Authors: Audrey Lopez, Audrey Sarah Cohen, Francine Vega, Tiffany Champagne-Langabeer
Categories: Review, Opioid use disorder, Screening tools, Documentation, OUD, Provider practices
Source: Drug and Alcohol Dependence Reports
Authors: Audrey Lopez, Audrey Sarah Cohen, Francine Vega, Tiffany Champagne-Langabeer
As the prevalence of opioid use disorder (OUD) continues to rise, early detection by medical professionals can often be the first step in linking individuals to treatment. This systematic review was designed to identify implemented OUD screening and assessment tools with studies published from January 2019 through June 2024, uncover common themes associated with implementation, and determine if these tools were recommended in clinical practice.
A systematic literature search was conducted within PubMed, EMBASE, and Web of Science using the keywords ‘opioid use disorder,’ ‘documentation,’ and ‘screening assessment tool.’ Three investigators independently reviewed titles, abstracts, and full-text articles for inclusion and exclusion criteria and inclusion within the study. The Johns Hopkins Evidence-Based Practice model was used to appraise evidence level, quality, and common themes.
The initial literature search yielded 914 articles for review, with 19 remaining in the final selection. Of the 19 articles, 15 provided quantitative results of an implemented OUD screening or assessment tool, and 4 offered qualitative results about the value of implemented tools within a clinical practice. 15 of the articles recommended a screening or diagnostic assessment tool. Key themes included the insufficiency of a single assessment tool, inconsistent documentation of OUD diagnoses and symptoms, and variable tool completion.
Diagnosing OUD is complex and dynamic. This review highlights the value of screening and assessment tools in identifying individuals and initiating opioid use-related care. Future research should explore implementing artificial intelligence and patient-centered care to assist with OUD screening and assessment.
Opioid use disorder (OUD) is a chronic health condition that can lead to impairment and respiratory distress (Eckert and Yaggi, 2022, Sharma et al., 2016). While OUD is one type of substance use disorder (SUD), it presents distinct clinical challenges that have led to the development of targeted care strategies. These include medications for opioid use disorder (MOUD), psychosocial interventions, and harm reduction approaches such as naloxone education and distribution (Kelley et al., 2022, McGinty et al., 2024). In addition, OUD is associated with a disproportionately high rate of overdose compared with other forms of SUD (Garnett and Miniño, 2024, Sharma et al., 2016). Out of the approximately 600,000 deaths worldwide due to drug use in 2019, about 80 % of those deaths were determined to be related to opioids (World Health Organization, 2023). In the United States alone, the economic cost of opioid use disorder and fatal opioid overdose in 2017 was estimated at $1.02 trillion (Florence et al., 2021). Given this impact, healthcare professionals face an urgent imperative to identify individuals with OUD, connect them to opioid use-related care, and mitigate the adverse effects this condition has on individuals, families, and communities (Bunting et al., 2023, Hutchison et al., 2023, Langabeer et al., 2020, Punches et al., 2024). However, identifying individuals with OUD is not always straightforward. Symptoms associated with OUD can vary, and stigma may prevent individuals from disclosing their opioid use concerns (Bunting et al., 2023, Ducharme and Moore, 2019). Medical providers may benefit from access to a concise, evidence-based list of OUD symptoms that can be discussed during clinical visits, documented within the patient’s medical record, and used to support timely referrals to care (Rossom et al., 2023).
An electronic health record (EHR) can store, record, and utilize screening and diagnostic assessment tools for various medical records and patient care purposes, such as providing documentation support for diagnoses and treatment recommendations (Maria et al., 2023). Screening and diagnostic assessment tools are also used to support diagnoses linked to International Classification of Diseases (ICD) codes to bill for medical services (Nielsen et al., 2020). To date, OUD screening and diagnostic assessment tools are not standardized within EHRs, and documenting symptoms that correspond to an OUD diagnosis varies based on the discretion of the medical provider (Skeer et al., 2023). The lack of standardized documentation can lead to inconsistencies and missed OUD diagnoses (Poulsen et al., 2023). These documentation inconsistencies reduce the ability to identify individuals with OUD and link these individuals to opioid use-related care (Osterhage et al., 2024).
