Health Technology Review

Drug-Checking Technologies to Detect Compositions of Unregulated Substance Samples

What Is the Issue?

• Drug toxicity and related death have been declared a public health emergency in Canada, driven largely by increasingly toxic unregulated fentanyl and opioid supplies contaminated with unexpected, high-potency opioids and other central nervous system and/or respiratory depressants (e.g., benzodiazepine, xylazine).

• Drug-checking technologies (DCTs) — referring here to the underlying analytical methods such as immunoassay or mass spectrometry — are used to determine the presence of specific substances or provide a comprehensive breakdown of the substance’s composition. DCTs are used as part of drug-checking services to support safer use, increase awareness of substance compositions, connect people with health and social services, and inform public health responses to the toxic drug supply.

• There is a need to synthesize the available DCT evidence on accuracy, limit of detection, repeatability, reproducibility, cost-effectiveness, and costs for detecting compositions of unregulated substance samples.

What Did We Do?

• Based on the results of an initial scoping exercise and experts’ feedback, we conducted a customized health technology review to inform decisions regarding the use of DCTs for detecting compositions of unregulated substance samples.

• We searched key resources, including journal citation databases, and conducted a focused internet search, with no publication date limits, for relevant evidence. We also sought costing information by contacting DCT manufacturers to obtain costing estimates in Canadian dollars.

What Did We Find?

• The body of evidence on the outcomes and capabilities of DCTs is small but growing.

• We found 18 test accuracy studies that evaluated the accuracy of any DCTs for detecting compositions of unregulated substance samples.

  • DCTs varied in accuracy outcomes depending on the substance and context. Immunoassay test strips (ITS) showed relatively high sensitivity and low to minimal false negative rates for target substances, whereas spectroscopic methods (e.g., Fourier transform infrared [FTIR] spectroscopy spectrometers) provided broader compositional information.

  • Combining or advancing technologies improved performance. Pairing methods, such as FTIR with ITS, or using emerging DCTs, including portable gas chromatography–mass spectrometry and surface-enhanced Raman spectroscopy, enhanced sensitivity and reduced false negative rates.

  • Colorimetric reagents and infrared spectroscopy were shown to have comparable high sensitivity against the confirmatory lab testing.

• We found 7 test accuracy studies that evaluated the limit of detection of any DCTs for detecting compositions of unregulated substance samples. Of these, 1 study also reported on the reproducibility of ITS. This review did not find any studies that reported results on the repeatability of DCTs that met our selection criteria for this review.

  • ITS (e.g., fentanyl test strips), FTIR, and hand-held Raman spectrometers were each able to detect the presence of target substances in samples, though their detection capabilities varied depending on the substance and concentration.

  • Xylazine test strips consistently detected xylazine in samples with low concentrations, across days.

• We did not find any studies on the relative cost-effectiveness of any DCTs for detecting compositions of unregulated substance samples that met our selection criteria for this review.

• We found 1 article that provided a cost description of a US pilot drug-checking service that employed an FTIR spectroscopy, fentanyl test strips, and confirmatory lab testing. The findings demonstrated the feasibility of drug checking in the evaluated context.

• Eleven manufacturers that we contacted provided cost information by completing surveys for 16 DCTs. Among the cost data received, 4 DCTs were technologies identified in the included test accuracy studies.

• Outcome measures and methods used for evaluation of DCTs were heterogeneous across the studies, and conclusions were often constrained by risks of bias and applicability. No studies were found that specifically address the performance of DCTs in distinct settings, such as rural or Indigenous communities. Well-designed, methodologically rigorous studies are needed to generate the evidence needed to confirm the accuracy and cost-effectiveness of DCTs.

What Does This Mean?

• This review includes several considerations for health care decision-makers, including those planning or implementing DCTs for use in clinical and community settings:

  • There is a trade-off between highly sensitive detection of a single substance (e.g., fentanyl) and broader screening capacity: ITS indicated high sensitivity and specificity in detecting target substances and may be used for early identification of toxic substances and their analogues; however, FTIR and Raman spectrometers offer the advantage of detecting multiple substances within a single sample.

  • A multitool approach may be preferable for drug checking and may better guide harm reduction messaging for people who use drugs: pairing point-of-care FTIR spectrometry and ITS combined with confirmatory lab testing can enhance sensitivity and reduce the false negative rate.

  • In the context of drug checking, certain test accuracy measures are more critical than others: studies most often reported on sensitivity, reflecting the importance of avoiding false negative results (i.e., failing to detect a harmful substance), and specificity, reflecting the problems with false positives, which can generate misinformation about the unregulated drug supply.

  • Innovations in emerging DCTs are specifically intended to balance trade-offs between sensitivity, specificity, cost, and ease of use. Most of these technologies are in the research, development, and evaluation stages, and some are in the field-testing stage.

  • Establishing national or regional guidelines for evaluation metrics, accuracy reporting, and service design specific for evaluation of DCTs may reduce variability in evidence.

  • Relevant costing estimates obtained from DCT manufacturers included in this report can be considered alongside operational and contextual factors (e.g., service delivery model, staff capacity, proximity to confirmatory testing facilities) when considering, planning, or implementing DCTs within their specific settings.

  • Along with the technical and cost considerations, ethics and equity considerations relevant to the context of DCTs, presented in this report, can be considered including those related to consent, access, legal requirements, and privacy.

Key Terminology

Accuracy: Ability of the test to correctly classify substance samples and their compositions, and produce the same results as the validated technology to which it is being compared.¹ Accuracy of a test can be evaluated using the following outcomes: sensitivity, specificity, positive predictive value, negative predictive value, true positive, true negative, false positive, and false negative.

Confirmatory lab testing: Highly sensitive and specific drug-checking technologies used to validate the accuracy of the emerging and point-of-care drug-checking technologies.

Drug checking: A harm reduction intervention to analyze the content of substances that individuals intend to consume, rather than testing biological samples from individuals (e.g., drug testing in blood or urine).² Drug checking also supports timely monitoring of the unregulated drug supply to inform public health policies and harm reduction strategies on a broader scale.³

Drug toxicity: The harmful effects of a drug on the body. It occurs when a drug is taken in excessive amounts (also known as “drug overdose” and “drug poisoning”) or when the body cannot properly metabolize or eliminate the drug, leading to damage to organs, tissues, or physiological functions.^4,5

Drug toxicity crisis: A widespread public health emergency caused by high rates of drug toxicity and related deaths. A drug toxicity crisis is typically driven by the increased presence of highly potent and toxic substances, such as fentanyl in unknown amounts that is heavily contaminated with other unexpected substances.^4,5

Index test: The index test refers to the diagnostic or accuracy tests (i.e., what is called “drug-checking technologies” in this review) being assessed to determine how accurately they identify participants (i.e., substance samples in this review) who do or do not have the target condition (i.e., substance compositions in this review).

Limit of detection (LOD): The smallest amount or concentration of a substance or composition that the test can reliably detect. LOD defines the lowest detectable level at which the substance can be identified but not necessarily quantified.⁶ LOD can be evaluated by preparing reference standards with known concentrations of the target substances (e.g., xylazine) and analyzing multiple replicates at each concentration to determine the lowest concentration at which the composition is reliably detected in 95% of replicates (or based on a signal at least 3 times or 10 times the background noise).⁶

People who use drugs: Individuals who consume psychoactive substances, including prescription medications, alcohol, or illicit (nonpharmaceutical) drugs.⁷

Reference standard: The most accurate and reliable method available for determining whether the target condition (i.e., called “substance compositions” in this review) is present or absent. Reference standards are referred to as “confirmatory lab testing” in this review.

Repeatability or intraday (assay) precision: The consistency of the test when repeated under the same conditions (i.e., same sample, same instrument, same operator, short time frame). A precise test gives similar results each time it is used on the same sample.⁸

Reproducibility or interday (assay) precision: The consistency of the test when repeated in different conditions (i.e., different days, instruments, operators, or sites). Reproducibility captures how robust the method is in practice.⁸

Sensitivity: The ability of the test to correctly identify the presence of a substance when it is actually present. Sensitivity is sometimes referred to as the true positive fraction (called “true positive rate” in this review).¹ A highly sensitive test will detect even small amounts of the target substance.

Specificity: The ability of the test to correctly identify the absence of a substance when it is not present. Specificity is occasionally referred to as the true negative fraction (called “true negative rate” in this review).¹ A highly specific test will rarely give a false positive for a substance that is not actually in the sample.

Target substance(s): Specific drug(s) or chemical composition(s) that a technology or test is designed to detect and analyze in a sample.

Note. The language around substance use health has evolved significantly in recent years to reduce stigma and reflect a more compassionate, evidence-based understanding of substance use health and addiction. Terms that were once commonly used — such as “addict” or “drug abuser” — have been largely replaced with person-first, nonjudgmental language like “people who use drugs” or “person with a substance use disorder.” This shift acknowledges substance use as a health issue rather than a moral failing and helps reduce the discrimination that often prevents individuals from seeking support. The terminology presented here aligns with the current, stigma-reducing approach; however, language continues to evolve, and future changes may further refine how we speak about substance use health to promote dignity and inclusivity. The language used in this report aligned with the Canada’s Drug Agency (CDA-AMC) style guide,⁹ the internal Equity Considerations Resource for CDA-AMC Researchers, and external resources.^7,9,10

Acknowledgements

In Partnership

The Canadian Centre on Substance Use and Addiction (CCSA) provides national leadership on substance use health issues and on strengthening public and community safety. Canada’s Drug Agency (CDA-AMC) is a not-for-profit organization responsible for providing Canada’s health care decision-makers with objective evidence to help make informed decisions about the optimal use of drugs and medical devices in the health care system. Recognizing the need for a formal evaluation of drug-checking technologies, CDA-AMC partnered with CCSA to undertake this health technology review, drawing on their complementary areas of expertise. CDA-AMC and CCSA partnered in conceptualization, including development of the objectives and research questions; CDA-AMC led the investigation, methodology, analyses, and writing and editing the project plan and report; and CCSA facilitated engagement with subject matter experts and provided input on the draft project plan and report. This successful collaboration reinforces the value of interagency partnerships in advancing evidence to support substance use health in Canada.

Peer Reviewers

The authors would like to thank the 2 content experts who externally reviewed this report.

Research Questions

1. What is the accuracy of any drug-checking technologies for detecting compositions of unregulated substance samples?

2. What is the limit of detection, repeatability, and reproducibility of any drug-checking technologies for detecting compositions of unregulated substance samples?

3. What is the cost-effectiveness of any drug-checking technologies for detecting compositions of unregulated substance samples?

4. What are the costs associated with any drug-checking technologies for detecting compositions of unregulated substance samples?

Context and Policy Issues

Drug Toxicity in Canada

The Government of Canada has reported a total of 50,544 deaths related to opioid toxicity between January 2016 and December 2024 (an average of 20 lives per day).⁵ Drug toxicity and related death continue to be declared a public health emergency in several provinces.⁵ The drug toxicity crisis is complex and multifaceted. However, the crisis is driven largely by increasingly toxic, unregulated fentanyl and opioid supplies, particularly when they are contaminated with unexpected, high-potency opioids and other central nervous system and/or respiratory depressants (e.g., benzodiazepine, xylazine).⁵ Of all deaths related to opioid toxicity reported in 2024, 84% involved nonpharmaceutical opioids, 74% involved fentanyl analogues, and 70% involved a stimulant, underscoring the severity of the ongoing emergency.⁵ Hence, many of these deaths involve polysubstance use, with stimulants and benzodiazepines frequently detected alongside opioids. Ontario, Alberta, and British Columbia have experienced the highest numbers of drug-related mortality.⁵ There may be regional differences in drug supply, disparities in service access, and socioeconomic determinants of health that impact the severity of the issue across jurisdictions. The burden of disease related to drug toxicity in Canada is profound, affecting individuals, families, health care systems, and society at large.¹¹ Drug toxicity, particularly from high-potency opioids, has led to a significant loss of life-years and has become 1 of the leading causes of premature death among adults living in Canada.¹¹ In addition to mortality, nonfatal drug toxicity results in substantial morbidity, including brain injury, infectious diseases, and mental health complications.¹² The health care costs associated with emergency responses, hospitalizations, and long-term care continue to rise, while the broader social and economic impacts include lost productivity and increased demand on social services.¹¹

There is no single solution to the complex and multifaceted nature of the drug toxicity crisis.¹³ People experiencing harms associated with substance use may benefit from access to comprehensive and coordinated services and supports, integrating harm reduction, prevention, treatment, and social supports that put the person and their needs at the centre.¹³ The unpredictable potency and composition of the unregulated fentanyl and opioid supply have prompted calls for harm reduction services and support, such as access to naloxone, supervised consumption sites, and expanded drug-checking services (DCSs), to reduce preventable deaths.¹⁴

What Is Drug Checking?

DCSs are harm reduction services often integrated into community and public health agencies and harm reduction sites, such as substance use services and support programs, supervised consumption sites, mobile harm reduction units, and music and other festivals. Drug-checking technologies (DCTs) are used as part of DCSs to analyze the content of substances that individuals intend to or have consumed, rather than testing biological samples from individuals (e.g., drug testing in blood or urine).² Depending on the method, these technologies may test for the presence of specific substances (such as xylazine) or provide a comprehensive breakdown of the substance’s composition, including the various components and their respective concentrations.² Whereas some DCTs can be used directly by people who use drugs (PWUD), DCSs often pair the results with personalized harm reduction advice from front-line health service providers according to the individual's needs.¹⁵ Beyond supporting individual decision-making, by collecting and analyzing unregulated drug samples, DCTs help track trends in the unregulated drug supply and inform public health policies and harm reduction strategies on a broader scale.³ When potentially dangerous substances are identified, DCTs can trigger public health alerts,¹⁶ extending their impact to the wider community of PWUD and service providers. Prompt and accessible drug market monitoring facilitated by robust DCSs can also inform first responders, clinicians, policy-makers, coroners, and researchers of the drug market trends in real time.¹⁷

DCSs can help detect the presence of harmful compositions and unexpected substances, reducing the risk of drug toxicity and related death.¹⁸ These services connect people to settings providing information about the compositions of drugs and the associated risk of their use through a supportive and stigma-free approach.^17,19 Therefore, they can be particularly helpful for equity-deserving groups such as individuals at high risk of drug toxicity, people without stable housing, and individuals with co-occurring mental illnesses.^20,21 Evidence suggests that access to DCS supports safer use, facilitates engagement with health and social services, and increases awareness of substance composition.^16,17 DCSs may also be effective in behaviour change; for example, PWUD may choose to discard harmful substances, use smaller doses or avoid mixing substances, and seek other harm reduction services after drug checking.¹⁶ These results are affected by the population, setting, and the type of substances found in the sample tested.¹⁶

DCTs in Canada

DCTs are increasingly being implemented as a harm reduction tool in response to the ongoing drug toxicity crisis in Canada. These technologies range in complexity from fentanyl test strips (FTS) to more advanced methods, like Fourier transform infrared (FTIR) spectroscopy and mass spectrometry (MS). Learning from European harm reduction efforts, DCSs in Canada began gaining attention in the 1990s. It initially focused on pill testing in rave and festival settings using colorimetric methods. The field expanded further in 2015 with the formation of the National Drug-Checking Working Group (NDCWG) and intensified after the onset of the opioid crisis in 2016.²² Between 2017 and 2019, 3 pilot projects were launched in Canada:

• The British Columbia Centre on Substance Use and the City of Vancouver received support from Health Canada’s Substance Use and Addictions Program (SUAP) to test FTIR spectroscopy, demonstrating the feasibility of community DCSs.²³

• Toronto DCS received support from Health Canada’s SUAP to pilot an offsite drug-checking model using laboratory-based gas chromatography–mass spectrometry (GC-MS), and high-resolution liquid chromatography–mass spectrometry (LC-MS).¹⁷

• The University of Victoria and Vancouver Island University researchers also received support from Health Canada’s SUAP to pilot various DCTs, including FTS, FTIR spectroscopy, GC-MS, Raman spectroscopy, surface-enhanced Raman spectroscopy (SERS), and paper spray–mass spectrometry (PS-MS) to provide real-time harm reduction in Victoria.²⁴

The growing presence of highly potent and toxic substances in the unregulated drug supply and the opioid toxicity crisis have driven the recent expansion of DCSs across Canada.²⁵ From 2020 to 2024, DCSs have increasingly been offered across Canada, with at least 1 service now operating in each of 9 provinces and territories using various technologies.²²

DCSs are offered at fixed sites in the community, such as supervised consumption sites, in mobile units, or at events (e.g., music festivals) across several provinces. While regulatory requirements may vary by province, all DCS operate by way of a section 56 exemption from the federal Controlled Drugs and Substances Act.²⁶ These services face challenges related to accessibility, legal frameworks, and standardization across jurisdictions.²²

In Canada, inequitable access to DCSs results from geographic disparities in the availability of DCSs and from limitations associated with how DCSs are structured. Despite the high rates of fatal and nonfatal drug toxicity associated with the use of unregulated drugs across all geographic regions of Canada, nationwide, there is considerable variability in access to DCSs.²⁷ As of 2023, 30 organizations across Canada provide community DCSs, but the number of DCSs in each of Canada’s provinces and territories varies widely.²² Access to DCSs in Canada is also concentrated in large urban areas, with limited access in smaller urban areas and rural and remote areas.^27,28 Additionally, disparities in access to DCSs arise due to limitations associated with how specific DCSs are structured, including limited hours of operation, DCSs being located in areas outside of where PWUD live and/or consume drugs, and the lack of services tailored to the needs of specific demographic groups — including Indigenous Peoples, racialized people, and 2SLGBTQIA+ individuals, in some areas.²⁷

Overview of DCT Outcomes

Different DCTs have varying advantages and limitations depending on the setting in which they are used. There are several types of DCTs, which differ in terms of their analytical techniques, complexity, accuracy, limit of detection (LOD), quantification capabilities, cost, and feasibility.^18,22,29 For example, some low-cost and rapid tests, such as FTS, offer identification of a limited range of substances but can be useful for quick screening.¹⁴ Midrange technologies, such as FTIR spectroscopy, are more expensive but provide better accuracy and can identify a broader range of substances.¹⁴ Advanced laboratory-based methods, such as GC-MS, LC-MS, quantitative nuclear magnetic resonance (qNMR) spectroscopy, and PS-MS, offer high sensitivity and the ability to quantify substances, but they are more complex and costly to implement.¹⁴ They are also used as reference standards to validate the accuracy of other DCTs and to confirm the presence and concentration of specific substances in samples.^18,22,29 Based on well-established principles in test accuracy evaluation¹ and harm reduction and DCT literature,^30-32 previous studies have used several outcomes to evaluate various DCTs in terms of their accuracy, LOD, repeatability, reproducibility, and other considerations (refer to the Key Terminology section for definitions and elaboration for the important terms used throughout this report). In the context of drug checking, certain test accuracy measures may be more critical than others. For example, reporting sensitivity is often prioritized because a false negative (FN) (failing to detect a harmful substance) carries greater risk. Specificity, reflecting the problems with false positives (FP), is also an important outcome to report because FP can generate misinformation about the unregulated drug supply and undermine harm reduction messaging.

