Urine Toxicology Profiles of Emergency Department Patients with Untreated Opioid Use Disorder: A Multi-Site View

Abstract

Background:

Opioid overdose deaths in 2021 were the highest ever, driven by fentanyl and polysubstance use.

Objective:

Characterize drug use, assessed by urine drug screens (UDS), in patients with untreated opioid use disorder (OUD) presenting to 28 Emergency Departments (EDs) nationally and by region.

Methods:

We analyzed UDS from patients enrolled in CTN-0099 ED-INNOVATION (Emergency Department-INitiated bupreNOrphine VAlidaTION) between July 12, 2020 and March 9, 2022. Participants were adult ED patients with OUD not engaged in addiction treatment with a UDS positive for an opioid, but negative for methadone. Sites were divided into “East” and “West” regions.

Results:

A UDS was available for all 925 enrolled participants, 543 from East and 382 from West. Fentanyl was in 702(76%) specimens [485(89%) East versus 217(57%) West, p<0.01] and was the only opioid in 269(29%). After fentanyl, the most common opioids were morphine (presumably heroin) 411(44%) [192(35%) East versus 219(57%) West, p<0.01] and buprenorphine 329(36%) [186(35%) East versus 143(37%) West, p=0.32)]. The most common drugs found with opioids were stimulants 545(59%), THC 417(45%) and benzodiazepines 151(16%). Amphetamine type stimulants were more common in West 209(55%) versus East 125(23%). Cocaine was more common in East 223(41%) versus West 82(21%). The presence of multiple drugs was common (759; 82%).

Conclusions:

Most participants had UDS specimens containing multiple substances with a high proportion having fentanyl, stimulants and buprenorphine. Regional differences were noted. Given the increased risk of death with fentanyl and polysubstance use, ED providers should be providing risk reduction counseling, treatment and referral.

Introduction

Opioid overdose deaths in 2021 were the highest ever recorded, driven by fentanyl and polysubstance use.¹ A recent Centers for Disease Control and Prevention (CDC) report on overdose deaths in the United States (US) from 2019–2020 demonstrated that 64% of these deaths involved a synthetic opioid other than methadone, primarily illicitly manufactured fentanyls.² There are also drug seizure data from the Drug Enforcement Agency (DEA) and other sources demonstrating that fentanyl is disproportionately distributed across the US with notable regional differences between sites East and West of the Mississippi River.^{3, 4} East of the Mississippi fentanyl has largely overtaken heroin markets, although there has been a notable recent expansion into western drug markets.⁵ Overdose deaths involving multiple substances, particularly stimulants, have also accelerated. In a recent report on trends in and characteristics of drug overdose deaths involving fentanyl, four in ten overdose deaths involved an opioid with a stimulant.²

While ongoing surveillance of drugs involved in fatal overdose deaths is well established through the CDC’s State Unintentional Drug Overdose Reporting System (SUDORS) there is less robust and systematic surveillance on overall drug use in untreated individuals with opioid use disorder (OUD) more generally. Furthermore, despite a critical need for ongoing surveillance, there are no national data on drug use profiles of patients with untreated OUD presenting to Emergency Departments (EDs). Given the high risk of overdose with fentanyl and polysubstance use, understanding drug use patterns generally and regional differences specifically could help target harm reduction messaging and highlight the need for EDs to offer buprenorphine, a medication shown to both reduce the risk of death from overdose and increase engagement in addiction treatment after an ED encounter.^{6, 7} ED initiated buprenophine can also help narrow the large treatment gap for patients with OUD.⁸

We sought to characterize drug use as assessed by urine drug screens (UDS) in patients with untreated OUD presenting to 28 geographically and practice diverse EDs, with a focus on comparing eastern and western regions.

