Skip to main content
. 2018 Jul 5;13:27. doi: 10.1186/s13011-018-0164-3

Table 2.

Bivariate and multivariate GEE analysis of factors associated with use of withdrawal management services in people who use drugs in Vancouver, Canada (n = 2001)

Unadjusted Adjusted
Characteristic Odds Ratio
(95% CI)
p – value** Odds Ratio
(95% CI)
p – value**
Male
 (yes vs. no) 1.26 (0.97–1.64) 0.084 1.62 (1.17–2.24) 0.004
White Ethnicity
 (yes vs. no) 1.07 (0.84–1.36) 0.599
Age
 (per 10 year increase) 0.66 (0.59–0.74) < 0.001 0.81 (0.70–0.94) 0.005
Homelessness*
 (yes vs. no) 2.47 (2.01–3.04) < 0.001 1.86 (1.45–2.38) < 0.001
In a stable relationship
 (yes vs. no) 0.82 (0.67–1.01) 0.064 0.91 (0.72–1.14) 0.402
Incarceration*
 (yes vs. no) 1.54 (1.16–2.05) 0.003
Regular Employment*
 (yes vs. no) 0.98 (0.78–1.24) 0.878
Sex Work*
 (yes vs. no) 1.58 (1.18–2.12) 0.002 1.42 (0.95–2.11) 0.085
Daily Heroin Use*
 (yes vs. no) 1.58 (1.25–2.01) < 0.001
Daily Cocaine Use*
 (yes vs. no) 1.97 (1.45–2.67) < 0.001 1.40 (0.98–1.99) 0.066
Daily Methamphetamine Use*
 (yes vs. no) 1.33 (0.88–2.01) 0.179
Daily Prescription Opioid Use*
 (yes vs. no) 1.58 (1.11–2.24) 0.011 1.37 (0.92–2.04) 0.119
High Risk Alcohol Use*
 (yes vs. no) 1.21 (0.94–1.56) 0.143
Binge Use (any substance)
 (yes vs. no) 1.63 (1.35–1.96) < 0.001 1.34 (1.08–1.67) 0.009
Injection Drug Use (any substance)*
 (yes vs. no) 2.19 (1.68–2.85) < 0.001
Overdose*
 (yes vs. no) 1.70 (1.19–2.44) 0.004
Use at Supervised Injection Facility*
 (yes vs. no) 2.24 (1.81–2.77) < 0.001 1.66 (1.30–2.11) < 0.001
Hospital Admission*
 (yes vs. no) 1.40 (1.13–1.73) 0.002 1.29 (0.99–1.67) 0.061
Mental Health illness
 (yes vs. no) 1.23 (0.96–1.57) 0.101
Opioid Agonist Therapy
 (yes vs. no) 0.94 (0.75–1.17) 0.556
Other Addiction Treatment/Supports*
 (yes vs. no) 3.64 (2.94–4.52) < 0.001 3.34 (2.64–4.22) < 0.001

* in the last 6 months

as per NIAAA guidelines (> 3 drinks/day or 7 drinks/week for females and > 4 drinks/day or 14 drinks/week for males)

Other than withdrawal management services or opioid agonist therapy

** A bivariate and multivariate generalized estimating equation (GEE) model was created based on examination of the quasi-likelihood under the independence model criterion (QIC) for GEE and p-values