Why Is It Important To Do This Review?

Substance use exists on a spectrum that includes abstinence (no use), beneficial use (e.g., prescribed medications used appropriately), lower-risk use, higher-risk use, and addiction (i.e., substance use disorder).⁴ This continuum model underlines the importance of tailoring services and support to an individual’s goals and needs, whether they seek to manage, reduce, or eliminate their use.¹⁰

DCTs can be used as a harm reduction and substance use service to determine the content of substances that individuals intend to consume.² DCSs provided at community and public health agencies can also connect individuals to other drug-related, health, and social services and supports according to their needs.¹⁵ Studies have produced evidence on DCTs, including their accuracy, LOD, effectiveness, and harms, as well as implementation guides for setting up DCSs.^16,18 Notably, the effectiveness of DCSs on the behaviour of PWUD, including intent to use, actual use and disposal of the drug, and outcomes related to models of DCSs (e.g., barriers and facilitators to use), has been addressed in previous studies.¹⁶ There have also been studies on various drug-checking tools and technologies which can specifically respond to issues around the drug supply in Canada.^22,29 However, there is an identified need for evidence synthesis of various DCTs regarding their accuracy, LOD, and repeatability and reproducibility to detect multiple toxic components at trace levels and unexpected compositions (e.g., carfentanil [a potent fentanyl analogue] in heroin or xylazine and benzodiazepines in fentanyl), in unregulated substance samples and their cost-effectiveness and overall costs along with ethical considerations. In particular, there is a need for summarizing studies that examine the same substance samples using multiple DCTs to better understand specific capabilities and outcomes of various technologies. A comprehensive assessment of DCTs through a customized health technology review is essential to ensure that these various technologies are equipped to adequately detect unexpected high-potency compositions of unregulated substance samples while balancing costs and ethical and equity considerations for people living in Canada.

Objectives

In response to an external request to support decision-making about DCTs, we prepared this customized health technology review to summarize and, when studies met the criteria for the selected appraisal tools, critically appraised the available evidence about:

• the accuracy of DCTs for detecting compositions of unregulated substance samples

• the LOD, repeatability, and reproducibility of DCTs for detecting compositions of unregulated substance samples

• the cost-effectiveness of DCTs for detecting compositions of unregulated substance samples

• the costs associated with DCTs for detecting compositions of unregulated substance samples.

Methods

An information specialist conducted customized literature searches, balancing comprehensiveness with relevancy, of multiple sources and grey literature on May 26, 2025.

Following Cochrane rapid review methods guidance,³³ at least 2 reviewers were involved in each level of screening citations and selecting studies based on the inclusion criteria presented in Table 1 and critically appraised included studies that met the criteria for appraisal using 2 critical appraisal tools, including the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) checklist³⁴ for test accuracy studies and relevant domains of Drummond checklist³⁵ for cost description studies.

The DCT manufacturers within and outside Canada were identified via search engines, by reviewing DCT literature, and through discussions with DCT experts. Results were reviewed and summarized by 2 reviewers.

Appendix 1 presents a detailed description of methods and selection criteria for included studies. We integrated ethics and equity considerations throughout the report.

Table 1: Selection Criteria

DCT = drug-checking technology.

^aBased on the request, for questions 1 to 3, we included studies that evaluated more than 1 DCT (at point-of-care, and/or laboratory facilities) on the same substance samples. This should not be confused with comparative test accuracy studies in clinical research, given that the clinical review questions are noncomparative and no comprehensive assessment of comparative bias was conducted.

^bReference standards are particularly used for outcomes including sensitivity, specificity, and limit of detection.

^cHighly sensitive and specific DCTs used to validate the accuracy of other DCTs (i.e., emerging and point-of-care DCTs).

^dThe outcomes of interest were informed by principles in test accuracy evaluation, the DCT literature, and findings of the survey with National Drug-Checking Working Group responses.

^eThe targeted literature search for costing estimates was supplemented by contacting DCT manufacturers to obtain relevant Canadian costing estimates.

Summary of Evidence

Quantity of Research Available

This report includes 20 studies^23,30,36-53 that met our inclusion criteria, including 19 test accuracy studies and 1 cost description study. Eighteen studies^23,30,36-51 addressed question 1 (accuracy), 7 studies^{30,36,42,45,46,51,53} addressed question 2 (LOD, repeatability, and reproducibility), and 1 study⁵² addressed question 4 (costs). No studies were identified that addressed question 3 (cost-effectiveness). We reported on the characteristics and results from the included studies.

Appendix 2 presents the PRISMA⁵⁴ flow chart of the study selection.

Summary of Study Characteristics

Summaries of study characteristics are organized by research question. Appendix 2 provides the list of included studies along with the list of excluded studies and the reasons for exclusion. Appendix 3 provides details regarding the characteristics of included studies.

Included Studies for Question 1: Accuracy of Any DCTs for Detecting Compositions of Unregulated Substance Samples

We identified 18 test accuracy studies addressing this question.^23,30,36-51 Of these,12 studies examined the accuracy outcomes of more than 1 DCTs against the confirmatory lab testing,^{23,30,36-38,41-43,47,49,50,53} whereas 6 studies^{39,40,45,46,48,51} examined the outcome of 1 DCT against the confirmatory lab testing.

All included studies were cross-sectional studies.^23,30,36-51 Nine studies^{23,36,37,41-43,47,49,50} were conducted in Canada, 7 studies^{30,38,44-46,48,51} in the US, 1 study³⁹ in Italy, and 1 study in France.⁴⁰

All 18 studies reported 1 or more outcomes on accuracy. Of these, 10 studies reported on sensitivity,^{23,30,36,37,39-42,45,46} 7 studies on specificity,^{23,30,36,40,41,45,46} 3 studies on positive predictive value (PPV) and negative predictive value (NPV),^23,36,40 10 studies on true positives (TPs),^{38,47-49,51,55-60} 2 studies on true negatives (TNs),^38,41 7 studies on FPs,^{30,38,40,41,43,45,51} 10 studies on FNs,^{23,30,36,40,41,43,45,47,48,51} and 2 studies on concordance (i.e., agreement between the results of 2 tests).^23,44 In addition to the accuracy outcomes, 1 study also reported outcomes on reproducibility;⁴⁶ however, no included studies reported results on repeatability of DCTs.

Fifteen studies^{23,36-44,47-51} included drug samples submitted by PWUD to various substance use services and supports (e.g., supervised consumption sites, harm reduction settings at night events), 2 studies^30,45 included drug samples obtained by police departments or law enforcement, and 1 study⁴⁶ included drug paraphernalia residue samples obtained through health departments. Of the included studies, 3 studies^38,40,44 reported information about the characteristics of participants who provided the substance samples for drug checking.

Fourteen studies^{23,36-44,47,48,50} conducted drug checking at both point-of-care and laboratory facilities, whereas 4 studies^30,45,46,51 conducted all drug checking of samples exclusively at laboratory facilities. For 13 studies,^23,30,38-48 all included samples underwent confirmatory lab testing, whereas for 5 studies,^36,37,49-51 part of the samples that were tested at the point of care also underwent confirmatory lab testing.

Fourteen studies focused on specific target substances in any drug sample:

• fentanyl and its analogues (5 studies)^{23,30,37,38,45}

• benzodiazepine and analogues (5 studies)^36,41-43,49

• xylazine (4 studies)^44,46,48,50

• synthetic cannabinoid (1 study).⁴⁷

In addition, 2 studies^39,40 focused on detecting any compositions in the drug samples submitted, whereas the focus of 1 study was on the analysis of 22 illicit drugs and cutting agents.⁵¹

The 18 test accuracy studies reporting on the accuracy of DCTs used the following DCTs as index tests or intervention(s):

• immunoassay test strips (ITS) (16 studies), including FTS (10 studies),^{23,30,37,38,44,45,47,49-51} benzodiazepine test strips (BTS) (4 studies),^36,41,43,49 and xylazine test strips (XTS) (2 studies)^46,48

• FTIR spectroscopy (11 studies)^{23,30,36-38,42-44,47,49,50}

• colorimetric reagent (1 study)³⁹

• portable GC-MS (1 study)⁴²

• hand-held Raman spectrometer (1 study)³⁰

• infrared spectroscopy (IR) spectroscopy (1 study)⁴⁰

• SERS (1 study).⁴¹

All 18 test accuracy studies used highly sensitive and specific DCTs (confirmatory lab testing) as reference standards (some studies used more than 1 type of confirmatory lab testing as their reference standard):

• GC-MS (11 studies)^{23,30,36-39,43,44,47,49,50}

• qNMR spectroscopy (5 studies)^{23,36,37,43,49}

• LC-MS (5 studies)^{36,37,43,47,49}

• liquid chromatography–tandem mass spectrometry (LC-MS-MS) (2 studies)^45,51

• liquid chromatography–quadrupole time-of-flight mass spectrometry (3 studies)^38,44,48

• PS-MS (2 studies)^41,42

• direct analysis in real time mass spectrometry (1 study)⁴⁶

• tandem gas chromatography–mass spectrometry (1 study)⁴⁶

• ultra-performance liquid chromatography–high-resolution mass spectrometry (UPLC-HRMS) (1 study).⁴⁰

Included Studies for Question 2: LOD, Repeatability, and Reproducibility of Any DCTs for Detecting Compositions of Unregulated Substance Samples

We identified 7 test accuracy studies^{30,36,42,45,46,51,53} addressing this question (6 studies^{30,36,42,45,46,51} also addressed outcomes related to research question 1). All were cross-sectional studies^{30,36,42,45,46,51,53} and reported on LOD. Of these studies, 1 study also reported on reproducibility.⁴⁶ Three studies^36,42,53 were conducted in Canada, and 4 studies^30,45,46,51 were conducted in the US.

Four studies^36,42,51,53 included drug samples submitted by PWUD to various substance use services and supports (e.g., harm reduction sites), 2 studies^30,45 included drug samples obtained by police departments or law enforcement, and 1 study included⁴⁶ drug paraphernalia residue samples obtained through health departments. In 4 studies,^30,36,42,53 drug checking of some samples was conducted at the point of care and some others at laboratory facilities; in 3 studies,^45,46,51 all drug checking of samples was conducted at laboratory facilities. For 6 studies,^{30,42,45,46,51,53} all samples underwent confirmatory lab testing, whereas some of the samples of 1 study³⁶ underwent confirmatory lab testing.

Five studies focused on specific target substances in any drug sample:

• fentanyl and its analogues (3 studies)^30,45,53

• benzodiazepine and analogues (1 study)³⁶

• xylazine (1 study).⁴⁶

The focus of the other 2 studies was on the detection of carfentanil and etizolam in opioid samples⁴² and the analysis of 22 illicit drugs and cutting agents.⁵¹

Of 7 test accuracy studies^{30,36,42,45,46,51,53} reporting on LOD, the following DCTs were tested:

• ITS (6 studies), including FTS (4 studies),^30,45,51,53 BTS (1 study),⁵³ and XTS (1 study)⁴⁶

• FTIR spectroscopy (3 studies)³⁰

• portable GC-MS (1 study)⁴²

• hand-held Raman spectrometer (1 study).³⁰

The authors of 1 study reported reproducibility⁴⁶ for XTS.

Three studies used confirmatory lab testing, including qNMR spectroscopy^36,53 and PS-MS,⁴² as reference standards. Four studies used various concentrations of target substances, including laboratory-grade fentanyl and analogues⁴² and xylazine,⁴⁶ to calculate the minimum cut-off points at which target substances can be consistently detected (i.e., LOD). One study that assessed reproducibility⁴⁶ used pure xylazine as its reference standard.

Included Studies for Question 3: Cost-Effectiveness of Any DCTs for Detecting Compositions of Unregulated Substance Samples

This review did not identify relevant studies that addressed research question 3; therefore, no summary can be provided.

Included Studies for Question 4: Costs Associated With Any DCTs for Detecting Compositions of Unregulated Substance Samples

We identified 1 study⁵² that provided information on the cost of launching and sustaining a community-based drop-in DCS program in Rhode Island, US, from January 2023 to May 2023, which included the implementation of FTIR spectroscopy and FTS at the point of care in addition to confirmatory lab testing. The authors of the study reported that the participants who provided the drug samples were PWUD.

The authors used a microcosting approach to estimate total service costs during the implementation period from a health care payer perspective. All cost inputs required to conduct their drug-checking program were included, regardless of whether they were paid or donated. Unit prices were obtained from various sources, including financial records, itemized bills or receipts, equipment catalogues, and expenditure reports (e.g., salaries and fringe benefits). Program staff were interviewed to estimate the number of hours spent on training, receiving technical assistance and troubleshooting to acquire the required technical expertise for conducting FTIR spectroscopy.⁵² To estimate the cost per drug checked (i.e., unit costs), both fixed (e.g., equipment) and recurrent (e.g., personnel) costs were included.⁵²

Given that the evaluated drug-checking program benefited from in-kind training support and had existing technical expertise in FTIR spectroscopy, a sensitivity analysis was conducted to replace the study’s estimated training costs with the assumed training costs to start up such a program if sites do not receive additional government funding or in-kind support.⁵²

Summary of Critical Appraisal

Appendix 4 provides additional details about the risk of bias and applicability of the included studies.

Included Studies for Question 1: Accuracy of Any DCTs for Detecting Compositions of Unregulated Substance Samples

Overall, the test accuracy studies were judged to be at risk of bias: all of the test accuracy studies were assessed as having a high or unclear risk of bias in at least 1 domain, with many studies demonstrating multiple sources of potential bias.^23,30,36-51 If a study included more than 1 index test, the appraisal conducted and judgment applies to all of them.

Concern regarding applicability was low for all studies,^23,30,36-51 which indicates that the included samples matched the review question, the index test and its conduct or interpretation were aligned with the review question, and the target condition as defined by the reference standard was consistent with the review question.

Risk of Bias and Applicability in Test Accuracy Studies

All studies^23,30,36-51 relied on voluntary or convenience sampling methods and were rated as having a high risk of bias in sample selection. There were no applicability concerns about the sample selection for any of the studies: study samples, such as samples submitted by PWUD or law enforcement, aligned with the real-world DCT settings.

Eight studies^{30,36,37,39,43-45,48} were judged to have a low risk of bias for the conduct and interpretation of the index test or intervention, where drug-checking procedures followed standardized protocols and thresholds were clearly prespecified. The remaining 10 studies^{23,38,40-42,46,47,49-51} had high or unclear risk, often due to insufficient reporting on whether index test results were interpreted without knowledge of confirmatory test results and/or without prespecified thresholds or cut-offs. There were no applicability concerns about the index test or intervention for any of the studies, and they all reflected real-world drug-checking practices.^23,30,36-51

All studies^23,30,36-51 used valid and appropriate reference standards (e.g., GC-MS, LC-MS, liquid chromatography–quadrupole time-of-flight MS, PS-MS, qNMR spectroscopy), and concern regarding applicability was low across all studies in this domain. Despite a lack of clarity on knowledge of index test results in most studies, the reference methods were well-established and objective and not likely influenced by index test or intervention results.

Nine studies^{30,39-42,44-46,48} were judged to have a low risk of bias for the flow and timing domain, with appropriate intervals between the index test or intervention and reference testing, and uniform application of reference standards. In contrast, 9 studies^{23,36-38,43,47,49-51} were judged to have high or unclear risk, often due to partial confirmatory testing or selective inclusion of samples in the confirmatory analysis.

Table 2 presents a summary of the risk-of-bias and applicability assessments for the 18 test accuracy studies.

Table 2: Summary of Risk of Bias in the Included Test Accuracy Studies Using QUADAS-2³⁴

BTS = benzodiazepine test strips; FTIR = Fourier transform infrared; FTS = fentanyl test strips; GC-MS = gas chromatography–mass spectrometry; QUADAS-2 = Quality Assessment of Diagnostic Accuracy Studies 2; SERS = surface-enhanced Raman spectroscopy.

^aIf a study includes more than 1 index test or intervention, the appraisal conducted and judgment made applies to all of them.

^bWe changed the word “patient” in QUADAS-2 to “sample” because the populations studied in this review are substance samples tested using DCTs.

Included Studies for Question 2: LOD, Repeatability, and Reproducibility of Any DCTs for Detecting Compositions of Unregulated Substance Samples

Seven studies^{30,36,42,45,46,51,53} reported LOD outcomes (6 studies^{30,36,42,45,46,51} addressing this question also reported results on accuracy, and 1 study⁵³ reported exclusive outcomes on LOD) and were not appraised with any specific tool, given that there is no known tool for appraising such outcomes. However, key sources of bias and applicability concerns were noted. This study⁵³ exhibited a high risk of bias regarding sample selection, as samples were selected based on convenience or technician suspicion rather than consecutive or randomized sampling, which may have introduced selection bias and limited representativeness of the local drug supply. No major applicability concerns were identified in this domain, given that the study population (i.e., samples submitted to a supervised consumption site DCS) aligned well with the context of evaluating point-of-care DCTs for harm reduction.

There was a low concern for bias in the application of the index test in the study,⁵³ given that both FTIR spectrometry and FTS were implemented using standardized procedures and commercially available technologies, although thresholds for detection were not prespecified. The reference standard was valid and appropriate, with no applicability concerns; however, blinding to index test results was not explicitly reported.

Regarding flow and timing, the study carried out confirmatory testing on a subset of samples, introducing potential risk of bias due to selective verification and lack of reporting on the time interval between the index test or intervention and the reference testing.⁵³

Included Studies for Question 3: Cost-Effectiveness of Any DCTs for Detecting Compositions of Unregulated Substance Samples

This review did not identify relevant studies that addressed research question 3; therefore, no appraisal can be provided.

Included Studies for Question 4: Costs Associated With Any DCTs for Detecting Compositions of Unregulated Substance Samples

This review identified 1 relevant study that addressed research question 4.⁵² The authors of the cost description study⁵² clearly stated their research question, justified its economic relevance in the context of the opioid crisis, and described with sufficient details the microcosting approach to assess the costs of implementing a point-of-care drug-checking program in Rhode Island, US. The authors clearly described the analytic perspective as that of the payer and the cost categories included reflected this analytical perspective, and the authors specified a short-term time horizon (January 2023 to May 2023), focusing on implementation costs. Costs considered were comprehensive in that they reflected expected costs associated with FTIR spectrometry, FTS, and confirmatory laboratory testing. The authors clearly reported on all assumptions, including the approach to annualize the cost of the FTIR machine given its substantial upfront start-up costs. Sensitivity analyses were conducted that varied the approach to derive training costs using cost ranges. The primary outcome-cost per drug checked was presented with a clear disaggregation of cost components, and the discussion was well supported by the data and accompanied by appropriate caveats.

