Primary outcome
The primary outcome is the development of opioid misuse or opioid use disorder during 24 weeks follow up, as measured by the Composite International Diagnostic Interview-Substance Abuse Module (CIDI-SAM) and/or the current opioid misuse measure (COMM).
The CIDI-SAM is a computer-assisted interview (average test–retest reliability
Κ = 0.84) that identifies opioid use disorder presence and severity, based on symptom clusters, according to the 5th Edition of the Diagnostic and Statistical Manual of Mental Disorders. [
10,
12,
13] The COMM is a 17-item paper survey with 5-point Likert scale response options, which will be used to identify the presence and severity opioid misuse behaviors. In order to determine changes in these behaviors over time, COMM will be administered at each follow up. The COMM has acceptable one-week test–retest reliability and internal validity (Intra-class Correlation Coefficient [ICC] = 0.86; 95% CI 0.77–0.92) [
14,
15]. A raw score of greater than or equal to 9 will be considered positive for opioid misuse.
The composite outcome will be positive for participants with positive scores at week 24 on either the CIDI-SAM (opioid use disorder) or the COMM (opioid misuse), and will be negative for participations with neither.
Independent variables
A number of independent variables will be collected throughout the 24 weeks follow up. The inclusion of these measures are based on evidence from literature, and the surveys and interviews with Wisconsin trauma center staff and coordinators that were conducted as an earlier part of this study. The follow up time points were determined by the need to regularly evaluate for ongoing opioid use.
The baseline measures will be assessed for the development of a screening tool that could be administered at the point of care for a traumatic injury. Measures administered at follow up time points will be assessed for the development of a screening tool that could be administered by trauma staff at post-discharge follow up appointments or by primary care after a handoff from a trauma service to a clinical service.
Sociodemographic information (including age, sex, race, place of birth, work status, marital status, and household income and inhabitants) are collected at baseline. Additional baseline measures include: depression screen (PHQ-9); anxiety screen (GAD-7); Opioid Risk Tool (ORT); Social Support Questionnaire (ISEL-12); description of mechanism of injury; whether or not the injury was work related; number of procedures and/or complications associated with the injury; medication allergies; and time spent on an intensive care unit.
Ongoing opioid use after hospital discharge is recorded at each visit in order to establish a Morphine Equivalent Daily Dose (MEDD) per time point, as well as to determine the date of last opioid use.
The Pain Medication Questionnaire (PMQ) [
16,
17] is a 23-item instrument assessing prescription use behaviors, whose score is associated with physician assessments of opioid medication risk. The instrument has excellent test–retest reliability (coefficient of stability 0.086); and acceptable internal consistency (Cronbach’s alpha = 0.73). This instrument is collected at each visit in an attempt to determine the optimal time for administering an opioid risk screening tool.
The Post-Traumatic Stress Disorder (PTSD) Check List (PCL-5) is a 20-item checklist that measures symptoms on a 5-point Likert scale. The instrument provides an overall score, and domain scores based on symptom clusters. The PCL-5 is used once at baseline, to evaluate for PTSD symptoms that may have existed at the time of the injury, and again at week 12 to evaluate specifically for PTSD symptoms that may have developed and persisted as a result of the event that caused the traumatic injury.
The Pain Catastrophizing Scale (PCS) [
18,
19] asks participants to indicate the intensity of each of 13 thoughts or feelings associated with past pain episodes on 5-point scales. The PCS yields a total score and three subscale scores assessing rumination, magnification and helplessness. The PCS has been shown to have adequate to excellent internal consistency (coefficient alphas: total PCS = 0.87, rumination = 0.87, magnification = 0.66, and helplessness = 0.78). The PCS is administered at baseline to assess pain catastrophizing as close to the time of the injury as possible, and again at week 24 to assess for changes that might occur to pain catastrophizing one the participant is no longer in acute pain.
In addition to the CIDI-SAM, which constitutes the primary outcome, the Alcohol Use Disorders Identification Test (AUDIT-C) [
20] will be collected at both baseline and week 24. The AUDIT-C consists of 3 items that are validated to assist in identifying hazardous alcohol consumption patterns.
The Adverse Childhood Experience (ACE) Questionnaire consists of 10 yes/no items that describe scenarios, including childhood abuse, neglect, and household dysfunction. These types of adverse childhood experiences have been strongly correlated with substance misuse and use disorder in adulthood. [
21] Since all of the participants in this study are adults, this instrument is administered at week 4, rather than baseline, in an effort to reduce survey fatigue at baseline.
