Perceptions of neighborhood social environment and drug dependence among incarcerated women and men: a cross-sectional analysis

Table 4 The logistic regression results of the mediating effect of mental health problem history ^a and intimate partner violence ^b to the relationship between drug dependence and fear of neighborhood by gender

Gender	Step and variable	B	S.E.	Wald	OR	p	Dependent
Female	Step 1						Drug dependence
	Age	0.01	0.02	0.60	1.01	0.44
	Fear of neighborhood	0.56	0.19	9.00	1.75	0.01
	Constant	−1.01	0.59	2.97	0.37	0.09
	Step 2^a						Mental health problem history
	Age	0.01	0.01	0.45	1.01	0.50
	Fear of neighborhood	0.53	0.18	8.83	1.71	0.01
	Constant	0.69	0.52	1.75	0.50	0.19
	Step 3^a						Drug dependence
	Age	0.01	0.02	0.82	1.01	0.37
	Mental health problem history	1.13	0.29	15.52	3.08	< 0.001
	Constant	−0.97	0.55	3.13	0.38	0.08
	Step 4^a						Drug dependence
	Age	0.01	0.02	0.61	1.01	0.44
	Fear of neighborhood	0.48	0.19	6.47	1.61	0.01
	Mental health problem history	1.06	0.31	11.97	2.88	0.001
	Constant	−1.55	0.63	6.07	0.21	0.01
	Step 2^b						Intimate partner violence
	Age	0.01	0.01	0.10	1.00	0.76
	Fear of neighborhood	0.52	0.16	10.88	1.68	< 0.001
	Constant	−1.10	0.51	4.66	0.33	0.03
	Step 3^b						Drug dependence
	Age	0.01	0.02	0.28	1.01	0.60
	Intimate partner violence	1.01	0.28	13.03	2.73	< 0.001
	Constant	−0.57	0.52	1.18	0.57	0.28
	Step 4^b						Drug dependence
	Age	0.01	0.02	0.32	1.01	0.57
	Fear of neighborhood	0.45	0.19	5.35	1.56	0.02
	Intimate partner violence	0.93	0.30	9.45	2.54	0.002
	Constant	−1.19	0.60	3.92	0.30	0.05

Note: Age is the control variable. B s were tested through the Wald χ² statistic with d.f. = 1. ^a: The tests of mediating effects caused by mental health problem history. ^b: The tests of mediating effects caused by intimate partner violence. The test results for the male group were omitted, as the results were not significant.