Reasons for Benzodiazepine Use Among Persons Seeking Opioid Detoxification

Abstract

Background

Over the past decade, patients admitted to addiction treatment programs have reported increasing rates of concurrent opioid and benzodiazepine (BZD) use. This drug combination places individuals at high risk for accidental overdose. Little is known about reasons for BZD use among individuals seeking treatment for opioid use disorders.

Methods

We surveyed consecutive persons initiating inpatient opioid detoxification and identified 176 out of 438 who reported BZD use in the past 30 days and/or had a positive toxicology.

Results

Forty percent of persons surveyed used a BZD in the month prior to admission, and 25% of these met criteria for BZD dependence (DSM IV). BZD users averaged 32.0 years of age, 63.6% were male, 85.2% used heroin, and reported, on average, 13.3 (± 11.2) days of BZD use during the past month. Alprazolam (Xanax) was the most commonly used BZD (52%), and buying it on the street the most common source (48%). The most commonly reported reason for BZD use was ‘to manage anxiety’ (42.6%), followed by ‘to get or enhance a high’ (27.7%), ‘to help with sleep’ (11.4%), and ‘to decrease opioid withdrawal’ (10.2%). The most common reason for BZD use was significantly associated (p < .001) with most likely source of BZDs, with persons who got their BZDs from a prescriber (23%) more likely to report BZD anxiety as their primary reason for use, while persons who bought BZDs on “the street” (48%) had the highest likelihood of reporting using BZD to get or enhance a high. Participants using BZDs most commonly for anxiety did not endorse lower anxiety than those using BZDs for other reasons.

Conclusions

Two in five persons seeking detoxification for an opioid use disorder used a BZD in the prior month. Anxiety was the most common reason patients reported using a benzodiazepine, but they also reported using BZDs to enhance a ‘high’ and manage opioid withdrawal. Evidence-based discussions about the risks of combining BZDs and opioids, and alternatives to BZDs should be a high priority in detoxification settings.

1.0 INTRODUCTION

Use of benzodiazepines (BZDs) is prevalent in patients with opioid use disorders (Bleich, et al., 1999; Brands, et al., 2008; Lavie, Fatseas, Denis, & Auriacombe, 2009), but there has been a dramatic increase in concurrent use in the recent past. Patients admitted to addiction treatment programs reporting combined use of opioids and BZDs increased 570% between 2000 and 2010 (Substance Abuse and Mental Health Services Administration Center for Behavioral Health Statistics and Quality, 2012). Concurrent use of benzodiazepines in persons receiving opioid replacement therapy such as buprenorphine or methadone can be as high as 70% (Nielsen, Dietze, Lee, Dunlop, & Taylor, 2007).

Serious health problems may arise for people using opioids and BZDs in combination. BZDs’ sedative actions target the GABA_A receptor and adverse effects depend on the presence of impaired hepatic function, and complex drug-drug interactions (via CYP system metabolism), which can produce increased opioid serum concentrations (Jann, Kennedy, & Lopez, 2014). Concurrent use of both can slow heart rate and breathing, and increases the risk of accidental overdose (Gudin, Mogali, Jones, & Comer, 2013; Jones, Mogali, & Comer, 2012; Park, Saitz, Ganoczy, Ilgen, & Bohnert, 2015). According to the Centers for Disease Control and Prevention (CDC), opioid overdose related deaths nearly quadrupled between the years 1999 to 2011 (Volkow, Frieden, Hyde, & Cha, 2014); an estimated 31% of opioid related deaths were associated with concurrent BZD use (Chen, Hedegaard, & Warner, 2014). Among veterans prescribed opioids, BZD use is associated with greater mortality in a dose-dependent fashion (Park, et al., 2015).

Few studies to date have explored the reasons why persons with opioid use disorders use benzodiazepines. Several studies have found that patients receiving chronic opioid agonist treatment (OAT) often use BZDs to help with anxiety and sleep (Gelkopf, Bleich, Hayward, Bodner, & Adelson, 1999; Posternak & Mueller, 2001; Vogel, et al., 2013). Other work in OAT populations has suggested patients in methadone maintenance treatment take BZDs to get high (Chen, et al., 2011; Fatseas, Lavie, Denis, & Auriacombe, 2009; Gelkopf, et al., 1999; Iguchi, Handelsman, Bickel, & Griffiths, 1993). Fewer studies exploring reasons for use have focused on treatment-seeking opioid users. Inciardi et al reported that individuals using BZDs non-medically, may use BZDs to decrease the discomfort of withdrawal in situations when opioids are unavailable, as BZDs may be easier or less expensive to obtain than opioids in certain regions (Inciardi, Surratt, Cicero, & Beard, 2009).