Accurate documentation within a medical record leads to several beneficial outcomes, such as improved patient care plans, improved communication and continuity of care, and decreased medical errors (Demsash et al., 2023). Screening and diagnostic assessment tools were designed to be quickly and efficiently used within EHRs to document symptoms and support diagnoses that can also be used as supporting evidence for healthcare insurance companies to approve additional treatment for an individual (Busis et al., 2024, Graber et al., 2017). Applying these principles to OUD, a screening or diagnostic assessment tool within an electronic medical record can be used to ask an individual questions regarding specific symptoms associated with the diagnosis. The affirmative responses are calculated, and based on the number of affirmative responses, an individual can then be referred for formal diagnostic evaluation, or, in the case of a diagnostic assessment tool, matched with an OUD diagnosis of severe, moderate, or mild (Palumbo et al., 2020). These diagnoses are then linked to ICD codes that can be processed through healthcare insurance companies and used as reporting metrics for various governmental assistance requests or quality improvement measures (Lagisetty et al., 2021).
Medical providers report a degree of uncertainty when initiating difficult conversations with patients about possible opioid use due to the social stigma that a documented OUD diagnosis can cause for an individual (Austin et al., 2023, Dela Cruz et al., 2023, Hooker et al., 2024). Ensuring that the symptoms leading to an OUD diagnosis are well documented within a medical record can provide clinicians with confidence to document a diagnosis of OUD. This diagnosis can create a pathway for opioid use-related care, such as medication-assisted treatment (MAT) or other patient-centered interventions, which can improve health outcomes for those affected by OUD (Kelley et al., 2022, Pendergrass et al., 2019).
Screening and diagnostic assessment tools allow clinicians to identify factors associated with OUD, such as increased tolerance, cravings, and medication-seeking behaviors. These tools serve as a guide for practitioners, helping them to gain confidence in documenting a diagnosis. Current research on OUD screening and diagnostic tools offers a roadmap for best practices in assessment, documentation, and diagnosis. Our review aimed to identify the tools currently implemented to detect opioid use disorder, reveal common themes in their implementation, and evaluate their adoption and recommendations in clinical practice.
In order to capture a wide range of literature that studied the implementation of both screening and diagnostic assessment tools for opioid use disorder, three search engines were PubMed, Embase, and Web of Science. Search terms were selected based on the indexing system for each search engine to capture articles relevant to the search objective. The expertise of a medical librarian was used to identify search terms appropriate for each search engine. A PubMed search was conducted using the MeSH terms ‘opioid use disorder’ and ‘documentation.’ Because Embase uses its own controlled vocabulary, Emtree, rather than MeSH, which is used by PubMed, search terms were translated accordingly. Specifically, the MeSH term ‘documentation’ was refined to the Emtree term ‘assessment screening tool’ to better align with Embase indexing. This adjustment reduced the screening workload while still capturing the relevant articles (Cadwell et al., 2017, Elsevier, 2023). Web of Science search terms included ‘opioid use disorder’ AND ‘screening assessment tool’ OR ‘documentation.’ Specific search terms are listed in Appendix A. Following this search, the article citations and abstracts were exported from the respective databases into Rayyan, a research collaboration forum, to remove duplicates and review articles for inclusion and exclusion of our search topic (Ouzzani et al., 2016).
Two reviewers independently assessed the eligibility of each article based on its title and abstract. The criteria for inclusion included original research studies published between January 2019 and June 2024 that investigated the implementation of an OUD screening or diagnostic assessment tool. Articles were included if they were published in English and involved human subjects. The title and abstract were examined to exclude studies with populations or outcome measures that were outside the intended scope. Articles were excluded if they were retrospective studies that analyzed existing medical records to identify OUD diagnoses already associated with corresponding ICD codes. Algorithms and natural language processing techniques for screening and diagnosing OUD were not included, as this study focused on assessment techniques that do not employ these advanced technologies. We also excluded articles that did not examine the implementation of a screening or diagnostic assessment tool for OUD. Furthermore, we excluded articles where the primary outcome of the study was out of scope, such as those dealing with criteria for prescribing opioids, Hepatitis C, pain assessments, health outcomes for patients with existing OUD diagnoses, OUD treatments such as medications for OUD (MOUD), effects of MOUD, and opioid overdose.