Methods

Study Design

We present an analysis of entry UDS specimens from participants enrolled in the CTN-0099 Emergency Department-INitiated bupreNOrphine VAlidaTION (ED-INNOVATION) randomized controlled trial (RCT) between July 12, 2021 and March 9, 2022 (ClinicalTrials.gov: NCT04225598). The ED-INNOVATION protocol has been previously published.⁹ Briefly, the study is a Hybrid Type 1 Effectiveness-Implementation study, currently enrolling in the RCT component, comparing the effectiveness of two BUP formulations, sublingual and a 7-day extended-release injectable BUP among ED patients with OUD on the primary outcome of engagement in formation addiction treatment at 7-days post ED-visit. To be eligible, participants had to be English speaking adults 18 or older with untreated OUD. All participants met Diagnostic and Statistical Manual of Mental Disorders Five (DSM-5) criteria for moderate to severe OUD and had a positive UDS for opioids, but negative for methadone. Patients with a UDS positive for methadone were excluded because buprenorphine initiation is generally contraindicated in patients using a long acting opioid such as methadone. Patient were also excluded if they were pregnant, required hospitalization at the index visit, required opioids for ongoing pain management, were a prisoner or in police custody, were engaged in comprehensive addiction treatment within the 14-days prior to their ED visit, were unable to provide reliable locator information or were previously enrolled in the study.

Participant’s enrollment UDS was tested for opiates, oxycodone, buprenorphine, cocaine, benzodiazepine, barbiturates, phencyclidine, tetrahydrocannabinol (THC), Methylenedioxy-methamphetamine (MDMA), amphetamine, and methamphetamine using instant immunoassay tests (CLIAwaived, Inc. IDTCII). Fentanyl testing was performed using a separate immunoassay (CLIawaived, Inc. FEN20). The cutoff values for detection of each drug are shown in supplemental table A.

A sample size was determined for the overall ED INNOVATION trial (N=2000) but not for this descriptive analysis.⁹

The study was approved by the Western Institutional Review Board and all participants signed written informed consent for study participation.

Site Selection

ED-INNOVATION sites were chosen in June 2019 based on: (1) the prevalence of patients with International Classification of Diseases, 10th revision codes for the prior 12 months related to overdose, OUD and other opioid-related diagnoses; (2) ability to provide ED initiated BUP and (3) ability to link patients to an addiction program able to provide medications for OUD (buprenorphine or methadone) within 7-days. Sites were selected to reflect a mix of rural/urban/suburban; community/academic; geographic location; and size.

Analysis

Sites were divided into two regions based on their relation to the Mississippi River reflecting prior analysis showing significant historical differences in fentanyl related mortality East and West of the Mississippi River,⁴ and the division was relevant due to the study site distribution (figure 1). Demographic characteristics of study participants were summarized using counts and percentages for categorical variables. Continuous variables were summarized using mean and standard deviation or median and interquartile range (IQR). We compared demographic characteristics and UDS results between participants enrolled at East versus West sites using Fisher’s exact or Chi-square tests for categorical variables, as appropriate based on cell counts. Continuous variables were compared using a Student’s T test or Mann-Whitney Test depending on their distribution. We report the self-reported racial/ethnic makeup of the population as this was an a-priori variable of interest. Multivariable logistic regression models were used to adjust for differences in baseline characteristics between regions, including age, sex, race, ethnicity, route of use, and severity of use. All statistical analysis was performed using SAS Version 9.4 (SAS Institute Inc, Cary, North Carolina). An α=0.05 level of significance was used for all statistical testing.

Figure 1:

ED INNOVATION RCT Sites

Results

Site Characteristics

A total of 28 sites participated in the RCT (figure 1). Six sites were eliminated due to low enrollment but contributed data to this analysis. Sites contributed a median of 30 UDS results (range 4–76) (figure 2).

Figure 2:

Contribution of Urine Drug Screens by RCT Site

Population Characteristics

All 925 participants enrolled in the ED-INNOVATION RCT between July 12, 2020 and March 9, 2022 had a UDS available for analysis. East sites enrolled 543 participants and West sites enrolled 382 participants. Demographics are shown in Table 1. There were 100 patients excluded from study participation for a UDS positive for methadone. The population was predominantly male (619; 67%), white (538; 58%) and not Hispanic or Latino (788; 85%). Thirty percent (30%) of the population self-reported a race of Black or African American with no significant difference by region. A higher proportion of Hispanic individuals were enrolled in West (77; 20%) than East sites (47; 9%). The most frequent route of opioid use was nasal (322; 35%) followed by intravenous (253; 27%) and smoking (156; 17%). Reported route of opioid use differed by region with more participants enrolled in East sites reporting intravenous and nasal use whereas participants enrolled in West sites more commonly reporting smoking their drugs. Half (458; 50%) of all participants reported daily opioid use.