Short-term inpatient medical detoxification is a common initial site of care for treatment-seeking persons with opioid use disorders (Bailey, Herman, & Stein, 2013; Carrier, et al., 2011; Mark, Dilonardo, Chalk, & Coffey, 2002; Stein, Anderson, Thurmond, & Bailey, 2015; Substance Abuse and Mental Health Services Administration, 2004), and offers a window on BZD use among opioid users in the community prior to seeking care. The purposes of this study were 1) to report the prevalence of BZD use among patients seeking treatment for an opioid use disorder, 2) to explore the most common reasons for and sources of BZDs in this population, and 3) to determine if these reasons were associated with BZD use severity or measures of anxiety and depression.

2.0 METHODS

2.1 Recruitment

Between September 2014 and May 2015, persons seeking opioid detoxification were approached within the first 24 hours of admission to Stanley Street Treatment Addiction and Recovery, Inc. (SSTAR) in Fall River, Massachusetts to participate in a survey research study. SSTAR’s program, one of the largest in Southeastern New England, has 38 beds and is a 24-hour medically supervised treatment facility that provides evaluation and withdrawal management with a mean length-of-stay of 4.9 days using a methadone protocol (as well as individual and group counseling and case management).

Of patients admitted to SSTAR during the recruitment period, 452 were opioid users who were 18 years or older, English-speaking, and able to provide verbal informed consent as approved by the Butler Hospital Institutional Review Board. Fourteen refused study participation or were discharged before staff could interview them. The remaining 438 persons completed a face-to-face interview and were not incentivized. All surveys were administered by non-treating research staff and required approximately 15 minutes. Persons were defined as BZD users if they responded to the question, “How many days in the last 30 have you used benzos (for instance, Ativan, Klonopin, Librium, Xanax, Valium)?” by reporting at least one day of use, or if they were positive for BZD on urine toxicological testing at entry to detoxification. This left a final sample of 176 for the current analysis.

2.2 Measures

Sample descriptors included age, gender, race/ethnicity, employment (part- or full-time vs. unemployed), and years of education. Regarding previous opioid treatment, we asked if participants had ever been in opioid detoxification in the past, or had ever received methadone or been prescribed buprenorphine. Participants were classified as recent cocaine users if they reported any use during the past 30 days. Participants also reported frequency and usual quantity of alcohol use during the past 30 days; hazardous drinking was defined as > 7 drinks / week for females or > 14 drinks / week for males (National Institute on Alcohol Abuse and Alcoholism, 2005). Participants were asked they were receiving professional treatment for depression or for anxiety, and if they were, what type they were receiving (counseling, medication, both counseling and medication). We asked “Which is the benzo you use most often?” Responses included Klonopin, Librium, Xanax, Valium, and ‘I don’t know.’ We also asked, “Where were you most likely to get the benzos that you used in the last 30 days?” Responses included: “bought them from someone else such as a dealer or on the street,” “bought or borrowed from a friend,” and “prescribed to you by a physician.” We asked participants: “Benzos can be used in lots of ways. Please indicate how often in the last month you have used benzos for the following reasons.” The list of reasons was derived from previous literature (Chen, et al., 2011; Fatseas, et al., 2009; Gelkopf, et al., 1999; Lintzeris & Nielsen, 2010) and included “To help you sleep,” “Because they make you high,” “To help manage your pain,” “To help manage your anxiety,” “To help manage your depression,” “To balance the effects of opiates,” “To balance the effects of alcohol,” “To balance the effects of cocaine,” “To help you drink less alcohol,” “To increase the high of opiates,” “To decrease the effects of opiate withdrawal,” or “Other reasons (please list).” The response options included 1 = never/almost never, 2=sometimes, 3=often, and 4 = always/almost always. We then asked, “What is the most common reason that you use benzos?” Depressive and anxiety symptoms were measured using the Patient Health Questionnaire (PHQ)-4 (Lowe, et al., 2010). Scores potentially range from 0–6 on the PHQ-2 Depression and PHQ-2 Anxiety Indices; persons screening positive for depression or anxiety have scores of 3 or higher on each, respectively. BZD use severity was measured using the Severity of Dependence Scale (Gossop, et al., 1995). This 5-item scale (each with a potential range 0–3) had high internal consistency reliability (alpha = .89), and the mean index score was 3.95 (SD = 4.85). A score of ≥ 7 is considered the cut-off for dependence (Gossop, et al., 1995).