The generalizability of the findings may be limited to well-supported pilot contexts and may not translate directly to broader health care settings or other jurisdictions. Key model inputs were mainly derived from US data, and differences in drug use patterns, service delivery infrastructure, and the costs associated with implementing point-of-care drug checking between health care systems in the US and Canada could impact the relevance of the cost findings to the context in Canada. Detailed reporting of results was provided on unit costs (e.g., salaries, equipment, training), total costs, and cost per drug checked. Given the transparency with which results are reported, it theoretically could permit the resource amount per drug checked to be calculated. If this value is expected to be similar across settings in Canada and in the US, these values may be transferable to the setting in Canada.

Although the authors refer to their work as a cost analysis, the study could be more accurately described as a cost description, given that it estimates the implementation costs of a single DCT without comparing it to alternative interventions. Several additional considerations could improve the robustness and generalizability of the findings. For example, costs related to the number and type of FTIR libraries purchased were not explored. Given that library expansion may improve the accuracy and sensitivity of drug checking, but at an additional cost, this could significantly affect implementation budgets and limit generalizability to other settings. Moreover, potential long-term maintenance costs such as software updates, calibration, or technical support were not included, likely due to the short time horizon. The study also relied solely on deterministic sensitivity analysis. Incorporating probabilistic methods could have provided more insight into the impact of simultaneous variation in input parameters. While the authors did examine the effect of in-kind training support, which was not found to substantially influence overall costs, this does not account for other plausible implementation scenarios. Finally, although a health care payer perspective is appropriate for many policy decisions, the analysis did not capture potential out-of-pocket or opportunity costs for service users (e.g., transportation), which may be relevant in some public health contexts.

Summary of Findings

Appendix 5 presents additional details regarding the main study findings.

Question 1: Accuracy of Any DCTs for Detecting Compositions of Unregulated Substance Samples

Evidence regarding the accuracy of any DCTs for detecting compositions of unregulated substance samples was available from 18 test accuracy studies.^23,30,36-51 Table 3 summarizes the test accuracy outcomes (i.e., sensitivity, specificity, and FP and FN rates) of any DCTs for detecting compositions of unregulated substance samples.

Test Accuracy for the Detection of Fentanyl and Analogues

Seven studies^{23,30,37,38,42,44,45} reported on the accuracy of any DCTs for detecting fentanyl and its analogues:

• The sensitivity of:

  • FTS ranged between 87.5% to 100% for detecting fentanyl in 4 studies,^23,30,37,45 and between 77% to 100% for detecting fentanyl analogues (1 study)³⁷

  • FTIR spectroscopy ranged between 58% and 91% for detecting fentanyl in 4 studies.^23,30,37,42 FTIR spectroscopy was not sensitive enough to detect carfentanil (i.e., a fentanyl analogue) at relevant concentrations (2 studies)^37,42

  • portable GC-MS was 95% for detecting fentanyl and 62% for carfentanil (1 study)⁴²

  • hand-held Raman spectrometers was 61.1% and 38.5% for detecting fentanyl against the confirmatory lab testing in 2 different facilities (1 study).³⁰

• The specificity of:

  • FTS ranged between 89.2% and 98.1% (3 studies)^23,30,45

  • FTIR spectroscopy ranged between 90.2% and 99% (2 studies)^23,30

  • portable GC-MS was not reported⁴²

  • hand-held Raman spectrometers was 91.5% and 92.3% against the confirmatory lab testing in 2 different facilities (1 study).³⁰

• The FP rate of:

  • FTS ranged between 1.9% and 10.9% (3 studies)^23,30,45

  • FTIR spectroscopy ranged between 1% and 9.3% (2 studies)^23,30

  • portable GC-MS was not reported⁴²

  • hand-held Raman spectrometers was 8.5% and 7.7% against the confirmatory lab testing in 2 different facilities (1 study).³⁰

• The FN rate of:

  • FTS ranged between 0% and 12.5% for the detection of fentanyl in 4 studies,^23,30,37,45 and between 0% and 23% for fentanyl analogues (1 study)³⁷

  • FTIR spectroscopy ranged between 9% and 42% for detecting fentanyl in 4 studies,^23,30,37,42 and was 100% for detecting fentanyl analogues (2 studies)^37,42

  • portable GC-MS was 5% for detecting fentanyl and 38% for carfentanil (1 study)⁴²

  • hand-held Raman spectrometers was 38.9% and 61.5% for detecting fentanyl against the confirmatory lab testing in 2 different facilities (1 study).³⁰

• Other accuracy outcomes:

  • One study⁴⁴ that examined fentanyl detection using FTIR spectrometry and confirmatory lab testing (GC-MS and LC-QTOF-MS) indicated that the results based on samples that tested positive for fentanyl were consistent across the 2 methods (33.9% and 36.8%, respectively) and that they showed a high pairwise concordance (96%).

  • FTS was able to correctly detect 2 fentanyl analogues (acetyl fentanyl and furanyl fentanyl) alone or in the presence of another drug, in both powder and pill forms, in 1 study.³⁰ However, it could not detect carfentanil and several novel fentanyl analogues (e.g., fluorobutyrfentanyl [FBF], fluoroisobutyrfentanyl [FIBF]) at low concentrations in 1 other study.⁴⁵

  • FTIR and Raman spectrometers could produce other relevant information, such as the percentage of fentanyl and the presence of other drugs and cutting agents (1 study).³⁰

Test Accuracy for the Detection of Benzodiazepine and Analogues

The results from 5 test accuracy studies^36,41-43,49 reported on the detection of benzodiazepine and its analogues:

• The sensitivity of:

  • BTS ranged between 8% and 100% (4 studies);^36,41,43,49 the relatively low sensitivity (8%) in 1 study⁴¹ was attributed to limitations of BTS in trace detection of etizolam in opioids

  • FTIR spectroscopy ranged between 8.3% and 26% (4 studies)^36,42,43,49

  • BTS and FTIR combined was 70.8% and 75% (2 studies)^36,43

  • portable GC-MS was 36% in all samples and 78% in samples with concentrations of more than 3% (1 study)⁴²

  • SERS was 96% (1 study).⁴¹

• The specificity of:

  • BTS ranged between 67% and 100% (4 studies)^36,41,43,49

  • FTIR spectroscopy ranged between 96.1% and 100% (3 studies)^36,43,49

  • BTS and FTIR combined was 82% and 82.2% (2 studies)^36,43

  • portable GC-MS was not reported⁴²

  • SERS was 86% (1 study).⁴¹

• The FP rate of:

  • BTS ranged between 0% and 33% (4 studies)^36,41,43,49

  • FTIR spectroscopy ranged between 0% and 3.9% (3 studies)^36,43,49

  • BTS and FTIR combined was 17.8% and 18% (2 studies)^36,43

  • portable GC-MS was not reported⁴²

  • SERS was 14% (1 study).⁴¹

• The FN rate of:

  • BTS ranged between 0% and 92% in (4 studies);^36,41,43,49 the relatively high FN rate (92%) in 1 study⁴¹ was attributed to the limitations of BTS in trace detection of etizolam in opioids

  • FTIR spectroscopy ranged between 74% and 91.7% (4 studies)^36,42,43,49

  • BTS and FTIR combined was 25% and 29.2% (2 studies)^36,43

  • portable GC-MS was 64% in all samples and 22% in samples with concentrations of more than 3% (1 study)⁴²

  • SERS was 4% (1 study).⁴¹

Test Accuracy for the Detection of Xylazine and Analogues

The results from 4 test accuracy studies^44,46,48,50 reported on the detection of xylazine:

• The sensitivity of:

  • XTS was 22.2% and 97.4% (2 studies);^46,48 the relatively low sensitivity (22.2%) in 1 study⁴⁸ was attributed to limitations of XTS in trace detection of xylazine, given that its sensitivity increased to 100% when xylazine was a major active component of the tested sample⁴⁸

  • FTIR spectroscopy was 0% (i.e., unable to detect xylazine) and 54% (2 studies).^44,50

• The specificity of:

  • XTS was 100% (2 studies)^46,48

  • FTIR spectroscopy was not reported (2 studies).^44,50

• The FP rate of:

  • XTS was 0% (2 studies)^46,48

  • FTIR spectroscopy was not reported (2 studies).^44,50

• The FN rate of:

  • XTS was 2.6% and 77.8% (2 studies);^46,48 the relatively high FN rate (77.8%) in 1 study⁴⁸ was attributed to the limitations of XTS in trace detection of xylazine.

  • FTIR spectroscopy was 46% and 100% (2 studies).^44,50

Test Accuracy for the Detection of Other Compositions

Two studies^39,40 reported the accuracy of DCTs for detecting any compositions in drug samples submitted for drug checking. The additional 2 studies reported on the analysis of 22 illicit drugs and cutting agents⁵¹ and on synthetic cannabinoid compounds.⁴⁷

• One study³⁹ indicated that colorimetric reagents had a moderate to high sensitivity (ranging between 78% and 100%) with regard to samples being contaminated (e.g., LSD, amphetamine) or samples with very a small quantity; however, colorimetric reagents failed to identify drug mixtures (with approximately equal presentation of different drugs) and methamphetamine samples identified using the confirmatory lab testing (GC-MS analysis) and the contaminant present in them.

• One study⁴⁰ that examined the sensitivity and specificity of IR spectroscopy against confirmatory lab testing (LC-QTOF) indicated that IR spectrometry showed high sensitivity (ranged between 82.4% to 100%), and low FN rates (ranged between 0% to 17.6%), with the exception of heroin, and high specificity (100%) for detection of all main substances (i.e., cocaine, 3,4-methylenedioxymethamphetamine (MDMA), heroin, and amphetamine).

• One study⁵¹ that focused on the analysis of 22 illicit drugs and cutting agents showed that FTS has a high sensitivity (95%) and specificity (100%), low FN rates (i.e., 5%), and no record of FPs.

• One study⁴⁷ that assessed the FN rate and sensitivity of FTIR spectroscopy against confirmatory lab DCTs (GC-MS or LC-MS) for detecting the synthetic cannabinoid compounds showed that FTIR spectroscopy had a sensitivity of 52% and FN rate of 48%.

Table 3: Summary of Findings for the Test Accuracy of Any DCTs for Detecting Compositions of Unregulated Substance Samples

BTS = benzodiazepine test strips; DART-MS = direct analysis in real time mass spectrometry; DCT = drug-checking technology; FN = false negative; FP = false positive; FTIR = Fourier transform infrared; FTS = fentanyl test strips; GC-MS = gas chromatography–mass spectrometry; GC-MS-MS = gas chromatography–tandem mass spectrometry; IR = infrared spectroscopy; LC-MS = liquid chromatography–mass spectrometry; LC-MS-MS = high-precision liquid chromatography–tandem mass spectrometry; LC-QTOF-MS = liquid chromatography quadrupole time-of-flight mass spectrometry; LSD = lysergic acid diethylamide ; MDMA = 3,4-methylenedioxymethamphetamine; NR = not reported; PS-MS = paper spray–mass spectrometry; PWUD = people who use drugs; qNMR = quantitative nuclear magnetic resonance; SERS = surface-enhanced Raman spectroscopy; XTS = xylazine test strips.

^aCalculated by authors of this report based on the available data in the included studies.

^bThe authors of this study reported a portion of the FTS and FTIR results (i.e., 801) for all positives obtained by the confirmatory lab testing (i.e., 855). We calculated the sensitivity and FNs based on 801 samples that were explicitly reported in the study.

^cLowest and highest sensitivity and FN rates identified for examining various types of fentanyl analogues.

^dSensitivity and specificity of the FTS against the confirmatory lab testing in the Rhode Island laboratory.

^eSensitivity and specificity of the FTS against the confirmatory lab testing in the Baltimore laboratory.

^fResults reported here are related to the Raman spectrometer SERS kit. For results reported on the Raman spectrometer point-and-shoot mode, please refer to Table 26.

^gSensitivity calculated in all samples.

^hSensitivity calculated in samples with concentrations of more than 3%.

ⁱIn this study,⁵¹ FTS are used in the context of testing any drugs, and not specifically fentanyl as the target composition. This study also used BTS, but the author stated that “a comparison of the results from BTS to LC-MS-MS could not be made because all samples but 1 tested negative on BTS. For the one sample that showed a positive BTS, none of the 4 benzodiazepines included in this LC-MS-MS method were quantified in that sample.” This information is available in Table 29.

^jMIX: samples with mixtures of substances, in which the substances were present in approximately equal quantities and had similar pharmacological activities.

^kDebris: a sample that was found to be free of active substances.

Question 2: LOD, Repeatability, and Reproducibility of Any DCTs for Detecting Compositions of Unregulated Substance Samples

Limit of Detection

Seven test accuracy studies^{30,36,42,45,46,51,53} reporting on LOD of DCTs:

• Two studies^36,53 that tested the LOD of ITS (FTS and BTS) and FTIR spectroscopy against confirmatory lab testing (qNMR spectroscopy) indicated that FTS and BTS were able to detect the presence of fentanyl and benzodiazepines in samples at concentrations between 2% and 5%. The LOD of FTIR spectroscopy in both studies was reported to be more than 10%.

• Two studies^45,51 that evaluated the LOD of FTS using laboratory-grade fentanyl and high-priority fentanyl analogues found the LOD to be 100 ng/mL⁵¹ and 200 ng/mL⁴⁵ for fentanyl and 1,000 ng/mL or less⁴⁵ for 13 analogues (among 17 analogues tested).

• One study⁴⁶ that examined the LOD of XTS using pure xylazine laboratory samples found a low LOD of 2.5 mcg/mL or more (2,500 ng/mL).

• One study⁴² that examined the LOD of portable GC-MS and FTIR against PS-MS found that, at concentrations of more than 3%, the portable GC-MS was able to detect 78% of the etizolam samples that were confirmed by PS-MS, whereas its LODs for the detection of carfentanil was between 0.13% and 0.63% by weight. The LOD of FTIR for detection of samples containing etizolam was not clearly stated. In addition, FTIR was not able to detect samples containing carfentanil that were detected by PS-MS.

• One study³⁰ that examined the LOD of FTS, hand-held Raman, and FTIR spectrometers, and GC-MS using powder fentanyl standards as reference, found LODs of 0.100 mcg/mL, 3.1 mcg/mL, 3% to 4% by weight, and 25 mcg/mL for FTS, GC-MS, FTIR spectroscopy, and hand-held Raman spectrometer, respectively.

Repeatability

We did not identify any study that reported results on the repeatability of DCTs.

Reproducibility

One test accuracy study⁴⁶ that examined the reproducibility for XTS, including consistency of results across days (2 tests per day) over a 6-week period and consistent detection near LOD, indicated that the XTS consistently produced clear positive results with the 5,000 mcg/mL xylazine solution, and clear negative results with xylazine-free water. The approximate LOD, where consistent positive results could be obtained, was found to be 2.5 mcg/mL (2,500 ng/mL).⁴⁶

Question 3: Cost-Effectiveness of Any DCTs for Detecting Compositions of Unregulated Substance Samples

We found no relevant evidence regarding the cost-effectiveness of any DCTs for detecting compositions of unregulated substance samples; therefore, no summary can be provided.

Question 4: Costs Associated with Any DCTs for Detecting Compositions of Unregulated Substance Samples

The findings from 1 cost description study⁵² that was conducted from a payer perspective in the US, indicated that the total cost of establishing and operating a DCS pilot was US$71,044 and the cost per drug checked was US$474, including confirmatory lab testing (US$150 per sample). More than half (approximately 54%) of this cost was attributed to the FTIR machine, software, libraries, and technical training.

Sensitivity analyses indicated that the cost could increase to between US$78,058 and US$83,058, or US$544 to US$593 per sample checked, for programs needing to pay for at-cost training supports, including specialized training in FTIR spectroscopy.⁵² The study authors further estimated that real-world implementation may require 10% to 25% of samples to be checked for ongoing quality assurance, which would increase costs. It is noteworthy that this study was conducted from January 2023 to May 2023 and the costs may fluctuate from 1 year to another; their costing represents the first year of program implementation. We note that prices are expected to vary based on the number of tests conducted, given that costs substantially decrease with higher testing volumes.

Summary of DCT Manufacturer Information

Table 4 provides a consolidated summary of findings, and Appendix 6 presents the detailed findings from surveys completed by DCT manufacturers.

We identified and contacted 21 DCT manufacturers within and outside Canada. Of these, 18 responded and were sent a series of questions via a survey about their respective DCTs, including costs and capabilities (e.g., portability, time to result). Of those manufacturers, 11 completed 16 independent surveys related to different types of DCTs (i.e., certain manufacturers sell more than 1 relevant DCT).

The responses include information about the following DCT types:

• ITS (2 manufacturers)

• Raman, infrared (IR), and near-infrared DCTs (6 manufacturers)

• Mass spectrometry (MS) DCTs (3 manufacturers, 1 overlap with the previous group)

Among the survey data received, 4 DCTs were for technologies included in the test accuracy studies:

• 1 manufacturer produces ITS that were evaluated in 14 studies;^{30,36,37,41,43-51,53}

• 1 manufacturer provided information on LC-MS that was evaluated in 1 test accuracy study;⁴⁵ and

• 1 manufacturer provided information on FTIR spectrometer that were reportedly used in 10 studies,^{30,36-38,43,44,47,49,50,53} and related to IR stereoscopy which was reportedly used in 1 study.⁴⁰

Of 11 manufactures who responded to the survey, 4 are Canadian, and 2 of the 7 international manufacturers have office locations in Canada. All the international manufacturers sell to buyers outside their country.

Immunoassay Test Strips

Two manufacturers of ITS responded to our survey. These ITS can detect substances, including fentanyl, xylazine, benzodiazepines, medetomidine, and nitazenes, typically within a few minutes. These strips are low-cost (i.e., a few dollars per strip depending on the type and quantity) and do not require any ongoing operational costs. These strips are designed for “lay users” (nonexperts), are highly portable, and can be widely used in community-based settings (e.g., mobile harm reduction vans).

Raman, Infrared, and Near-Infrared Technologies

We received 7 completed surveys from 6 manufacturers for this category, including:

• 2 manufacturers of FTIR;

• 3 manufacturers of SERS and Raman Spectroscopy (1 completed 2 surveys on 2 different models); and

• 1 manufacturer of near-infrared spectroscopy.