The initial literature search resulted in 914 articles for review. After eliminating 51 duplicate records, two investigators reviewed the article titles and abstracts and excluded 771 articles outside the study scope. Three investigators then conducted a full-text article review of the remaining 92 articles and applied the same exclusion and inclusion criteria to narrow down the search. A majority vote resolved disagreements in article selection. The final analysis included the results of 15 qualitative articles and four quantitative articles. The Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) diagram depicted in Fig. 1 describes the review process for this literature review (Page et al., 2021).Fig. 1PRISMA Diagram for Identification of Studies via Databases.Fig. 1
Following the selection of the articles, the Johns Hopkins Evidence-Based Practice Model (JHEBP) was utilized to appraise each article's evidence level and quality (Dang et al., 2022). OUD implementation tools were identified and assessed by study location and treatment focus. Using the JHEBP model, the key findings and study limitations were synthesized into themes and used as best evidence recommendations for implementing an OUD screening or diagnostic assessment tool (Dang et al., 2022). The articles were also assessed to determine if a diagnostic assessment tool for OUD is recommended for clinical practice.
Each article was appraised by the research study design as the type of evidence using the JHEBP framework, as shown in Appendix B. Of the nineteen articles identified, one article was appraised at the highest level of evidence ranking as a level I, thirteen articles were appraised as evidence level II, four articles at level III, and one article ranked at an evidence level of IV (Dang et al., 2022). Among each level of evidence, the quality of the articles was identified with five articles categorized as an “A” strong quality level. The majority of the articles held a “B” or good quality ranking, with thirteen articles identified. One article was identified as a “C” or low-quality study (Dang et al., 2022).
The Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (DSM-5), was implemented in three studies. One study implemented the DSM-5 along with the National Institute on Drug Abuse (NIDA) Modified Alcohol, Smoking, and Substance Involvement Screening Test (NM-ASSIST). One study implemented the DSM-5 with the World Health Organization (WHO) ASSIST. This resulted in five studies implementing the DSM-5. The following tools were implemented within two articles The Assessment for Opioid Overuse, Worrying, Losing Interest, and Feeling Slowed Down (OWLS), Tobacco, Alcohol, Prescription medication and other Substance use (TAPS), Current Opioid Misuse Measure (COMM), and Computer-Administered Routine Opioid Outcome Monitoring (ROOM) tools. The remaining assessment tools were each implemented in a single study. These tools were NIDA-ASSIST, POMI, the Substance Use Symptom Checklist modified from DSM-5, the Unintended Use, Neglected Role Obligations, Cut Down, Object to use, Preoccupation with Use, and Emotional Discomfort Screening tool (UNCOPE), Rapid Opioid Use Disorder Assessment (ROUDA), and the Psychiatric Research Interview for Substance and Mental Disorders DSM-5 Opioid Version (PRISM-5) pain adjusted prescription opioid use disorder (POUD). Table 1 lists the assessment tools implemented from the selected articles and the number of times they were implemented in the literature.Table 1Assessment Tools Implemented from Systematic Review Articles.Table 1Implemented toolNumber of articles with implemented toolFirst Author, YearDiagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM−5)5Edmond et al., (2022); Kufeld et al., (2024); Punches et al., (2024); Sullivan et al., (2020); Staton et al., (2024)Assessment for Opioid Overuse, Worrying, Losing Interest, and Feeling Slowed Down (OWLS)2Nielsen et al., (2020); Picco et al., (2020a)Tobacco, Alcohol, Prescription medication and other Substance use (TAPS)2Bunting et al., (2023); Carter et al., (2022)Current Opioid Misuse Measure (COMM)2Mallery Lankford et al., (2022); Wilson et al., (2020)National Institute on Drug Abuse (NIDA) Modified Alcohol, Smoking, and Substance Involvement Screening Test (ASSIST)2Austin et al., (2023); Staton et al., (2024)Computer-Administered Routine Opioid Outcome Monitoring (ROOM) Tool2Lam et al., (2023); Picco et al., (2020b)Substance Use Symptom Checklist modified from DSM−51Williams et al., (2024)Unintended Use, Neglected Role Obligations, Cut Down, Object to use, Preoccupation with Use, and Emotional Discomfort Screening tool (UNCOPE)1Kopak et al., (2024)World Health Organization (WHO) ASSIST1Punches et al., (2024)Rapid Opioid Use Disorder Assessment (ROUDA)1Di Paola et al., (2023)Prescription Opioid Misuse Index (POMI)1Laporte et al., (2022)Psychiatric Research Interview for Substance and Mental Disorders DSM−5 Opioid Version (PRISM−5-OP) pain adjusted prescription opioid use disorder (POUD)1Hasin et al., (2022)
The United States was the most common location with thirteen identified articles. Three studies took place in Australia, and one in Germany, Canada, and France. Two of the three studies in Australia utilized the OWLS tool in the primary care setting. This tool was unique to Australia. Similarly, the POMI was unique to France.
