Racial Disparities at Baseline in NIH Toolbox Motor Normative Data
Deborah Hall1, Lisa Shulman2, Kayli Hilliard1, Jacqueline Vanegas1, Rebecca Weimer2, Michelle Tosin3, Bichun Ouyang1, David Gonzalez3
1Rush University, 2University of Maryland School of Medicine, 3Rush University Medical Center
Objective:
To determine whether healthy control data on motor measures of the NIH Toolbox are similar in Black and White populations. 
Background:
The NIH Toolbox gait speed, fine motor dexterity, strength, and balance tests have been used as a set of quantitative measures in movement disorders research. It is unclear if there are baseline differences in the normative healthy control data among those racialized as Black and White. 
Design/Methods:
Relationships among demographics (self-reported race) and motor measures from the NIH Toolbox Norming Study were analyzed. T-tests were performed on the motor measures comparing the African American (AA) and White groups, with Holm correction for multiple tests. 
Results:

Demographics of the sample were: n=812 AA (55% women, average age 19.5±17.3 yrs) and n=4055 White (54% women, 22.1± 21.4yrs). There was a difference in 4-meter walk (p=0.001) and 4-meter walk fast (p=0.03), with AA faster than White individuals. Grip strength was also dissimilar in the dominant hand (p<.001) and nondominant hand (p=<0.001), with AA stronger than Whites. Balance was better in AA (p=0.001). There was no difference in 2-meter walk (p=0.07), 9-hole pegboard (p=0.7), knee strength (p=0.64) and balance sway (p=0.82). When the data was restricted to individuals over 40 (n=812 AA, 66% women, 56.3±11.6 yrs) and n=4055 White (56% women, 61.4±13.1yrs), knee extension strength was weaker in the AA compared to White individuals in the dominant (42.2 vs. 66.4, p=0.03) and non-dominant leg (42.2 vs. 55.4, p=0.02). Balance and balance sway were better in the AA compared to Whites (0.7 vs. 0.4, p<0.001) and (0.37 vs. 0.35, p=0.03).

Conclusions:

Researchers using the NIH toolbox motor measures in various racial groups should be cautious to adjust for baseline differences in the data. Without adjustment, differences seen in disease populations may be due to normative data differences rather than attributable to the disease under study.

10.1212/WNL.0000000000216454
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.