2.3 Analytical Methods

We present descriptive statistics to summarize the characteristics of the sample. T-tests for differences in means and χ²-tests for differences in frequencies were used to compare males and females with respect to background characteristics, substance use behaviors, mood, and most common reason for BZD use. Because of non-normal distributions and small expected cell sizes, we used the nonparametric Kruskal-Wallis one-way ANOVA on ranks and Fisher’s exact test to compare benzodiazepine dependence severity, anxiety and depression by most common reason for BZD use.

3.0 RESULTS

Participants averaged 32.0 (± 8.8) years of age, 63.6% were male, 88.6% were non-Latino Caucasian, and mean educational attainment was 12.0 (± 1.8) years (Table 1). Eighty-two (46.6%) had seen a primary care physician in the past year, 68.8% reported a prior detox, and most (85.2%) were detoxing from heroin. On average they reported 13.3 (± 11.2) days of BZD use during the past month, the mean BZD severity dependence score was 3.8 (± 4.7), and 44 (25.0%) met screening criteria for BZD dependence. Over two-thirds (71.6%) reported injection drug use, 45.5% reported cocaine use, 39.8% met NIAAA criteria for hazardous drinking, and 80.1% and 75.0% met screening criteria for depression and anxiety, respectively, on the PHQ-2. About 39.2% had a history of methadone maintenance treatment and 46.6% of buprenorphine treatment.

Table 1.

Background Characteristics and Primary Reason for BZD Use (n = 176).

	n (%)	Mean (± SD)
Years Age		32.0 (± 8.8)
Gender (Male)	112 (63.6%)
Non-Latino Caucasian (Yes)	156 (88.6%)
Years Education		12.0 (± 1.8)
Seen Prim. Care Physician (Yes)	82 (46.6%)
Prior Detox (Yes)	121 (68.8%)
Detox From Heroin (Yes)	150 (85.2%)
Days of BZD Use (Past 30 Days)		13.3 (± 11.2)
BZD Severity Dependence		3.8 (± 4.7)
BZD Severity Dependence. ≥ 7	44 (25.0%)
Injection Drug Use (Yes)	126 (71.6%)
Any Cocaine Use (Yes)	80 (45.5%)
Hazardous Drinking (Yes)	70 (39.8%)
Mood (PHQ-4)		8.1 (± 3.2)
PHQ-2 (Depression Positive)	141 (80.1%)
PHQ-2 (Anxiety Positive)	132 (75.0%)
Ever Methadone RX (Yes)	69 (39.2%)
Ever Buprenorphine RX (Yes)	82 (46.6%)
Ever OAT (MMT or Bup) (Yes)	107 (60.7%)
Benzodiazepine Used
Alprazolam	92 (52.3%)
Clonazepam	53 (30.1%)
Diazepam	19 (10.8%)
Other	12 (6.8%)
Main Source of Benzodiazepines
Prescribed by Physician	41 (23.3%)
Friend or Family	50 (28.4%)
Purchased on Street	85 (48.3%)
Main Reason for BZD Use
To Help Manage Anxiety	75 (42.6%)
To Get or Enhance a High	43 (24.4%)
To Help Sleep	20 (11.4%)
To Decrease Opiate Withdrawal	18 (10.2%)
To Help Manage Depression	4 (2.3%)
Other Reasons	16 (9.1%)

Alprazolam (52.3%), clonazepam (30.1%), and diazepam (10.8%) were the most commonly reported benzodiazepines (Table 1). Over half (48.3%) reported purchasing on the street, 28.4% reported buying or borrowing from friends, and 23.3% said a prescription from a physician was their most likely source of BZDs. “To help manage anxiety” was the most frequently cited (42.6%) primary reason for BZD use; 24.4% stated they used BZDs to get or enhance a high, 11.4% said they used BZDs to help them sleep, and 10.2% said they used BZDs to help manage opioid withdrawal. Only 2.3% reported that managing depression was their primary reason for BZD use. Other cited reasons (9.1%) included managing pain and drinking less alcohol. Twenty-three persons (13.1%) reported taking a prescribed medication for depression or anxiety with 19 of these persons receiving medication for both anxiety and depression.

