Introduction
The role of urine drug screening (UDS), sometimes referred to as A Uds in clinical shorthand, is crucial in managing patients with opioid use disorder (OUD), particularly those undergoing buprenorphine-naloxone treatment in primary care settings. Federal guidelines from the Substance Abuse and Mental Health Services Administration mandate regular drug testing in opioid treatment programs, suggesting at least eight random tests annually per patient in maintenance therapy [15]. Similarly, the American Society of Addiction Medicine advocates for UDS during treatment to monitor patient adherence to prescribed medications and to detect the use of alcohol, illicit substances, and controlled drugs [3]. However, specific guidelines on the frequency of a uds testing are less defined, with recommendations suggesting that testing frequency should be tailored to individual patient factors such as stability, treatment type, and setting [3]. Clinicians rely on the interpretation of a uds results as a vital tool in patient care and treatment adjustments [11].
Despite the recognized importance of a uds, there’s a notable gap in the literature providing clear direction on optimal testing frequency across different clinical environments [5]. A comprehensive review of existing research on the impact and recommendations for UDS frequency in OUD patients receiving opioid agonist therapy revealed a significant lack of evidence supporting the association between frequent urine drug screening and improved health outcomes. This review highlighted an “urgent gap in research evidence underpinning an area of clinical importance,” emphasizing the need for more data to guide clinical practice [10]. While consensus guidelines from the American Society of Addiction Medicine Board of Directors have suggested weekly UDS in early recovery, potentially decreasing to monthly in stable recovery, they also acknowledged in their 2017 statement on drug testing in addiction medicine that further research in this area is warranted [1]. In pain management, UDS has shown some efficacy in identifying opioid misuse, potential drug diversion, and prognostic indicators for suicidality and overdose risk, but these findings also call for more in-depth investigation to determine optimal UDS frequency [4].
The financial implications of widespread urine drug screening, especially in the context of the opioid epidemic, have become increasingly apparent. Payers, patients, and healthcare providers are all concerned about the escalating costs associated with a uds [13]. For example, Vermont Medicaid experienced a dramatic increase in spending on urine drug screen testing, rising from over two million dollars in 2016 to over fourteen million dollars in 2018, coinciding with a tripling in the number of members undergoing UDS for OUD [17]. Highly publicized cases of potentially unnecessary and excessive testing have further raised questions about the true clinical utility and value of routine a uds practices [12].
As the approach to treating OUD with medications for opioid use disorder (MOUD) shifts towards more patient-centered models that prioritize reducing stigma and enhancing privacy, healthcare providers need to be equipped to adjust the frequency of urine testing based on its actual utility in supporting ongoing patient recovery [9, 14]. Therefore, understanding the value of a uds and determining how often it yields unexpected results is paramount [8]. This study aimed to assess the value of a uds by focusing on the frequency of expected and unexpected results in patients receiving buprenorphine-naloxone for OUD in primary care settings. Our primary goal was to evaluate the outcomes of a uds, differentiating between expected and unexpected findings in this patient population within adult primary care practices affiliated with the University of Vermont Medical Center (UVMMC).
Methods
Clinical Practice Setting and Infrastructure
This retrospective study was conducted within office-based opioid treatment programs across nine medical homes associated with UVMMC. These medical homes encompassed 35 healthcare providers across five family medicine and four internal medicine practices located in Chittenden County, Vermont. This clinical framework operates within Vermont’s recognized hub-and-spoke model, which provides support to patients through specialized hubs staffed by addiction medicine experts and spokes consisting of buprenorphine-waivered physicians or advanced practice providers, a dedicated registered nurse, and a licensed drug and alcohol counselor [2]. Complementing this model, UVMMC established the Addiction Treatment Program (ATP) in 2016, a “super spoke” functioning as a multidisciplinary bridge clinic within the hub-and-spoke system. The ATP offers additional support to both patients and their primary care providers (PCPs) who serve as spoke providers.
Patients entering these office-based opioid treatment programs, or spokes, are generally considered to be in a more stable phase of recovery. Typically, they have demonstrated consistent expected a uds results and are well-managed on their current buprenorphine-naloxone dosage at the time they transition to their medical home for ongoing MOUD. Stable patients are characterized by a history of consecutive expected urine drug screens, maintenance on a stable dose of buprenorphine-naloxone, consistent appointment attendance, and a determination by the ATP clinic that they are ready for spoke-level care.
All providers within the UVMMC spoke system adhere to a standardized protocol designed to guide patient management, including specific procedures for urine drug testing. According to the UVMMC protocol, urine drug tests are ordered on a random basis, varying from weekly to monthly depending on the patient’s stability. Patients deemed most stable are monitored with monthly urine drug tests in the primary care setting, while those considered less stable undergo weekly testing. Patients are notified randomly to present for a uds and are required to provide an observed urine sample within 24 hours of notification.