Their time to results ranges from a few seconds to 15 minutes. These DCTs were reported to be at a midrange cost. In most cases, the manufacturers reported that services, including software updates, database or library upgrades, online support, and a limited-time warranty, are included in the purchase price. All manufacturers reported no or minimal ongoing operational costs (e.g., consumables, reagents, test cartridges, maintenance) for these devices. All manufacturers provide remote and/or onsite DCT servicing as required and access to DCT support by phone and email. Expected service and maintenance costs were reported to be $0 or nominal for all devices. The training and qualification required for operating these devices vary from minimal to basic knowledge around FTIR spectroscopy, spectral interpretation, sampling preparation, and optimization. Except for 1 FTIR device, these devices were reported as portable and can be used outdoors and in a vehicle. Except for the FTIR device, other manufacturers recommended using ITS in combination with their technologies for the detection of small quantities and specific substances (ITS must be purchased separately).

MS Technologies

We received 7 completed surveys from 3 manufacturers of different types of MS technologies, including:

• 1 from a manufacturer of a Mini MS (small size and portable)

• 1 from a manufacturer of direct ionization MS

• 5 from a manufacturer of high-resolution LC-MS and GC-MS, triple quadrupole LC-MS and GC-MS, and an ion source paper spray mass spectrometry (PS-MS).

The time required for these DCTs to provide results ranges from 1 to 20 minutes, depending on the device. These DCTs were at the highest price range; however, in most cases, other expenses such as software updates or database or library upgrades, perpetual licence, online support, shipping, installation, familiarization, limited-time warranty, as well as onsite training of staff in 1 case were included in the purchase price. All provide remote and onsite DCT servicing as required, service and maintenance support based on the warranty or contract, and technical or user support included in the base purchase price. The training and qualification required for operating these devices vary from minimal to knowledge around running analytical instrumentation (MS is preferred, related to either GC or LC). One device can be used outdoors, and 2 devices can be used in a vehicle. Manufacturers reported there is no need to use ITS in combination with these DCTs.

Table 4: Consolidated Findings From Surveys With DCT Manufacturers

FTIR = Fourier transform infrared; ITS = immunoassay test strips; LC/GC-MS = liquid chromatography/gas chromatography–mass spectrometry; MS = mass spectrometry; NA = not applicable; NIS = near-infrared spectroscopy; PS-MS = paper spray mass spectrometry; SERS = surface-enhanced Raman spectroscopy; TSQ = triple stage quadrupole.

Limitations

Evidence Gap

There is a significant evidence gap in drug-checking research, particularly regarding the accuracy, LOD, cost-effectiveness, and real-world performance of different DCTs.^44,48 This review did not find any study that met the inclusion criteria of reporting results on the repeatability of DCTs. Many studies rely on small-scale pilots, limiting the generalizability of their findings to diverse and evolving drug markets.⁴⁸ Research often focuses on the detection of a narrow and specific range of substances (e.g., fentanyl, benzodiazepines), with limited evaluation of emerging synthetic analogues and complex and ultra-high-potency mixtures (e.g., nitazenes, highly potent opioid analogues).³⁶ There is also a need for more studies on the accuracy and other outcomes of many emerging DCTs and the multimethod approaches for drug checking (i.e., combining different DCTs for drug checking). Furthermore, there is a lack of standardized outcome measures and consistent reporting of sensitivity, specificity, and FN and FP rates in the studies evaluating DCTs, making cross-study comparisons challenging. Many studies are conducted as pilot projects or conducted with limited or nonrepresentative samples of the local drug market, which may not accurately reflect the complexity of evolving unregulated drug markets. Robust, large-scale studies that evaluate the effectiveness of multimethod approaches are needed.⁴⁸

This review concentrated on the technical performance and evaluation of DCTs. A key limitation of inconsistent evidence on the accuracy of DCT is the potential for FNs and FPs. FNs that allow highly potent and toxic substances to go undetected may create a false sense of safety among PWUD, potentially increasing the likelihood of use or higher-risk consumption.^36,39,44 Additionally, FP are also highly problematic, as they can generate misinformation about the unregulated drug supply and undermine harm reduction messaging — for example, by reinforcing myths that fentanyl is present in all substances. While most studies focus on detecting the presence or absence of target substances, less attention has been given to the assessment of the concentration of those detected substances. Yet, concentration levels are of significant interest and concern for regional health authorities and their community partners because they have implications for both individual and public health risk and may necessitate the use of different tools depending on the setting and purpose of testing. These limitations may undermine the harm reduction intent of DCT and highlight the need for clear communication about the technologies’ limitations, including their detection thresholds and the possibility of undetected or misidentified compositions.³⁶

Another limitation in the existing literature is that the definition of DCTs often focuses exclusively on the hardware of the device (e.g., spectrometer or test kit) without adequately considering the software components that process, interpret, or visualize the resulting data. This narrow framing overlooks how software contributes to the accuracy, usability, and adaptability of these technologies. In many cases, these software tools — whether embedded in the device or external analysis programs — play a critical role in determining the sensitivity and specificity of the technology. Different software systems can employ distinct algorithms, databases, or update schedules, which may lead to variations in how substances are identified and reported. Even for devices with onboard software, differences between versions can influence performance, as updates may modify reference libraries or analytical algorithms. However, these distinctions are rarely reported or evaluated in the current literature, making it difficult to compare results across studies or over time. Furthermore, the literature has not yet addressed the growing interest in applying machine learning approaches to enhance drug analysis, which represents an important gap in understanding future directions for the field.

Additional research to better integrate the priorities and needs of PWUD into program development and evaluation would be of benefit. DCSs are most effective when they are designed and evaluated in collaboration with PWUD because their insights are critical for understanding real-world barriers to access, trust in testing results, and the usability of different technologies. These perspectives provide key contextual factors such as how results are interpreted; whether technologies meet the needs of diverse groups; and the role of stigma, criminalization, or cultural safety in shaping service uptake.

This review did not identify any relevant evidence regarding the cost-effectiveness of DCTs for detecting compositions of unregulated substance samples. This limits the ability to evaluate their economic value. Without evidence on costs relative to health and/or social outcomes, policy-makers face challenges in how to allocate scarce resources and how to prioritize DCT implementation over other interventions. This gap also hinders the development of funding models and scalability plans because the return on investment and potential health care savings (e.g., reduced drug toxicity and related death, hospitalizations) remain unclear.

One cost description study⁵² was identified that may have limited generalizability to the setting in Canada. Given that this was a pilot program that reflected on establishing and operating a DCS in the US over 9 months (of which 3 months were operational for drug checking), the long-term sustainability within harm reduction and public health programs remains unknown. The DCT manufacturers we surveyed provided some of the information regarding the costs (e.g., upfront and ongoing costs, staff training requirements, and manufacturer support) and capabilities (e.g., portability, time to result) of various DCTs. However, other cost implications, such as the cost associated with implementing and maintaining a drug-checking program within comprehensive substance use services and support, in particular in the context in Canada, remain largely unanswered. While the scope of our costing exercise mostly focused on device-related costs, it is important to note that there may be other implementation costs that have not been considered (e.g., facility costs, labour costs). While we acknowledge that the costs of implementing and scaling a lab-based drug-checking program for regional health authorities and their community partners may be prohibitive (i.e., costs are so high that they prevent or severely limit something from happening), emerging DCTs are specifically intended to help overcome some of these financial and resource barriers (refer to the Emerging Technologies section).

Generalizability

The generalizability of findings from drug-checking studies is often limited due to variations in sample types, testing conditions, limited collection radius and time frame, and the nonrandom nature of the collection procedure.^37,44,48 Some of the evaluations are conducted in controlled laboratory settings or focus on specific substances, which may not reflect the complexity and variability of real-world drug markets.⁴⁸ Regional differences in drug supply, prevalence of contaminants, and service use and specific needs of PWUD further challenge the applicability of study results to other settings.^36,37 Additionally, the performance of technologies, such as FTIR spectroscopy, FTS, or portable GC-MS, can vary depending on drug form, concentration, and matrix, making it difficult to generalize findings to broader populations or contexts.^41,42

Out of 20 included studies, 10 studies,^{30,38-40,44-46,48,51,52} including 1 cost description,⁵² were conducted in countries other than Canada. The generalizability of the findings to settings in Canada may be limited due to differences in health care systems (e.g., point of care, harm reduction infrastructure), regulatory frameworks, and characteristics of the unregulated drug supply compared to the settings where the studies were conducted.

It should also be acknowledged that the field of drug checking is relatively new. While technologies such as FTIR and GC-MS have long been used in laboratory settings with well-established evidence to support their application, there is limited evidence regarding their use in community settings with the unregulated drug supply. Further research is needed to strengthen the evidence base through efforts such as field testing of unregulated drug supply in real-world contexts, such as DCSs or music festivals.²⁹

Heterogeneity of the Evidence

The evidence on DCTs is highly heterogeneous, reflecting substantial variability in the substances tested and the performance metrics reported.^37,48 Different DCTs, such as FTIR spectroscopy, FTS, and colorimetric tests, are evaluated under diverse conditions, with varying confirmatory lab testing like GC-MS or LC-MS-MS. Sensitivity, specificity, and FP and FN rates are inconsistently measured and reported, making direct comparisons between studies challenging.³⁹ This heterogeneity limits the ability to synthesize results into clear conclusions about the accuracy of DCTs across settings.

Conclusions and Implications for Decision- or Policy-Making

This review evaluated the literature regarding accuracy, LOD, repeatability, and reproducibility, cost-effectiveness, and cost associated with DCTs for detecting compositions of unregulated substance samples. We identified 18 test accuracy studies^23,30,36-51 on accuracy, 7 studies^{30,36,42,45,46,51,61} on LOD, and 1 study⁴⁶ on reproducibility of any DCTs for detecting compositions of unregulated substance samples. We did not find any study on the repeatability of DCTs that meet the review’s inclusion criteria. We found 1 cost description study⁵² reporting on the costs of DCTs. We found no relevant evidence regarding the cost-effectiveness of any DCTs for detecting compositions of unregulated substance samples. The context of DCTs also raises several ethics and equity considerations that are critical to the design and implementation of DCTs and DCSs.

Summary of Evidence

This review provides a synthesis of the current evidence on the accuracy, LOD, and reproducibility and cost associated with DCTs, highlighting both their strengths and limitations. The findings reinforce that although certain technologies, such as FTS and SERS, demonstrate high sensitivity for detecting target substances like fentanyl and benzodiazepine analogues, other commonly used methods like FTIR spectroscopy and colorimetric tests show more variable accuracy, especially in detecting low-concentration compounds, complex mixtures, and emerging synthetic analogues. However, a challenge with ITS is their narrow focus on 1 target substance and their potential issues with cross-reactivity (i.e., they provide positive results for most fentanyl analogues and do not provide in-class drug discrimination and quantitative data). In practice, broader-spectrum technologies can help identify overall trends and contaminants in the drug supply, while targeted tools such as ITS can be deployed in settings closer to PWUD (e.g., festivals or supervised consumption sites) to rapidly detect priority substances of immediate concern. The review adds value by summarizing recent data on validated and emerging DCTs and indicating that using multimethod approaches (e.g., combining FTIR spectroscopy with ITS) can improve detection accuracy and reliability in harm reduction settings. This finding was consistent with the findings from our survey with the manufacturers of Raman, IR, and near-infrared technologies, which recommended using ITS in combination with their technologies to detect small quantities and specific substances. However, there remains considerable uncertainty due to limited studies on the evaluation of combined DCTs, heterogeneity in reported results, and reliance on small samples, which limit the generalizability and strength of conclusions.

Significant gaps persist in the published evidence, particularly regarding the cost-effectiveness and costs of these technologies in real-world contexts.³⁷ Despite our review’s intention to address economic evaluations, no evidence was found on cost-effectiveness or resource implications. As a first step in understanding the costing implications of these technologies, we obtained cost information from DCT manufacturers. However, in comparing this cost description study³⁷ with the information obtained directly from manufacturers, several differences emerged. While manufacturers provided baseline costing information for devices, consumables, and training, the study incorporated additional operational factors such as confirmatory lab testing and quality assurance protocols, which substantially increased overall costs. Together, these complementary sources highlight both the direct and downstream costs of DCT implementation, underscoring the importance of considering real-world operational requirements beyond device acquisition alone.

This review did not identify any studies that specifically addressed the performance of DCTs in distinct settings, such as Indigenous or rural communities. This highlights a need for more inclusive and context-sensitive studies that reflect the unregulated drug market in various locations. Notably, this review did not capture the perspectives of PWUD because qualitative or participatory research was beyond its scope, representing an important avenue for future inquiry.

The current review aligns with the research and recommendations about the strengths of confirmatory lab testing and the limitations of rapid, point-of-care devices for detecting complex drug mixtures.^37,51 Differences in findings, such as the improved sensitivity of newer portable GC-MS devices, likely reflect technological advancements and the expansion of spectral libraries since earlier studies.⁴⁵ In addition, while evidence for some technologies is well established, other new and emerging DCTs remain in early stages of evaluation. These gaps emphasize the importance of ongoing research and validation efforts.

Considerations for Future Research

Future research on DCTs should focus on improving both methodological rigour and real-world applicability. Comparative studies using standardized protocols and confirmatory lab testing (e.g., LC-MS-MS or GC-MS) are needed to accurately assess the sensitivity, specificity, and LOD of various technologies across different drug types, matrices, and concentrations. Larger multisite studies that capture the diversity of unregulated drug markets would help improve the generalizability of findings. Research should also explore the performance of multimethod approaches, such as pairing FTIR spectroscopy with ITS or confirmatory lab testing, to overcome the limitations of single devices. Additionally, studies should evaluate the cost-effectiveness, scalability, and operational feasibility of DCTs in community settings. Finally, research must address the detection of emerging synthetic analogues and contaminants, which requires the continuous updating of spectral libraries, reagent tests, and device capabilities to keep pace with rapidly evolving drug supplies.

Emerging Technologies

This report included peer-reviewed studies plus costing information provided by manufacturers who responded to our query. A key limitation of the existing literature is its heavy emphasis on ITS and FTIR testing, reflecting the early adoption and availability of these methods rather than the full spectrum of new and emerging technologies. Although these approaches are well studied, they present analytical limitations, particularly for multicomponent or untargeted analyses. MS-based systems, which offer higher sensitivity and broader detection capabilities, are comparatively less represented in the current literature. Overall, the research landscape has been shaped by early adopted DCTs, with more recent advancements, including Canadian contributions, yet to gain comparable visibility or integration into mainstream DCT studies. Our review of grey literature and other resources highlights several of the emerging DCTs that have been developed and are being piloted in Canada and internationally and that have not yet been fully validated or widely published in the scientific literature.²⁹ These innovations often aim to address limitations of existing methods by improving portability, speed, sensitivity, or ease of use. Two of these new technologies, SERS⁴¹ and portable GC-MS,⁴² were evaluated in the included studies of this review. Other notable emerging technologies include ion mobility spectrometry,⁶² Scatr’s Series One,⁶³ Spectra Plasmonics Amplifi IDTM,²⁹ and Waters’ Radian ASAP,⁶⁴ portable robotic high performance liquid chromatography,²⁹ portable near-infrared technology,⁶⁵ miniature MS,^66,67 portable PS-MS,^68,69 DoseCheck technologies (voltammetry and electrochemical analysis),²⁹ and mobile-IR.²⁹

All these technologies share the common goal of enabling rapid, onsite chemical analysis of substances outside traditional laboratory settings.²⁹ They are designed to provide timely and actionable information about the composition of substances, often at trace levels and with minimal sample preparation. Whether based on spectroscopy (e.g., near-infrared, Raman, and SERS) or mass spectrometry (e.g., ion mobility spectrometry, miniature MS, PS-MS, and Radian ASAP), each technology relies on comparison with spectral or mass reference libraries to identify unknown or mixed substances. Their portability and increasingly user-friendly interfaces suggest potential suitability for use in community or clinical settings. A key aim for these emerging technologies is to balance trade-offs between sensitivity, specificity, cost, and ease of use. Importantly, these tools are increasingly being piloted in harm reduction settings in Canada and internationally, where they can detect contaminants in unregulated drug supplies and provide both individual-level feedback and population-level early warning data. There is little evidence of formal evaluation for most of these new instruments. Most are in the research, development, and evaluation stages, and some are in the field-testing stage.²⁹

Given the rapidly evolving landscape of DCTs, ongoing research, validation, and pilot implementation studies are warranted to determine their accuracy, effectiveness, and practical suitability for various harm reduction settings. While laboratory testing is essential, field testing is particularly important, as it evaluates technologies in real-world contexts, such as in community-based drug-checking programs or at music festivals using samples from the unregulated drug supply.²⁹ For example, Toronto’s DCS and Ontario's Drug Checking Community leads performance assessments of various DCTs available for sale in Canada using unregulated substance samples collected throughout Ontario and gold standard mass spectrometry instrumentation.⁷⁰ The British Columbia Centre on Substance Use collaborates with service delivery partners, laboratory services such as Health Canada’s Drug Analysis Service),⁷¹ and manufacturers to conduct field-testing pilots and verify the performance of new and emerging DCTs when used in real-world settings in Ontario⁷⁰ and British Columbia.²⁹ In recent years, drug-checking sites across Canada have been collaborating with their partners and Health Canada to field test a number of emerging technologies and develop recommendations for their use in their respective jurisdictions.⁷¹ Although many of the field-test results have not yet been published in academic outlets, they draw on community-generated knowledge and structured pilot studies, highlighting the importance of both conducting field-based evaluations and sharing findings to inform broader adoption of various DCTs and best practices.²⁹

Considerations for Substance Use Services and Support and Policy-Making

When implementing DCTs in clinical or community settings, decision-makers must balance analytical accuracy with operational feasibility. No single technology can detect the full range of substances in today’s rapidly evolving drug market, particularly low-concentration compounds or emerging synthetic analogues, such as carfentanil or novel fentanyl derivatives.^41,51 Consequently, multimethod approaches, combining point-of-care devices like FTIR spectroscopy and ITS (e.g., BTS) with confirmatory lab testing (e.g., GC-MS or LC-MS-MS), are often required to ensure accuracy of detection.^41,51 This layered strategy allows front-line services to provide quick, actionable information while still having access to confirmatory lab testing for complex or ambiguous samples. However, 2 included studies^36,43 have evaluated this approach.

Cost-effectiveness and sustainability are also central considerations. High-performance technologies, like portable GC-MS, can provide detailed and accurate results but come with high costs, significant maintenance requirements, and the need for specialized technical expertise.^42,45 In contrast, low-cost tools, such as ITS or colorimetric reagents, may be more feasible for widespread implementation but have narrower detection capabilities and may miss critical contaminants or analogues. Policy-makers must assess resource allocation, including training needs, infrastructure requirements, and supply chains, to ensure that DCSs remain reliable and scalable over time.

