Comparison of Three Measures of Upper Extremity Function in Multiple Sclerosis

Background: Upper extremity (UE) dysfunction is common in people with multiple sclerosis (pwMS), and can have a marked impact on activity and participation. It is important for clinicians to use outcome measures (OM) that not only provide information regarding clinical status, but can predict useful correlates, and have high clinical utility (i.e. they are simple and inexpensive to conduct). Three such OMs are the 9-Hole Peg Test (9HPT), Box and Block Test (BBT), and Finger-to-Nose Test (FTN).

Objectives: To establish cut-off scores and classification accuracy of the 9HPT, BBT, and FTN in predicting the Disabilities of Arm, Shoulder and Hand (DASH), Functional Status Index- Assistance portion (FSI-A), Test d’Evaluation de la performance des Membres Supérieurs des Personnes Âgées (TEMPA), and Employment Status (ES) in a sample of pwMS.

Methods: The MS-CUE data set, established as part of a large study designed to characterize UE involvement in a random sample of pwMS (MS-CUE) was utilized.  Participants (n=267) from a random sample of pwMS were included in this study. Due to missing scores, sample size varied (n=177-267) among each analysis. The DASH, FSI-A, TEMPA, and ES scores were dichotomized, with impairment defined as needing assistance on one or more UE task (FSI-A), scoring above 2 standard deviations (DASH and TEMPA), and no current employment. Receiver-operating-characteristic (ROC) curve analysis was used to identify cut-off scores that yielded optimal sensitivity and specificity for the three OMs in predicting each of these four dichotomized variables.

Results: The FTN was comparable to the BBT in predicting the DASH, FSI-A, and TEMPA, as well as to the 9HPT in predicting the DASH. While the FTN had higher specificity (82%) compared to the 9HPT in predicting the FSI-A (51%), its sensitivity was significantly lower (52% vs 80%). The 9HPT also had higher sensitivity (84%) in predicting the TEMPA compared to the FTN (69%). With ES, although the FTN had higher specificity than the BBT (76% vs 65%), it had significantly lower sensitivity (62%) than both the 9HPT (77%) and the BBT (83%).

Conclusions:

As each of these OMs has its strengths and weaknesses relevant to classification accuracy, clinicians should consider clinical utility when selecting OMs for the upper extremity. The FTN has very high clinical utility, and predicts certain UE functions as well as the more time consuming 9HPT and BBT. However, if employment is the outcome of interest, then clinicians should consider using the BBT or 9HPT as they had superior psychometric properties in predicting this outcome.