Abstract
Introduction:
Community pharmacies are emerging as a valuable setting to identify patients with substance use. Few tools have been specially validated to screen patients in these settings, particularly among those prescribed opioid medications. The goal of this study was to validate the performance of the Tobacco, Alcohol, Prescription medication, and other Substance use (TAPS) tool in community pharmacy settings compared to a reference-standard substance use assessment.
Methods:
Participants were recruited while receiving opioid medications (not solely buprenorphine) from 19 pharmacies from a large national chain in Ohio and Indiana. Adults who were not involved in the criminal justice system or receiving cancer treatment were invited to participate in a one-time, cross-sectional, self-administered, health survey which included the TAPS tool. Substance use risks calculated from the TAPS tool were compared with the reference standard, World Health Organization Alcohol, Smoking, and Substance Involvement Screening Test (ASSIST) using confusion matrices. We calculated Areas Under the Curve (AUC) of Receiver Operating Characteristics Curves (ROC) to evaluate the TAPS tool’s validity.
Results:
The TAPS tool showed fair or better discrimination between moderate-risk use and high-risk use for tobacco, alcohol, and prescription opioids (AUCs: 0.75–0.97 and fair or better discrimination between low-risk and moderate-risk use in five of eight subscales, including tobacco, alcohol, marijuana, stimulants, and heroin (AUCs: 0.70–0.92).
Conclusion:
The TAPS tool detected clinically relevant problem substance use in several drug classes and likely would be a valuable assessment for screening illicit drug use among community pharmacy patients prescribed opioid medications.
Keywords: Substance use disorders, screening, unhealthy substance use, community pharmacy intervention, opioid use
Introduction
Substance use in the US continues to be a significant burden on public health, with 20.8% of Americans reporting illicit drug use in the past year in 2019 (SAMHSA, 2020). While substance use disorder (SUD) rates remained stable between 2015–2019, all states reported a spike in overdose-related deaths during 2020 (American Medical Association, 2021). These increases are particularly poignant amid the continued US opioid epidemic—with polysubstance overdose, including opioids, having a significant negative impact across the US (Cicero, Ellis, & Kasper, 2020; Compton, Valentino, & DuPont, 2021; Ferries et al., 2017; Paulozzi, Strickler, Kreiner, & Koris, 2015).
Screening and intervention are critical in detecting substance-related problems and evaluating starting points for treatment. One measure emerging as a valuable tool for assessing substance use risk in healthcare settings is the Tobacco, Alcohol, Prescription Medication, and other Substance use (TAPS) tool, a brief 4-item screener (McNeely et al., 2016). While maintaining brevity to meet medical provider workflow demands, the TAPS tool provides information about specific substances used and patient risk level to guide clinical decisions for care.
Prior studies validating the TAPS tool have primarily taken place in primary care settings. In addition to primary care, community pharmacies are emerging as valuable service settings to identify patients with SUD (Compton, Jones, Stein, & Wargo, 2019; Dhital, Whittlesea, Norman, & Milligan, 2010; Hattingh & Tait, 2017), with these care settings being highly accessible, as most Americans (>90%) live within a five-minute drive (National Chain Drug Stores, 2011), and pharmacists are among the most trusted professionals in the US (Crossley, 2019). Moreover, pharmacists have specialized training and skill in counseling patients in safe and effective use of medications, and prior research shows promising evidence for pharmacy led or involved SUD interventions (Afzal, Pogge, & Boomershine, 2017; Cochran et al., 2019; Dhital, Whittlesea, Norman, & Milligan, 2010). Indeed, pharmacists are a last point of care before dispensation of medications that may result in harmful interactions, such as opioids, benzodiazepines, muscle relaxants, and stimulants with other substances such as illicit opioids and alcohol. Such screening and intervention may be particularly pertinent among patients prescribed opioid medications, given heightened risk for adverse events, such as addiction and overdose (Cicero et al., 2020; Compton et al., 2021; Ferries et al., 2017; Paulozzi et al., 2015).
Given these advancements for community pharmacy, validation of screening tools, including the TAPS tool, in community pharmacies has potential to provide pharmacists with actionable data to better provide care for individuals prescribed opioid medications who also use other substances. This study aimed to validate the TAPS tool using the World Health Organization Alcohol, Smoking, and Substance Involvement Screening Test (ASSIST) as the reference-standard in this population.
