Skip to main content
. 2018 Jun 18;13(6):e0198853. doi: 10.1371/journal.pone.0198853

Table 3. Logistic regression models for predictors of transactional sex.

Variables Weighted univariate Weighted multivariate
OR (95% CI) P-value aOR (95% CI) P-value
Has other sources of income
No 1.0 1.0
Yes 1.97 (1.04–3.76) 0.037 2.07 (0.94–4.59) 0.070
Current living arrangements
Alone 1.0 1.0
With someone 0.54 (0.31–0.97) 0.041 0.51 (0.23–1.09) 0.083
Number of life time sexual partners
0–9 1.0 1.0
10–19 1.06 (0.53–2.13) 0.848 0.80 (0.36–1.78) 0.592
20+ 4.19 (1.83–9.63) 0.001 2.92 (0.69–12.0) 0.142
Age at sexual debut 0.89 (0.78–1.02) 0.095 1.05 (0.85–1.29) 0.625
Had sex with a male customer in the past 6 months
No 1.0 1.0
Yes 7.38 (4.11–13.26) <0.001 7.84 (3.63–16.98) <0.001
Last sexual intercourse
< 1 week ago 1.0 1.0
1–2 weeks ago 0.72 (0.38–1.37) 0.235 1.34 (0.56–3.18) 0.502
3–4 weeks ago 0.35 (0.11–1.13) 0.079 1.29 (0.36–4.66) 0.686
More than 4 weeks ago 0.33 (0.15–0.72) 0.005 0.26 (0.08–0.84) 0.024
Used a condom during last sex
No 1.0
Yes 1.02 (0.57–1.81) 0.941
Had unprotected sex under the influence of alcohol in the past 6 months
No 1.0 1.0
Yes 2.52 (1.14–4.46) 0.001 2.42 (1.18–4.96) 0.016
Frequency of alcohol consumption
Non-frequent 1.0 1.0
Frequent 2.87 (1.56–5.27) 0.001 2.06 (1.04–4.10) 0.038

Notes: Age at sexual debut was included as a continuous variable; CI = Confidence Interval; OR = Odds Ratio; aOR = Adjusted Odds ratio.