Neither mean benzodiazepine dependence severity scores (Kruskal-Wallis χ² = 9.16, df = 5, p = .103) nor the percentage meeting criteria for benzodiazepine dependence (Fisher’s exact p = .228) varied significantly by most common reason for BZD use (Table 2), although persons who used to manage anxiety or depression tended to have the highest mean dependence severity score and likelihood of meeting dependence criteria. Days of BZD use was significantly correlated with the continuous measure of BZD dependence severity (r= .58; p < .001) and associated with BZD dependence (p< .01). There was no significant difference between type of BZD most often used and the most common reason for use (p= .52).

Table 2.

Benzodiazepine Dependence Severity (Continuous And Dichotomized at ≥ 7) by Most Common Reason for Benzodiazepine Use (n = 176).

Most Common Reason for Benzodiazepine Use	BZD Dependence Severity	BZD Dependence Severity ≥ 7
	Mean (± SD)	n Yes (%)	n (No) (%)
To Help Manage Anxiety	4.53 (± 4.97)	24 (32.0%)	51 (68.0%)
To Get/Enhance High	4.30 (± 5.07)	9 (20.9%)	34 (79.1%)
To Help Sleep	1.50 (± 3.72)	2 (10.0%)	18 (90.0%)
To Reduce Opioid Withdrawal	3.00 (± 3.99)	3 (16.7%)	15 (83.3%)
To Help Manage Depression	4.50 (± 4.20)	2 (50.0%)	2 (50.0%)
Other Reason	3.06 (± 4.14)	4 (25.0%)	12 (75.0%)
	χ2 = 9.16, df = 5, p = .103a	Fisher’s Exact p = .228

^aKruskal-Wallis one-way analysis of variance on ranks.

There was no significant statistical difference in the percentage screening positive for either anxiety (Fisher’s exact p = .418) or depression (Fisher’s exact p = .822) by most common reason for BZD use (Table 3). Additional analysis (not reported in Table 3) found that mean PHQ-4 scores did not vary significantly by most common reason for using BZDs (Kruskal-Wallis χ² = 4.40, df = 5, p = .493). Differences in mean PHQ-2 anxiety scores were very similar (3.98 vs 4.04) among daily and non-daily BZD users (t = −0.18, p = .856).

Table 3.

Meeting PHQ-2 Screening Criteria for Anxiety and Depression by Most Common Reason for Benzodiazepine Use (n = 176).

Most Common Reason for Benzodiazepine Use	PHQ-2 Anxiety (Positive)		PHQ-2 Depression (Positive)
	n (%) Yes	n (%) No	n (%) Yes	n (%) No
To Help Manage Anxiety	59 (78.7%)	16 (21.3%)	60 (80.0%)	15 (20.0%)
To Get/Enhance High	34 (79.1%)	9 (20.9%)	36 (83.7%)	7 (16.3%)
To Help Sleep	14 (70.0%)	6 (30.0%)	16 (80.0%)	4 (20.0%)
To Reduce Opioid Withdrawal	10 (55.6%)	8 (44.4%)	15 (83.3%)	3 (16.7%)
To Help Manage Depression	3 (75.0%)	1 (25.0%)	3 (75.0%)	1 (25.0%)
Other Reason	12 (75.0%)	4 (25.0%	11 (68.8%)	5 (31.3%)
	Fisher’s Exact p = .418		Fisher’s Exact p = .822

Most common reason for BZD use was associated significantly (Fisher’s exact p < .001) with most likely source of BZDs (Table 4). About 70.3% of persons who reported that a physician prescription was their most likely source of BZD said anxiety was their primary reason for use; an additional 4.9% cited depression. Managing anxiety was also relatively frequently cited by those who said they got BZDs from family or friends (36.0%) or who purchased BZDs on the street (32.9%). Persons who purchased BZDs on the street had the highest likelihood (34.1%) of saying their primary reason for BZD use was to get or enhance a high.

Table 4.

Most Common Reason for Benzodiazepine Use by Most Likely Source (n = 176).

	Most Likely Source of Benzodiazepines			Row Total
Reason for Use	Physician Rx	Friend/Family	Street Purchase
To Manage Anxiety	29 (70.3%)	18 (36.0%)	28 (32.9%)	75 (42.6%)
To get/enhance a high	1 (2.4%)	13 (26.0%)	29 (34.1%)	43 (24.3%)
To help you sleep	5 (12.2%)	6 (12.0%)	9 (10.6%)	20 (11.4%)
To manage withdrawal	2 (4.9%)	5 (10.0%)	11 (12.9%)	18 (10.2%)
To Manage Depression	2 (4.9%)	1 (2.0%)	1 (1.2%)	4 (2.3%)
Other Reasons	2 (4.9%)	7 (14.0%)	7 (8.2%)	16 (9.1%)
Column Total	41 (100.0%)	50 (100.0%)	85 (100.0%)	176 (100.0%)
	Fisher’s Exact p < .001

Meeting screening criteria for BZD dependence was associated significantly (χ² = 8.40, df = 2, p = .015) with most likely BZD source. Rates of a positive BZD dependence screen were 31.7% among persons with physician prescribed BZDs, 30.6% among persons purchasing BZDs on the street, and 10.0% among those whose most likely source of BZDs was family or friends.

