How Can Routine Clinical Laboratories Keep Up with the Opioid Crisis?
Bibliographic record
Abstract
Opioid-related overdoses and deaths have grown to epidemic proportions throughout the United States and Canada, and are increasing elsewhere in the world as well. Despite targeted efforts such as naloxone distribution and prescription drug monitoring programs, progress toward reversing these trends has been slow. On the contrary, many areas are reporting staggering increases in overdoses and drug-related deaths during the COVID-19 pandemic. Laboratory testing such as urine drug screening plays an important role in clinical management of both licit and illicit use of opioids and other controlled substances. Indications for drug of abuse testing range from urgent needs such as identifying substances involved in intoxication or overdose, to diverse settings including assessing compliance with prescribed medications or abstinence in workplace or high-risk populations. As need for this testing grows, laboratories are under increasing pressure to expand support for emergency departments, pain management, rehabilitation, primary care, and other practices that require drug testing for patient care. Meeting this demand can be particularly challenging for routine clinical laboratories, which often lack the resources available to specialized or reference laboratories. These latter institutions can respond to changes in drug abuse trends by developing tests for novel drugs and bringing in newer technology such as high-resolution mass spectrometry (HRMS). Yet, with the opioid crisis widespread in North America and elsewhere, it is routine laboratories with fewer resources and less technical expertise who bear a substantial share of the clinical burden. So how can laboratories facing resource constraints best support their clinical practices in combating this crisis? The panelists addressing this question have extensive experience with drug testing and interpretation, in laboratory settings ranging from large and sophisticated to small and basic. All understand the limitations facing routine laboratories, and the pressure to optimize support for practices managing all aspects of licit and illicit opioid use. Jessica Boyd, Leland Baskin, Hossein Sadrzadeh: The type of testing required really does depend on the circumstances of the request and the patient population. For emergency and urgent care requests, clinicians should not be waiting for a laboratory result to start treatment (e.g., administration of naloxone in suspected opioid poisoning or N-acetylcysteine for acetaminophen toxicity, since the side effects are minor). In these cases, the laboratory urine drug screen results may be more useful in confirming the initial diagnosis made by the physicians based on the presenting toxidrome or explaining what the causative agent was. Having mass spectrometry available for emergency department cases, particularly untargeted analysis, can be very helpful especially in cases where the causative agent is not what we would consider a typical drug of abuse (e.g., baclofen, gabapentin) and would not necessarily be picked up on an immunoassay drug screen. For pain management clinics, mass spectrometry is still recommended because it provides a definitive result. However, a major difference in testing for pain management compared to other requests is the requirement to check for medication compliance. Where in most cases physicians are looking for a negative urine drug screen, in pain management a negative test would be abnormal since it is expected that the patient should be positive for the drug they are prescribed. Therefore, it is important to include the drugs commonly prescribed by pain management physicians on the drug screen panel. In Alberta, we are seeing more and more of our primary care physicians prescribing Suboxone® (buprenorphine plus naloxone) for opioid dependency. Currently, approximately 70% of our total urine drug screen volume in Calgary is for opioid dependency patients. Because of the large volume of samples received, we have streamlined our current process to provide a liquid chromatography-tandem mass spectrometry (LC-MS/MS) panel consisting of common drugs of abuse as well as methadone and buprenorphine and their metabolite. Brian Kelly: The needs for each type of practice are different. In the emergency department or urgent care, patient triage, stabilization, and treatment are the priority. Rapid turnaround time (TAT) is vital, and identification of drug class, rather than specific drugs, is sufficient. A fast immunoassay screen is reasonable for this setting and will guide the clinician toward an accurate diagnosis quickly, as well as help prevent use of potentially contraindicated medications. In rural areas, primary care physicians are more likely than their colleagues in urban areas to treat chronic pain conditions due to the absence of nearby pain specialists. Thus, there is potential for substantial overlap in the drug testing needs for primary care and pain management. Primary care, in general, is less likely to need a rapid TAT compared to the emergency department, and having a larger immunoassay panel that can identify specific drugs rather than drug classes, including illicit compounds, is acceptable for this setting. Pain specialists are focused on pain control, on either an acute or chronic basis, and typical treatment involves higher initial doses of analgesics with an eventual taper and discontinuation. For these needs, quantitative testing (e.g., LC-MS/MS) becomes more valuable than a qualitative screen so that the provider can be certain the patient is taking their prescribed medication by detection of prescribed drug metabolites. The clinic and laboratory should coordinate so that the test menu reflects the compounds prescribed in the clinic; in this way the laboratory is able to efficiently test for relevant metabolites as well