Application of data mining techniques and logistic regression to model drug use transition to injection: a case study in drug use treatment centers in Kermanshah Province, Iran

Table 5 Logistic regression model based on the six important variables of the SVM model

Variables	Odds ratio (OR)	95% confidence interval for OR		P-value
Occupational status (reference = Housewife)
Unemployed	1.495	0.40	5.54	0.547
Self-employed	0.782	0.22	2.83	0.708
Employee	0.362	0.84	1.57	0.174
Motivation for starting the drug use (reference = unemployment)
Pleasure	0.483	0.25	0.94	0.032
Drug use of friends	0.932	0.48	1.81	0.836
Curiosity	1.478	0.70	3.13	0.307
Use as a pain reliever	0.122	0.09	0.21	< 0.001
Emotional distress and mental	0.634	0.27	1.49	0.297
Others	0.324	0.12	0.89	0.029
History of taking heroin
Yes	12.23	8.33	17.81	< 0.001
History of taking cocaine
Yes	5.92	3.71	9.48	< 0.001
History of taking hallucinogens
Yes	19.01	43.38	8.24	< 0.001
History of prison
Yes	2.43	1.70	4.35	< 0.001

ISSN: 1747-597X