Methods
Design, Sites, and Participants
Data for the current study was derived from a cross-sectional one-time, self-administered survey (surveys administered sequentially) among adult patients filling opioid prescriptions for pain at 19 pharmacies in Ohio and Indiana within a large US chain from November 2019 to October 2020 (NCT: 03936985); details of which have been reported elsewhere (Cochran et al., In press). At point-of-dispensing, pharmacy staff provided study information to individuals receiving opioid medications. Interested patients entered contact information using a study-provided computer tablet or received a flyer with information on initiating the survey from personal electronic devices. Once patients submitted their contact information, they received an automatic email including a brief study overview and a link to the consent document. After reviewing study information and completing informed consent, participants were directed to complete a study eligibility self-screening assessment. The study excluded individuals <18 years old, non-English speaking, or solely filling buprenorphine or buprenorphine combination products (given some patients may use only for addiction treatment and others only for pain, which distinction may not be clear cut for patients or for dispensing pharmacists without medical record access). Additionally, the study excluded individuals receiving current cancer treatment, those who had previously completed the survey, or had current involvement with the criminal justice system (self-reported). Individuals meeting all study inclusion/exclusion criteria were advanced to the study survey. Those who completed the survey were provided with a $50 gift card. The University of Cincinnati and the University of Utah Institutional Review Boards approved this project.
Measures
Experimental measure.
The TAPS tool is a two-step screening and assessment instrument (McNeely et al., 2016). First, participants are asked frequency of use of tobacco, alcohol, illicit drugs (marijuana, cocaine, heroin, methamphetamine, hallucinogens, and ecstasy), and non-medical use of prescription medications (sedatives, opioids, and attention-deficit/hyperactivity disorder [ADHD] medications) in the past 12 months. Second, participants are asked frequency of use within the past three months as well as 2–3 substance-specific items. Item responses are summed to calculate substance-specific risk scores. The scored categories included low=0, moderate=1, and high ≥2 risk.
Reference measure.
The ASSIST was the reference-standard against which the TAPS tool was compared. The ASSIST has been found to have criterion, construct, concurrent, and discriminant validity (Humeniuk & Ali, 2006; Humeniuk et al., 2008). The ASSIST used herein was the version with expanded medication categories, validated by McNeely et al. (2016), which includes an initial screening question inquiring about lifetime use of tobacco, alcohol, cannabis, cocaine, prescription stimulants, methamphetamine, inhalants, sedatives, hallucinogens, prescription opioids, street opioids, and other drugs. Respondents then answer Likert-scale items regarding substance use behaviors within the past 3 months. Responses were calculated into three discrete ASSIST risk categories. The categories included low (0–3, all drugs; 0–10, alcohol), moderate (4–26, all drugs; 11–26, alcohol), and high (27–33) risk.
Demographics.
Participant demographics reported herein include age (years), sex (male vs female), ethnicity (Hispanic vs non-Hispanic), race (White, Black, other, see Limitations), marital status (married vs. not married), employment status (full-time vs. not full-time employment), and insurance status (insured vs. not insured).
Statistical Analyses
We summarized baseline demographic variables using means and counts. We used frequency counts and percentages to create confusion matrices reflecting agreement between the TAPS tool and ASSIST. We calculated Area Under the Curve (AUC) of Receiver Operating Characteristics Curve (ROC), which evaluated the ability of the TAPS tool in discriminating different drug use risk levels (low, moderate, high), using ASSIST-defined risk levels as reference. Additionally, we calculated AUC values for each subdomain to discriminate low vs moderate/high-risk use cases and high vs low-/moderate-risk use cases. Evaluation of all curves used the following scale: <0.70=poor, ≥0.70=fair, ≥0.80 good ≥0.9=excellent (Hajian-Tilaki, 2013; Swets, Dawes, & Monahan, 2000; Youngstrom, 2014).
Results
Participants
There were 1,523 participants who completed the assessment. Participants were on average 49 years (SD=14.9), and most were female (61.9%). The study population was 92.5% White, 4.5% Black/African American, and 0.9% Hispanic. Most participants were married (54.6%), insured (93.8%), and were not employed full-time (64.6%).