Participants reported using BZDs for multiple reasons (Table 5). Among persons who said that anxiety was the most common reason for BZD use, 45.3% said they sometimes or more often used BZDs to get or enhance a high, 77.3% said they sometimes used BZDs to help them sleep, 52.0% to help manage opioid withdrawal, 61.3% to help manage depression, and 38.7% said they sometimes used BZDs for other reasons. Overall, only 24 (13.6%) persons reported they sometimes or more often used BZDs for only 1 reason, 22 (12.5%) endorsed 2 reasons, 35 (19.9%) endorsed 3 reasons, 31 (17.6%) endorsed 4 reasons, 33 (18.8%) endorsed 5 reasons, and 31 (17.6%) said they sometimes or more often used BZDs for all 6 reasons.

Table 5.

Endorsement of Any BZD Use for Specific Reasons by Reason Endorsed as the Most Common Reason for Using BZDs.

	Most Common Reason for Benzodiazepine Use (Number Endorsing)
Endorsed Any BZD Use for This Reason	1. Anxiety (n = 75)	2. Get High (n = 43)	3. Help Sleep (n = 20)	4. Withdrawal (n = 18)	5. Depression (n = 4)	6. Other (n = 16)
1. Anxiety	75 (100%)	26 (60.5%)	14 (70.0%)	14 (77.8%)	3 (75.0%)	10 (62.5%)
2. To get/enhance a high	34 (45.3%)	43 (100.0%)	3 (15.0%)	8 (44.4%)	1 (25.0%)	11 (68.8%)
3. To help you sleep	58 (77.3%)	22 (51.2%)	20 (100%)	12 (66.7%)	2 (50.0%)	9 (56.3%)
4. To manage withdrawal	39 (52.0%)	31 (72.1%)	7 (35.0%)	18 (100.0%)	2 (50.0%)	10 (62.5%)
5. Depression	46 (61.3%)	19 (44.2%)	10 (50.0%)	8 (44.4%0	4 (100.0%)	7 (43.8%)
6. Other Reasons	29 (38.7%)	26 (60.5%)	2 (10.0%)	11 (61.1%)	0 (0.0%)	16 (100.0%)

4.0 DISCUSSION

In this sample of 438 persons with opioid use disorders seeking opioid detoxification, 176 (40%) used BZDs in the month prior to entering the detoxification program. This finding is noteworthy, because it is one of only a few studies that illustrates the prevalence of BZD use among treatment-seeking opioid-addicted individuals, many of whom are using heroin (85%) and injecting drugs (72%). We found that the mostly commonly used BZD was alprazolam (Xanax) (52%), followed by clonazepam (Klonopin) (30%), followed by diazepam (Valium) (11%). Alprazolam is commonly preferred among addicted persons due to its quick onset of action and subjective high (Lader, 2011). Clonazepam, with its slower onset of action and long duration, may be preferred for insomnia or opioid withdrawal.

Members of this already high-risk population are compounding their risk of overdose with concurrent use of benzodiazepines. Most public health campaigns highlighting the risk of overdose due to combining opioids and BZDS are directed at prescribers. Only 23% of our sample got their BZDs from a prescriber, lower than that reported in populations of opioid users in drug treatment (Chen, et al., 2011; Jann, et al., 2014). The majority got BZDs from “the street” (48%) or a friend or family member (28%). Prescribers continue to need education on the risks of combining opioids and benzodiazepines, but another important target audience is drug users themselves, who prior to seeking treatment may never cross paths with a health care provider in their pursuit and self-administration of opioids and BZDs and therefore may be missing out on the diagnosis of psychiatric symptoms and alternative treatments for anxiety or depression. Self-medication and peer-medication are critical issues to discuss during detoxification.

Most participants in our sample reported using BZDs for multiple reasons. The most common reason was to help manage anxiety (43%). The second most common reason was to get or enhance a high (28%). More than 45% of participant who reported using BZDs to manage anxiety, were also using BZDs to get high.