Table 1:

Demographic and Drug Use Characteristics of Study Participants by Region

	Region		Overall (N = 925)
	West (n = 382)	East (n = 543)
Age(years) (Mean (SD))	37.70 (11.3)	38.88 (12.5)	38.40 (12.0)
Days Since Last Opioid Use (Median (Range))	1.0 (1.0 – 7.0)	1.0 (1.0 – 7.0)	1.0 (1.0 – 7.0)
Sex
Female	131 (34.3%)	175 (32.2%)	306 (33.1%)
Male	251 (65.7%)	368 (67.8%)	619 (66.9%)
Race
Black or African American	106 (27.8%)	168 (30.9%)	274 (29.6%)
Other/Unknown	061 (36.0%)	052 (09.6%)	113 (12.2%)
White	215 (56.3%)	323 (59.5%)	538 (58.2%)
Ethnicity
Hispanic or Latino	077 (20.2%)	047 (08.7%)	124 (13.4%)
Not Hispanic or Latino	300 (78.5%)	488 (89.9%)	788 (85.2%)
Unknown	005 (01.3%)	008 (01.5%)	013 (01.4%)
Route of Use
IV	083 (21.7%)	170 (31.3%)	253 (27.4%)
Nasal	090 (23.6%)	232 (42.7%)	322 (34.8%)
Oral	038 (09.9%)	070 (12.9%)	108 (11.7%)
Other/NA/Missing	039 (10.2%)	047 (08.7%)	086 (09.3%)
Smoking	132 (34.6%)	024 (04.4%)	156 (16.9%)
Severity of Use
Daily	203 (53.1%)	255 (47.0%)	458 (49.5%)
Less than Daily	169 (44.2%)	283 (52.1%)	452 (48.9%)
Missing	010 (02.6%)	005 (00.9%)	015 (01.6%)

SD=Standard Deviation

Urine Drug Screen Analysis

The results of the UDS analysis are shown in Table 2. The presence of multiple substances in the UDS was common (759; 82%). Fentanyl was detected in 702(76%) specimens [485(89%) East versus 217(57%) West, p<0.01] and was the only opioid in 269(29%) (294(54%) East versus 139(36% West), p<0.01]. After fentanyl, the most common opioids were morphine (presumably heroin) 411(44%) [192(35%) East versus 219(57%) West, p<0.01] and buprenorphine 329(36%) [186(35%) East versus 143(37%) West, p=0.32)]. The most common drugs found with opioids were stimulants 545(59%) [294(54%) East versus 251(66%) West, p<0.01], THC 417(45%) [248(46%) East versus 169(44%) West, p=0.67] and benzodiazepines 151(16%) [105(19%) East versus 46(12%) West, p=0.003]. Among the stimulant class, amphetamine type stimulants were more common in the West 209(55%) versus East 125(23%), p<0.01 whereas Cocaine was more common in the East 223(41%) versus West 82(21%), p<0.01. Specimens where fentanyl was the only drug were rare, 52(6%) [East 38(7%) versus West 14(4%), p=0.03] as were specimens where opioids were the only drug 166(18%) [East 106(20%) versus West 60(16%), p=0.14]. After adjustment for differences in baseline covariates including age, sex, race, ethnicity, route of use and severity of use between the regions, the only statistically significant change was with benzodiazepine use. The statistical significance of all other unadjusted comparisons were unchanged after adjusting for differences in baseline covariates between the regions (Table 2).