Comparing substance use risk levels
Confusion matrices for the agreement between the TAPS tool and the ASSIST showed tobacco, marijuana, stimulants, heroin, and ADHD medication subscales had TAPS low-risk marginal percentages comparable to their respective ASSIST low-risk marginal percentages (69% vs 63%, 89% vs 89%, 99% vs 97%, 99% vs 99%, and 100% vs 96%, respectively; Table 1). However, low-risk marginal percentages between the TAPS tool and ASSIST differed more dramatically for alcohol (69% vs 90%), prescription opioids (96% vs 55%), and sedatives (97% vs 83%).
Table 1:
Frequency of Self-reported Drug Use based on TAPS tool and WHO ASSIST
| ASSIST Risk Level | ||||
|---|---|---|---|---|
| TAPS Risk Level | Low Risk | Moderate Risk | High Risk | Total |
| Tobacco A | ||||
| Low Risk | 911 (62%) | 103 (7.0%) | 2 (0.1%) | 1,016 (69%) |
| Moderate Risk | 12 (0.8%) | 80 (5.4%) | 2 (0.1%) | 94 (6.4%) |
| High Risk | 7 (0.5%) | 323 (22%) | 40 (2.7%) | 370 (25%) |
| Total | 930 (63%) | 506 (34%) | 44 (3.0%) | 1,480 (100%) |
| Alcohol B | ||||
| Low Risk | 1,016 (68%) | 14 (0.9%) | 0 (0%) | 1,030 (69%) |
| Moderate Risk | 159 (11%) | 29 (1.9%) | 0 (0%) | 188 (13%) |
| High Risk | 173 (12%) | 74 (5.0%) | 25 (1.7%) | 272 (18%) |
| Total | 1,348 (90%) | 117 (7.9%) | 25 (1.7%) | 1,490 (100%) |
| Marijuana C | ||||
| Low Risk | 1,297 (87%) | 25 (1.7%) | 0 (0%) | 1,322 (89%) |
| Moderate Risk | 25 (1.7%) | 55 (3.7%) | 1 (<0.1%) | 81 (5.4%) |
| High Risk | 3 (0.2%) | 77 (5.2%) | 8 (0.5%) | 88 (5.9%) |
| Total | 1,325 (89%) | 157 (11%) | 9 (0.6%) | 1,491 (100%) |
| Stimulant D | ||||
| Low Risk | 1,465 (97%) | 24 (1.6%) | 0 (0%) | 1,489 (99%) |
| Moderate Risk | 4 (0.3%) | 6 (0.4%) | 0 (0%) | 10 (0.7%) |
| High Risk | 0 (0%) | 5 (0.3%) | 5 (0.3%) | 10 (0.7%) |
| Total | 1,469 (97%) | 35 (2.3%) | 5 (0.3%) | 1,509 (100%) |
| Heroin E | ||||
| Low Risk | 1,478 (99%) | 11 (0.7%) | 1 (<0.1%) | 1,490 (99%) |
| Moderate Risk | 0 (0%) | 1 (<0.1%) | 0 (0%) | 1 (<0.1%) |
| High Risk | 0 (0%) | 2 (0.1%) | 5 (0.3%) | 7 (0.5%) |
| Total | 1,478 (99%) | 14 (0.9%) | 6 (0.4%) | 1,498 (100%) |
| RX Opioids F | ||||
| Low Risk | 801 (54%) | 593 (40%) | 18 (1.2%) | 1,412 (96%) |
| Moderate Risk | 9 (0.6%) | 23 (1.6%) | 5 (0.3%) | 37 (2.5%) |
| High Risk | 0 (0%) | 9 (0.6%) | 15 (1.0%) | 24 (1.6%) |
| Total | 810 (55%) | 625 (42%) | 38 (2.6%) | 1,473 (100%) |
| Sedatives G | ||||
| Low Risk | 1,232 (82%) | 221 (15%) | 7 (0.5%) | 1,460 (97%) |
| Moderate Risk | 6 (0.4%) | 9 (0.6%) | 0 (0%) | 15 (1.0%) |
| High Risk | 3 (0.2%) | 16 (1.1%) | 6 (0.4%) | 25 (1.7%) |
| Total | 1,241 (83%) | 246 (16%) | 13 (0.9%) | 1,500 (100%) |
| ADHD Medication H | ||||
| Low Risk | 1,438 (96%) | 61 (4.1%) | 0 (0%) | 1,499 (100%) |
| Moderate Risk | 1 (<0.1%) | 3 (0.2%) | 0 (0%) | 4 (0.3%) |
| High Risk | 0 (0%) | 2 (0.1%) | 0 (0%) | 2 (0.1%) |
| Total | 1,439 (96%) | 66 (4.4%) | 0 (0%) | 1,505 (100%) |
TAPS Tobacco compared to ASSIST Tobacco.
