Factors influencing women’s sex work in a Lebanese sample: results of a case-control study

BMC Women's Health

Table 2 Multivariable analysis: Logistic regression taking non-sex worker^a/sex worker as the dependent variable

Model 1: Independent variables: Sociodemographic characteristics
Variable	p-value	ORa	95% CI
Education level	< 0.001
Illiterate		1
Primary	0.683	1.50	0.21–10.52
Complementary	0.122	0.27	0.05–1.42
Secondary	0.359	0.46	0.09–2.41
University	0.001	0.09	0.02–0.35
Model 2: Independent variables: Sociodemographic characteristics and child and partner abuse.
Variable	p-value	ORa	95% CI
Education level	0.031
Illiterate		1
Primary	0.907	0.88	0.10–7.45
Complementary	0.266	0.35	0.05–2.25
Secondary	0.116	0.21	0.03–1.47
University	0.008	0.13	0.03–0.59
Partner non-physical abuse	< 0.001	1.03	1.02–1.05
Model 3: Independent variables: Sociodemographic characteristics, alcohol use, child and partner abuse and stress, depression and anxiety.
Variable	p-value	ORa	95% CI
Anxiety (HAMA scale)	0.002	1.08	1.03	1.13
Partner physical abuse	0.004	1.02	1.01	1.04

Hosmer and Lemeshow test (p = 0.745); Nagelkerke R² = 0.617
Variables entered in the final model: education level, socioeconomic status, stress, anxiety, depression, psychological child abuse, physical child abuse, sexual child abuse, partner physical abuse, partner nonphysical abuse, alcohol use and substance use
^a Reference group; numbers in bold indicate significant p-values

ISSN: 1472-6874