Table 2:

Unadjusted and Adjusted Urine Drug Screen Results by Region and Overall

	region			Overall (N = 925)	Adjusted Odds Ratio (95% CI) East compared to West**	Adjusted P Value
	West (N = 382)	East (N = 543)	Unadjusted P Value
Opioids* + Any Other Drug	322 (84.3%)	437 (80.5%)	0.14	759 (82.1%)	0.794 (0.536, 1.178)	0.2523
Fentanyl + Any Other Drug	203 (53.1%)	447 (82.3%)	<0.001	650 (70.3%)	4.978 (3.553, 6.974)	<0.001
Fentanyl + Any Other Opioid *	139 (36.4%)	294 (54.1%)	<0.001	433 (46.8%)	2.303 (1.696, 3.126)	<0.001
Fentanyl + No Other Opioid *	078 (20.4%)	191 (35.2%)	<0.001	269 (29.1%)	2.509 (1.748, 3.602)	<0.001
Fentanyl Only (No other drug)	014 (03.7%)	038 (07.0%)	0.030	052 (05.6%)
Opioids* + Any Stimulant (Cocaine, Amphetamine or Methamphetamine)	251 (65.7%)	294 (54.1%)	<0.001	545 (58.9%)	0.607 (0.443, 0.832)	<0.001
Opioids* + Amphetamine or Methamphetamine	209 (54.7%)	125 (23.0%)	<0.001	334 (36.1%)	0.201 (0.140, 0.290)	<0.001
Opioids + Other Drug Listed Below
Amphetamine	188 (49.2%)	106 (19.5%)	<0.001	294 (31.8%)	0.208 (0.144, 0.300)	<0.001
Barbiturate	002 (0.5%)	003 (0.6%)	1.0	005 (0.5%)	***	***
Benzodiazepines	046 (12.0%)	105 (19.3%)	0.003	151 (16.3%)	1.399 (0.927, 2.110)	0.109
Buprenorphine	143 (37.4%)	186 (34.3%)	0.32	329 (35.6%)	0.794 (0.579, 1.090)	0.153
Cocaine	082 (21.5%)	223 (41.1%)	<0.001	305 (33.0%)	2.337 (1.677, 3.258)	<0.001
Marijuana	169 (44.2%)	248 (45.7%)	0.67	417 (45.1%)	1.123 (0.829, 1.521)	0.454
MDMA	061 (16.0%)	031 (5.7%)	<0.001	092 (9.9%)	0.371 (0.220, 0.624)	<0.001
Methamphetamine	197 (51.6%)	105 (19.3%)	<0.001	302 (32.7%)	0.198 (0.137, 0.285)	<0.001
Opiates	219 (57.3%)	192 (35.4%)	<0.001	411 (44.4%)	0.300 (0.215, 0.419)	<0.001
Oxycodone	018 (04.7%)	044 (08.1%)	0.042*	062 (06.7%)	***	***
Phencyclidine	007 (1.8%)	008 (1.5%)	0.67	015 (1.6%)	***	***

^*Opioids includes Buprenorphine, Fentanyl, Opiates or Oxycodone

^**All models are adjusted for: age, sex, race, ethnicity, Route of Use, and Severity of Use

^***Not calculated due to failure of model convergence secondary to low cell counts

Discussion

This report highlights five key findings regarding the drug use patterns of ED patients with untreated OUD nationally: 1) polysubstance use is nearly ubiquitous; 2) regional differences exist in route of drug use; 3) fentanyl use is extremely common; 4) concomitant stimulant and opioid use is common, albeit stimulant type varies in the East versus the West and 5) there is widespread, likely non-prescribed, use of buprenorphine in this population.

This data has important epidemiological implications. It demonstrates that polysubstance use is extremely common among ED patients with untreated OUD. Not only can polysubstance use increase the risk of overdose it also indicates a need for multimodal interventions and harm reduction strategies in the ED to address concomitant use disorders. It furthermore suggests that ED providers should be cognizant that patients with OUD may have co-occurring use disorders, such as stimulant use disorder, that could require a referral to more specialized addiction services.

While more than one drug was found in 82% of all UDS specimens, stimulants were the most common substances found with opioids, although the type of stimulant differed by region. Amphetamine type stimulants (Amphetamine/Methamphetamine) were more common in West sites whereas cocaine was more common in East sites. These drug use patters mirror those found in other studies and in toxicology results from fatal overdoses.^{2, 10} The increase in amphetamine type stimulant use and overdoses has been a concerning trend nationally given that concomitant stimulant use with opioids is associated with increased risk of overdose.^{1, 11, 12} Due to the listing of all possible drugs in a decedents system at the time of death it is unclear if this increased risk comes from the pharmacologic combination of the drugs or is simply a marker of more severe use disorders.¹³ Furthermore, while there has been an understandable intense focus on ED patients with OUD given the epidemic of opioid overdose related deaths, the significant amount of concomitant stimulant use indicates that more interventions addressing stimulant use disorders in this population are urgently needed.