TAPS Alcohol compared to ASSIST Alcohol.
TAPS Marijuana. compared to ASSIST Cannabis.
TAPS Cocaine, Crack, Meth compared to ASSIST Cocaine, Meth.
TAPS Heroin compared to ASSIST Street Opiates.
TAPS Rx Opiates compared to ASSIST Rx Opiates.
TAPS Sedatives compared to ASSIST Sedatives.
TAPS ADHD Medication compared to ASSIST Rx Stimulant.
TAPS tool discriminating between individual risk levels
ROC analyses demonstrated the TAPS tool discriminated well between individual levels of risk use. It demonstrated ≥fair discrimination between moderate-risk use and high-risk use for tobacco (AUC=0.86, 95% CI=0.81–0.91), alcohol (AUC=0.97, 95% CI=0.96–0.99), marijuana (AUC=0.98, 95% CI=0.96–0.99), stimulants (AUC=0.99, 95% CI=1.00–1.00), heroin (AUC=0.91, 95% CI=0.75–1.00), prescription opioids (AUC=0.75, 95% CI=0.67–0.83), and sedatives (AUC=0.89, 95% CI=0.69–1.00). Similarly, the TAPS tool demonstrated ≥fair discrimination between low-risk and moderate-risk use in five of eight subscales, including tobacco (AUC=0.89, 95% CI=0.87–0.91), alcohol (AUC=0.88, 95% CI=0.85–0.91), marijuana (AUC=0.92, 95% CI=0.89–0.95), stimulants (AUC=0.70, 95% CI=0.62–0.78), and heroin (AUC=0.70, 95% CI=0.59–0.81). However, two TAPS subscales discriminated poorly between low-risk and moderate-risk use, including prescription opioids (AUC=0.53, 95% CI=0.54–0.58), sedatives (AUC=0.56, 95% CI=0.54–0.58), and ADHD medications (AUC=0.54, 95% CI=0.51–0.57). Additionally, the TAPS tool discriminated poorly between low-risk and moderate-risk use in the ADHD medication subscale (AUC=0.54, 95% CI=0.51–0.57).
TAPS tool for discriminating between dichotomous risk levels
Table 2 shows the TAPS tool, when comparing low-risk vs moderate/high-risk use, had fair or better discrimination for tobacco (AUC=0.71, 95% CI=0.67–0.75), alcohol (AUC=0.77, 95% CI=0.70–0.84), marijuana (AUC=0.84, 95% CI=0.78–0.89) use, and heroin (AUC 0.70, 95% CI=0.59–0.81). Similarly, when comparing low/moderate-risk vs high-risk use, the TAPS tool demonstrated ≥fair discrimination for tobacco (AUC=0.71 95% CI=0.67–0.75), alcohol (AUC=0.77, 95% CI=0.70–0.84), and marijuana (AUC=0.84, 95% CI=0.78–0.89). In the prescription opioid subclass, the TAPS tool performed better when discriminating low/moderate vs high-risk use (AUC=0.69, 95% CI: 0.62–0.77) than when discriminating only low vs moderate/high-risk use (AUC=0.53, 95% CI: 0.52–0.54). However, in the sedatives class, the TAPS tool discriminated similarly whether considering low-risk vs moderate-risk use or low-risk vs moderate/high-risk use (AUCs=0.56 and 0.56, 95% CIs: 0.54–0.58 and 0.54–0.58, respectively).
