Reliability of the “Ten Test” for Assessment of Sensation in Hand Trauma: Level 4 Evidence

Abstract

The purpose of this study was to determine reliability parameters (inter-rater reliability, response stability and responsiveness) of the “Ten Test” (TT), a clinical measure of discriminative sensation whereby the magnitude of abnormal sensation is normalized to an area of normal sensation on a 10-point Likert scale, in a sample of hand trauma patients. A secondary purpose was to compare the reliability of TT to that of the Weinstein Enhanced Sensory Test (WEST), a semi-quantitative monofilament test that is commonly used in this setting. We hypothesized that TT would display good reliability and that reliability parameters would be similar to those of WEST. Patients (n=29, mean age=37±12) with self-reported abnormal hand sensation, presenting to an outpatient hand trauma clinic within seven days of injury, underwent TT and WEST by two separate raters on the same day. The order of test and rater were randomized to reduce systematic error. Levels of WEST were transformed into a 5-point ordinal scale corresponding to the coloured monofilaments, which represent progressively impaired sensation. Bland-Altman plots were constructed to provide a visual representation of agreement. Inter-rater reliability, response stability and responsiveness of each test were determined by the intraclass correlation coefficient (ICC: 2, 1), standard error of measurement (SEM) with 95% confidence intervals (CI) and minimal clinical difference score (MCD), respectively. Inspection of the Bland-Altman plots revealed greater inter-rater agreement for TT than for WEST (Figure 1). TT displayed excellent inter-rater reliability (ICC=0.95, 95% CI 0.89-0.97) compared to good reliability for WEST (ICC=0.78, 95% CI 0.58-0.89). The range of true scores expected with 95% confidence based on the SEM (i.e. response stability), was ±1.05 for TT and ±1.10 for WEST. MCD scores reflecting test responsiveness were 1.50 and 1.60 for TT and WEST, respectively. •TT displayed excellent inter-rater reliability, response stability and responsiveness in a sample of patients presenting to an outpatient hand trauma clinic.•Reliability parameters were stronger for TT compared to WEST.•The range of scores expected on repeated measurements of WEST encompassed a wide range of sensory impairment (±1.10). For a patient with a score of 3, the true score would lie between 2-4, reflecting a range of sensory impairment from mild-to-severe.•These results provide support for the utility of TT in this population. Considering the ease of implementation and excellent reliability of TT, it may be a practical alternative to monofilament testing.