. Author manuscript; available in PMC: 2017 Dec 1.

Published in final edited form as: Mov Disord. 2016 Nov 7;31(12):1865–1873. doi: 10.1002/mds.26847

Table 4.

Race/ethnicity-based statistically significant DIF

Race/ethnicity-Based NU-DIF: Impact magnitude of identified significant DIF
	Item		R²	Magnitude

Part 1 N=5561	1.12 Lightheadedness	White vs. all others	0.0010	Negligible

Race/ethnicity-Based U-DIF: Impact magnitude of identified significant DIF
	Item		R²	Magnitude

Part 1 N=5561	1.2 Hallucinations
		White vs. all others	0.0020	Negligible
		Asian vs. all others	0.0071	Negligible
	1.7 Sleep problems	Asian vs. all others	0.0034	Negligible
	1.11 Constipation
		White vs. all others	0.0076	Negligible
		Asian vs. all others	0.0070	Negligible
		Hispanic vs. all others	0.0024	Negligible
	1.13 Fatigue	Asian vs. all others	0.0041	Negligible

Part 2 N=5559	2.2 Saliva/drooling
		White vs. all others	0.0012	Negligible
		Hispanic vs. all others	0.0014	Negligible
	2.3 Swallowing	Hispanic vs. all others	0.0010	Negligible
	2.10 Tremors
		White vs. all others	0.0039	Negligible
		Asian vs. all others	0.0058	Negligible
	2.13 Fatigue	Asian vs. all others	0.0011	Negligible

Legend: Three groups had sufficient item representation to be compared: Whites (non-Hispanic), Hispanics, and Asians. Each group was compared against the combined comparator groups. Most of the MDS-UPDRS items did meet the minimal statistical criteria for DIF (see text). The Table lists items with DIF identified by both lordif and MIMIC as independent approaches (p values shown in Supplemental Materials. McFadden's R² and impact Magnitude shown in columns.⁷