Concerning opioids, fentanyl is the predominant opioid identified in ED patients with untreated OUD across the US. Even in West sites where the percentage of fentanyl positive specimens were less than in East sites, fentanyl was still found in more than 50% of all UDS specimens. While regional differences in fentanyl exist in this study, recent data suggest that this could change as fentanyl overtakes drug markets in the West.⁵ Of note, fentanyl was the only opioid detected in 29% of study participants. This has practical implications for EDs that exclude fentanyl from their routine toxicology screen. While opioid intoxication and overdose can often be determined by history and examination, if there is need to confirm opioid use with a UDS, patients only using fentanyl would screen negative. In a recent report of patients presenting to EDs nationally with overdose, testing for opiates was performed nearly 50% of the time in comparison to only 5% for fentanyl testing.¹⁴

Buprenorphine was also found in the UDS of more than one-third of study participants. This was somewhat unexpected given that entry into the study required patients to have untreated OUD with no engagement in formal addiction treatment within 2-weeks of study entry. While it is not clear if this buprenorphine was obtained through non-medical sources, studies have shown that patients are frequently taking non-prescribed buprenorphine to prevent withdrawal symptoms or maintain abstinence.¹⁵ In one study of treatment seeking patients with OUD, 42% of participants reported illicit buprenorphine use, with more than 92% of these patients using buprenorphine to reduce opioid withdrawal cravings and not to achieve euphoria.¹⁶ Furthermore, the use of non-prescribed buprenorphine dropped by 70% once patients were initiated onto treatment.¹⁶ A comprehensive literature review identified similar trends across multiple studies with only a small percentage of individuals using illicit buprenorphine for reasons related to abuse.¹⁷

Lastly, route of drug use clearly differed by region with study participants in West sites reporting a higher proportion of smoking whereas participants enrolled in East sites reported a higher proportion of intravenous and nasal use. These use patterns also reflect national trends, especially with the increase of fentanyl into Western drug markets and local reports of increasing smoking of fentanyl in the West.^{5, 18}

Limitations

This study had several limitations. Site selection was not random, thus may not reflect the drug use prevalence in specific areas of the country. Only participants enrolled in ED-INNOVATION were included in this analysis so our results may not be representative of the broader population of ED patients with untreated OUD. Despite the limited sample, our data comports with national data on both drug seizures and non-ED based surveillance programs.^{2, 3} Site enrollment, and thus the number of UDS specimens contributed per site differed. It is possible that sites that contributed a large number of UDS specimens could have skewed the results for a given region. While we adjusted our analysis for differences in baseline characteristics we did not have sufficient sample size to adjust for individual sites.

Conclusions:

In this national sample of ED patients with untreated OUD participating in an RCT comparing two different formulations of BUP, most participants had UDS specimens containing multiple substances with a high proportion having fentanyl, stimulants and buprenorphine. Given the increased risk of overdose death with fentanyl and polysubstance use, ED providers should be providing risk reduction counseling, treatment and referral for ongoing care.

ARTICLE SUMMARY.

1. Why is this topic important?

Opioid overdose deaths are at an all-time high, driven by fentanyl and polysubstance use. Understanding the frequency of fentanyl and polysubstance use in a national sample of ED patients with untreated opioid use disorder can help guide risk reduction strategies.

2. What does this study attempt to show?

The study characterizes drug use as assessed by urine drug screens (UDS) in patients with untreated OUD presenting to 28 geographically and practice diverse EDs, with a focus on comparing eastern and western regions.

3. What are the key findings?

Most participants had UDS specimens containing multiple substances with a high proportion having fentanyl, stimulants and buprenorphine. Regional differences were found in fentanyl, stimulant and benzodiazepine use.

4. How is patient care impacted?

Given the high prevalence of fentanyl and polysubstance use among ED patients with untreated OUD and the known associated risk of overdose, providers should routinely provide risk reduction counseling, treatment and referral for ongoing care.