the device in the community/home environment. For the community based study, the device could distinguish the less active household walkers (e.g. subject 2 and 8) from the more active community walkers who took on average over 5000 steps per day. CONCLUSIONS The activPAL may be used to classify activity level in people with CP and therefore may provide valuable information on the rehabilitation progress and mobility status of children with CP by identifying the daily activity levels of the individual in their free-living environment.Tang, KT; Spence, WD; Simpson, D; Richardson, A; Maxwell, D; Stansfield, BW Multi-Day Physical Activity Monitoring In People With Cerebral Palsy 1st International Conference on Ambulatory Monitoring of Physical Activity and Movement 2008. http://www.umass.edu/sphhs/documents/ConferencebookICAMPAM2008.pdf.