Table 2.

Demographic and Clinical Characteristics of Patients Receiving Buprenorphine (January 2015 to December 2017) Related to Retention in December 2018 (n = 238)

Variable	Retained N = 82 n [%]		Dropped N = 156 n [%]		χ² ort	P
Age, years, mean (SD)	37.4	(11.5)	34.4	(10.6)	−1.98	.049
Female	40	[49]	58	[37]	2.99	.08
Marital status						.66
Single, never married	63	[78]	125	[81]
Married	14	[17]	24	[16]
Divorced, separated, or widowed	4	[5]	5	[3]
Race					5.60	.06
Black	9	[11]	37	[24]
White	64	[78]	104	[67]
Other	9	[11]	15	[10]
Insurance coverage						<.001
Medicaid	51	[62]	88	[56]
Other insurance	31	[38]	49	[32]
Self-pay/uninsured	0	[0]	19	[12]
Comorbid conditions
ADHD	11	[13]	16	[10]	0.55	.47
Anxiety	42	[51]	59	[38]	3.95	.047
Bipolar	6	[7]	8	[5]	0.47	.50
Chronic pain	17	[21]	17	[11]	4.25	.039
Depression	31	[38]	51	[33]	0.62	.43
Personality disorder	5	[6]	5	[3]	1.12	.29
Psychosis	3	[4]	3	[2]		.42
PTSD	6	[7]	13	[8]	0.08	.78
Medications
ADHD/Stimulants	8	[10]	10	[6]	0.86	.35
Antipsychotics	12	[15]	23	[15]	0.00	.98
Benzodiazepines	2	[2]	4	[3]		1.00
Mood stabilizers	6	[7]	9	[6]	0.21	.64
SSRI/SNRI	28	[34]	64	[41]	1.07	.30
Other antidepressants	17	[21]	24	[15]	1.08	.30
Smoking status (n = 237)					11.40	.003
Current smoker	50	[60]	125	[81]
Former smoker	16	[20]	14	[9]
Never smoker	16	[20]	15	[10]
Continuity of care, mean (SD)	0.70	(0.12)	0.61	(0.24)	−3.87	<.001
Seen by behavioral health	23	[28]	24	[15]	5.44	.019
Buprenorphine status at first visit					2.16	.34
New induction	17	[21]	38	[24]
Restart	11	[13]	30	[19]
Continuation	54	[66]	88	[57]

ADHD, attention deficit hyperactivity disorder; PTSD, posttraumatic stress disorder; SSRI, selective serotonin reuptake inhibitor; SNRI, selective norepinephrine reuptake inhibitor; SD, standard deviation.
Independent samples t-test were used to examine differences in continuous variables (signified by mean, SD) by retention status; χ² tests were used to examine differences in categorical variables by retention status. The retained category includes 6 patients who died during the study observation period and are censored in the survival analysis. For variables with cells < 5, a Fisher’s exact test was used to calculate the P-value (marital status, insurance type, benzodiazepine prescription, and psychosis). For tables larger than 2 × 2, SAS uses a Monte Carlo simulation to estimate the P-value in Fisher’s exact tests.