Skip to main content
. 2016 Nov 29;11:38. doi: 10.1186/s13011-016-0082-1

Table 3.

Adjusted risk ratios of substance use treatment enrollment outcomes vs. successfully enrolling: RAPiDS (n = 200)

Never attempted to enroll (n = 91) Unsuccessfully attempted to enroll (n = 39)
Adjusted risk ratio 95% Confidence Interval (CI) p - value Adjusted risk ratio 95% Confidence Interval (CI) p - value
Ethnicity
 Hispanic or Latino descent 0.30 (0.10, 0.95) 0.040 0.12 (0.01, 1.07) 0.058
 Non-Hispanic 1.00 (Ref) 1.00 (Ref)
Race
 White 1.00 (Ref) 1.00 (Ref)
 Non-white 3.16 (1.28, 7.83) 0.013 1.39 (0.44, 4.43) 0.578
Monthly income
  < $501 1.00 (Ref) 1.00 (Ref)
 $501 - $1500 3.93 (1.53, 10.12) 0.005 5.36 (1.79, 16.03) 0.003
  > $1500 2.16 (0.90, 5.80) 0.128 2.32 (0.74, 7.31) 0.151
Ever overdosed by accident
 Yes 0.50 (0.19, 1.34) 0.169 2.71 (1.06, 6.91) 0.037
 No 1.00 (Ref) 1.00 (Ref)
Drug-related discrimination by medical community
 Yes 0.25 (0.10, 0.62) 0.003 1.33 (0.55, 3.27) 0.527
 No 1.00 (Ref) 1.00 (Ref)
Ever incarcerated in jail or prison
 Yes 0.31 (0.14, 0.66) 0.003 0.99 (0.40, 2.41) 0.977
 No 1.00 (Ref) 1.00 (Ref)

Notes Model adjusted for recruitment source

The log likelihood of the model before stepwise removal is −151.39

The log likelihood of the model after stepwise removal is −157.95

The Nagelkerke R-squared of the model before stepwise removal is 0.478

The Nagelkerke R-squared of the model after stepwise removal is 0.432

The mean variance inflation factor for the model before stepwise removal is 1.33

The mean variance inflation factor for the model after stepwise removal is 1.15

The final model uses multinomial logistic regression and has 16 degrees of freedom