The treatment focus varied among studies, with the primary care setting used for eight studies, the emergency department, psychiatry, and incarcerated populations each had two studies. Pharmacy and rheumatology were the focus for one study each. Three articles had combined treatment pain management and primary care, pain management and inpatient substance use treatment, and a study with multiple treatment foci. Eighteen articles studied individuals within the adult population aged 18 years or older. One article did not mention the age of the population assessed, but also did not mention the involvement of minors in the study.
Across the eight studies in the primary care setting, seven unique tools were implemented. One study was unique to individuals with arthritis (Mallery Lankford et al., 2022). In this study, the Current Opioid Misuse Measure (COMM) was compared to DSM-5 diagnostic criteria and recommended as a future assessment tool (Mallery Lankford et al., 2022). However, when implemented in an emergency room setting, the COMM was not recommended (Wilson et al., 2020). The DSM-5 and the WHO ASSIST were also implemented in the emergency room in a study that compared the two tools (Punches et al., 2024). Both tools were recommended, with the WHO ASSIST preferred in some instances, such as in screening for substances in addition to opioids (Punches et al., 2024).
One study evaluated OUD assessment in incarcerated individuals using the Unintended Use, Neglected Role Obligations, Cut Down, Object to Use, Preoccupation with Use, and Emotional Discomfort Screening tool (UNCOPE) compared to the DSM-5 (Kopak et al., 2024). The authors found the 6-item UNCOPE to have an ROC of 0.93, with a 95 % CI of 0.91–0.95 for the diagnosis of OUD at any severity level (Kopak et al., 2024). Hasin et al. developed the Psychiatric Research Interview for Substance and Mental Disorders DSM-5 Opioid Version (PRISM-5-OP), pain-adjusted prescription opioid use disorder (POUD) to assess for POUD specifically in individuals with pain (2022). Based on the Rapid Opioid Dependence Screen, Di Paola et al. developed the Rapid Opioid Use Disorder Assessment (ROUDA) to be used by non-clinicians. It was designed to assess both current use (last 12 months) as well as short-term (3 months) and long-term (lifetime) remission (2023). Finally, Williams et al. evaluated the use of an 11-item checklist based on the DSM-5 to assess for OUD in the primary care setting (2024). Table 2 lists the identified articles, study location, treatment focus, implemented tools, and themes for the selected articles.Table 2List of selected articles for analysis along with treatment focus, implemented themes, and tools recommended.Table 2First Author, YearCountryTreatment Focus (N)Implemented toolScreening or Diagnostic Assessment Tool****ReferenceTool: Recommendor Not RecommendThemesAustin et al., (2023)United StatesPrimary Care (10 clinics)NIDA modified ASSISTScreeningN/A- Qualitative StudyNot RecommendedNAD, NS, ICBunting et al., (2023)United StatesPrimary Care (2000)TAPSScreeningComposite International Diagnostic Interview (CIDI)RecommendedNAD, NSCarter et al., (2022)United StatesPharmacy (1523)TAPSScreeningWHO ASSISTRecommendedNSDi Paola et al., (2023)United StatesPsychiatry (150)ROUDADiagnostic AssessmentMINI version 7RecommendedNAD, NS, ICEdmond et al., (2022)United Statesmultiple healthcare specialties (51 survey panelists)DSM−5Diagnostic AssessmentN/A- Qualitative StudyRecommendedNAD, NSHasin et al., (2022)United StatesPain Management and Inpatient Substance Treatment Facilities (606)PRISM−5-OP, pain-adjusted POUDDiagnostic AssessmentInterview assessing DSM−5 POUDRecommendedNSKopak et al., (2024)United StatesHealthcare for Incarcerated Individuals (717)UNCOPEScreeningComprehensive Addictionand Psychological Evaluation−5 (CAAPE) structured interview assessing DSM−5 