Table 2:
TAPS Discriminating Substance Risk Level Against the WHO ASSIST
| Comparison | Sensitivity, Specificity | AUC | SE | 95% CI | Agreement |
|---|---|---|---|---|---|
| High vs Moderate Risk | 0.20, 0.95 | 0.86 | 0.03 | (0.81, 0.91) | Good |
| Moderate vs Low Risk | 0.99, 0.44 | 0.89 | 0.01 | (0.87, 0.91) | Good |
| High vs Low/Moderate Risks | 0.44, 0.98 | 0.71 | 0.02 | (0.67–0.75) | Good |
| Low vs Moderate/High Risks | 0.44, 0.99 | 0.71 | 0.02 | (0.68–0.75) | Fair |
| High vs Moderate Risk | 0.28, 1.00 | 0.97 | 0.01 | (0.96, 0.99) | Excellent |
| Moderate vs Low Risk | 0.86, 0.67 | 0.88 | 0.02 | (0.85, 0.91) | Good |
| High vs Low/Moderate Risks | 0.67, 0.86 | 0.77 | 0.04 | (0.70–0.84) | Good |
| Low vs Moderate/High Risks | 0.67, 0.86 | 0.77 | 0.04 | (0.70–0.84) | Fair |
| High Risk vs Moderate Risk | 0.42, 0.89 | 0.98 | 0.01 | (0.96, 0.99) | Excellent |
| Moderate Risk vs Low Risk | 0.98, 0.69 | 0.92 | 0.01 | (0.89, 0.95) | Excellent |
| High Risk vs Low/Moderate Risks | 0.69, 0.98 | 0.84 | 0.03 | (0.78–0.89) | Good |
| Low Risk vs Moderate/High Risks | 0.69, 0.98 | 0.84 | 0.03 | (0.79–0.89) | Good |
| High Risk vs Moderate Risk | 0.54, 1.00 | 0.99 | 0.001 | (1.00, 1.00) | Excellent |
| Moderate Risk vs Low Risk | 0.99, 0.20 | 0.70 | 0.04 | (0.62, 0.78) | Fair |
| High Risk vs Low/Moderate Risks | 0.00, 0.99 | 0.50 | 0.001 | (0.49–0.50) | Poor |
| Low Risk vs Moderate/High Risks | 0.38, 0.99 | 0.69 | 0.07 | (0.55, 0.82 | Poor |
| High vs Moderate Risk | 0.33, 1.00 | 0.91 | 0.08 | (0.75, 1.00) | Excellent |
| Moderate vs Low Risk | 1.00, 0.08 | 0.70 | 0.06 | (0.59, 0.81) | Fair |
| High vs Low/Moderate Risks | 0.07, 0.99 | 0.54 | 0.04 | (0.47–0.67) | Poor |
| Low vs Moderate/High Risks | 0.40, 0.99 | 0.70 | 0.06 | (0.59, 0.81) | Fair |
| High vs Moderate Risk | 0.72, 0.75 | 0.75 | 0.04 | (0.67, 0.83) | Fair |
| Moderate vs Low Risk | 0.99, 0.04 | 0.53 | 0.01 | (0.52, 0.54) | Poor |
| High vs Low/Moderate Risks | 0.39, 0.99 | 0.69 | 0.04 | (0.62–0.77) | Poor |
| Low vs Moderate/High Risks | 0.08, 0.99 | 0.53 | 0.01 | (0.52–0.54) | Poor |
| High vs Moderate Risk | 0.36, 1.00 | 0.89 | 0.1 | (0.69, 1.00) | Good |
| Moderate vs Low Risk | 0.99, 0.04 | 0.56 | 0.01 | (0.54–0.58) | Poor |
| High vs Low/Moderate Risks | 0.04, 0.99 | 0.52 | 0.01 | (0.50–0.53) | Poor |
| Low vs Moderate/High Risks | 0.12, 0.99 | 0.56 | 0.01 | (0.54–0.58) | Poor |
| High vs Moderate Risk | Not calculated* | Not calculated* | |||
| Moderate vs Low Risk | 0.99, 0.5 | 0.54 | 0.02 | (0.51, 0.57) | Poor |
| High vs Low/Moderate Risks | Not calculated* | Not calculated* | |||
| Low vs Moderate/High Risks | 0.99, 0.5 | 0.54 | 0.02 | (0.51, 0.57) | Poor |
Not calculated given insufficient sample size
Discussion
This study assessed the validity of the TAPS tool as a screener for substance use risk among community pharmacy patients filling opioid medications. Similar to past research, results demonstrated good to excellent validity (AUC ≥ 0.80) of the TAPS tool for discriminating risk for tobacco, alcohol, and marijuana drug use (McNeely et al., 2016). Furthermore, TAPS had good to excellent discrimination (AUC ≥ 0.80) for determining high vs moderate and moderate vs low-risk levels for stimulants and had fair to excellent discrimination (AUC ≥ 0.70) for determining high vs moderate and moderate vs low-risk levels for heroin. Also similar to original validation analyses, TAPS had poor discrimination between most risk levels for prescription opioids, sedatives, and ADHD medication use, as noted by previous authors (McNeely et al., 2016; Schwartz et al., 2017). For these substances, TAPS tended to overestimate the number of participants who were low-risk according to the ASSIST. The TAPS tool also overestimated the number of moderate-risk participants in most subscales, and comparing between risk levels dichotomously rather than comparing between individual risk levels decreased efficacy of the TAPS Tool for most substances. Future research may seek to recalibrate items or rephrase instructions to respondents, particularly in health care settings, to ensure understanding of these items and repercussions of affirming use.