The Brief COPE Inventory (COPE), the Opioid Craving Scale (OCS), the Patient-Reported Outcomes Measurement Information System (PROMIS) Global Health Scale, Pain Intensity and Interference (PEG), and the Perceived Need Inventory (PNI) are all administered at each follow up time point in an attempt to capture changes in various areas that may correlate with opioid tapering and discontinuation. The COPE is a 26-item, 4-point Likert scale survey that measures the type and frequency of coping strategies in 13 different domains, including substance use, religion, emotional support, positive reframing, disengaging, humor, self-blame, etc. The OCS consists of three visual analog scales that measure current craving, craving in risky situations, likelihood of giving into craving. The OCS is a composite measure of craving, which exhibits good internal consistency (Spearman’s rho range from 0.85 to 0.92) and predictive validity of future opiate use [
22]. The PROMIS is a 10-item measure of general quality of life, which includes a physical health domain (Cronbach’s α = 0.81), and a mental health domain (Cronbach’s α = 0.86) [
23]. The PEG and PNI both use 11-item Likert scales. The PEG is a 3-item instrument that measures level of pain, and how that pain interacts with enjoyment of life and general activity. The PNI is a 5-item instrument that measures perceived problems with opioid use, including the need for opioid use treatment and the perceived potential effectiveness of that treatment [
24‐
27].
At the conclusion of the follow up period, data will be collected about each participant’s felony and misdemeanor record, and their prescription drug activity. Felony and misdemeanor records will be obtained from public court records from the Wisconsin Circuit Court Access Program (CCAP). A limitation to CCAP data is that it is only available for offenses that occurred within the state of Wisconsin. Prescription drug activity will be collected from the Wisconsin electronic Prescription Drug Monitoring Database (PDMP). Data on prescription opioid fills can be obtained from Wisconsin and 14 other states (Table
1).
Table 1
Aim 1 measures and follow-up timeline
Sociodemographic characteristics | X | | | |
Morphine Equivalent Daily Dose (MEDD) | X | X | X | X |
Pain Medication Questionnaire (PMQ) | X | X | X | X |
Post-Traumatic Stress Disorder Check List (PCL-5) | X | | X | |
Pain Catastrophizing Scale (PCS) | X | | | X |
Alcohol Use Disorders Identification Test (AUDIT-C) | X | | | X |
Composite International Diagnostic Interview-Substance Abuse Module for Opioid Use Disorder (CIDI-SAM) | X | | | X |
Patient Health Questionnaire (PHQ-9) | X | | | |
Generalized Anxiety Disorder (GAD-7) | X | | | |
Opioid Risk Tool (ORT) | X | | | |
Social Support Questionnaire (ISEL-12) | X | | | |
Adverse Childhood Experiences (ACE) | | X | | |
Brief COPE Inventory (COPE) | | X | | X |
Current Opioid Misuse Measure (COMM) | | X | X | X |
Opioid Craving Scale (OCS) | | X | X | X |
Quality of Life Questionnaire (PROMIS) | | X | X | X |
Perceived Need Inventory (PNI) | | X | X | X |
Pain Intensity and Interference (PEG) | | X | X | X |
WI Circuit Court Access Program (CCAP) | | | | X |
WI Prescription Drug Monitoring Database (PDMP) | | | | X |
Pilot implementation of the risk screening tool
Study Aim 3 relies upon a well-established approach to the implementation of evidence-based practices, based on fundamental principles of organizational change developed by the Network for the Improvement of Addiction Treatment (NIATx) [
28‐
30]. Of the approaches used to deliver the NIATx organizational change model, peer coaching has emerged as the most cost-effective option [
28]. Members of the research team have adapted the peer coaching model employed by NIATx and applied it to promoting adoption of clinical guidelines for opioid prescribing in primary care [
31]. Further adaptations will be made to tailor the approach for trauma settings. Peer coaches guide organizations through the change process by assessing workflows and identifying opportunities for improvement, and facilitating the plans, actions, analysis, and adoption of those improvements. This study will culminate with the pilot test of a strategy for implementing an opioid risk screening protocol in Wisconsin’s 4 largest trauma centers, using the novel opioid risk screening that was developed from the trauma patient data.
Implementation work will begin with observations of the UW Hospital Trauma Inpatient Service clinical workflows regarding the Alcohol Use Disorders Identification Test (AUDIT-C), a standardized alcohol use screening tool, a process which will inform potential implementation procedures for the opioid risk screen [
32,
33]. This observation period will focus on when, where, and who is administering the AUDIT-C, how the information obtained from the AUDIT-C is recorded and communicated to the clinical care team, if and how that information guides clinical care decisions, and if and how an intervention or referral to treatment is conducted with the patient. Interviews will be conducted with UW Hospital Trauma Inpatient Service staff to assess how opioid risk screening could fit within the trauma inpatient workflow and how opioid-related interventions might best dovetail with that workflow. Nuances of the implementation model that are specific to a trauma service will be shared from the UW site to the four other pilot testing sites. The implementation model will be continuously refined based on the experience of each subsequent implementation, with emphasis placed on adjustments that are needed to tailor the implementation model for local organizational contexts.