criteriaRecommendedMWKufeld et al., (2024)GermanyPrimary Care/Pain Management (14 telephone interviews)DSM−5 InterviewDiagnostic AssessmentN/A- Qualitative StudyNot RecommendedNAD, NSLam et al., (2023)CanadaPrimary Care (6 participants)Routine Opioid Outcome Monitoring (ROOM) ToolScreeningN/A- Qualitative StudyRecommendedMWLaporte et al., (2022)FrancePrimary Care (160)Prescription Opioid Misuse Index (POMI)ScreeningDSM−5 QuestionnaireNot RecommendedNAD, NS, IC, MWMallery Lankford, et al. (2022)United StatesRheumatology (318)Current Opioid Misuse Measure (COMM)- modified to COMM−11 PWDAScreeningDSM−5 OUD ScreenerRecommendedNAD, NS, ICNielsen et al., (2020)AustraliaPrimary Care (1134)OWLSScreeningModified CIDI assessing ICD−10, ICD−11 and DSM−5RecommendedNAD, NS, IC, MWPicco et al., (2020a)AustraliaPrimary Care (324)OWLSScreeningCIDI DSM−5 questions in an online moduleRecommendedNS, ICPicco et al., (2020b)AustraliaPrimary Care (324)Computer-Administered Routine Opioid Outcome Monitoring (ROOM) Tool - OWLSScreeningCIDI DSM−5 questions in an online moduleRecommendedNC, ICPunches et al., (2024)United StatesEmergency Department (1305)Adapted DSM−5 checklist and WHO ASSISTScreeningComparison studyRecommendedNS, ICStaton et al., (2024)United StatesHealthcare for Incarcerated Individuals (200)(NIDA Modified ASSIST) NM-ASSIST, DSM−5 ChecklistScreeningHigh risk drug use indicatorsRecommendedNAD, MWSullivan et al., (2020)United StatesInpatient Psychiatry (60)DSM−5 level 2Screeningchart abstracted toxicology screensRecommendedMWWilliams et al., (2024)United StatesPrimary Care (2007)Substance Use Symptom Checklist modified from 11 DSM−5 SUD criteriaScreeningDocumentation of OUD diagnosis, treatment or long term opioid therapyRecommendedNADWilson et al., (2020)United StatesEmergency Department (154)COMMScreeningdocumentation of concerning drug behaviorNot RecommendedMWNAD: OUD diagnosis not always documented; NS: One tool not sufficient; IC: Inconsistent completion of tool; MW: OUD symptoms documented in multiple ways⁎For quantitative articles, the implemented tool was identified as recommended if it was consistent or provided better results than the reference. For qualitative articles, the recommendation is based on the authors’ conclusions, not diagnostic performance.
Upon synthesis of the nineteen identified study results and conclusion, four common themes emerged. Twelve of the articles stated that one assessment tool may not be sufficient, ten studies acknowledged that the OUD diagnosis is not always documented, eight articles reported inconsistent completion of the implemented diagnostic assessment tool, and seven articles stated that OUD symptoms can be documented in multiple ways. These themes identify barriers that should be considered when implementing a diagnostic assessment tool for OUD within a medical practice.
Studies highlight the value of using multiple diagnostic assessment tools, as relying on a single assessment tool may have limitations. Patients may not feel comfortable with how the questions are phrased or the context, or how questions are asked (Mallery Lankford et al., 2022, Nielsen et al., 2020).
To address this barrier, several studies compared the Diagnostic and Statistical Manual of Mental Disorders, Version 5 (DSM-5) OUD diagnostic assessment with other assessment tools to understand if a secondary assessment tool could improve knowledge of OUD symptoms or predictive indicators (Di Paola et al., 2023, Laporte et al., 2022, Mallery Lankford et al., 2022, Nielsen et al., 2020, Picco et al., 2020a, Picco et al., 2020b, Punches et al., 2024). Utilizing a secondary assessment tool or modifying the DSM-5 could assist in understanding the severity of each symptom and isolating symptoms experienced by patients with chronic pain or long-term opioid use (Edmond et al., 2022, Hasin et al., 2022, Kufeld et al., 2024).