Noted by McNeely and colleagues, the TAPS tool has several characteristics making it attractive as a screening tool in primary care settings, such as being easily integrated into workflow (McNeely et al., 2016). Our results suggest benefits of the TAPS tool in primary care settings possibly could be extended to pharmacy settings among patients filling opioid medications for assessing non-prescription substance use. For instance, pharmacy staff may likewise integrate the TAPS tool into workflows for examining medication safey, such as when checking prescription drug monitoring programs (PDMP). As a complementary tool to PDMPs, the TAPS tool may aid pharmacists identifying concomitant, non-prescription substance use, making way for subsequent brief interventions and/or referrals to treatment. An additional benefit of utilizing the TAPS tool in pharmacies is it assesses all major drug classes in a single instrument and can be self-administered (McNeely et al., 2016)—limiting staff burden within busy pharmacy environments. Following the U.S. Preventive Services Task Force recommendations for universal screening and brief motivational conversations targeting low levels of use or potential treatment linkage (United States Preventive Services Task Force, 2020), responses to positive screenings could follow more traditional Screening, Brief Intervention, and Referral to Treatment models or potentially more substance specific interdisciplinary models specially tailored for community pharmacy settings (Cochran et al., 2019).
Limitations
While this study possesses strengths, results should be considered in relation to its limitations. This study had a robust sample size, but it did not have sufficient participants who had high-risk use of ADHD medications to assess validity. Additionally, while marijuana, stimulants, heroin, and sedatives classes had excellent validity in discriminating between medium and high-risk use (AUC≥0.89, respectively), their high-risk use populations according to the ASSIST were all small (Table 2). Further research should be conducted to assess validity of the TAPS tool for patients using these drugs, particularly regarding high-risk use. Furthermore, the full ASSIST was administered followed by the TAPS—thus we cannot rule out an order effect influencing our results. It could also be that a clinical diagnostic assessment of substance and other medication use may yield greater insight into actual use patterns and thus increase insight into the validity of the TAPS tool. An additional limitation is homogeneity of the racial/ethnic distribution of participants, most being White within the Ohio and Indiana pharmacy sites utilized herein. Next steps should examine racial/ethnic performance differences as well as additional states/regions. Performance differences by race would be especially important given disparities noted for people of color related to the opioid epidemic and pain management (Badreldin, Grobman, & Yee, 2019; Hewes, Dai, Mann, Baca, & Taillac, 2018). However, trends in our analyses are consistent with original validation studies, which speaks to external validity of these results (McNeely et al., 2016; Schwartz et al., 2017).
Conclusions
With increasing public health burden of substance misuse and related overdose mortality, detection and screening for substance-related disorders is vital. This study furthers support for use of the TAPS tool for non-prescription substance use, specifically in pharmacy-based settings. Results show promise for the TAPS tool as a potential instrument for integration into community pharmacies for substance use screening among patients filling opioid medications.
Highlights.
The TAPS tool is an attractive screening tool in pharmacy settings and can be integrated into existing workflows.
