Skip to main content
. Author manuscript; available in PMC: 2024 Sep 1.
Published in final edited form as: Womens Health Issues. 2023 Jul 20;33(5):541–550. doi: 10.1016/j.whi.2023.05.009

Table 2:

Multivariable logistic regression of intention to initiate PrEP on factors with significant bivariate associations

Variable Adjusted Odds Ratio Estimate 95% Wald Confidence Interval
Individual-level

Behavioral Exposure (past 12 months)
 Injection Drug Use (Yes vs. No) 2.23 1.004 4.95
 Casual Sex Partner (Yes vs. No) 1.15 0.72 1.83
 >2 Sex Partners (Yes vs. No) 1.29 0.79 2.11
 Transactional Sex (Yes vs. No) 2.59 1.03 6.48

Attitudes (1–5)1 1.56 1.13 2.15

Self-Efficacy (1–5) 1 1.32 1.02 1.72

Interpersonal-level

Global Injunctive Norm Score (1–5)1 1.15 0.90 1.47

Community-level

Global Descriptive Norm Score (1–5) 1 1.65 1.33 2.04

Health System-level

Heard about PrEP from Health Care Provider (Yes vs. No) 1.28 0.64 2.53
Discussed Taking PrEP with a Health Care Provider (Yes vs. No) 2.39 1.20 4.75

Structural-level

Race
 White (vs. African American) 1.04 0.45 2.45
 Other (vs. African American) 0.79 0.40 1.56

Education (Years of Schooling) 0.85 0.76 0.96

Income Level
 $15k–29k (vs. 0–$14k) 0.90 0.49 1.65
 $30k–49k (vs. $15k–29k) 1.63 0.87 3.07
 >= $50k (vs. $30k–49k) 0.40 0.17 0.94
1

The estimated odds ratio indicates the odds of behavioral intention to use for every one level increase in the score.