We detected an interesting relationship between reasons for BZD use and the primary source of BZDs. Nearly three in four persons who reported they use prescribed BZDs also reported anxiety as their primary reason for using BZDs. By contrast, only 3% of persons reporting using BZDs to get high got their BZDs from a prescription – the rest procured them from non-prescription sources (friends, family, the street). These data remind us that the source of a prescription medication plays an important role in a person’s self-narrative around why they are using that medication, which is why it is essential for clinicians to help patients who have become addicted to prescribed BZDs to unwind the anxiety-treatment narrative if it no longer applies. The proportion of persons fulfilling criteria for dependence on the Severity Dependence Scale (25%) is comparable here to that found in other opioid-using populations (Lavie, et al., 2009; Ross & Darke, 2000; Seivewright & Iqbal, 2002).

Participants using BZDs primarily for anxiety did not, of note, endorse lower overall anxiety than those using benzodiazepines for other reasons. As a group, our cohort of BZD users had much higher rates of anxiety than other reported samples of persons with opioid use disorders on OAT (Mark, Dilonardo, Vandivort, & Miller, 2013; Milby, et al., 1996; Rooney, Kelly, Bamford, Sloan, & O’Connor, 1999; Ross & Darke, 2000). One possible explanation for these findings is that BZDs used for anxiety in the context of an active, untreated opioid use disorder, are ineffective for that purpose, even when patients subjectively report anxiety-relief with their use. This would be consistent with the literature showing that psychiatric symptoms tend to get worse in the face of ongoing addictive substance use, even with concurrent psychotropic medications to treat that disorder (Brienza, et al., 2000). Another possible explanation is social desirability bias; participants who believe they need BZDs to treat anxiety, may endorse high level of anxiety to justify BZD use. An additional possibility is that most persons using BZD to manage anxiety were not procuring BZD from physicians and may be insufficiently treated, although we cannot deduce the effectiveness of BZDs for anxiety in this cross-sectional study. It is unknown whether the rates of anxiety would be higher or lower without BZDs. Finally, some patients may be experiencing anxiety as an effect of acute nicotine or alcohol withdrawal, or even opioid withdrawal, despite the site’s pharmacological management of these.

Our study had limitations. First, we relied on self-report of treatment histories. Second, recruitment was limited to individuals seeking inpatient opioid detoxification. Third, we recruited participants from only one location and our sample was primarily non-Latino Caucasian and male. Fourth, we did not inquire about the dose of BZD used, route of administration, or use patterns; future studies should examine these variables. Fifth, while it is likely that participants used more than one type of BZD, and indeed may have used different BZDs for different reasons, we were not able to perform this analysis. Qualitative research could shed light on this issue. Finally, we do not have clinical assessment of psychiatric disorders that may be associated with BZD use in this population, such as PTSD or borderline personality disorder, nor do we have clinical diagnoses of anxiety or depression.

5.0 CONCLUSIONS

This study offers additional information for clinicians and policy-makers to address the opioid use and opioid overdose epidemic, particularly as pertains to concurrent use of BZDs. Clinicians need to be aware that some persons with opioid use disorders use BZDs to get or augment the opioid high, in addition to using BZDs for symptoms such as anxiety, sleep, opioid withdrawal, and depression. Clinicians should educate their patients that although in the short term they may experience subjective relief of anxiety with BZDs, long-term use is likely to have limited effectiveness. Education on the risks, benefits and alternatives of BZDs need to be directed not just to clinicians, but also to opioid-addicted BZD users, many of whom who are more likely to be getting the BZD from a non-clinical source, and thus will not have the opportunity to gain knowledge of BZD risks from health care providers. The best option to manage anxiety and other symptoms commonly treated with BZDs remains unknown in this population. Whether the benefits of prescribing BZDs in the context of maintenance OAT outweigh the risks remains a debated issue. Either way, there is likely to be a fine line between therapeutic use and misuse, and the risk of developing a BZD use disorder may be high (Center for Substance Abuse Treatment, 2005; Seivewright & Iqbal, 2002) although we are not aware of prospective studies evaluating this possibility.

Highlights.

• Prior to entering detox, opioid users frequently used BZDs, most often alprazolam

• “To help manage anxiety” was the most frequently cited primary reason for BZD use

• Persons who bought BZDs on “the street” were most likely to report using BZD “to get or enhance a high.”

• 75.0% met screening criteria for anxiety despite BZD use