Table 3.

Association between everyday discrimination and diabetes-related distress: multiple linear regression results

Variable	All Participants (n = 156) ba (95 % CI)b,d Model 1	African Americans (n = 90) b (95 % CI) Model 2	Latinos (n = 66) b (95 % CI) Model 3
Discrimination	1.17 (0.85, 1.62)	0.96 (0.59, 1.57)	1.58 (1.08, 2.31)*
Malec	0.82 (0.52, 1.28)	0.98 (0.47, 2.04)	0.62 (0.39, 1.00)¶
Age (years, continuous)	0.98 (0.96, 1.00)	0.97 (0.94, 1.00)*	1.01 (0.99, 1.04)
High school graduatec	1.18 (0.68, 2.04)	1.12 (0.45, 2.78)	1.36 (0.78, 2.38)
Employedc	0.85 (0.55, 1.33)	0.69 (0.35, 1.38)	0.91 (0.56, 1.46)
African Americanc	0.58 (0.28, 1.20)
Neighborhoodc	0.74 (0.46, 1.21)	0.86 (0.46, 1.59)	0.48 (0.19, 1.26)
Large health systemc	0.69 (0.38, 1.25)	0.66 (0.32, 1.37)
Diabetes self-management	0.77 (0.58, 1.02)¶	0.68 (0.42, 1.10)	0.97 (0.73, 1.29)
≥ 15 years of U.S. residencec			1.21 (0.72, 2.04)
R2	0.23	0.18	0.20

Bold values are statistically significant at the alpha = 0.05 level

^aUnstandardized exponentiated beta-coefficients

^b95 % CI exponentiated 95 % confidence interval

^cReference groups: Male gender referenced to female gender. High school graduate referenced to less than high school education. Employed referenced to unemployed. African Americans referenced to Latinos. Neighborhood of residence: residence in Eastside Detroit or residence in a neighborhood outside of Eastside or Southwest Detroit referenced to residence in Southwest Detroit. Large health system referenced to federally qualified health center (FQHC) as REACH study site. Residence in the USA for 15 or more years referenced to <15 years of residence in the USA for Latino sample

^dp values based on log-transform of PAID

^¶p < 0.10;

^*p <0.05