Bradykinesia-akinesia incoordination test: validating an online keyboard test of upper limb function

Abstract

Background: The Bradykinesia Akinesia Incoordination (BRAIN) test is a computer keyboard-tapping task that was developed for use in assessing the effect of symptomatic treatment on motor function in Parkinson's disease (PD). An online version has now been designed for use in a wider clinical context and the research setting.

Methods: Validation of the online BRAIN test was undertaken in 58 patients with Parkinson's disease (PD) and 93 age-matched, non-neurological controls. Kinesia scores (KS30, number of key taps in 30 seconds), akinesia times (AT30, mean dwell time on each key in milliseconds), incoordination scores (IS30, variance of travelling time between key presses) and dysmetria scores (DS30, accuracy of key presses) were compared between groups. These parameters were correlated against total motor scores and sub-scores from the Unified Parkinson's Disease Rating Scale (UPDRS).

Results: Mean KS30, AT30 and IS30 were significantly different between PD patients and controls (p≤0.0001). Sensitivity for 85% specificity was 50% for KS30, 40% for AT30 and 29% for IS30. KS30, AT30 and IS30 correlated significantly with UPDRS total motor scores (r = -0.53, r = 0.27 and r = 0.28 respectively) and motor UPDRS sub-scores. The reliability of KS30, AT30 and DS30 was good on repeated testing.

Conclusions: The BRAIN test is a reliable, convenient test of upper limb motor function that can be used routinely in the outpatient clinic, at home and in clinical trials. In addition, it can be used as an objective longitudinal measurement of emerging motor dysfunction for the prediction of PD in at-risk cohorts.

Conflict of interest statement

Competing Interests: The authors have read the journal's policy and report the following conflicts: AJ Noyce declares the following: consultancy, Élan/Prothena Pharmaceuticals; Grants, Parkinson's UK Innovation Grant (reference number K-1006), and National Institute of Health Research Academic Clinical Fellowship. A Nagy, S Acharya, S Hadavi, JP Bestwick, and J Fearnley declare no competing interests. AJ Lees declares the following: board membership, Novartis, Teva, Meda, Boehringer Ingelheim, GSK, Ipsen, Lundbeck, Allergan, Orion, BIAL, Noscira, Roche; consultancy, Genus; employment, UCH/UCLH; grants/grants pending, PSP Association, Weston Trust–Reta Lila Howard Foundation; speaking fees, Novartis, Teva, Meda, Boehringer Ingelheim, GSK, Ipsen, Lundbeck, Allergan, Orion, BIAL, Noscira, Roche. G Giovannoni declares the following: consultancy fees, Merck-Serono, Biogen-Idec, Ironwood, Genentech, Teva, GSK, Sanofi-Aventis, Novartis, Roche, UCB Pharmaceuticals, Vertex, Eisai, Elan, Fiveprime, Bayer-Schering, Synthon BV, and Genyme; grants/grants pending from Biogen-Idec, Bayer-Schering, GW Pharma, Merck-Serono, Merz, Novartis, Teva, Sanofi-Aventis. This does not alter the authors' adherence to PLOS ONE policies on sharing data and materials.

Figures

Figure 1. Comparison of KS30, AT30 and IS30 in patients with PD and controls (average of score from each hand).

Distribution of KS30 (mean and standard deviation), (b) AT30 and (c) IS30 (medians and interquartile ranges). For IS30, 7 data points were out of the axis range. Receiver operating characteristic (ROC) curves for (d) KS30, (e) AT30 and (f) IS30.

Figure 2. Correlation of (a) KS30, (b) AT30, (c) IS30 but not (d) DS30 with total motor UPDRS in patients with PD.

Figure 3. Examples of repeat tests in 3 PD patients with predictable motor fluctuation.

Arrows denote times at which levodopa was taken.

Figure 4. Further examples of repeat tests in 3 patients with predictable fluctuation (patient 4) and unpredictable motor fluctuation (patients 5 and 6).

Arrows denote times at which levodopa was taken.