Skip to main content

Table 4 Associations between participants’ characteristics and race/ethnicity based discrimination within healthcare settings among non-White participants

From: Different faces of discrimination: perceived discrimination among homeless adults with mental illness in healthcare settings

Explanatory variables

Discrimination due to race or ethnicity or skin color (n = 329)

 

Unadjusted OR

Adjusted OR3

Age§, years, mean (SD)

.94 [.75-1.18]

-

Male£

1.07 [.61-1.90]

-

Race

Black (ref)

1.0

-

Other ethnic minority

.84 [.50-1.44]

-

Monthly income, CAD

<$500 (ref)

1.0

1.0

$500-$999

1.80 [.98-3.32]

1.54 [.79-3.00]

≥$1,000

1.96 [.94-4.09]

1.87 [.83-4.19]

Homeless for ≥3 years

1.57 [.92-2.68]

1.54 [.85-2.80]

CSI† ≥30

1.69 [.84-3.41]

1.67 [.75-3.70]

GAIN SDScrâ•‘

No/low (ref)

1.0

1.0

Moderate

3.61 [1.43-9.13] **

3.66 [1.39-9.61]**

High

2.82 [1.20-6.60] *

1.75 [.70-4.38]

Psychotic disorder‡

.97 [.57-1.67]

1.05 [0.58-1.89]

ED visits in the past 6 months

No visits (ref)

1.0

1.0

1 visit

2.68 [1.38-5.19] *

3.17 [1.56-6.48] **

 

≥ 2 visits

2.58 [1.34-4.96] *

2.27 [1.13-4.55] *

  1. The adjusted model includes variables that were associated with the outcome variable at the level of P ≤.20 in the univariate analysis and presence of psychotic disorder as a control variable.
  2. CI, confidence interval; OR, Odds Ratio (95% CI).
  3. §Age is measured in decades in logistic regression analyses.
  4. £Male category includes transgender/transsexual individuals.
  5. †CSI ≥30, score ≥30 on the Colorado Symptom Index indicates psychiatric disability and presence of clinically significant mental symptoms.
  6. â•‘GAIN SDScr; Substance Disorder Screener. Counts are triaged into low (0 symptoms), moderate (1-2 symptoms), and high (3 or more symptoms) severity symptoms.
  7. ‡MINI v. 6.0; the Mini International Neuropsychiatric Interview 6.0.
  8. Model coefficients: χ2 (9, n=329) = 27.24, P < .01; Hosmer&Lemeshow test: χ2 (8, n=329) = 4.28, p =.83.
  9. *p<.05; **p<.01; ***p<.001.