Table 4.

Factors associated with drop-out (fewer weeks in treatment), Cox proportional hazard models

Baseline or induction-related characteristic	All patients, n = 477				Inductions, n = 302
	Hazard ratio	95% CI	Adjusteda HR	95% CI	Hazard ratio	95% CI	Adjustedb HR	95% CI
Age (increasing by year)	0.99	0.97–1.00	0.98	0.97–1.00	0.99	0.97–1.00	0.99	0.97–1.00
Unemployed	0.87	0.68–1.10	–	–	0.72	0.54–0.95	0.74	0.54–1.01
Year of First Visit (ref. 2007)
2008	0.66	0.45–0.96	0.81	0.54–1.20	1.02	0.63–1.66	1.10	0.62–1.96
2009	0.65	0.48–0.88	0.67	0.49–0.91	0.78	0.55–1.11	0.94	0.64–1.38
2010	0.75	0.55–1.04	0.75	0.54–1.04	0.72	0.48–1.07	0.88	0.57–1.35
2011	0.67	0.47–0.97	0.70	0.49–1.01	0.83	0.54–1.28	1.00	0.63–1.58
2012	0.62	0.36–1.06	0.61	0.35–1.05	0.61	0.31–1.22	0.67	0.33–1.36
2013	0.64	0.29–1.38	0.62	0.29–1.34	0.60	0.24–1.50	0.70	0.28–1.78
Inducted	1.71	1.36–2.16	1.46	1.10–1.93	–	–	–	–
Prior buprenorphine	1.29	0.49–0.82	–	–	0.79	0.60–1.04	0.79	0.58–1.06
Heroin use, active	1.59	1.27–1.99	1.25	0.96–1.64	1.20	0.89–1.61	1.27	0.93–1.75
Cocaine use, active	1.32	1.03–1.71	1.18	0.90–1.54	1.22	0.91–1.63	1.15	0.83–1.59
Outpatient counseling, active	0.97	0.72–1.31	–	–	0.81	0.54–1.21	–	–
12-step attendance, active	1.01	0.79–1.29	–	–	1.02	0.77–1.37	–	–
Any induction-related AE	–	–	–	–	1.24	0.84–1.81	–	–
Methadone-to-buprenorphine induction	–	–	–	–	1.02	0.64–1.64	–	–

Factors not shown and not significantly associated with retention among all patients or inductions: gender, ethnicity, homelessness, uninsured, active benzodiazepine or cannabis use

^aHazard model adjusted for age, year of first visit, inducted, active baseline heroin and cocaine use. All transfer patients had prior buprenorphine experience; adding this term to the model reduces the significance of induction and vice versa

^bHazard model adjusted for age, unemployment, year of first visit, prior buprenorphine experience, active baseline heroin and cocaine use