UDS Analysis and Dataset
The study population included all UVMMC primary care patients who were prescribed buprenorphine-naloxone for opioid use disorder at their medical home and had a uds collected between January 19, 2016, and January 31, 2018. All collected samples were analyzed at Aspenti Health (South Burlington, VT), a specialized diagnostic laboratory focusing on substance use disorders. Testing involved a combination of immunoassay screening using enzyme-linked immunoassay (EIA) and confirmation testing via Liquid Chromatography-Tandem Mass Spectrometry. In cases where both test methods were available, the confirmation result was prioritized for analysis. Data collected included sample characteristics such as unique identification, specimen collection date, and provider, along with patient demographics like name, age, and gender.
Samples were assessed for a comprehensive panel of substances, including cocaine, heroin, methadone, general opiates, oxycodone, fentanyl, tapentadol, and tramadol. Methamphetamines were excluded from the panel due to their low regional prevalence in our community. Similarly, THC was not reported in the findings due to its common detection.
Laboratory results were classified as either expected or unexpected. The presence of buprenorphine in urine was considered an expected finding, while its absence was deemed unexpected. To address the possibility of specimen tampering through direct contamination with buprenorphine, confirmation testing that detected only buprenorphine without its metabolite norbuprenorphine was also categorized as unexpected [18]. Unexpected categories also included the detection of cocaine, heroin, methadone, and other opioids (codeine, fentanyl, hydrocodone, hydromorphone, morphine, oxycodone, oxymorphone, tapentadol, and tramadol). Confirmation testing results superseded initial screening results in the analysis. Although benzodiazepine presence was also examined in the a uds, it was not classified as an unexpected result due to the lack of medication lists to confirm prescriptions. The analytical plan assessed primary outcomes both at the patient level and per individual test. The dataset for this study was derived from Aspenti’s laboratory information system (LIS).
The LIS database provides a secure environment for recording, managing, and storing test orders, results, and protected health information for all patients tested by Aspenti. Descriptive statistical methods were used to analyze subject characteristics, using Stata 15.0 [16].
The University of Vermont Institutional Review Board reviewed and approved this study.
Results
The study included a total of 161 patients, from whom 2,588 urine drug tests were analyzed. The distribution of urine tests per patient is detailed in Table 1. Patients were monitored for up to 660 days, with an average study duration of 289.6 days per patient. The gender distribution was nearly balanced, with 77 men and 79 women, and gender was unrecorded for 5 patients.
Table 1. Demographic characteristics of the study population
| Individuals | 161 |
|---|---|
| Total test results | 2588 |
| Age at initial visit | 38 (SD 11, Median 35) |
| Gender at initial visit | M = 77(47.8%) F = 79(49%) Unknown 5 (3%) |
| Tests per patient | Average 16.1 (SD 11.1, Median 15) Range per patient 1–48 |
| Days Followed | 289.6 (SD 218.5 Median 255) Range 0–660 days |
| Patients with buprenorphine and no unexpected substances in the urine | 104 Patients (64.4%) (N = 161) |
Our findings indicated that 64.4% of the patient population (104 out of 161 patients) demonstrated both treatment adherence, evidenced by positive buprenorphine test results, and no unexpected test findings, characterized by negative tests for opioids, methadone, cocaine, and heroin. Conversely, 57 patients (35.4%) did not meet these criteria.
Table 2 presents unexpected results at both the patient and test levels. Patient-specific analysis revealed that 7 patients (4.3%) had at least one a uds lacking buprenorphine. Unexpected opioids were detected in at least one a uds from 20 patients (12.4%). Methadone was found in at least one a uds for 5 patients (3.1%). Cocaine was detected in at least one a uds from 21 patients (13%), and heroin in at least one a uds from 2 patients (1.2%). When considering the combined presence of unexpected substances (opioids, methadone, cocaine, and heroin), 39 out of 161 patients (24.2%) tested positive for at least one of these substances. Additionally, 7 patients (4.3%) tested positive for two types of unexpected substances, and one patient tested positive for three types. A significant majority, 122 patients (75.8%), never tested positive for any of these unexpected medications.
