Table 2

Logistic Regression analysis of factors associated with food insufficiency among injection drug users (n = 1,053)

Variable

Adjusted Odds Ratio (AOR)

95% Confidence Interval (95% CI)

Wald Chi-Square^{†, ‡}

p - value

Age

(per year increase)

0.994

(0.98 - 1.01)

0.65

0.420

DTES residency

(yes vs no)

1.34

(0.96 - 1.89)

2.89

0.089

Unstable housing

(yes vs no)

1.68

(1.20 - 2.36)

9.04

0.003

Money spent on drugs per day

(≥ $50 vs < $50)

1.43

(1.06 - 1.91)

5.64

0.018

Incarceration **

(yes vs no)

1.16

(0.79 - 1.69)

0.57

0.451

Symptoms of Depression*

(≥ 16 CES-D vs < 16 CES-D)

3.32

(2.45 - 4.48)

62.79

< 0.001

Daily injection heroin **

(yes vs no)

1.27

(0.92 - 1.75)

2.05

0.152

Daily non-injection crack/rock **

(yes vs no)

1.18

(0.86 - 1.61)

1.02

0.313

Any injection or non-injection drug binge **

(yes vs no)

1.15

(0.86 - 1.53)

0.90

0.342

* Values for 125 participants missing from univariate treated as a separate category

** In the past six months

^† All Chi-Square tests were performed with one degree of freedom (df = 1), with the exception of the

'symptoms of depression' variable, where df = 2.

‡ Chi-Squared Test were applied to all categorical variables. A Wald Test was applied to the continuous

'age' variable, yielding a Wald Chi-Square value.

Anema et al. Substance Abuse Treatment, Prevention, and Policy 2010 5:20   doi:10.1186/1747-597X-5-20