How many body locations need to be tested when assessing sensation after stroke? An investigation of redundancy in the Rivermead Assessment of Somatosensory Performance

Abstract

Objective: Assessing sensation after stroke is time consuming. This study aimed to identify how many body locations need to be tested to establish whether sensation is `intact', `impaired' or `absent' and to assess validity of that classification. Design: Participants' sensation was tested in a single assessment session. Agreement between the scores for individual anatomical sites and those for the whole limb was calculated using a weighted kappa and percentage agreement for each modality. High agreement between tests indicated redundancy, suggesting that the number of anatomical sites tested could be reduced. The Kruskal—Wallis test assessed the validity of classification by comparing a range of measures of functional ability in people with intact, impaired and absent sensation. Setting: Hospital-based stroke care. Subjects: One hundred and two patients tested 2—4 weeks after stroke. Main measures: Four proprioceptive and tactile modalities were measured using the Rivermead Assessment of Somatosensory Performance. Results: Agreement between the total limb score and individual anatomical sites was substantial to excellent for all modalities and anatomical sites. Agreement was greater than 90% when sensation was intact or absent. The comparison between patients with intact, impaired and absent sensation showed significant differences in functional mobility, independence in the activities of daily living, balance and weakness in people with stroke. Conclusions: Sensory impairment can be classified as `intact', `impaired' or `absent'. There is a high redundancy between anatomical sites when the patient's sensation is `intact' or `absent' and not all sites need to be tested. Reducing the number of sites tested will improve usability of the Rivermead Assessment of Somatosensory Performance in day-to-day clinical practice.