as common illicit drugs. Danyel Tacker: This depends somewhat on philosophy, and at West Virginia University (WVU) the differences in drug testing vary substantially by setting. In the pain management (oncology and free-standing) and comprehensive opioid addiction therapy (COAT) clinics, abstinence from other substances and compliance with prescribed therapy are key. Thus, those clinics use cup-based waived screens, send other screens not available in waived format to the laboratory, and order definitive testing when unexpected scenarios and/or suspected simulated compliance arise. Examples of other screens include fentanyl testing in the central laboratory, and ethanol biomarkers and gabapentin offered via mass spectrometry. Approximately half of patients coming to COAT clinics are cup-screened, and approximately 10% of patients are randomly selected for gabapentin, ethanol biomarkers, and fentanyl screening. In contrast, the WVU emergency department is a screen-only operation. Patients come through quickly and the goal is to evaluate and triage. Getting back a reflexed definitive result 3 days later is not useful to this group. Consequently, this service uses an 11-component panel screen from our central laboratory, which contains basic adulteration assessment—results are ready in an hour or less. In the middle, there is primary care. Because these providers increasingly use pain management contracts with select patients, cups are on-hand for spot-screening compliance in-clinic. However, these providers also use laboratory-based routine screens, and definitive testing is mostly used for compliance checks in patients with low-dose therapy commonly missed by standard screens. A final population pertinent here is the obstetrics population. Our COAT clinic extends to the obstetrics setting, and ALL mothers are screened during prenatal visits and at presentation to labor and delivery. Matthew Krasowski: For emergency/urgent care, the ability to perform basic urine drug screens for major drug classes (amphetamines, benzodiazepines, cocaine, opiates, tetrahydrocannabinol [THC]) with rapid TAT is important. For primary care and pain management, basic urine drug screens may suffice for some purposes (e.g., checking compliance with opioid therapy) provided the testing covers the intended purpose. Pain management especially may need at least periodic use of more detailed confirmatory type drug testing. Danyel Tacker: This too is likely to depend on region (and budget). Here in Appalachia, WVU is geographically very close to the epicenter of the opioid crisis. The standard screening panel offered at the WVU Hospital is thus amphetamine/methamphetamine, barbiturates, benzodiazepines, buprenorphine, cocaine metabolite, Ecstasy/MDMA (methylenedioxymethamphetamine), fentanyl, methadone, opiates (300 ng/mL cutoff), oxycodone, and THC. We also developed a definitive opioid panel in our mass spectrometry laboratory that results in 10 separate test offerings. We started with the high-volume captures (opiate and oxycodone confirmations) and added related lower-volume reference laboratory testing (heroin, fentanyl, meperidine, methadone, and tramadol) that we could easily put into our panel. Collectively, these also made sense for offering lower-cutoff compliance checks, to allow providers to directly get to definitive answers in patients who commonly get false-negative screens. Definitely, we do not have the capacity to test for designer drugs and every fentanyl derivative; these investigations are rarely needed, so are sent to a reference laboratory. Also, the other drug class confirmations will need to be transferred to in-house testing, to at least cover all on-site screens; we will use the same, grouped approach to pull them in over time. Jessica Boyd, Leland Baskin, Hossein Sadrzadeh: A common approach to provide urine drug testing locally is to have immunoassay urine drug screens on site and send presumptive positives to a reference laboratory for confirmation by LC-MS/MS. Of course, a major problem with using immunoassay drug screens is understanding the issues with false positives and false negatives. With regard to opioid testing, it is critical to understand that what is commonly termed a urine opiates screen only detects codeine and morphine use. Even in these assays, cross-reactivity with hydromorphone and hydrocodone can be limited, making them unreliable for monitoring pain management patients. To address these limitations, additional immunoassay screens are required specifically targeting fentanyl, hydromorphone, hydrocodone, oxycodone, methadone (and metabolite), and buprenorphine. Mass spectrometry-based testing helps alleviate the concerns over false positives and negatives, but may not be reasonable to put in every laboratory. In Alberta, due to the concerns with immunoassay drug screens, we are making the move to one-step urine drug testing by LC-MS/MS for all clinical requests, and immunoassay drug screens are being phased out. Two LC-MS/MS panels are offered: 1) an opioid dependency panel that consists of ∼25 drugs and metabolites, and 2) a panel for non-opioid dependency use (e.g., emergency department, pain management) that includes the drugs on the opioid dependency panel and adds several additional prescription and over-the-counter drugs. This streamlined approach allows for use of automated sample preparation and middleware connections to the laboratory information system to increase throughput and reduce TAT while providing definitive results on the first report. Brian Kelly: I advise clinical laboratories to be able to detect drugs commonly prescribed in the practice(s) that they provide laboratory services for as well as the inclusion of specimen validity test(s) and possibly illicit drugs, depending on pre-test probability of illicit drug use in the patient population. The makeup of the test menu may vary based upon geographical