Poster 94: Arm Motor Ability Test: Validity of a 9-Item Version

Abstract

Objective: To provide evidence for the construct validity of a 9 item version of the Arm Motor Ability Test (AMAT-9, Kopp et al. 1997 version without “lightswitch/doorknob” item). Design: Analysis of baseline data prior to an intervention. Setting: Outpatient rehabilitation medicine clinic. Participants: 28 community-dwelling persons ≥6 months poststroke with manual muscle strength grade of 1 to 4 in the paretic arm and able to follow commands. Mean age ± SD is 56±12.3 years, mean time poststroke ± SD was 4.3±4.7 years, mean NIHSS ± SD was 4.1±4.0, and mean AMAT ± SD was 1.3±1.2. 89% were right handed and equal numbers had right or left hemiparesis. Intervention: None. Main Outcome Measures: Series of baseline upper-extremity measures: AMAT-9, Wolf Motor Function Test (WMFT), Fugl-Meyer Assessment (FMA), Action Research Arm Test (ARAT), and Stroke Impact Scale (SIS). We hypothesized that correlations with AMAT-9 would be (1) strongest with WMFT, followed by FMA and ARAT (convergent validity), (2) stronger with the SIS self-care than SIS communication subscore (divergent validity). Correlations were evaluated using the Spearman's coefficient. Results: Spearman r between AMAT-9 and FMA=.86, ARAT=.83, and WMFT=.82, all P<.001 but not in the order hypothesized. Correlation between the AMAT-9 and SIS self care=.41 (P<.03) and communication=–.18 (not significant) and consistent with our hypothesis. Conclusions: This data are the first to demonstrate the construct validity of the AMAT-9 and among the first for the AMAT in such a severely impaired sample. By deleting the requirement of standing to manipulate a light switch and doorknob, the AMAT-9 now becomes available to nonambulatory and balance-impaired subjects. Further research should determine the reliability of the AMAT-9 and its sensitivity to detect change over time.