Eligible patients presenting to participating sites (ie, emergency departments and associated observation units) from four geographically distinct health systems have been randomized to one of three study arms:

Patients are stratified by complaint (ie, renal colic or back and neck pain) and hospital center (ie, the University of Pennsylvania, Northwell Health, the Mayo Clinic, or the University of Alabama at Birmingham [UAB]). Electronic consent and randomization within strata and hospital centers occur automatically through a password-protected, web-based data collection platform for behaviorally oriented randomized clinical trials; this occurs during the enrollment process [31] using computer-generated random numbers. All outcome data are collected through the web-based platform, which is a secure, electronic database accessible only to researchers.

The PRT was developed by the research team based on the validated ORT survey. The ORT was designed to assess risk of opioid dependency for patients for whom an opioid pain relief prescription was being considered in outpatient settings. The ORT considers multiple clinical and experiential factors in assessing risk of opioid dependency and reports risk on a scale of 0-26, divided into three probabilistic categories of risk for opioid misuse. In validation studies, 6% of patients in the lower-risk category (score 0-3) developed substance use disorder (SUD), 28% of patients in the moderate-risk category (score 4-7) developed SUD, and 91% of patients in the high-risk category (score 8-26) developed SUD [22]. It should be noted that the ORT was never specifically validated in the emergency department setting. Through an iterative process that drew on the team’s expertise in patient-provider communication and shared decision making, a visual aid of the ORT results was developed. The team partnered with experts in patient communication to create a tool that was easy to interpret and leveraged best practices of risk communication in settings where literacy and numeracy would vary [32]. Using an iterative process, we developed a visual tool that uses each patient’s assessed ORT score to demonstrate an absolute number and an overall risk category displayed on a visual scale; this was developed in collaboration with patient stakeholders, experts in the fields of patient education and communication, and clinical educators. Tool development followed best practices for decision aid development as defined by the International Patient Decision Aid Standards [33]. The prototype was also informed by our team’s previous work to identify patient views [34] and provider views [35,36] on this topic. The probabilities described above are expressed with consistent denominators of 100 (ie, 6/100, 28/100, or 91/100). A color-coded visual thermometer is used to deliver information in a comparative fashion that accounts for lower levels of health literacy and numeracy. Aware that no group, including the one at the lowest risk, had less than a 6% risk of aberrant behavior, the color green was eliminated from the thermometer so that the revised spectrum would go from yellow to orange to red as the patient’s individual risk increased. A probability of 6/100 corresponds to a yellow category of At Risk, 28/100 corresponds to an orange category of At High Risk, and 91/100 corresponds to a red category of At Highest Risk. A total of six iterations that included multiple rounds of meetings as a full study team, elicitation of patient investigator input, revision of the tool, and elicitation of further input was performed. The PRT was then piloted for 2 days in the emergency department for patients in acute pain, in order to seek feedback from patients, providers, and staff, and further refinements were made.

To develop the narrative videos, the research team recruited patients and caregivers with lived experiences to record their stories. Professional videographers edited the videos for clarity and salience. The team then tested the videos through an iterative process with a core group of patient investigators and community stakeholders to maximize factors such as identification with characters, perceived realism, normative values, and reinforcement. Each narrative is a 1-3-minute, first-person, interview-based vignette of a patient or caregiver sharing personal perspectives and experiences of acute pain and opioid use and misuse. Together, the eight narrative videos feature patients of diverse backgrounds and risk levels to offer a balanced, varied commentary. Originally, an earlier version of the narratives displayed individual risk scores and was designed to align closely with known mechanisms of narrative persuasion, such as transportation, by using scripted stories derived from real experiences and narrated by professional actors. Ultimately, because the investigation team and patient advisors felt that these stories were overly scripted and not realistic, we chose to film real patients with real stories sharing their stories. These stories were real but did not align perfectly to any specific risk scoring system or any specific mechanism of narrative persuasion (see Table 1 [23-30]).

