Skip to main content
. Author manuscript; available in PMC: 2016 Jan 1.
Published in final edited form as: Drug Alcohol Depend. 2014 Nov 26;0:75–80. doi: 10.1016/j.drugalcdep.2014.11.010

Table 3.

Association between having documented crystal methamphetamine dependence and 1) HIV seroconversion and 2) returning for >1 PEP course

HIV seroconversion1 Repeat NPEP Use2
HR (95% CI) aHR (95% CI) OR (95% CI) aOR (95% CI)
CM dependence 2.49 (1.18 to 5.26) 4.04 (1.71 to 9.53) 5.87 (3.38 to 10.2) 5.30 (2.87 to 9.79)
Age at first PEP visit 0.95 (0.91 to 0.99) 1.01 (0.99 to 1.03)
In primary care at Fenway Health 4.00 (0.53 to 30.3) 2.92 (1.75 to 4.88)
Any health insurance 1.20 (0.57 to 2.52) 1.72 (1.09 to 2.69)
Race/ethnicity
White/Caucasian Ref Ref
Hispanic/Latino 2.42 (0.98 to 5.99) 0.97 (0.51 to 1.83)
African American 4.40 (1.20 to 16.1) 1.02 (0.40 to 2.60)
Asian NA 1.43 (0.60 to 3.40)
Other NA 0.99 (0.33 to 2.90)
History of homelessness 0.79 (0.57 to 2.52) 0.53 (0.14 to 1.93)
History of sex work 0.85 (0.09 to 7.84) 0.65 (0.14 to 2.90)
Depression at time of PEP 0.89 (0.38 to 2.10) 1.21 (0.76 to 1.94)
Anxiety at time of PEP 1.02 (0.43 to 2.43) 0.91 (0.56 to 1.48)
ADHD at time of PEP 0.54 (0.13 to 2.34) 1.72 (0.94 to 3.17)
Bipolar I/II at time of PEP NA 1.35 (0.54 to 3.38)
Year of 1st NPEP Prescription NA 0.92 (0.87 to 0.96)

Abbreviations: HR: Hazards Ratio (for Cox models); OR: Odds Ratio (for logistic models); aHR: adjusted Hazards Ratio (for multivariable Cox models); aOR: adjusted Odds Ratio (for multivariable logistic models); CI: confidence interval; Italics indicates statistical significance (P<0.05)

1

Cox proportional hazards model

2

Logistic regression model