Drug-checking data and services are an important component of the public health response to the toxic drug crisis.^16,27 Much of what we now know about the variability in the drug supply comes from the data made available through these services.^27,71 Investment in research, cross-jurisdictional data sharing, and partnerships between harm reduction services and laboratories can help ensure that DCTs remain responsive and evidence-informed. However, existing legislative requirements, such as permits and exemptions for handling, transporting, storing, and testing controlled substances, can vary by jurisdiction and setting, and they may influence the feasibility of such studies. Acknowledging these regulatory contexts is important while planning and advancing research and implementation efforts. Policy-makers may also consider how data from DCSs can be integrated into public health surveillance systems to monitor trends, enhance community awareness of toxic and highly potent drugs, and inform proactive interventions.⁴⁴

Implications for First Nations, Inuit, and Métis Peoples

In Canada, First Nations, Inuit, and Métis Peoples experience higher rates of unregulated substance use and experience a disproportionate burden of the harms associated with unregulated substance use as compared to non-Indigenous people due to documented barriers, including racism, discrimination, and marginalization within health care systems in Canada.⁷² Drug checking as a harm reduction approach remains underexplored from Indigenous, decolonial, and culturally centred perspectives.⁷³ This includes considerations of data sovereignty and the value of such data for First Nations, Inuit, and Métis Peoples to monitor and respond to the unregulated drug supply. One recent publication offers guiding principles and foundational insights to (re)conceptualize drug checking in alignment with First Nations, Inuit, and Métis cultural values, ancestral knowledge, and community-based knowledge and worldviews that emphasize wellness, healing, and (re)connection.⁷³ Additionally, Tla’amin First Nation’s community-based, Indigenous-led harm reduction program, ʔaǰɩmɛt, incorporates the Tla’amin First Nation’s language, culture, and ways of being into all aspects of its programming related to the use of unregulated drugs. including its DCSs.²² As of 2023, it was the only community-owned and Indigenous-led harm reduction intervention in Canada and could provide a basis for other Indigenous communities to draw upon when developing similar interventions in their communities.

Ethics and Equity Considerations

Reducing Harms and Informed Consent

The overarching ethical aim of harm reduction interventions, including DCSs, is to reduce the negative consequences of certain activities.⁷⁴ In connection with the use of DCTs, this means DCTs should provide highly accurate findings — that is, findings should provide comprehensive results in terms of detecting contaminants and have low FN rates. Research has shown that the accuracy of DCT test results is a concern for PWUD who access DCSs.³¹ Our findings identified that the DCTs reviewed for this report had variable accuracy and LOD, which may have implications for the potential of these DCTs to achieve the overall goal of harm reduction.

The findings in this report also have implications for informed consent. To respect human autonomy, people have the right to be informed of the potential risks and benefits of interventions.⁷⁵

The communication of results must be accurate, transparent, and tailored to ensure PWUD can make informed decisions about their substance use.⁴³ The risk of FN or incomplete results (e.g., when certain contaminants or analogues are undetectable by a given technology) must be clearly communicated to prevent harm or a false sense of safety.^39,43

Access to DCSs

In the context of Canada, recent research has highlighted that PWUD consider lack of access to be a major barrier to using DCTs and DCSs.⁷⁶ Access to DCSs is often uneven, with rural, remote, and underserved communities, which are disproportionately impacted by the toxic drug supply, facing significant barriers. These include limited availability of equipment, lack of trained personnel, and logistical challenges in transporting samples for confirmatory lab testing. Collectively, this differential access to DCSs likely contributes to health inequities, given that some populations of PWUD will bear a disproportionate risk of adverse health outcomes because of disparities in access to DCSs. This highlights the need for technologies that can support broader access, particularly across geographical regions. Ensuring equitable access requires investment in mobile or decentralized testing services as well as culturally safe and stigma-free environments that encourage PWUD seek these services without fear of criminalization or discrimination.

Ethical implementation also requires considering the affordability and sustainability of DCTs so that communities are not left without essential services due to resource constraints. This includes ensuring that PWUD living in rural and/or remote communities have the same access to advanced DCTs with high accuracy as urban or better-funded regions. Collaboration with PWUD, community-based organizations, and First Nations, Inuit, and Métis communities is essential for designing services that are equitable, respectful of lived experiences, and aligned with culturally informed and harm reduction principles.

Privacy

Privacy refers to the right of individuals to be free from intrusion or interference by others.⁷⁷ Individuals have privacy interests “in relation to their bodies, personal information, expressed thoughts and opinions, personal communications with others, and the spaces they occupy.”⁷⁷ In the context of Canada, ensuring privacy requires abiding by the relevant provincial or territorial data privacy laws and regulations (e.g., Ontario’s Personal Health Information Protection Act, Nova Scotia’s and British Columbia’s Freedom of Information and Protection of Privacy Acts). Additional privacy considerations are articulated in First Nations, Inuit, and Métis data sovereignty principles — including the First Nations principles of OCAP (ownership, control, access, and possession), Manitoba Métis principles of OCAS (ownership, control, access, and stewardship), Inuit Qaujimajatuqangit, and the Engagement, Governance, Access and Protection (EGAP) Framework that guides the collection, storage, and use of race-based data from Black communities.⁷⁸ The use of DCTs and DCSs can give rise to data privacy considerations and physical privacy considerations associated with accessing DCS receiving results.

In some jurisdictions, the use of DCSs may require the collection of personal information to inform individuals of the contents of their drug supply.²² Additionally, this information is often aggregated and used for broader purposes, including identifying trends and informing program and policy decisions.²² The use of cloud servers with subscription-based access may pose additional challenges, particularly when negotiating data agreements with regional health authorities and community partners, and in relation to the potential commercial use of these data. Consent conversations with people accessing DCSs should include information on the steps that will be taken to ensure that users’ personal information is collected, stored, used, transmitted, and disposed of according to the relevant provincial privacy standards and other guiding frameworks and principles when relevant. Additionally, programs must balance individual privacy with public health surveillance, ensuring that data collected from DCSs is anonymized and used in ways that benefit communities rather than perpetuate stigma or punitive responses.

The use of DCSs and DCTs also requires safeguarding people’s physical privacy in connection with accessing DCSs. Recent research has highlighted that PWUD are concerned about being identified as a person who uses unregulated drugs due to accessing DCSs.⁷⁹ These concerns may be mitigated if DCSs are based in locations offering multiple services;⁷⁹ however, it also may be more challenging to protect people’s physical privacy in these contexts.⁸⁰ People may also prefer to receive the results of their tests in an individualized manner rather than having to access their results through a website or other public channel.⁸¹ For example, British Columbia has expanded its client results portal to allow people to access their own results anonymously and privately.^82-84

Legal Considerations

In addition to the importance of adherence to relevant privacy legislation, legal risks have been identified as among the most common barriers to accessing DCSs by PWUD.³¹ In the context of Canada, PWUD have expressed concern about the potential that possessing illegal drugs and/or drug testing equipment could lead to criminal charges.⁸⁵ To reduce the risk of criminalization and support the use of DCSs, some Canadian jurisdictions have applied for and have been granted legal exemption from Canada’s Controlled Drugs and Substance Act to decriminalize the possession of small amounts of unregulated drugs for personal use (e.g., the province of British Columbia), and other jurisdictions have applied for this exemption (e.g., the City of Toronto). Establishing supportive relationships between law enforcement bodies and DCSs has been suggested as an additional means to reduce the fear of criminalization and, therefore, to facilitate the use of DCSs by PWUD.⁸⁵

Conclusion

This review underscores both the strengths and limitations of current DCTs. While point-of-care approaches show variable accuracy, particularly for complex or low-concentration substances, confirmatory laboratory-based DCTs demonstrate high sensitivity. In Canada, provinces like Ontario and British Columbia have piloted and expanded community drug-checking programs that integrate several of these technologies,^{36,37,41,43,47,49,50,53} often combining ITS and portable IR, FTIR, and Raman spectrometers at the point of care while sending complex or inconclusive samples to centralized laboratories equipped with highly sensitive and specific devices. Evidence gaps — especially regarding multimethod approaches, cost-effectiveness, diverse settings, and user perspectives — limit current conclusions. Continued research and validation are needed to ensure these technologies are accurate, feasible, and contextually relevant.

References

1.Deeks J, Bossuyt P, Leeflang M, Takwoingi Y. Cochrane Handbook for Systematic Reviews of Diagnostic Test Accuracy. Version 2.0 Cochrane; 2023. Accessed April 22, 2025. https://training.cochrane.org/handbook-diagnostic-test-accuracy/current

2.Kerr T, Tupper K. Drug Checking as a Harm Reduction Intervention. 2017. Accessed August 4, 2025. https://www.bccsu.ca/wp-content/uploads/2017/12/Drug-Checking-Evidence-Review-Report.pdf

3.Schroers A. Drug checking: Monitoring the contents of new synthetic drugs. Journal of Drug Issues. 2002;32(2):635-646. doi:10.1177/002204260203200219

4.Government of Canada. About Substance Use. Health Canada. Accessed April 22, 2025, https://www.canada.ca/en/health-canada/services/substance-use/about-substance-use.html

5.Government of Canada. Opioid- and Stimulant-Related Harms in Canada. Public Health Agency of Canada. Accessed April 22, 2025, https://health-infobase.canada.ca/substance-related-harms/opioids-stimulants/

6.Armbruster DA, Pry T. Limit of blank, limit of detection and limit of quantitation. Clin Biochem Rev. Aug 2008;29 Suppl 1(Suppl 1):S49-52.

7.Public Health Agency of Canada. Communicating About Substance Use in Compassionate, Safe and Nonstigmatizing Ways: A Resource for Canadian Health Professional Organizations and Their Membership. Public Health Agency of Canada; 2020. Accessed May 15, 2025. https://www.canada.ca/content/dam/phac-aspc/documents/services/publications/healthy-living/communicating-about-substance-use-compassionate-safe-non-stigmatizing-ways-2019/guilding-rinciples-eng.pdf

8.Chesher D. Evaluating assay precision. Clin Biochem Rev. Aug 2008;29 Suppl 1(Suppl 1):S23-6.

9.Canada's Drug Agency. Equity Considerations: Resource for MDCI Researchers. Ottawa, ON: Canada's Drug Agency; 2025.

10.Canadian Centre on Substance Use and Addiction. Substance Use and Stigma: Why Language Matters. 2021. Accessed August 4, 2025. https://www.ccsa.ca/substance-use-and-stigma-why-language-matters

11.Ledlie S, Juurlink DN, Tadrous M, Mamdani M, Paterson JM, Gomes T. Opioid-related deaths between 2019 and 2021 across 9 Canadian provinces and territories. CMAJ. Apr 14 2024;196(14):E469-E476. doi:10.1503/cmaj.231339 PubMed

12.Gomes T, Kitchen SA, Tailor L, et al. Trends in hospitalizations for serious infections among people with opioid use disorder in Ontario, Canada. J Addict Med. Jul-Aug 01 2022;16(4):433-439. doi:10.1097/ADM.0000000000000928

13.Taha S. Best Practices Across the Continuum of Care for Treatment of Opioid Use Disorder. Canadian Centre on Substance Use and Addiction.; 2018. Accessed April 30, 2025. https://www.ccsa.ca/sites/default/files/2019-04/CCSA-Best-Practices-Treatment-Opioid-Use-Disorder-2018-en.pdf

14.Daly P, de Villa E, Drouin M, Lysyshyn M, Tockman J. Harm reduction, treatment, and prevention-essential pillars to Canada's response to the drug poisoning crisis. Can J Public Health. Mar 31 2025;doi:10.17269/s41997-025-01029-1 PubMed

15.Leece P. Evidence Brief: Drug Checking Services as a Harm Reduction Intervention. 2017. Accessed August 4, 2025. https://www.substanceuse.ca/evidence-brief-drug-checking-services-harm-reduction-intervention

16.Maghsoudi N, Tanguay J, Scarfone K, et al. Drug checking services for people who use drugs: a systematic review. Addiction. Mar 2022;117(3):532-544. doi:10.1111/add.15734 PubMed

17.McDonald K, Thompson H, Werb D. 10 Key Findings Related to the Impact of Toronto’s Drug Checking Service. Centre on Drug Policy Evaluation; 2024. Accessed June 06, 2025. https://drugchecking.community/resource/key-findings-related-to-impact/

18.Harper L, Powell J, Pijl EM. An overview of forensic drug testing methods and their suitability for harm reduction point-of-care services. Harm Reduct J. Jul 31 2017;14(1):52. doi:10.1186/s12954-017-0179-5 PubMed

19.Maghsoudi N, McDonald K, Stefan C, et al. Evaluating networked drug checking services in Toronto, Ontario: study protocol and rationale. Harm Reduct J. Jan 28 2020;17(1):9. doi:10.1186/s12954-019-0336-0 PubMed

20.Long V, Arredondo J, Ti L, et al. Factors associated with drug checking service utilization among people who use drugs in a Canadian setting. Harm Reduct J. Dec 14 2020;17(1):100. doi:10.1186/s12954-020-00454-4 PubMed

21.Moniruzzaman A, Rezansoff S, Somers J. The Effects of Drug Checking Services on the Reduction Of Harm Among Marginalized People Who Use Drugs: A Rapid Review. Simon Fraser University; 2021. Accessed April 30, 2025. https://www.sfu.ca/content/dam/sfu/carmha/resources/bcdcs/SFU_MoH_RapidReview_DCS.pdf

22.Giwa A, Payer D. An Environmental Scan of Drug-Checking Services in Canada: Final Report. Canadian Centre on Substance Use and Addiction; 2024. Accessed April 21, 2025. https://www.ccsa.ca/en/environmental-scan-drug-checking-services-canada-final-report

23.Ti L, Tobias S, Lysyshyn M, et al. Detecting fentanyl using point-of-care drug checking technologies: A validation study. Research Support, Non-U.S. Gov't Validation Study. Drug Alcohol Depend. 07 01 2020;212:108006. doi:10.1016/j.drugalcdep.2020.108006

24.Wallace B, van Roode T, Pagan F, Hore D, Pauly B. The potential impacts of community drug checking within the overdose crisis: qualitative study exploring the perspective of prospective service users. Research Support, Non-U.S. Gov't. BMC Public Health. 06 16 2021;21(1):1156. doi:10.1186/s12889-021-11243-4

25.Maghsoudi N, McDonald K, Stefan C, et al. Evaluating networked drug checking services in Toronto, Ontario: study protocol and rationale. Harm Reduct J. Jan 28 2020;17(1):9. doi:10.1186/s12954-019-0336-0 PubMed

26.Government of Canada. Subsection 56(1) Class Exemption in Relation to Urgent Public Health Need Sites in the Provinces and Territories. Health Canada. Accessed June 06, 2025, https://www.canada.ca/en/health-canada/services/health-concerns/controlled-substances-precursor-chemicals/policy-regulations/policy-documents/subsection-56-1-class-exemption-urgent-public-health-needs-sites-provinces-territories.html

27.Hutchison A, Wallace B, Hore D. Community Drug Checking Services in British Columbia/Yukon: An Environmental Scan. Canadian Institute for Substance Use Research.; 2023. Accessed August 19, 2025. https://substance.uvic.ca/files/resources/Drug-Checking-in-BC-Env-Scan.pdf

28.Parker J, Jackson L, Dykeman M, Gahagan J, Karabanow J. Access to harm reduction services in Atlantic Canada: implications for non-urban residents who inject drugs. Health Place. Mar 2012;18(2):152-62. doi:10.1016/j.healthplace.2011.08.016 PubMed

29.British Columbia Centre on Substance Use. New Drug Checking Instruments in Canada: A Summary of Drug Checking Technology Developments. 2024. Accessed April 21, 2025. https://drugcheckingbc.ca/wp-content/uploads/sites/4/2024/01/BCCSU_New_drug_checking_technologies_2024.pdf

30.Green TC, Park JN, Gilbert M, et al. An assessment of the limits of detection, sensitivity and specificity of three devices for public health-based drug checking of fentanyl in street-acquired samples. Int J Drug Policy. Mar 2020;77:102661. doi:10.1016/j.drugpo.2020.102661 PubMed

31.Grace Rose C, Kulbokas V, Carkovic E, Lee TA, Pickard AS. Contextual factors affecting the implementation of drug checking for harm reduction: a scoping literature review from a North American perspective. Harm Reduct J. Sep 4 2023;20(1):124. doi:10.1186/s12954-023-00856-0 PubMed

32.Palamar JJ, Salomone A, Barratt MJ. Drug checking to detect fentanyl and new psychoactive substances. Curr Opin Psychiatry. Jul 2020;33(4):301-305. doi:10.1097/YCO.0000000000000607 PubMed

33.Garritty C, Hamel C, Trivella M, et al. Updated recommendations for the Cochrane rapid review methods guidance for rapid reviews of effectiveness. BMJ. Feb 6 2024;384:e076335. doi:10.1136/bmj-2023-076335 PubMed

34.Whiting PF, Rutjes AW, Westwood ME, et al. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med. Oct 18 2011;155(8):529-36. doi:10.7326/0003-4819-155-8-201110180-00009 PubMed

35.Drummond MF, Jefferson TO. Guidelines for authors and peer reviewers of economic submissions to the BMJ. The BMJ Economic Evaluation Working Party. BMJ. Aug 3 1996;313(7052):275-83. doi:10.1136/bmj.313.7052.275 PubMed

36.Crepeault H, Tobias S, Angelucci J, et al. Detection of benzodiazepines in the unregulated drug supply using point of care and confirmatory drug checking technologies: A validation study. Validation Study. Drug Alcohol Depend. Jun 01 2025;271:112631. doi:10.1016/j.drugalcdep.2025.112631 PubMed

37.Crepeault H, Socias ME, Tobias S, et al. Examining fentanyl and its analogues in the unregulated drug supply of British Columbia, Canada using drug checking technologies. Research Support, Non-U.S. Gov't Research Support, N.I.H., Extramural. Drug Alcohol Rev. 03 2023;42(3):538-543. doi:10.1111/dar.13580

38.Estrada Y, Sauer J, Dominguez L, et al. The prevalence of fentanyl in New York City's unregulated drug supply as measured through drug checking offered at syringe service programs. Drug Alcohol Depend. Mar 01 2025;268:112578. doi:10.1016/j.drugalcdep.2025.112578 PubMed

39.Fregonese M, Albino A, Covino C, et al. Drug checking as strategy for harm reduction in recreational contests: Evaluation of two different drug analysis methodologies. Front Psychiatr. 2021;12:596895. doi:10.3389/fpsyt.2021.596895 PubMed

40.Goncalves R, Titier K, Latour V, et al. Suitability of infrared spectroscopy for drug checking in harm reduction centres. Int J Drug Policy. 02 2021;88:103037. doi:10.1016/j.drugpo.2020.103037

41.Gozdzialski L, Rowley A, Borden SA, et al. Rapid and accurate etizolam detection using surface-enhanced Raman spectroscopy for community drug checking. Research Support, Non-U.S. Gov't. Int J Drug Policy. 04 2022;102:103611. doi:10.1016/j.drugpo.2022.103611