The TAPS tool has good to excellent validity for tobacco, alcohol, and marijuana use identification.
The TAPS tool had fair validity for stimulant and heroin drug use identification.
Future research should work to improve the TAPS for prescription opioid, sedative, and prescription stimulant identification.
Funding:
This study was supported by the National Institute on Drug Abuse (UG1DA013732; UG1DA049444).
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflicts: None
CRediT author statement
Gentry Carter: Validation, Writing – Original Draft, Visualization, Ziji Yu: Formal analysis, Writing – Review and Editing, M. Aryana Bryan: Project administration, Data Curation, Writing – Review and Editing, Jennifer L. Brown: Conceptualization, Methodology, Investigation, Writing – Review and Editing, Funding acquisition, T. Winhusen: Conceptualization, Methodology, Investigation, Writing – Review and Editing, Supervision, Funding acquisition, Gerald Cochran: Conceptualization, Methodology, Investigation, Writing – Review and Editing, Supervision, Funding acquisition
References
- Afzal Z, Pogge E, & Boomershine V (2017). Evaluation of a Pharmacist and Nurse Practitioner Smoking Cessation Program. Journal of Pharmacy Practice, 30(4), 406–411. doi: 10.1177/0897190016659221 [DOI] [PubMed] [Google Scholar]
- American Medical Association. (2021). Issue brief: Drug overdose epidemic worsened during COVID pandemic. Retrieved from Chicago, IL: https://www.ama-assn.org/system/files/2020-12/issue-brief-increases-in-opioid-related-overdose.pdf
- Badreldin N, Grobman WA, & Yee LM (2019). Racial Disparities in Postpartum Pain Management. Obstetrics & Gynecology, 134(6), 1147–1153. doi: 10.1097/aog.0000000000003561 [DOI] [PMC free article] [PubMed] [Google Scholar]
- National Chain Drug Stores. (2011). 2011–2012 Chain Pharmacy Industry Profile. National Chain Drug Stores, Alexandria, VA. [Google Scholar]
- Cicero TJ, Ellis MS, & Kasper ZA (2020). Polysubstance Use: A Broader Understanding of Substance Use During the Opioid Crisis. American Journal of Public Health, 110(2), 244–250. doi: 10.2105/AJPH.2019.305412 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cochran G, Brown J, Z. Y, Frede S, Bryan A, Ferguson A, … Winhusen T (In press). Validation and Threshold Identification of a Prescription Drug Monitoring Program Clinical Opioid Risk Metric with the WHO Alcohol, Smoking, and Substance Involvement Screening Test. Drug And Alcohol Dependence. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cochran G, Chen Q, Field C, Seybert AL, Hruschak V, Jaber A, … Tarter R (2019). A community pharmacy-led intervention for opioid medication misuse: A small-scale randomized clinical trial. Drug and Alcohol Dependence, 205, 107570. doi: 10.1016/j.drugalcdep.2019.107570 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Compton WM, Jones CM, Stein JB, & Wargo EM (2019). Promising roles for pharmacists in addressing the U.S. opioid crisis. Research in Social and Administrative Pharmacy, 15(8), 910–916. doi: 10.1016/j.sapharm.2017.12.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Compton WM, Valentino RJ, & DuPont RL (2021). Polysubstance use in the U.S. opioid crisis. Molecular Psychiatry, 26(1), 41–50. doi: 10.1038/s41380-020-00949-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Crossley K (2019). Public Perceives Pharmacists as Some of the Most Trusted Professionals. Pharmacy Times. Retrieved from https://www.pharmacytimes.com/publications/career/2019/CareersWinter19/public-perceives-pharmacists-as-some-of-the-most-trusted-professionals
- Dhital R, Whittlesea CM, Norman IJ, & Milligan P (2010). Community pharmacy service users’ views and perceptions of alcohol screening and brief intervention. Drug and Alcohol Review, 29(6), 596–602. doi: 10.1111/j.1465-3362.2010.00234.x [DOI] [PubMed] [Google Scholar]