Table 2. Number of unexpected patient and UDS results (percentages are in ())
| Category | No bup | +Other opioid | +Methadone | +Cocaine | +Heroin | +Any unexpected substance |
|---|---|---|---|---|---|---|
| Patients* 161 | 7 (4.34) | 20 (12.4) | 5 (3.1) | 21 (13.0) | 2 (1.2) | 47** (29.2) |
| Urine Tests 2588 | 38 (1.46) | 28 (1.08) | 8 (0.30) | 39 (1.50) | 2 (0.07) | 77 (2.97) |
*The same patient and the same UDS could be positive for more than one substance and negative for buprenorphine
**This could be more than one unexpected substance
Analysis at the a uds level showed that out of 2,588 test results, 38 (1.5%) lacked buprenorphine. Furthermore, 28 tests (1.1%) were positive for opioids, 8 (0.3%) for methadone, 39 (1.5%) for cocaine, and 2 (0.1%) for heroin.
Discussion
Urine drug screening has become an essential component of managing patients with OUD in recovery, serving as a tool for prescribers to routinely monitor for unexpected substances in urine [7]. Beyond detection, a uds can also play a supportive role in patient recovery, helping individuals maintain their treatment course [6]. Clinical utility has been evidenced in pain management settings, including risk mitigation for suicidality and overdose, detection of potential drug diversion, and identification of initial misuse [4]. However, robust evidence on how a uds frequency impacts outcomes for patients with OUD remains limited.
Given that the majority of patients in our study demonstrated expected urine results, our findings suggest that reducing the frequency of urine testing may be a reasonable and more patient-centered approach, particularly for stable patients. Patients who consistently show expected a uds results, are maintained on a stable dose of buprenorphine-naloxone, adhere to scheduled appointments, and are progressing well in their recovery could be considered for less frequent UDS testing [1]. Decreasing testing frequency for appropriate patients can lead to reduced costs for both patients and the healthcare system, and minimize disruption to employed patients’ work schedules [18]. While a comprehensive economic analysis is beyond the scope of this study, it represents an important direction for future research.
Our study has several limitations. Its retrospective, descriptive design meant we lacked direct access to patient medical charts and medication lists, preventing us from confirming if benzodiazepine or opioid detections were medically indicated. Our patient cohort may also represent a more stable population receiving MOUD, potentially limiting the generalizability of our findings to all primary care settings. The availability of resources, such as easy referral pathways back to our ATP clinic for unstable patients, might not be universally available in other practice locations. The clinical response to an unexpected a uds result can vary and is likely clinician-dependent; however, it often serves as an alert that a patient may require a higher level of care than primary care can provide.
This descriptive study contributes valuable data on a uds results in patients managed by primary care physicians using buprenorphine-naloxone for MOUD. Considering the relative stability of our office-based setting, these results may establish a baseline expectation for the frequency of unexpected test results in similar populations. Further research, correlating findings with patient medication lists and long-term outcomes, is needed to refine best practices for patient-centered care in OUD recovery with buprenorphine-naloxone. This includes further investigation into optimizing the use of a uds in this context (Additional file 1).
Conclusion
Patients undergoing treatment for opioid use disorder with buprenorphine-naloxone within an academic primary care spoke program exhibited low rates of unexpected urine drug screens. Further research is warranted to explore the optimal frequency of testing and its impact on patient outcomes in this stable population. Such studies will be crucial in developing evidence-based guidelines for a uds monitoring in MOUD, enhancing patient care and resource utilization.
Supplementary Information
13722_2021_264_MOESM1_ESM.docx (92.7KB, docx) Additional file 1. The opioid use disorder protocol at the author’s institution.
Acknowledgements
None
Abbreviations
ATP Addiction Treatment Program
MOUD Medication for opioid use disorder
PCPs Primary care physicians
OUD Opioid use disorder
UVMMC University of Vermont Medical Center
UDS Urine Drug Screen
Authors’ contributions
Drs. Sobel, Warrington, and Berger designed the study; Dr. Warrington and SF-F developed and cleaned the data set; Dr. Crocker performed the statistical analysis and contributed to the study design. All authors participated in the manuscript writing process.
Funding
None.
Availability of data and materials
The dataset utilized in this project is accessible to the study authors.
Declarations
Ethics approval and consent to participate
Not applicable. This study was reviewed and approved by the University of Vermont Institutional Review Board, which deemed it not to be human subject research.
Consent for publication
Not applicable.
Competing interests
None.
Footnotes
Publisher’s Note
Springer Nature maintains a neutral stance regarding jurisdictional claims in published maps and institutional affiliations.
References
Associated Data
This section includes data citations, data availability statements, and supplementary materials referenced in this article.
Supplementary Materials
13722_2021_264_MOESM1_ESM.docx (92.7KB, docx) Additional file 1. The opioid use disorder protocol at the author’s institution.
Data Availability Statement
The dataset used for this project is available to the authors of this study.