42.Gozdzialski L, Aasen J, Larnder A, et al. Portable gas chromatography-mass spectrometry in drug checking: Detection of carfentanil and etizolam in expected opioid samples. Research Support, Non-U.S. Gov't. Int J Drug Policy. 11 2021;97:103409. doi:10.1016/j.drugpo.2021.103409

43.Laing MK, Ti L, Marmel A, et al. An outbreak of novel psychoactive substance benzodiazepines in the unregulated drug supply: Preliminary results from a community drug-checking program using point-of-care and confirmatory methods. Research Support, Non-U.S. Gov't. Int J Drug Policy. 07 2021;93:103169. doi:10.1016/j.drugpo.2021.103169

44.Park JN, Serafinski R, Ujeneza M, et al. Xylazine awareness, desire, use and exposure: Preliminary findings from the Rhode Island community-based drug checking cohort study. Drug Alcohol Depend Rep. Jun 2024;11:100247. doi:10.1016/j.dadr.2024.100247 PubMed

45.Park JN, Sherman SG, Sigmund V, Breaud A, Martin K, Clarke WA. Validation of a lateral flow chromatographic immunoassay for the detection of fentanyl in drug samples. Research Support, N.I.H., Extramural. Drug Alcohol Depend. 11 01 2022;240:109610. doi:10.1016/j.drugalcdep.2022.109610

46.Sisco E, Nestadt DF, Bloom MB, et al. Understanding sensitivity and cross-reactivity of xylazine lateral flow immunoassay test strips for drug checking applications. Drug Test Anal. Sep 2024;16(9):942-947. doi:10.1002/dta.3612 PubMed

47.Ti L, Tobias S, Maghsoudi N, et al. Detection of synthetic cannabinoid adulteration in the unregulated drug supply in three Canadian settings. Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov't. Drug Alcohol Rev. 05 2021;40(4):580-585. doi:10.1111/dar.13237

48.Thompson E, Tardif J, Ujeneza M, et al. Pilot findings on the real-world performance of xylazine test strips for drug residue testing and the importance of secondary testing methods. Drug Alcohol Depend Rep. Jun 2024;11:100241. doi:10.1016/j.dadr.2024.100241 PubMed

49.Tobias S, Shapiro AM, Grant CJ, Patel P, Lysyshyn M, Ti L. Drug checking identifies counterfeit alprazolam tablets. Research Support, Non-U.S. Gov't. Drug Alcohol Depend. 01 01 2021;218:108300. doi:10.1016/j.drugalcdep.2020.108300

50.Tobias S, Shapiro A, Wu H, Ti L. Xylazine Identified in the Unregulated Drug Supply in British Columbia, Canada. Can J Addict. 01 Sep 2020;11(3):28-32. doi:10.1097/cxa.0000000000000089

51.Whitehead HD, Hayes KL, Swartz JA, et al. Validated method for the analysis of 22 illicit drugs and their metabolites via liquid chromatography tandem mass spectrometry (LC-MS/MS) in illicit drug samples collected in Chicago, IL. Forensic Chem. May 2023;33doi:10.1016/j.forc.2023.100475

52.Cepeda JA, Thompson E, Ujeneza M, et al. Costing analysis of a point-of-care drug checking program in Rhode Island. Drug Alcohol Depend. Dec 01 2023;253:111028. doi:10.1016/j.drugalcdep.2023.111028 PubMed

53.McCrae K, Tobias S, Grant C, et al. Assessing the limit of detection of Fourier-transform infrared spectroscopy and immunoassay strips for fentanyl in a real-world setting. Research Support, Non-U.S. Gov't. Drug Alcohol Rev. 01 2020;39(1):98-102. doi:10.1111/dar.13004

54.Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. Mar 29 2021;372:n71. doi:10.1136/bmj.n71 PubMed

55.Fregonese M, Albino A, Covino C, et al. Drug Checking as Strategy for Harm Reduction in Recreational Contests: Evaluation of Two Different Drug Analysis Methodologies. Front Psychiatr. 2021;12:596895. doi:10.3389/fpsyt.2021.596895 PubMed

56.Goncalves R, Titier K, Latour V, et al. Suitability of infrared spectroscopy for drug checking in harm reduction centres. International Journal of Drug Policy. 02 2021;88:103037. doi:10.1016/j.drugpo.2020.103037

57.Gozdzialski L, Rowley A, Borden SA, et al. Rapid and accurate etizolam detection using surface-enhanced Raman spectroscopy for community drug checking. Research Support, Non-U.S. Gov't. International Journal of Drug Policy. 04 2022;102:103611. doi:10.1016/j.drugpo.2022.103611

58.Monti MC, Bauer M, Koch K, Scheurer E, Schlotterbeck G. Evaluation of ATR-FTIR, HPLC-DAD, GC-MS, and GC-IR for the Analysis of 145 Street Drug Samples From Drug Checking Services. Drug Test Anal. Dec 26 2024;26:26. doi:10.1002/dta.3843 PubMed

59.Park JN, Serafinski R, Ujeneza M, et al. Xylazine Awareness, Desire, Use and Exposure: Preliminary Findings from the Rhode Island CUTS Drug Checking Cohort. Preprint. medRxiv. 2024;02doi:10.1101/2024.02.29.24303571

60.Tobias S, Shapiro AM, Wu H, Ti L. Xylazine Identified in the Unregulated Drug Supply in British Columbia, Canada. Canadian Journal of Addiction. 01 Sep 2020;11(3):28-32. doi:10.1097/cxa.0000000000000089

61.McCrae K, Hayashi K, Bardwell G, et al. The effect of injecting alone on the use of drug checking services among people who inject drugs. Int J Drug Policy. Apr 15 2020;79:102756. doi:10.1016/j.drugpo.2020.102756 PubMed

62.Gabelica V, Marklund E. Fundamentals of ion mobility spectrometry. Curr Opin Chem Biol. Feb 2018;42:51-59. doi:10.1016/j.cbpa.2017.10.022 PubMed

63.Scatr One. Scatr Series One Basic Limit of Detection Evaluation 2025. Accessed September 24, 2025. https://scatr.ca/documents/2024_09_05_lod_technical_note.pdf

64.Wood M. Corporation W, ed. RADIAN ASAP with LiveID – Fast, Specific, and Easy Drug Screening. 2021. Accessed September 24, 2025. https://www.waters.com/content/dam/waters/en/app-notes/2021/720007125/720007125-en.pdf

65.Fursman H, Morelato M, Chadwick S, et al. Development and evaluation of portable NIR technology for the identification and quantification of Australian illicit drugs. Forensic Sci Int. Sep 2024;362:112179. doi:10.1016/j.forsciint.2024.112179 PubMed

66.Fan X, Jiao B, Zhou X, Zhang W, Ouyang Z. Miniaturization of mass spectrometry systems: An overview of recent advancements and a perspective on future directions. Anal Chem. May 6 2025;97(17):9111-9125. doi:10.1021/acs.analchem.5c01223 PubMed

67.Laxton J-C, Monaghan J, Wallace B, Hore D, Wang N, Gill CG. Evaluation and improvement of a miniature mass spectrometry system for quantitative harm reduction drug checking. Int J Mass Spectrom. 2023/02/01/ 2023;484:116976. doi:https://doi.org/10.1016/j.ijms.2022.116976

68.Borden SA, Saatchi A, Vandergrift GW, Palaty J, Lysyshyn M, Gill CG. A new quantitative drug checking technology for harm reduction: Pilot study in Vancouver, Canada using paper spray mass spectrometry. Research Support, Non-U.S. Gov't. Drug Alcohol Rev. 02 2022;41(2):410-418. doi:10.1111/dar.13370

69.Borden SA, Saatchi A, Krogh ET, Gill CG. Rapid and quantitative determination of fentanyls and pharmaceuticals from powdered drug samples by paper spray mass spectrometry. Anal Sci Adv. Aug 2020;1(2):97-108. doi:10.1002/ansa.202000083 PubMed

70.Toronto’s Drug Checking Service. What is Toronto's Drug Checking Service? Accessed October 29, 2025, https://drugchecking.community/about

71.Government of Canada. Drug Analysis Service. Accessed September 24, 2025, https://www.canada.ca/en/health-canada/services/health-concerns/controlled-substances-precursor-chemicals/drug-analysis-service.html

72.Statistics Canada. Health Care Access and Experiences Among Indigenous People. 2024. Accessed August 8, 2025. https://www150.statcan.gc.ca/n1/daily-quotidien/241104/dq241104a-eng.htm

73.Littlechild S, Wallace B, Hore D. Visioning Towards a Decolonized, Indigenous-Centered Service Model for Drug Checking: An AutoEthnographic Exploration. Canadian Institute for Substance Use Research; 2025. Accessed July 25, 2025. https://www.uvic.ca/research/centres/cisur/assets/docs/littlechild-visioning-towards-a-decolonized-model.pdf

74.Hawk M, Coulter RWS, Egan JE, et al. Harm reduction principles for healthcare settings. Harm Reduct J. Oct 24 2017;14(1):70. doi:10.1186/s12954-017-0196-4 PubMed

75.Eyal N. Informed Consent. In The Stanford Encyclopedia of Philosophy. Edward N. Zalta, editor. 2025. Accessed August 19, 2025. https://plato.stanford.edu/archives/spr2025/entries/informed-consent/

76.Tobias S, Ferguson M, Palis H, et al. Motivators of and barriers to drug checking engagement in British Columbia, Canada: Findings from a cross-sectional study. Int J Drug Policy. Jan 2024;123:104290. doi:10.1016/j.drugpo.2023.104290 PubMed

77.Canadian Institutes of Health Research NSaERCoC, and Social Sciences and Humanities Research Council of Canada. Tri- Council Policy Statement: Ethical Conduct for Research Involving Humans. 2022. Accessed August 19, 2025. https://ethics.gc.ca/eng/documents/tcps2-2022-en.pdf

78.Black Health Equity Working Group. Engagement, Governance, Access, and Protection (EGAP): A Data Governance Framework for Health Data Collected from Black Communities. Black Health Equity Working Group; 2021. Accessed August 19, 2025. https://blackhealthequity.ca/wp-content/uploads/2021/03/Report_EGAP_framework.pdf

79.Measham F. City checking: Piloting the UK's first community-based drug safety testing (drug checking) service in 2 city centres. Research Support, Non-U.S. Gov't. Br J Clin Pharmacol. 03 2020;86(3):420-428. doi:10.1111/bcp.14231

80.Olding M, Ivsins A, Mayer S, et al. A low-barrier and comprehensive community-based harm-reduction site in Vancouver, Canada. Research Support, N.I.H., Extramural. Am J Public Health. 06 2020;110(6):833-835. doi:10.2105/ajph.2020.305612

81.Barratt MJ, Bruno R, Ezard N, Ritter A. Pill testing or drug checking in Australia: Acceptability of service design features. Research Support, Non-U.S. Gov't. Drug Alcohol Rev. 02 2018;37(2):226-236. doi:10.1111/dar.12576

82.SubstanceUVIC. Substance Drug Checking. Accessed October 29, 2025. https://substance.uvic.ca/

83.Island Health. Drug Checking Services. Accessed October 29, 2025. https://www.islandhealth.ca/our-services/mental-health-substance-use-services/harm-reduction-substance-use/drug-checking-services

84.British Columbia Centre for Substance Use. Drug Checking BC. Accessed October 29, 2025, https://drugcheckingbc.ca/

85.Wallace B, van Roode T, Pagan F, et al. What is needed for implementing drug checking services in the context of the overdose crisis? A qualitative study to explore perspectives of potential service users. Evaluation Study Research Support, Non-U.S. Gov't. Harm Reduction Journal. 05 12 2020;17(1):29. doi:10.1186/s12954-020-00373-4

86.Garritty C, Hamel C, Trivella M, et al. Updated recommendations for the Cochrane rapid review methods guidance for rapid reviews of effectiveness. Bmj. Feb 6 2024;384:e076335. doi:10.1136/bmj-2023-076335 PubMed

87.Khangura S, Konnyu K, Cushman R, Grimshaw J, Moher D. Evidence summaries: the evolution of a rapid review approach. Syst Rev. Feb 10 2012;1:10. doi:10.1186/2046-4053-1-10 PubMed

88.O'Blenis P. One Simple Way To Speed Up Your Screening Process. DistillerSR. Accessed June 5, 2025, https://www.distillersr.com/resources/blog/one-simple-way-to-speed-up-your-screening-process#:~:text=How%20Liberal%20Screening%20works%3A,the%20reference%20automatically%20gets%20excluded

89.Atkins D, Best D, Briss PA, et al. Grading quality of evidence and strength of recommendations. BMJ. Jun 19 2004;328(7454):1490. doi:10.1136/bmj.328.7454.1490 PubMed

90.Guyatt G, Oxman AD, Akl EA, et al. GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables. J Clin Epidemiol. Apr 2011;64(4):383-94. doi:10.1016/j.jclinepi.2010.04.026 PubMed

91.Schünemann HJ, Mustafa RA, Brozek J, et al. GRADE guidelines: 21 part 2. Test accuracy: inconsistency, imprecision, publication bias, and other domains for rating the certainty of evidence and presenting it in evidence profiles and summary of findings tables. J Clin Epidemiol. Jun 2020;122:142-152. doi:10.1016/j.jclinepi.2019.12.021 PubMed

92.Yang B, Mustafa RA, Bossuyt PM, et al. GRADE Guidance: 31. Assessing the certainty across a body of evidence for comparative test accuracy. J Clin Epidemiol. Aug 2021;136:146-156. doi:10.1016/j.jclinepi.2021.04.001 PubMed

93.Kristensen FB, Lampe K, Wild C, Cerbo M, Goettsch W, Becla L. The HTA core model(®)-10 years of developing an international framework to share multidimensional value assessment. Value Health. Feb 2017;20(2):244-250. doi:10.1016/j.jval.2016.12.010 PubMed

94.McInnes MDF, Moher D, Thombs BD, et al. Preferred Reporting Items for a Systematic Review and Meta-analysis of Diagnostic Test Accuracy Studies: The PRISMA-DTA Statement. Jama. Jan 23 2018;319(4):388-396. doi:10.1001/jama.2017.19163 PubMed

Appendix 1: Detailed Methods and Selection of Included Studies

Please note that this appendix has not been copy-edited.

Canada’s Drug Agency (CDA-AMC) received an external request on the topic of DCTs. Based on this request, an informal scoping exercise, including searching and screening evidence from preliminary literature searches, was conducted to identify key priorities for an evidence review on DCTs about their various considerations (e.g., accuracy, LOD, costs). We also developed a questionnaire to gather expert input from the NDCWG on specific areas that informed the content of this report. Based on the results of this scoping exercise and experts’ feedback, and balancing rigour and feasibility considerations, we developed a prespecified project plan and conducted a customized health technology review to address the research questions, specifically a rapid review of primary studies according to Cochrane standards.⁸⁶ Ethics and equity considerations were integrated throughout the report using prespecified analytic categories developed by an internal advisor with ethics expertise.

We made adaptations to the project plan, as necessary, to ensure that we addressed the needs of the request while maintaining methodological rigour and considering the quality of the available evidence. The comparative evidence (i.e., studies including more than 1 index test [i.e., DCT] assessed versus a reference standard [i.e., confirmatory lab testing]) was weak, as all comparisons were made only narratively within the included studies. No comparative accuracy statistics were directly reported. Given the rapid nature of the review, it was not feasible to perform new statistical comparisons of the tests. Therefore, we adopted a noncomparative question which allowed us to include studies that assessed the accuracy of a single index test (i.e., there was not a requirement for a comparator). We did not make formal comparisons within this review and did not appraise the risk of bias in the comparisons made within the included studies. Therefore, readers are cautioned that within-study comparisons may be biased, and between-study comparisons are at high risk of bias due to probable confounding.

To better reflect the realities of drug-checking implementation and practice, we also made several adaptations in how we described the included test accuracy studies on DCTs. Although we selected and appraised the included studies based on the methods used in test accuracy research (e.g., distinguishing index tests and reference standards), we presented the results in language that aligns with the drug-checking practice (e.g., point-of-care and confirmatory lab testing). Drug-checking studies often combine point-of-care and confirmatory lab testing as complementary strategies, rather than as distinct methods, that is, all testing procedures are part of a single approach rather than separate measures. For consistency with this context, we used the term “index test/Intervention” in place of “index test” to reflect that, in the context of drug checking, the testing process itself is not only a diagnostic activity but also constitutes a service that is implemented as part of harm reduction practice. In this sense, “intervention” meant the operational role of drug checking, not an evaluation of its effectiveness. By using the term “index test/Intervention,” we acknowledged that the DCTs reviewed were both ‘tests’ and, at the same time, ‘interventions’ in service delivery. When reporting the results, we described where and how drug checking was conducted (i.e., at point of care, in laboratory facilities, or both), rather than categorizing the methods as index tests versus reference standards. This allowed us to represent testing locations and approaches in a way that is more consistent with how drug checking is implemented and evaluated in the real world. We also replaced the term “patient” with “substance sample” throughout the report and appraisal tools. This adaptation was made to better reflect the context of drug-checking studies, where the unit of analysis is the substance sample rather than the patient.

What Is a Rapid Review?

Rapid reviews are based on accelerated and abbreviated systematic review methods, balancing timeliness with rigour, to allow for timely decision-making. Due to these abbreviated methods, rapid reviews have some limitations. For example, unlike in systematic reviews, for which at least 2 independent reviewers are needed to screen studies to reduce selection bias, we followed a liberal-accelerated approach.¹ In this approach, selection by a single reviewer was required to include a study, and exclusion by 2 reviewers was needed to exclude a study.⁸⁷ Similarly, no calculations were done to homogenize the effect measures. Though we used data from within the studies to calculate sensitivity, specificity, and FP and FN rates, we did not calculate confidence intervals due to time constraints inherent in the rapid review process. As a result, the degree of uncertainty around the estimates is not quantified, and readers should interpret the findings with appropriate caution.

National Drug Checking Working Group Survey

We conducted a survey with 8 targeted questions, distributed to members of the NDCWG, to gather expert input on priority areas for our upcoming review of DCTs. The goal was to better understand which technologies and issues—such as accuracy, LOD, costs, and ethical implications—should be central to the review. We received responses from 14 individuals, including front-line workers, harm reduction/psychosocial workers, lab technicians, researchers, project managers, and individuals with lived or living experience. They represented a range of organizations across Canada.

Respondents emphasized the importance of reviewing both emerging technologies and validated ones, with a particular focus on their ability to detect high-priority substances like opioids (most frequently ranked first), benzodiazepines, and stimulants. Contaminants of concern included fentanyl analogues and nitazenes (ultra-high-potency, synthetic opioids). The most important criteria for evaluating technologies were LOD, ease of use, time to result, sensitivity, specificity, and portability. Most participants preferred a review of comparing DCTs against each other, rather than comparing them to clinical testing methods (e.g., urine sample). They also underscored the need to address key ethical considerations, such as unintended consequences (e.g., FNs), accountability and liability, data privacy, and unequal access to services. Additional feedback emphasized the need to include practical factors, such as upfront and ongoing costs, staff training requirements, manufacturer support, and sample-destroying limitations that may impact service delivery.