Practitioners may not always document an OUD diagnosis, even when identifying individuals with OUD using supporting evidence such as prior drug use, a previous diagnosis, patient self-report, or requests for medication-assisted treatment (Bunting et al., 2023, Di Paola et al., 2023). Patients seeking early opioid prescription refills or multiple prescriptions can also serve as indicators of OUD (Laporte et al., 2022, Mallery Lankford et al., 2022). Identifying predictors for patients who are at a higher risk for developing OUD can also factor into the screening process, such as assessing for OUD within incarcerated populations or inpatient psychiatric units (Kopak et al., 2024, Staton et al., 2024, Sullivan et al., 2020). Formally documenting an OUD diagnosis can be challenging due to the subtleties of the symptoms, repercussions for documenting an incorrect diagnosis, and patient deceptiveness when self-reporting symptoms due to the stigma surrounding the diagnosis (Austin et al., 2023, Edmond et al., 2022, Kufeld et al., 2024, Mallery Lankford et al., 2022).
The existence of a diagnostic assessment tool does not ensure consistent completion. Several studies noted the importance of having trained staff and sufficient time and space to complete assessments, as well as limiting the number of questions in the tool (Di Paola et al., 2023, Mallery Lankford et al., 2022, Nielsen et al., 2020, Picco et al., 2020a, Picco et al., 2020b, Punches et al., 2024). One study developed the Rapid Opioid Use Disorder Assessment (ROUDA) and the Rapid Stimulant Use Disorder Assessment (RSUDA) to shorten assessment time compared to longer assessments like the DSM-5 (Di Paola et al., 2023). These tools also enable non-clinical staff to complete the assessments and refer patients to care opportunities as needed (Di Paola et al., 2023). Picco et al. developed the Computer-Administered Routine Opioid Outcome Monitoring (ROOM) Tool that could be self-administered on a tablet or computer in the waiting room before appointments, thus saving time (2020b).
A clinical provider might locate previously documented OUD symptoms in multiple places within a patient’s medical record and not within a centralized area, such as a completed diagnostic assessment tool (Lam et al., 2023, Wilson et al., 2020). Documented OUD symptoms can be found in preceding medical record notes that were not transferred over to current medical record notes or in other specialty clinic notes that were scanned in and within the Prescription Drug Monitoring Program, a database for controlled substance prescriptions in the United States, which can also provide context for an OUD diagnosis (Lam et al., 2023, Wilson et al., 2020). A patient’s demographic or medical history of injury, pain, or pain medication-seeking activities could prompt a provider to assess the probability of a patient developing OUD. However, if there is no screening tool to tie all these symptoms together, the current patient context and demographic information alone might not prompt a provider to screen for OUD (Laporte et al., 2022, Nielsen et al., 2020).
Fifteen out of the nineteen identified studies offered a conclusion that tools used for screening or diagnostic assessment of OUD are recommended. Eight studies indicated that assessment tools could be utilized by medical providers for diagnosing OUD and for identifying patient risk indicators for OUD or early signs of misuse (Laporte et al., 2022, Nielsen et al., 2020, Picco et al., 2020a, Picco et al., 2020b, Punches et al., 2024, Staton et al., 2024, Sullivan et al., 2020, Williams et al., 2024). When studying a cohort of 1305 patients who received care in Midwestern United States emergency departments, 14.5 % received a positive DSM-5 OUD assessment, and 10.0 % received a positive OUD diagnosis from the WHO ASSIST survey, which is higher than the national percentage of 6.5–9 % (Punches et al., 2024). These results indicate that if assessment tools were put in place in a standardized manner within medical settings, patients with OUD could be better identified and referred for opioid use-related care.
Four articles reported the unsuccessful implementation of a diagnostic tool. The Prescription Opioid Misuse Index (POMI) did not reach a statistical strength to be identified as useful (Laporte et al., 2022). The Opioid Misuse Measure (COMM) OUD study indicated that the diagnostic assessment tool alone was insufficient for identifying OUD (Wilson et al., 2020). Recommendations included additional follow-up questions from providers in addition to an assessment tool and not relying solely on the results of an assessment tool (Kufeld et al., 2024, Wilson et al., 2020). In addition, the diagnostic assessment tool might not be right for patients managing chronic pain or overly complicated in a primary care setting (Austin et al., 2023, Kufeld et al., 2024).
