Skip to main content

Table 4 Results from design-agnostic, model-based and design-based logistic regression analyses in which trust in healthcare system was regressed on self-reported race/ethnicity, low income, age, gender, rural living and education

From: Survey design and analysis considerations when utilizing misclassified sampling strata

  Design-agnostic Model-based Design-based
Variable OR (95% CI) OR (95% CI) OR (95% CI)
Race/ethnicity    
White 1.00 1.00 1.00
Black 1.11 (0.66, 1.86) 1.42 (0.60, 3.33) 0.71 (0.26, 1.92)
Asian 0.80 (0.46, 1.38) 1.69 (0.71, 4.02) 0.69 (0.33, 1.44)
Other 0.76 (0.43, 1.35) 1.19 (0.61, 2.31) 1.23 (0.37, 4.07)
Hispanic 0.67 (0.41, 1.09) 0.78 (0.42, 1.47) 0.24 (0.08, 0.76)
Low income    
No (Income ≥$30,000) 1.00 1.00 1.00
Yes (Income <$30,000) 1.25 (0.81, 1.95) 1.24 (0.79, 1.95) 1.46 (0.56, 3.83)
Age in years    
≤35 1.03 (0.69, 1.52) 1.54 (0.58, 4.10) 0.87 (0.39, 1.94)
>35 1.00 1.00 1.00
Gender    
Male 1.00 1.00 1.00
Female 0.69 (0.49, 0.99) 1.23 (0.34, 4.45) 1.03 (0.49, 2.16)
Rural living    
No (Suburban/Urban) 1.00 1.00 1.00
Yes (Rural) 0.83 (0.59, 1.18) 0.82 (0.57, 1.17) 0.63 (0.31, 1.27)
Education    
Less than HS 0.96 (0.48, 1.92) 0.90 (0.43, 1.86) 0.90 (0.13, 6.37)
HS to some college 0.95 (0.64, 1.39) 0.93 (0.62, 1.39) 1.30 (0.58, 2.89)
At least college graduate 1.00 1.00 1.00