Skip to main content

Advertisement

Table 4 Summary of hierarchical logistic regression analysis using predisposing, enabling and need factors as predictors of substance abuse treatment utilization a, b

From: Inequitable access to substance abuse treatment services in Cape Town, South Africa

Predictor variables Model 1 Model 2 Model 3
  Wald (df) OR c (95% CI d ) Wald(df) OR c (95% CI d ) Wald(df) OR c (95% CI d )
Need for treatment       
Seriousness of drug use: reference "not at all serious")       
Slightly serious 3.55(1) 0.55 (0.29-1.02) 3.34(1) 0.56 (0.30-1.04) 0.15(1) 1.36 (0.30-6.22)
Moderately serious 18.39(1)*** 0.24 (0.13-0.46) 18.59(1)*** 0.24 (0.12-0.45) 2.13(1) 0.38 (0.11-1.39)
Considerably serious 0.01(1) 1.01 (0.54-1.89) 0.02(1) 1.05 (0.56-1.97) 5.45(1)* 5.11 (1.30-20.09)
Extremely serious 10.87(1)** 2.92 (1.54-5.89) 10.67(1)** 2.92 (1.54-5.54) 8.79(1)** 8.51 (2.07-35.03)
Others suggest treatment need (yes) 14.12(1)*** 2.36 (1.51-3.70) 13.83(1)*** 2.35 (1.50-3.68) 1.97(1) 1.99 (0.76-5.22)
SOCRATES-problem recognition 3.03(1) 0.96 (0.92-1.01) 3.33(1) 0.96 (0.91-1.00) 2.99(1) 0.92 (0.83-1.01)
SOCRATES-ambivalence 9.33(1)** 1.13 (1.05-1.22) 10.01(1)** 1.14 (1.05-1.23) 20.39 (1)*** 1.42(1.22-1.65)
SOCRATES-taking steps 105.02*** 1.16 (1.13-1.19) 101.46(1)*** 1.16 (1.13-1.19) 2.90(1) 1.06 (0.99-1.13)
Predisposing variables       
NES - - 5.18(1)* 0.95 (0.91-0.99) 0.11(1) 0.99 (0.90-1.08)
Enabling variables       
Number of known treatment centres - - - - 72.33(1)*** 5.80 (3.85-8.75)
Travelling time to treatment - - - - 64.35(1)*** 0.07 (0.04-0.14)
Competing financial priorities (No) - - - - 20.23(1)*** 5.31 (2.56-10.98)
Affordability barriers - - - - 30.02(1)*** 0.85 (0.81-0.91)
Delays in accessing treatment - - - - 7.16(1)** 0.93 (0.88-0.98)
TCU depression - - - - 1.61(1) 1.05 (0.97-1.14)
TCU anxiety - - - - 3.35(1) 0.93 (0.86-1.01)
Abstinence support - - - - 4.66(1)* 0.94 (0.87-0.99)
Treatment concerns      32.08 (1)*** 1.17 (1.11-1.23)
Stigma consciousness      0.45(1) 1.08 (0.86-1.34)
  1. *α < .05; ** α < .01; *** α < .001;
  2. a Controlling for confounding effects of gender and race
  3. b Blank spaces indicate that the variable did not enter the equation
  4. c Estimated odds ratio
  5. d 95% confidence interval based on the Wald's test