- Ferries EA, Gilson AM, Aparasu RR, Chen H, Johnson ML, & Fleming ML (2017). The Prevalence of and Factors Associated With Receiving Concurrent Controlled Substance Prescriptions. Substance Use & Misuse., 1–7. doi: 10.1080/10826084.2017.1298617 [DOI] [PubMed] [Google Scholar]
- Hajian-Tilaki K (2013). Receiver Operating Characteristic (ROC) Curve Analysis for Medical Diagnostic Test Evaluation. Caspian Journal of Internal Medicine, 4(2), 627–635. [PMC free article] [PubMed] [Google Scholar]
- Hattingh HL, & Tait RJ (2017). Pharmacy-based alcohol-misuse services: current perspectives. Integrated Pharmacy Research and Practice, 7, 21–31. doi: 10.2147/iprp.S140431 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hewes HA, Dai M, Mann NC, Baca T, & Taillac P (2018). Prehospital Pain Management: Disparity By Age and Race. Prehospital Emergency Care, 22(2), 189–197. doi: 10.1080/10903127.2017.1367444 [DOI] [PubMed] [Google Scholar]
- Humeniuk R, & Ali R (2006). Validation of the Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) and Pilot Brief Intervention: A Technical Report of Phase II Findings of the WHO ASSIST Project. Retrieved from Geneva:
- Humeniuk R, Ali R, Babor TF, Farrell M, Formigoni ML, Jittiwutikarn J, … Simon S (2008). Validation of the Alcohol, Smoking And Substance Involvement Screening Test (ASSIST). Addiction, 103(6), 1039–1047. doi: 10.1111/j.1360-0443.2007.02114.x [DOI] [PubMed] [Google Scholar]
- McNeely J, Strauss SM, Rotrosen J, Ramautar A, & Gourevitch MN (2016). Validation of an audio computer-assisted self-interview (ACASI) version of the alcohol, smoking and substance involvement screening test (ASSIST) in primary care patients. Addiction, 111(2), 233–244. doi: 10.1111/add.13165 [DOI] [PMC free article] [PubMed] [Google Scholar]
- McNeely J, Wu L-T, Subramaniam G, Sharma G, Cathers LA, Svikis D, … Schwartz RP (2016). Performance of the Tobacco, Alcohol, Prescription Medication, and Other Substance Use (TAPS) Tool for Substance Use Screening in Primary Care Patients. Annals of Internal Medicine, 165(10), 690. doi: 10.7326/m16-0317 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Paulozzi LJ, Strickler GK, Kreiner PW, & Koris CM (2015). Controlled Substance Prescribing Patterns--Prescription Behavior Surveillance System, Eight States, 2013. Morbidity And Mortality Weekly Report. Surveillance Summaries (Washington, D.C.: 2002), 64(9), 1–14. doi: 10.15585/mmwr.ss6409a1 [DOI] [PubMed] [Google Scholar]
- SAMHSA. (2020). Key substance use and mental health indicators in the United States: Results from the 2019 National Survey on Drug Use and Health (HHS Publication No. SMA 18–5068, NSDUH Series H-53). Retrieved from Rockville, MD: https://www.samhsa.gov/data/report/2017-nsduh-annual-national-report
- Schwartz RP, McNeely J, Wu LT, Sharma G, Wahle A, Cushing C, … Subramaniam GA (2017). Identifying substance misuse in primary care: TAPS Tool compared to the WHO ASSIST. Journal of Substance Abuse Treatment, 76, 69–76. doi: 10.1016/j.jsat.2017.01.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Swets JA, Dawes RM, & Monahan J (2000). Psychological Science Can Improve Diagnostic Decisions. Psychological Science in the Public Interest, 1(1), 1–26. doi: 10.1111/1529-1006.001 [DOI] [PubMed] [Google Scholar]
- United States Preventive Services Task Force. (2020). Screening for Unhealthy Drug Use: US Preventive Services Task Force Recommendation Statement. Journal of the American Medical Association, 323(22), 2301–2309. doi: 10.1001/jama.2020.8020 [DOI] [PubMed] [Google Scholar]
- Youngstrom EA (2014). A Primer on Receiver Operating Characteristic Analysis and Diagnostic Efficiency Statistics for Pediatric Psychology: We Are Ready to ROC. Journal of Pediatric Psychology, 39(2), 204–221. doi: 10.1093/jpepsy/jst062 [DOI] [PMC free article] [PubMed] [Google Scholar]