location, provider prescribing practices, budget, and space. For those infrequent analyte(s) or specimen type(s) not offered in the clinical laboratory, specimens can be sent to a reference laboratory. Matthew Krasowski: Basic urine drug screens should be available on-site. Confirmatory testing, which typically uses chromatography/mass spectrometry, can be sent to a reference laboratory. Very large medical centers may find it cost-effective and clinically advantageous to have confirmatory testing on site. Brian Kelly: By having both instrumentation and well-trained laboratory staff available, as changes in drug abuse trends occur, the laboratory can change its test menu by adding or removing tests on an as-needed basis. For example, if a drug, such as heroin, is uncommonly used by the geriatric patient population being cared for, removing heroin from the test menu and replacing with THC would be reasonable to maximize resources as medical marijuana is becoming commonly prescribed. The same idea can be used for clinical laboratories in general. Matthew Krasowski: Laboratories should be aware of trends in drug usage and periodically review their test menu to make sure it is appropriate for the patient population. In this regard, a recent publication examining trends in drug of abuse proficiency testing over the last decade indicates that the drug of abuse testing menu for many clinical laboratories may be out of date with current trends in drug usage patterns. For example, this analysis that specific for oxycodone, fentanyl, and buprenorphine may be laboratories that pain management or abuse services may need specific testing that not be offered would be use of buprenorphine urine drug screens if a that has a buprenorphine clinic to treat opioid Danyel Tacker: could start by the for their may be to For menu looking into for the is increasingly are several these days for an of drugs, but some and may be to find the for the and The to more screens, can be more a Jessica Boyd, Leland Baskin, Hossein Sadrzadeh: is very important to what the drug trends are in since we that drugs of abuse trends can change with that we use in our laboratory is a pull from the laboratory information system to on what drugs are being in our For the we have to at trends in and drug from our we have for opioid dependency clinic which are with information from our laboratory information system and allow close monitoring of Jessica Boyd, Leland Baskin, Hossein Sadrzadeh: Our clinical colleagues in are a of information for drugs of abuse trends and into 3 1) we are in with the medical at our who on relevant information what they are seeing in emergency the 2) our colleagues at the medical who often on information on drugs they are seeing in are of we also often important information on suspected drugs from our physicians the opioid dependency programs, who to a drug screen result that may be with what patients have them they are not necessarily is our laboratory The has of and in Canada, we in through the of Matthew Krasowski: resources such as the for and and are helpful with and emergency can be helpful in more Brian Kelly: are several resources available on the for review opioid and other drug including the and the on and the For centers may be to provide on they have will drugs during investigations and the laboratory can to identify these the or class of the drugs will be by Danyel Tacker: and/or primary care can often provide a The on also has some but it can be out of are increasingly more that do and more Also, the and is and are these often of and from drug use and Brian Kelly: laboratories or reference laboratories should be a of information for their In information specimen sample test and an easily are important. should be in a that is for the on laboratory information and patient medical this may be or to the patient care Danyel Tacker: the needs of We the idea of drug cups at clinics those we to understand the at the clinic and that cups the only way to for screening in several these clinics and to find a for them screening and confirmatory testing with the process has a and to testing. Matthew Krasowski: In and of confirmatory drug testing can be a for clinicians and also clinical laboratories. This is an not well in such as medical and of a useful resource by laboratory would be information such as or that issues such as and drug of abuse detection laboratories also provide some of the Jessica Boyd, Leland Baskin, Hossein Sadrzadeh: with is of laboratories can provide support in to select appropriate tests and the if the needs on site. can also provide on required and for these The central laboratory also should be involved in of what a for testing and support for of the Matthew Krasowski: the experience of at a very large medical system that a comprehensive and drug monitoring testing laboratory and more at a that does not have this In having this technology on-site would be The are of the instrumentation often more need for clinical laboratory staff to and the This is very challenging to up in a and if clinical laboratory is Danyel Tacker: such as an that is not common to routine laboratories. The best of come 1) 2) that the is in appropriate for the technology and has a support staff who are to and in their appropriate for the and and support from the can to a mass spectrometry laboratory, and more time to get the menu this reference laboratory testing will and are for can to in-house this is a Jessica Boyd, Leland Baskin, Hossein Sadrzadeh: LC-MS/MS testing should not be on especially by laboratories that have experience in this laboratories using LC-MS/MS would have the system and developed their as a This substantial expertise and time on the of the laboratory. In our laboratory, it typically several to and an LC-MS/MS for drugs of are available for urine drugs of abuse testing, but they still require substantial and support to up automated LC-MS/MS are on the