We shared and discussed the survey results with the internal CDA-AMC team. The findings helped guide our approach to this customized health technology review.

Literature Search Methods

Published literature was identified by searching the following bibliographic databases: MEDLINE and Embase via Ovid. Ovid searches were run simultaneously as a multifile search, retrieval was not limited by publication date, and was limited to the English language. Duplicates were removed using Ovid deduplication for multifile searches, followed by manual deduplication in EndNote. The search strategy comprised both controlled vocabulary, such as the National Library of Medicine’s MeSH (Medical Subject Headings), and keywords. Search concepts were developed based on the elements of the PICOS (population, intervention, comparison, outcomes, and study) framework and research questions.

To address research questions 1 and 2, the search was completed in 2 sections on May 26, 2025. The first section was focused on the concept of drug checking, with no filters applied to limit the retrieval by study type. To supplement this search, the second section was completed wherein the main concepts were DCTs and drug contaminants, and filters were applied to this search to limit retrieval to randomized controlled trials, controlled clinical trials, or any other type of clinical trial.

To address research question 3, a filter was applied to the same search concepts as just described to limit retrieval to health economic literature on May 30, 2025. The full search strategy is available upon request.

Selection Criteria and Methods

The literature review management software DistillerSR (Evidence Partners, Ottawa, Ontario) was used to facilitate screening and study selection. In the first level of screening, a pilot round for reviewing abstracts was not conducted; rather, 3 reviewers screened the titles and abstracts retrieved from the literature searches: 1 reviewer with content expertise independently screened all the titles and abstracts retrieved from the literature searches; the 2 other reviewers each screened half of the retrieved titles and abstracts. We followed a liberal-accelerated approach,^87,88 in which selection by a single reviewer was required to include a study, and exclusion by both reviewers (i.e., reviewer with content expertise and 1 of the other 2 reviewers) was needed to exclude a study.

Full texts of titles and abstracts that were judged to be potentially relevant were retrieved. Following Cochrane rapid review methods guidance,⁸⁶ 2 reviewers independently assessed 20% of the full texts for inclusion using a standard full-text screening form based on the selection criteria outlined in Table 1. They then met to discuss the discrepancies and calibrate the form. Reviewers then split the remaining full texts and each reviewed half of them. If fewer than 5 potentially relevant full texts were identified to address a specific research question, all retrieved full texts were independently assessed by both reviewers. A list of studies excluded after full-text review was provided, along with the reasons for their exclusion. A third reviewer with content expertise reviewed all excluded full texts.

Eligible studies from all countries were included. The final selection of full-text articles was based on the inclusion criteria presented in Table 1.

Exclusion Criteria

Articles were excluded if they:

• did not meet the selection criteria outlined in Table 1,

• were duplicate publications,

• were of any design published only as abstracts, conference proceedings, presentations, thesis documents, or preprints,

• were health technology assessments, systematic reviews, meta-analyses, case studies, and case reports,

• were published in languages other than English (for feasibility reasons),

• compared different manufacturers of the same DCT,

• included drug checking that was not completed by the research team or trained staff (e.g., PWUD, take-home tests).

Inquiring Information From DCT Manufacturers

Based on the initial scoping exercise, we did not anticipate comprehensive information about the costs associated with DCTs from the published literature. In addition to the targeted literature search for costing estimates related to question 4, the DCT manufacturers within and outside of Canada were identified via search engines, by reviewing DCT literature, and through discussions with DCT experts. DCT manufacturers were contacted via email or manufacturer websites to obtain relevant Canadian costing estimates. Working with our internal expert in health economics, we developed a list of questions regarding costing information and capabilities (e.g., portability, time to result) of DCTs to obtain relevant information from the manufacturers. This included 9 major categories and a total of 43 items with structured questions in the English language. Surveys for DCT manufacturers were contacted via email. Video calls with manufacturers were optional depending on their preferences. A pilot test of the instrument was conducted with the first 5 manufacturers; an email with clear instructions was drafted; and 3 follow-up reminders were sent to increase the response rate. Results were reviewed and summarized by 2 reviewers. Surveys were conducted from June 6, 2025 to August 11, 2025 via email.

Critical Appraisal of Individual Studies

The risk of bias and applicability of the included studies for research question 1 were evaluated using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) checklist³⁴ for test accuracy studies, for each index test within an included study. Published guidance for this tool was followed. The included studies about outcomes specific to research question 2 were not appraised using a specific tool, because no tool was identified that facilitates a comprehensive appraisal of the relevant outcome measures. The appraisal of studies related to research question 4 was guided by the relevant domains of the Drummond checklist³⁵ and by an assessment of the study methodology (i.e., validity of the assumptions, strength of the data), the overall robustness of the findings (e.g., sufficient sensitivity analyses to adequately inform the uncertainty and key drivers of results), and the generalizability of these studies to the Canada’s health care system (e.g., relevancy of cost information from countries other than Canada, including different health care models of care).

Two reviewers independently piloted the appraisals across at least 1 study and met to resolve any disagreements, to ensure a mutual understanding of the tool. After piloting, 1 reviewer completed the appraisals for the remaining studies. The second reviewer verified all judgments and justifications. Any disagreements were resolved through discussion, with the involvement of a third reviewer if consensus could not be reached.

Summary scores were not calculated for the included studies rather, the strengths and limitations (risk of bias and applicability for accuracy studies) of each included study were described narratively. Specifically, tables were developed to present the answers to the items within the appraisal tools. Studies were not excluded from this review based on the critical appraisal results.

Data Extraction

The data were extracted directly into tables created in Microsoft Word, which was developed and modified, as necessary. Piloting of data extraction tables was performed to ensure only the most important data fields for addressing the research questions were included. In the pilot round, 2 reviewers independently extracted data from 3 included studies, then met to resolve disagreements through discussion until they were satisfied with the content and usability of the tables. Formal data extraction was then performed by 1 reviewer, and a second reviewer independently verified the most important data items (i.e., outcome data, characteristics of the included studies) for accuracy and completeness. Both reviewers independently extracted the data from the full text of 1 study⁵² addressing question 4 and later compared them. Disagreements were resolved through discussion until consensus was reached or through adjudication by a third reviewer, if required.

Relevant information that was extracted included study characteristics, methodology (e.g., study design), index test(s) or intervention, comparator, reference standard (i.e., confirmatory lab testing), and results regarding the outcomes of interest. Where available, the characteristics of the samples used for comparison of DCTs (e.g., brought in by PWUD, submitted by police departments from their unregulated drug samples) were provided. When data were available, certain populations and settings of interest were presented, as defined by study authors and reported in tables in Appendix 3.

Per our project plan, if relevant data conflicted in the included studies (e.g., discrepancies between values reported in the abstract and the main text of a study), we reported all values and described the inconsistency. Due to the project’s timeline, no attempts to contact the corresponding authors of included studies were made to obtain missing information or to clarify conflicting information. We did not extract data presented only as figures or graphs that would require manual estimation or extraction using image processing software.

Data Analysis and Synthesis

Detailed descriptions of study characteristics from eligible studies were provided in tables, together with a narrative summary in the main text for an overview. The study and sample characteristics (when the information was available) were considered in the synthesis of the accuracy, LOD, repeatability, reproducibility, cost-effectiveness, and costs within and across the studies, and to identify relevant ethics and equity considerations (e.g., legal considerations, FP/FN rates). When the required data were provided in the studies, we conducted calculations of sensitivity and specificity, as follows, for studies that did not explicitly report them:

• Sensitivity = TP / all positives on the reference standard (i.e., confirmatory lab testing or TP + FN)

• Specificity = TN / all negatives on the reference standard (i.e., confirmatory lab testing or TN + FP)

• FN rate = 1 – Sensitivity or FN / all positives on the reference standard (i.e., confirmatory lab testing or TP + FN)

• FP rate: 1 – Specificity or FP / all negatives on the reference standard (i.e., confirmatory lab testing or TN + FP)

We used the term FN and FP ‘rates’ throughout the review to remain consistent with the terminology used in the included studies, although according to Cochrane guidance, these should more accurately be described as FN and FP “fractions.”¹

A narrative synthesis was conducted as per existing guidance by Popay et al.,⁴¹ with recognition that some adaptation was required due to the nature of the review questions within the context of drug-checking practices. For research question 1, for example, we first grouped studies similar in their population (i.e., substance sample type), intervention or index test, reference standard (i.e., confirmatory lab testing), and outcomes (e.g., accuracy). Next, we developed a preliminary synthesis by organizing the results and identifying patterns in the results. We evaluated within- and between-study relationships and discussed our findings about any differences observed. Outcomes were reported as reported by the study authors.

We planned to assess the overall certainty of the evidence for question 1 using the methods of the Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group.^89-92 However, due to heterogeneity in the methods and reporting of the results of included studies, lack of data needed for this assessment, and time constraints of conducting a rapid review (e.g., contacting authors of the included studies to request additional data), it was not feasible to conduct a certainty of evidence appraisal.

Ethics and Equity Considerations

An identification of ethics and equity considerations relevant to the context of DCTs was informed by the EUnetHTA Core Model 3.0 Ethics Domain,⁹³ ethical considerations identified through the survey conducted with members of the NDCWG (e.g., privacy and confidentiality, unintended consequences [e.g., FNs], inequities in access), and by additional harm reduction and ethics literature identified through handsearching. A reviewer with ethics expertise developed prompts and sensitizing concepts to identify and reflect on ethics and equity considerations associated with the use of DCTs. Responses to these prompts, findings from the targeted literature search, and ethical implications of the results of this review were synthesized to identify relevant ethics and equity implications for PWUD and DCTs. A reviewer with ethics expertise assisted with developing analytic categories to represent the key ethical and equity considerations arising from the identified evidence and with developing the content associated with the ethical considerations associated with the use of DCTs.

Reporting of Findings

This Rapid Review was prepared in consideration of relevant reporting guidelines (i.e., the PRISMA 2020,⁵⁴ and PRISMA-DTA statements).⁹⁴

External Review

Peer Review

Before the review phase began, 1 content expert with expertise in Canada’s DCS reviewed the project plan. The final report was externally reviewed by 2 content experts: the same content expert as the project plan and a content expert with expertise in DCTs.

Feedback Opportunity

This draft version of the report was posted on the CDA-AMC website to allow interested parties the opportunity to provide feedback on the draft report. We considered all feedback and revised the report accordingly. Therefore, this final version of the report, reflects the comments and suggestions from peer reviewers and from the feedback opportunity.

Appendix 2: Selection of Included Studies

Please note that this appendix has not been copy-edited.

Figure 1: Selection of Included Studies — PRISMA Flow Chart⁵⁴ of Selected Reports

PRISMA = Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

List of Included Studies

The citations provided in this list are the studies that were included in this customized health technology review (in reverse chronological and alphabetical order).

Crepeault H, Tobias S, Angelucci J, et al. Detection of benzodiazepines in the unregulated drug supply using point of care and confirmatory drug checking technologies: A validation study. Validation Study. Drug & Alcohol Dependence. Jun 01 2025;271:112631. doi:10.1016/j.drugalcdep.2025.112631 PubMed

Estrada Y, Sauer J, Dominguez L, et al. The prevalence of fentanyl in New York City's unregulated drug supply as measured through drug checking offered at syringe service programs. Drug & Alcohol Dependence. Mar 01 2025;268:112578. doi:10.1016/j.drugalcdep.2025.112578 PubMed

Park JN, Serafinski R, Ujeneza M, et al. Xylazine awareness, desire, use and exposure: Preliminary findings from the Rhode Island community-based drug checking cohort study. Drug Alcohol Depend Rep. Jun 2024;11:100247. doi:10.1016/j.dadr.2024.100247 PubMed

Sisco E, Nestadt DF, Bloom MB, et al. Understanding sensitivity and cross-reactivity of xylazine lateral flow immunoassay test strips for drug checking applications. Drug Testing & Analysis. Sep 2024;16(9):942-947. doi:10.1002/dta.3612 PubMed

Thompson E, Tardif J, Ujeneza M, et al. Pilot findings on the real-world performance of xylazine test strips for drug residue testing and the importance of secondary testing methods. Drug Alcohol Depend Rep. Jun 2024;11:100241. doi:10.1016/j.dadr.2024.100241 PubMed

Cepeda JA, Thompson E, Ujeneza M, et al. Costing analysis of a point-of-care drug checking program in Rhode Island. Drug & Alcohol Dependence. Dec 01 2023;253:111028. doi:10.1016/j.drugalcdep.2023.111028 PubMed

Crepeault H, Socias ME, Tobias S, et al. Examining fentanyl and its analogues in the unregulated drug supply of British Columbia, Canada using drug checking technologies. Research Support, Non-U.S. Gov't Research Support, N.I.H., Extramural. Drug & Alcohol Review. 03 2023;42(3):538-543. doi:10.1111/dar.13580

Whitehead HD, Hayes KL, Swartz JA, et al. Validated method for the analysis of 22 illicit drugs and their metabolites via liquid chromatography tandem mass spectrometry (LC-MS/MS) in illicit drug samples collected in Chicago, IL. Forensic Chem. May 2023;33doi:10.1016/j.forc.2023.100475

Gozdzialski L, Rowley A, Borden SA, et al. Rapid and accurate etizolam detection using surface-enhanced Raman spectroscopy for community drug checking. Research Support, Non-U.S. Gov't. International Journal of Drug Policy. 04 2022;102:103611. doi:10.1016/j.drugpo.2022.103611

Park JN, Sherman SG, Sigmund V, Breaud A, Martin K, Clarke WA. Validation of a lateral flow chromatographic immunoassay for the detection of fentanyl in drug samples. Research Support, N.I.H., Extramural. Drug & Alcohol Dependence. 11 01 2022;240:109610. doi:10.1016/j.drugalcdep.2022.109610

Fregonese M, Albino A, Covino C, et al. Drug Checking as Strategy for Harm Reduction in Recreational Contests: Evaluation of Two Different Drug Analysis Methodologies. Front Psychiatr. 2021;12:596895. doi:10.3389/fpsyt.2021.596895 PubMed

Goncalves R, Titier K, Latour V, et al. Suitability of infrared spectroscopy for drug checking in harm reduction centres. International Journal of Drug Policy. 02 2021;88:103037. doi:10.1016/j.drugpo.2020.103037

Gozdzialski L, Aasen J, Larnder A, et al. Portable gas chromatography-mass spectrometry in drug checking: Detection of carfentanil and etizolam in expected opioid samples. Research Support, Non-U.S. Gov't. International Journal of Drug Policy. 11 2021;97:103409. doi:10.1016/j.drugpo.2021.103409

Laing MK, Ti L, Marmel A, et al. An outbreak of novel psychoactive substance benzodiazepines in the unregulated drug supply: Preliminary results from a community drug checking program using point-of-care and confirmatory methods. Research Support, Non-U.S. Gov't. International Journal of Drug Policy. 07 2021;93:103169. doi:10.1016/j.drugpo.2021.103169

Ti L, Tobias S, Maghsoudi N, et al. Detection of synthetic cannabinoid adulteration in the unregulated drug supply in three Canadian settings. Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov't. Drug & Alcohol Review. 05 2021;40(4):580-585. doi:10.1111/dar.13237

Tobias S, Shapiro AM, Grant CJ, Patel P, Lysyshyn M, Ti L. Drug checking identifies counterfeit alprazolam tablets. Research Support, Non-U.S. Gov't. Drug & Alcohol Dependence. 01 01 2021;218:108300. doi:10.1016/j.drugalcdep.2020.108300

Green TC, Park JN, Gilbert M, et al. An assessment of the limits of detection, sensitivity and specificity of three devices for public health-based drug checking of fentanyl in street-acquired samples. Int J Drug Policy. Mar 2020;77:102661. doi:10.1016/j.drugpo.2020.102661 PubMed

McCrae K, Tobias S, Grant C, et al. Assessing the limit of detection of Fourier-transform infrared spectroscopy and immunoassay strips for fentanyl in a real-world setting. Research Support, Non-U.S. Gov't. Drug & Alcohol Review. 01 2020;39(1):98-102. doi:10.1111/dar.13004

Ti L, Tobias S, Lysyshyn M, et al. Detecting fentanyl using point-of-care drug checking technologies: A validation study. Research Support, Non-U.S. Gov't Validation Study. Drug & Alcohol Dependence. 07 01 2020;212:108006. doi:10.1016/j.drugalcdep.2020.108006

Tobias S, Shapiro AM, Wu H, Ti L. Xylazine Identified in the Unregulated Drug Supply in British Columbia, Canada. Canadian Journal of Addiction. 01 Sep 2020;11(3):28-32. doi:10.1097/cxa.0000000000000089

List of Excluded Publications and Reasons for Exclusion

The citations provided in this list are publications that were excluded after full-text review as part of the rapid review (in reverse chronological and alphabetical order).

Duplicate Record (n = 3)

Borden SA, Saatchi A, Vandergrift GW, et al. A new quantitative drug checking technology for harm reduction: pilot study in Vancouver, Canada using paper spray mass spectrometry. 2022;41(2):410-418.

Park JN, Serafinski R, Ujeneza M, et al. Xylazine awareness, desire, use and exposure: Preliminary findings from the Rhode Island community-based drug checking cohort study. Drug Alcohol Depend Rep. Jun 2024;11:100247. doi:10.1016/j.dadr.2024.100247 PubMed

Tobias S, Shapiro AM, Wu H, Ti L. Xylazine Identified in the Unregulated Drug Supply in British Columbia, Canada. Canadian Journal of Addiction. 01 Sep 2020;11(3):28-32. doi:10.1097/cxa.0000000000000089

Irrelevant Intervention or Index Test (n = 23)

Abruzzi LR, Laxton J-C, Zarkovic TM, Gill CG. Internal standard utilization strategies for quantitative paper spray mass spectrometry. 2025;36(5):1167-1174.

Liu X, Liu X, Li B, Zhang X, Hu BJAC. Lab-on-Robot: Unmanned Mass Spectrometry Robot for Direct Sample Analysis in Hazardous and Radioactive Environments. 2025;97(17):9126-9130. PubMed

Miskulin A, Wallace B, Hore DK, Gill CJA. Development of a high-resolution paper-spray mass spectrometry method using street drugs for the early detection of emerging drugs in the unregulated supply. 2025;150(9):1872-1883.

Teal T, Wallace B, Hore DJJoPHM, Practice. Evaluation of a Drug Checking Training Program for Frontline Harm Reduction Workers and Implications for Practice. 2025;31(1):43-50.