Within this systematic review, the DSM-5 was implemented most frequently out of the identified articles. The DSM-5 has been used in assessment for other substance use disorders, such as alcohol use disorder and stimulant use disorder (Hallgren et al., 2022, Tiet and Moos, 2021). The number of possible assessment tools is not unique to OUD. A systematic review of depression assessment tools identified 40 unique tools, with the Patient Health Questionnaire-9 being the most frequently utilized (Miller et al., 2021). Given the number of tools and the number of tools recommended by at least one study, finding the right tool for the given setting requires consideration of many factors.
In reviewing studies that implemented an OUD assessment tool, four themes emerged. A majority of the articles indicate that one assessment tool is not sufficient to diagnose OUD and that the OUD diagnosis is not always documented within the EHR. Additional but less common themes were an inconsistent completion of the diagnostic assessment tool, and that OUD symptoms can be documented in multiple ways. However, despite the flaws, the articles indicated that diagnostic assessment tools provide a roadmap for medical providers to follow when assessing the possibility of OUD.
Several studies noted the limitations of a single screening tool and the value of secondary screening. The Tobacco, Alcohol, Prescription Medication, and Other Substances (TAPS) screening tool asks a patient to self-report using non-prescription opioids. However, the TAPS would not capture patients who used their medication as prescribed but felt that the opioids were affecting school, work, or interpersonal relationships. This symptom would be captured on the DSM-5 OUD screening, but the DSM-5 OUD screening would not evaluate the use of other substances. Utilizing the TAPS as a primary screening and, if positive, conducting a secondary screening with another tool may aid in both identifying individuals with OUD and understanding symptoms and severity (Bunting et al., 2023, Carter et al., 2022). The concept of primary and secondary screening is seen elsewhere in the medical field. When screening for breast cancer, an initial screening occurs, and if concerning factors are identified, additional screening is recommended (Barba et al., 2021). This principle can also be applied when screening for OUD. Utilizing machine learning algorithms in EHRs can also assist providers with OUD screening, possibly eliminating the need for multiple screening techniques (Afshar et al., 2025).
Although providers may screen for OUD, documentation of an OUD diagnosis was frequently lacking from the electronic health record (Mallery Lankford et al., 2022, Nielsen et al., 2020). Underrepresentation of diagnoses is not specific to opioid use disorder and can be found with other diseases. Obesity is a chronic disease that can be effectively managed in primary care. Yet, practitioners do not always document a diagnosis of obesity due to hesitation when addressing the diagnosis with individuals or focusing on other patient health issues (Mawardi et al., 2019). Alcohol use disorder is another example of a disease that relies on self-reporting, yet individuals may be hesitant to accurately disclose alcohol consumption for fear of stigma or unwillingness to change, which can lead to fewer documented diagnoses and treatment plans (DiMartini et al., 2022). The lack of a formally documented disease diagnosis may limit the opportunity for opioid use-related care (Staton et al., 2024, Williams et al., 2024). Focusing on patient-centered care techniques and dialogue can assist providers when communicating care plans for patients with OUD (Carpenter et al., 2023, Kelley et al., 2022)
Several studies noted challenges with completing OUD screening, which can result from the busy nature of a practice or if the medical facility is short-staffed or not trained on the diagnostic assessment tool (Austin et al., 2023, Blackstone et al., 2022, Laporte et al., 2022). Medical providers may be busy, especially when triaging patients or evaluating other care concerns (Blackstone et al., 2022, Punches et al., 2024). A study of depression screening in five primary care clinics identified that only 61.03 % of adult patients were up to date on their depression screening (Blackstone et al., 2022). Finding ways to seamlessly implement OUD assessment tools in clinical care by identifying the right amount and types of questions and innovative delivery modes is an ongoing consideration when striving for a high OUD diagnostic assessment completion rate.