of the clinical laboratory, are not available for urine drugs of abuse testing. routine urine drug testing by LC-MS/MS is by testing is of untargeted to how spectrometry have been used for the These are more than and really should be in the of the reference laboratory for use with or Brian Kelly: The of specimens expected will a to LC-MS/MS In at approximately specimens of testing it becomes reasonable to consider LC-MS/MS. This is due to test and higher throughput of LC-MS/MS compared to the of specimens in a and test on a typical laboratory immunoassay In most cases, will not be appropriate for clinical laboratories due to that increase is on compounds in a central or reference laboratory that has staff and Danyel Tacker: This is our with started 10 with somewhat on-site cup-based screens. We have more appropriate in-house testing to I to and drug testing and at their use I in for as-needed to help providers at testing results and of the to The WVU has come to the laboratory to the and to the Meeting in to clinical the clinical I could not at WVU Jessica Boyd, Leland Baskin, Hossein Sadrzadeh: the laboratory and clinicians is in the testing and is appropriate for the needs of the clinicians and patient population. As with laboratory test urine drugs of abuse testing be with the for request and patient population in the testing may not be or appropriate for the clinician to their patients our laboratory and opioid dependency physicians critical to our to the opioid epidemic in In when fentanyl becoming a problem we started to increases in requests for urine drug screens. Our specimen volume in the first our drug screen testing which at that time of an immunoassay screen and untargeted spectrometry analysis on each By the of we samples and our TAT to which to treatment of patients. We our process required major changes to the rapid increase in testing We with our opioid dependency clinicians and developing a LC-MS/MS panel for them based on their The physicians when they that over of the drugs they not positive in their patient population. We setting up a panel for each clinic that the more commonly drugs of abuse based on of The panels of the same (amphetamines, but with of certain drugs that commonly in each clinic population (e.g., panel has a of which allows to our analysis through use of targeted LC-MS/MS automated sample preparation and middleware it provides our physicians with results on the drugs that most common in their patient population. the in their clinics, our physicians would not have our With their support we able to make these changes to laboratory and the panels in we have not issues up with samples increases in urine drug screen Brian Kelly: The laboratory plays a critical role in patient care and Laboratory results can or of patient of prescribed drugs as well as use of illicit Therefore, the laboratory and providers the clinic or is as there may be differences in test menu depending on the patient population. For example, a primary care provider chronic pain may not be in if their patient a of with but for an addiction of ethanol use may treatment of drug screens, there is test menu and is important for the laboratory to that the needs of the clinical practice are Matthew Krasowski: practices often have experience with drug testing. The laboratory can a role in to the technology and its a very basic can help that the routine testing offered is the needs of clinical practices and also such as false and false Even more clinical practices a who can that arise. I have that I can address a of drug abuse testing through or Matthew Krasowski: Even the laboratory testing has an important role in clinicians the opioid laboratories can a positive role by providing accurate information and being a resource for drug abuse testing Even being very with the basic urine screens offered on-site can be a help to For example, a common from clinicians I have is that the screen will detect opioids such as methadone or buprenorphine when in those drugs will likely not Danyel Tacker: by understanding that use of is a does care or and in the laboratory information with a a with and/or pertinent providers can only help laboratories the when providers the and be Jessica Boyd, Leland Baskin, Hossein Sadrzadeh: We the same for laboratory testing or clinical 1) with appropriate is important. up with clinical such as the emergency department and opioid dependency clinics is a to 2) The best way to is to to physicians and for their will be patients. are patients, depending on the of laboratory may a more them than and experience with the clinical colleagues the of the laboratory and may not understand the best approach to with them will provide a for information and in operation. Brian Kelly: primary would be to understand the needs of the laboratory and patient care staff can a way in the laboratory services a of to the providers what drugs are on the test TAT specimen sample to the provider on how they can use laboratory testing to patient care. out which laboratory are and have additional instrumentation or test menu may tests from time to which clinicians would may that includes the thus the initial The opioid epidemic is likely here to and clinical laboratories can patient care. liquid mass comprehensive opioid addiction high-resolution mass spectrometry. All they have to the of this and have the to the and of or analysis and of or the for final of the and to be for all aspects of the thus that related to the or of of the are and all the and/or potential of Baskin, Boyd, West Virginia
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.041 | 0.201 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.006 | 0.007 |
| Scholarly communication | 0.017 | 0.025 |
| Open science | 0.004 | 0.010 |
| Research integrity | 0.020 | 0.025 |
| Insufficient payload (model declined to judge) | 0.020 | 0.009 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".