TABLE 4.
Reported Discrimination, % |
Crude ORa (95% CI) |
Adjusted ORb (95% CI) |
|
---|---|---|---|
VA health care utilization | |||
VA user | 5.4 | 2.08 (1.04, 4.18) | 1.30 (0.54, 3.13) |
VA nonuser (Ref) | 2.7 | 1.00 | 1.00 |
Race/ethnicity | |||
White (Ref) | 1.6 | 1.00 | 1.00 |
Hispanic | 3.0 | 1.93 (0.48, 7.73) | 1.43 (0.25, 8.23) |
African American | 13.0 | 9.23 (5.26, 16.20) | 4.33 (1.70, 11.0) |
Racial saliencec | |||
≤Once a mo (Ref) | 2.0 | 1.00 | 1.00 |
Once a wk | 4.1 | 2.11 (0.80, 5.56) | 1.43 (0.30, 6.67) |
≥Once a d | 11.5 | 6.46 (3.67, 11.39) | 2.57 (1.02, 6.49) |
Gender | |||
Women (Ref) | 3.1 | 1.00 | 1.00 |
Men | 3.1 | 0.98 (0.42, 2.29) | 2.10 (0.56, 7.90) |
Age, y | |||
18−24 | 7.3 | 4.70 (0.59, 37.45) | |
25−34 | 6.8 | 4.31 (1.69, 10.98) | 4.21 (1.12, 15.9) |
35−44 | 3.1 | 1.88 (0.85, 4.18) | 1.81 (0.66, 4.96) |
45−54 | 3.3 | 2.05 (1.02, 4.12) | 1.90 (0.72, 4.99) |
55−64 | 3.2 | 1.96 (1.03, 3.73) | 1.93 (0.77, 4.85) |
≥65 (Ref) | 1.7 | 1.00 | 1.00 |
Income, $ | |||
< 15 000 | 9.0 | 6.40 (2.77, 14.78) | 3.55 (1.26, 10.0) |
15 000−24 999 | 4.3 | 2.93 (1.32, 6.47) | 1.33 (0.47, 3.76) |
25 000−34 999 | 3.4 | 2.28 (1.00, 5.20) | 1.81 (0.70, 4.68) |
35 000−49 999 | 2.6 | 1.71 (0.72, 4.09) | 1.21 (0.45, 3.25) |
> 50 000 (Ref) | 1.5 | 1.00 | 1.00 |
Education | |||
< High school | 2.9 | 1.56 (0.55, 4.44) | 0.83 (0.17, 3.96) |
High school graduate | 4.2 | 2.33 (1.22, 4.44) | 1.83 (0.78, 4.29) |
Some college | 3.3 | 1.82 (0.91, 3.65) | 1.26 (0.49, 3.24) |
College degree (Ref) | 1.9 | 1.00 | 1.00 |
Health care coverage | |||
Yes (Ref) | 2.6 | 1.00 | 1.00 |
No | 8.5 | 3.55 (1.79, 7.01) | 0.85 (0.37, 1.93) |
Cost of medical care prohibitive in past 12 mo | |||
No (Ref) | 2.4 | 1.00 | 1.00 |
Yes | 9.9 | 4.44 (2.47, 7.99) | 1.82 (0.77, 4.33) |
Health status | |||
Excellent, very good, or good (Ref) | 2.4 | 1.00 | 1.00 |
Fair or poor | 6.2 | 2.64 (1.61, 4.34) | 1.66 (0.81, 3.38) |
Note. CI = confidence interval. Ellipsis indicates too few respondents with complete data to estimate in adjusted model.
Unadjusted ORs reflect the bivariate associations between perceived discrimination and each variable (weighted n = 1 727 967).
Adjusted ORs reflect the association between perceived discrimination and each variable, after adjustment for all the other variables. State was included as an additional covariate in the adjusted model (weighted n = 1 293 909).
Defined as frequency of thoughts about one's own race.