In addition to screening tool completion, results must be documented to be useful. Within an electronic health record (EHR), patient care documentation can be duplicated within multiple visit notes, or patient care information can be scattered throughout the EHR, creating provider fatigue and missed opportunities to locate all relevant information associated with patient care (Steinkamp et al., 2022). This review found OUD symptoms and diagnoses to be similarly documented in multiple ways (Kopak et al., 2024, Lam et al., 2023, Laporte et al., 2022, Nielsen et al., 2020, Staton et al., 2024, Sullivan et al., 2020, Wilson et al., 2020). Integrating multiple documentation sources with the assistance of artificial intelligence may be helpful to medical providers when diagnosing OUD (Gabriel et al., 2025)
Overall, screening or diagnostic assessment tools for OUD were recommended in various settings, including primary care, the emergency room, and psychiatry. In these same settings, screenings for depression can also occur successfully (Beard et al., 2016, Ford et al., 2020, Shyman et al., 2021). Like depression screenings, screenings for OUD can help identify individuals who may have OUD and prompt clinicians to connect them to additional resources. Although barriers to screening and documentation of OUD symptoms and diagnosis exist, this systematic review found several assessment tools and the process of screening to be beneficial. Creating clinical guidelines for OUD screening and referral that include a standardized opioid use disorder screening or diagnostic assessment tool could help to identify more individuals with OUD and connect them to opioid use-related care.
Areas for future study that should be considered are reviewing best practices for collecting previous medical history, including the emerging field of artificial intelligence for assistance with OUD diagnosis, and researching patient-centered care in diagnosing OUD. An EHR is equipped to provide different assessments for varying symptoms (Hooker et al., 2024). With limited time, providers can become fatigued with the number, or the time involved in completing assessments. Researching how assessment fatigue can be minimized by utilizing artificial assistance and machine learning techniques for OUD screening could help in reducing the barriers to the successful implementation of a diagnostic assessment tool.
In addition, collecting previous medical and social history to support an OUD diagnosis was highlighted within this literature review as beneficial to use in conjunction with a self-reported diagnostic assessment tool. Researching the stigma behind OUD, as well as reasons why medical providers do not document an OUD diagnosis, such as fear of the breakdown of the provider-patient relationship or how an OUD diagnosis will affect an individual professionally or personally, should be considered (Fortney et al., 2023). Acknowledging that stigma exists and incorporating patient-centered care approaches to assist individuals when receiving opioid use-related care without negative personal or professional consequences should be explored (Carpenter et al., 2023, Kelley et al., 2022).
Within this review, OUD diagnostic assessment tools were implemented in different health provider locations, such as primary care, emergency departments, psychiatric care, and pharmacies. Although primary care was the most studied clinical area for implementing an OUD screening or diagnostic assessment tool, the selected articles did not identify one type of medical setting as better or more appropriate than another medical setting. Although the literature review articles consistently recommended that an OUD diagnostic assessment tool should be concise for effective completion and collection within a clinical setting, the best method for collecting completed assessment surveys was not identified. The sustainability of collecting and documenting an OUD diagnostic assessment tool was not discussed within this literature review; rather, the articles focused solely on whether the assessment tools were helpful when identifying individuals at risk for OUD. In addition, the search criteria for this systematic review included original research articles only, which limited the authors’ ability to back-search other article types for referenced articles that might fall into this study’s inclusion criteria.
The articles identified within this literature review support the conclusion that diagnosing opioid use disorder is a multifaceted and complicated undertaking. The majority of OUD diagnostic assessments rely on patients’ self-reporting symptoms of OUD. Due to this, the nature in which the assessments are collected and collecting the assessments in a nonjudgmental manner within a safe space is important for individuals to feel comfortable when disclosing symptoms to a healthcare provider. Overall, OUD diagnostic assessment tools within this literature review were recognized as a value-added endeavor to identify individuals with OUD and introduce those individuals to opioid use-related care.
The authors of the above-named manuscript, submitted for publication to Drug and Alcohol Dependence, have no conflict of interests to declare and nothing to disclose. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Tiffany Champagne-Langabeer: Writing – review & editing, Supervision. Audrey Sarah Cohen: Writing – review & editing, Visualization, Validation. Audrey Lopez: Writing – original draft, Methodology, Formal analysis, Conceptualization. Francine Vega: Writing – review & editing, Validation.
This work did not receive external funding.
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.