Yang X, Wang X, Chen Y, et al. Rapid on-site identification of pyrrolizidine alkaloids in herbal medicines using miniature mass spectrometry. 2025;150(9):1929-1938.

Fursman H, Morelato M, Chadwick S, et al. Development and evaluation of portable NIR technology for the identification and quantification of Australian illicit drugs. 2024;362:112179.

Gozdzialski L, Hutchison A, Wallace B, Gill C, Hore DJDt, analysis. Toward automated infrared spectral analysis in community drug checking. 2024;16(1):83-92.

Gozdzialski L, Louw R, Kielty C, et al. Beyond a spec: assessing heterogeneity in the unregulated opioid supply. 2024;21(1):63.

Hua W, Zhang W, Brown H, et al. Rapid detection of IDH mutations in gliomas by intraoperative mass spectrometry. 2024;121(23):e2318843121.

Monti MC, Bauer M, Koch K, Scheurer E, Schlotterbeck G. Evaluation of ATR-FTIR, HPLC-DAD, GC-MS, and GC-IR for the Analysis of 145 Street Drug Samples From Drug Checking Services. Drug Test Anal. Dec 26 2024;26:26. doi:10.1002/dta.3843 PubMed

Pereira I, Sboto JN, Robinson JL, Gill CGJA. Paper spray mass spectrometry combined with machine learning as a rapid diagnostic for chronic kidney disease. 2024;149(9):2600-2608.

Saatchi A, Zarkovic TM, Borden SA, Palaty J, Gill CGJJoMS, Lab AitC. Therapeutic drug monitoring of clozapine in human serum by high-throughput paper spray mass spectrometry. 2024;32:41-46.

Coppey F, Schelling C, Veuthey JL, Esseiva PJHCA. Cloud‐Enabled Handheld NIR Spectroscopy: A Transformative Approach for Real‐Time Forensic Analysis of Cannabis Specimens. 2023;106(8):e202300052.

Hore D. Toward automated infrared spectral analysis in community drug checking. 2023.

Ti L, Grant CJ, Tobias S, Hore DK, Laing R, Marshall BDL. Development of a neural network model to predict the presence of fentanyl in community drug samples. PLoS ONE. 2023;18(7):e0288656. doi:10.1371/journal.pone.0288656 PubMed

Wermelinger M, Coppey F, Gasté L, Esseiva PJFsi. Exploring the added value of portable devices such as near infrared spectrometer in the field of illicit drugs analyses. 2023;348:111605.

Borden SA, Saatchi A, Palaty J, Gill CGJA. A direct mass spectrometry method for cannabinoid quantitation in urine and oral fluid utilizing reactive paper spray ionization. 2022;147(13):3109-3117.

Gozdzialski L, Wallace B, Noda I, Hore D. Exploring the use of infrared absorption spectroscopy and two-trace two-dimensional correlation analysis for the resolution of multi-component drug mixtures. Spectrochim Acta A Mol Biomol Spectrosc. Dec 05 2022;282:121684. doi:10.1016/j.saa.2022.121684 PubMed

Larnder A, Saatchi A, Borden SA, et al. Variability in the unregulated opioid market in the context of extreme rates of overdose. 2022;235:109427.

Wallace B, Gozdzialski L, Qbaich A. A distributed model to expand the reach of drug checking. DHSP. 2022; 23 (3): 220-231. doi: 10.1108.

Deidda R, Coppey F, Damergi D, et al. New perspective for the in-field analysis of cannabis samples using handheld near-infrared spectroscopy: A case study focusing on the determination of Δ9-tetrahydrocannabinol. 2021;202:114150.

Jain R, Singh M, Kumari A, Tripathi RM. A rapid and cost-effective method based on dispersive liquid-liquid microextraction coupled to injection port silylation-gas chromatography-mass spectrometry for determination of morphine in illicit opium. Anal Sci Adv. Aug 2021;2(7-8):387-396. doi:10.1002/ansa.202000121 PubMed

Coppey F, Bécue A, Sacré P-Y, Ziemons EM, Hubert P, Esseiva PJFsi. Providing illicit drugs results in five seconds using ultra-portable NIR technology: An opportunity for forensic laboratories to cope with the trend toward the decentralization of forensic capabilities. 2020;317:110498.

Irrelevant Comparator (n = 4)

Borden SA, Saatchi A, Vandergrift GW, Palaty J, Lysyshyn M, Gill CG. A new quantitative drug checking technology for harm reduction: Pilot study in Vancouver, Canada using paper spray mass spectrometry. Research Support, Non-U.S. Gov't. Drug Alcohol Rev. 02 2022;41(2):410-418. doi:10.1111/dar.13370

Deconinck E, Ait-Kaci C, Raes A, et al. An infrared spectroscopic approach to characterise white powders, easily applicable in the context of drug checking, drug prevention and on-site analysis. Comparative Study Validation Study. Drug Test Anal. Mar 2021;13(3):679-693. doi:10.1002/dta.2973 PubMed

Gerace E, Seganti F, Luciano C, et al. On-site identification of psychoactive drugs by portable Raman spectroscopy during drug-checking service in electronic music events. Research Support, Non-U.S. Gov't. Drug Alcohol Rev. 01 2019;38(1):50-56. doi:10.1111/dar.12887

McCrae K, Tobias S, Tupper K, et al. Drug checking services at music festivals and events in a Canadian setting. Research Support, Non-U.S. Gov't. Drug Alcohol Depend. 12 01 2019;205:107589. doi:10.1016/j.drugalcdep.2019.107589

Irrelevant Outcome or Design (n = 21)

Gonzalez-Nieto P, Wallace B, Kielty C, et al. Not just fentanyl: Understanding the complexities of the unregulated opioid supply through results from a drug checking service in British Columbia, Canada. Int J Drug Policy. Apr 2025;138:104751. doi:10.1016/j.drugpo.2025.104751 PubMed

Pereira MB, Familia C, Martins D, et al. Drug-Checking and Monitoring New Psychoactive Substances: Identification of the U-48800 Synthetic Opioid Using Mass Spectrometry, Nuclear Magnetic Resonance Spectroscopy, and Bioinformatic Tools. Int J Mol Sci. Feb 28 2025;26(5):28. doi:10.3390/ijms26052219 PubMed

Tobias S, Angelucci J, Wood E, Buxton JA, Ti L. Novel adulterants in unregulated opioids and their associations with adverse events. Can J Public Health. Feb 24 2025;24:24. doi:10.17269/s41997-024-00990-7 PubMed

Wallace B, Shkolnikov I, Kielty C, et al. Is fentanyl in everything? Examining the unexpected occurrence of illicit opioids in British Columbia's drug supply. Harm Reduct J. Mar 10 2025;22(1):28. doi:10.1186/s12954-025-01189-w PubMed

Barratt MJ, Ball M, Wong GTW, Quinton A. Adulteration and substitution of drugs purchased in Australia from cryptomarkets: An analysis of Test4Pay. Research Support, Non-U.S. Gov't. Drug Alcohol Rev. May 2024;43(4):969-974. doi:10.1111/dar.13825 PubMed

Fabresse N, Papias E, Heckenroth A, Martin V, Allemann D, Roux P. Implementation of a community-based LC-UV drug checking service: promising preliminary findings on feasibility and validity. Validation Study. Harm Reduct J. Oct 18 2024;21(1):185. doi:10.1186/s12954-024-01098-4 PubMed

Martens RR, Gozdzialski L, Newman E, Gill C, Wallace B, Hore DK. Trace Detection of Adulterants in Illicit Opioid Samples Using Surface-Enhanced Raman Scattering and Random Forest Classification. Anal Chem. Jul 17 2024;17:17. doi:10.1021/acs.analchem.4c01271 PubMed

Miskulin A, Wallace B, Gill C, Hore D. A strategy for the detection of benzodiazepine drugs using low-resolution paper-spray mass spectrometry for harm reduction drug checking. Drug Test Anal. Oct 2024;16(10):1085-1093. doi:10.1002/dta.3630 PubMed

Cruz SL, Bencomo-Cruz M, Medina-Mora ME, Vazquez-Quiroz F, Fleiz-Bautista C. First drug-checking study at an electronic festival and fentanyl detection in the central region of Mexico. Research Support, Non-U.S. Gov't. Harm Reduct J. 12 06 2023;20(1):174. doi:10.1186/s12954-023-00905-8

Gozdzialski L, Wallace B, Hore DJHRJ. Point-of-care community drug checking technologies: an insider look at the scientific principles and practical considerations. 2023;20(1):39.

Laxton J-C, Monaghan J, Wallace B, Hore D, Wang N, Gill CG. Evaluation and improvement of a miniature mass spectrometry system for quantitative harm reduction drug checking. 2023;484:116976.

Mullin A, Scott M, Vaccaro G, et al. Handheld Raman Spectroscopy in the First UK Home Office Licensed Pharmacist-Led Community Drug Checking Service. Research Support, Non-U.S. Gov't. Int J Environ Res Public Health. 03 08 2023;20(6):08. doi:10.3390/ijerph20064793

Nielsen S, Barratt M, Hiley S, et al. Monitoring for fentanyl within Australian supervised injecting facilities: Findings from feasibility testing of novel methods and collaborative workshops. Research Support, Non-U.S. Gov't. Int J Drug Policy. 05 2023;115:104015. doi:10.1016/j.drugpo.2023.104015

Green TC, Olson R, Jarczyk C, et al. Implementation and Uptake of the Massachusetts Drug Supply Data Stream: A Statewide Public Health-Public Safety Partnership Drug Checking Program. Research Support, N.I.H., Extramural Research Support, U.S. Gov't, P.H.S. J Public Health Manag Pract. Nov-Dec 01 2022;28(Suppl 6):S347-S354. doi:10.1097/phh.0000000000001581

Klaire S, Janssen RM, Olson K, et al. Take-home drug checking as a novel harm reduction strategy in British Columbia, Canada. Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov't. Int J Drug Policy. 08 2022;106:103741. doi:10.1016/j.drugpo.2022.103741

Bowles JM, McDonald K, Maghsoudi N, et al. Xylazine detected in unregulated opioids and drug administration equipment in Toronto, Canada: clinical and social implications. Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov't. Harm Reduct J. 10 13 2021;18(1):104. doi:10.1186/s12954-021-00546-9

Karch L, Tobias S, Schmidt C, et al. Results from a mobile drug checking pilot program using three technologies in Chicago, IL, USA. Research Support, U.S. Gov't, Non-P.H.S. Research Support, U.S. Gov't, P.H.S. Drug Alcohol Depend. 11 01 2021;228:108976. doi:10.1016/j.drugalcdep.2021.108976

Patel P, Guzman S, Lysyshyn M, et al. Identifying Cocaine Adulteration in the Unregulated Drug Supply in British Columbia, Canada. Canadian Journal of Addiction. 11 Jun 2021;12(2):39-44. doi:10.1097/cxa.0000000000000112

Kang M, Zhang W, Dong L, et al. On-site testing of multiple drugs of abuse in urine by a miniature dual-LIT mass spectrometer. 2020;1101:74-80.

Tupper KW, McCrae K, Garber I, Lysyshyn M, Wood E. Initial results of a drug checking pilot program to detect fentanyl adulteration in a Canadian setting. Research Support, Non-U.S. Gov't. Drug Alcohol Depend. 09 01 2018;190:242-245. doi:10.1016/j.drugalcdep.2018.06.020

Sondermann N, Kovar KA. Screening experiments of ecstasy street samples using near infrared spectroscopy. Forensic Sci Int. Dec 20 1999;106(3):147-56. doi:10.1016/s0379-0738(99)00195-4 PubMed

Irrelevant Format (i.e., Conference Proceeding or Abstract) (n = 12)

Fan X, Jiao B, Zhou X, Zhang W, Ouyang ZJAC. Miniaturization of Mass Spectrometry Systems: An Overview of Recent Advancements and a Perspective on Future Directions. Anal Chem. 2025;97(17):9111-9125. PubMed

Molina C, Romero R, Sisco E, et al. Evaluating Xylazine Immunoassay Test Strips in Street-Based Samples of Rocks, Powder, Pills, and Tar in Los Angeles, California. Conference Abstract. Drug and Alcohol Dependence. 01 Feb 2025;267111955. doi:10.1016/j.drugalcdep.2024.111955

Monti MC, De Vrieze LM, Vandeputte MM, et al. Detection of N-desethyl etonitazene in a drug checking sample: Chemical analysis and pharmacological characterization of a recent member of the 2-benzylbenzimidazole “nitazene” class. J Pharm Biomed Anal. Dec 15 2024;251:116453. doi:10.1016/j.jpba.2024.116453 PubMed

Philip L, Black C, Powell G, Clark L, Johnson C, Johnson O. The Effect of External Influences on Drug Trends at Music Festivals in New Zealand, 2018 to 2023. Conference Abstract. Emerging Trends in Drugs, Addictions, and Health. 01 Dec 2024;4100108. doi:10.1016/j.etdah.2023.100108

Strathmann F, Gillan G, Toohey J, et al. B-282 Cost and Benefits of Utilization of Ion Mobility Mass Spectrometry Testing for Drugs of Abuse Compared to Immunoassay Screening in a Large Urban Hospital. Conference Abstract. Clinical Chemistry. 01 Oct 2024;70(Supplement_1):i243-i244. doi:10.1093/clinchem/hvae106.639

Augsburger M, Déglon J, Widmer C, Thomas A, Morger Mégevand R, Longère S. Interest of drug checking supply: Results of 3 years of activity in Geneva. Conference Abstract. Toxicologie Analytique et Clinique. 01 Sep 2022;34(3):S32-S33. doi:10.1016/j.toxac.2022.06.027

Borden SA, Saatchi A, Palaty J, Gill CGJA. A direct mass spectrometry method for cannabinoid quantitation in urine and oral fluid utilizing reactive paper spray ionization. 2022;147(13):3109-3117.

Borden SA, Saatchi A, Gill CG, Wijeratne NRJTN. Quantitation of drugs of abuse and their metabolites in urine using PaperSpray tandem mass spectrometry for clinical research and forensic toxicology. 2020;73467

Vandergrift GW, Gill CG. Paper spray mass spectrometry: A new drug checking tool for harm reduction in the opioid overdose crisis. Review. J Mass Spectrom. Sep 2019;54(9):729-737. doi:10.1002/jms.4431 PubMed

Bardwell G, Kerr T. Drug checking: a potential solution to the opioid overdose epidemic? Letter Research Support, Non-U.S. Gov't. Subst Abuse Treat Prev Policy. 05 25 2018;13(1):20. doi:10.1186/s13011-018-0156-3

Glick JL, Christensen T, Park JN, McKenzie M, Green TC, Sherman SG. Stakeholder perspectives on implementing fentanyl drug checking: Results from a multi-site study. Research Support, N.I.H., Extramural Research Support, Non-U.S. Gov't. Drug Alcohol Depend. 01 01 2019;194:527-532. doi:10.1016/j.drugalcdep.2018.10.017

Zhang C, Manicke NEJAc. Development of a paper spray mass spectrometry cartridge with integrated solid phase extraction for bioanalysis. 2015;87(12):6212-6219.

Appendix 3: Characteristics of Included Publications

Please note that this appendix has not been copy-edited.

Table 5: Characteristics of the Included Test Accuracy Studies

BC = British Columbia; BTS = benzodiazepine test strips; CDC = Centers for Disease Control and Prevention; DART-MS = direct analysis in real time mass spectrometry; FN = false negative; FP = false positive; FTIR = Fourier transform infrared; FTS = fentanyl test strips; GC-MS = gas chromatography–mass spectrometry; GC-MS-MS = gas chromatography–tandem mass spectrometry; IR = infrared spectroscopy; LC-MS = liquid chromatography–mass spectrometry; LC-MS-MS = high-precision liquid chromatography–tandem mass spectrometry; LC-QTOF-MS = liquid chromatography–quadrupole time-of-flight mass spectrometry; LOD = limit of detection; NIH-NIDA = National Institutes of Health-National Institute on Drug Abuse; NPV = negative predictive value; NR = not reported; PPV = positive predictive value; PS-MS = paper spray mass spectrometry; PWUD = people who use drugs; qNMR = quantitative nuclear magnetic resonance; SCS = supervised consumption site; SD = standard deviation; SERS = surface-enhanced Raman spectroscopy; SUAP = Substance Use and Addictions Program; TN = true negative; TP = true positive; UPLC-HRMS = ultraperformance liquid chromatography–high-resolution mass spectroscopy; XTS = xylazine test strips.

^a.All included studies had cross-sectional design.

^bWhen reporting on sex, gender, race, or ethnicity in this Rapid Review, we retained the language used by the original study authors, and, whenever possible, we referred to these groups based on guidance from the CDA-AMC Style Guide⁹ at the time this Rapid Review was conducted, with an understanding that language is constantly evolving.

Table 6: Characteristics of the Included Cost Description Study

FTIR = Fourier transform infrared; FTS = fentanyl test strips, NA = not applicable; NR = not reported; PWUD = people who use drugs.

Appendix 4: Critical Appraisal of Included Publications

Please note that this appendix has not been copy-edited.

Table 7: Risk of Bias and Applicability Assessment of Crepeault et al. (2025)³⁶ — QUADAS-2

BTS = benzodiazepine test strips; FTIR = Fourier transform infrared; GC-MS = gas chromatography–mass spectrometry; LC-MS = liquid chromatography–mass spectrometry; qNMR = quantitative nuclear magnetic resonance; QUADAS-2 = Quality Assessment of Diagnostic Accuracy Studies 2.

^aThe appraisal applies to all the index tests or interventions in the study.

Table 8: Risk of Bias and Applicability Assessment of Crepeault et al. (2023)³⁷ — QUADAS-2

 FTIR = Fourier transform infrared; FTS = fentanyl test strips; QUADAS-2 = Quality Assessment of Diagnostic Accuracy Studies 2.

^aThe appraisal applies to all the index tests or intervention(s) in the study.

Table 9: Risk of Bias and Applicability Assessment of Estrada et al. (2025)³⁸ — QUADAS-2

FTIR = Fourier transform infrared; FTS = fentanyl test strips; GC-MS = gas chromatography–mass spectrometry; LC-QTOF-MS = liquid chromatography–quadrupole time-of-flight mass spectrometry; QA = quality assurance; QUADAS-2 = Quality Assessment of Diagnostic Accuracy Studies 2.

^aThe appraisal applies to all the index tests or interventions in the study.

Table 10: Risk of Bias and Applicability Assessment of Fregonese et al. (2021)³⁹ — QUADAS-2

GC-MS = gas chromatography–mass spectrometry; QUADAS-2 = Quality Assessment of Diagnostic Accuracy Studies 2.

Table 11: Risk of Bias and Applicability Assessment of Goncalves et al. (2021)⁴⁰ — QUADAS-2