The primary objective of this study is to determine whether risk-informed communication with or without a narrative-enhanced tool can improve functional outcomes and patient-centered outcomes in the domains of knowledge and opioid use. Specifically, the study will measure the effectiveness of different risk communication strategies using the following outcomes: risk awareness as well as opioid and treatment preferences for fewer opioids, particularly among those at higher risk for addiction. See Table 2 for a full list of primary and secondary measures, response types, collection points, and their relevance to patients and other stakeholders.

We hypothesize that patients receiving narrative-enhanced risk communication will demonstrate greater knowledge as determined by awareness of risk for opioid dependency and will request and take fewer opioids for fewer days, while achieving the same degree of pain relief and improved functional status, as compared to study patients receiving only a generalized risk information sheet or information sheet plus an individualized, visual probabilistic risk communication tool.

Patients are recruited from acute care settings (ie, emergency department or observation units) in four US academic hospital centers: (1) the University of Pennsylvania in Philadelphia, Pennsylvania, located on the east coast with an urban patient population, (2) Northwell Health (ie, Long Island Jewish Hospital and North Shore Hospital) in Long Island, New York, located on the east coast with a suburban patient population, (3) the Mayo Clinic in Rochester, Minnesota, located in the Midwest with a diverse rural, suburban, and small-city patient population, and (4) the UAB, Alabama, located in the Southeast with an urban patient population. These centers were selected to capture geographically and ethnically diverse patient populations and clinical practices.

Our sample population includes patients presenting to participating study sites with a chief complaint suggestive of acute renal colic or musculoskeletal neck or back pain. A trained project manager, research coordinator, or research associate identifies eligible patients based on chart review and collaborates with the treating clinician to confirm patient eligibility for the study. Patients are concurrently enrolled at all participating study sites 5-7 days per week, whenever a trained enroller is available.

The study inclusion and exclusion criteria are listed in Textbox 1.

Additionally, patients who display aberrant drug use behavior, as determined by the lead provider, or have used opioid medications in the past 30 days—excluding opioids taken for the current condition within 48 hours of this acute care visit—are ineligible. Patients with a known history of chronic kidney disease (glomerular filtration rate <60) are also ineligible, because they may not be able to take nonsteroidal anti-inflammatory drugs (NSAIDs).

After providing informed consent, participants answer questions from a series of surveys, including the following: (1) an informational profile of demographic data (ie, gender, race, ethnicity, and education level) [37-39], (2) the Revised American Pain Society Patient Outcome Questionnaire [40], (3) the 20-Item Short Form Survey from the Medical Outcomes Study (MOS-20) [41], (4) a tobacco use survey [38], (5) the ORT [22], (6) the Screener and Opioid Assessment for Patients with Pain-Revised [42], and (7) a pain relief preference survey. Participant responses to survey questions are recorded in real time on a secure, password-protected iPad or tablet owned by the study. Study staff will administer surveys (1) through (3) and (7), whereas participants will complete surveys (4) through (6) directly on the study tablet computer due to the sensitive nature of these questions.

All participants are given—in addition to standard institution-specific discharge instructions—a generalized, fact-based risk information sheet for renal colic or neck and back pain, based on chief complaint. This information sheet was developed by consensus among the study team, including clinicians, patients, and nurse educators. During enrollment, patients are given 3 minutes to familiarize themselves with the risk information sheet and ask questions to study staff.

Regarding the ORT, for participants randomized into either intervention arm, the tablet computer will display one of three ORT images—At Risk, At High Risk, or At Highest Risk for opioid misuse—based on participant responses to the ORT survey (see Figures 3-5).

Regarding video narratives, for participants randomized to the narrative-enhanced PRT arm, also displayed is a menu with a choice of eight video narratives (see Figure 6 and Table 1). The narratives are displayed in a menu format on a tablet computer and participants can watch any story by selecting the picture of the storyteller. The participants may view as many videos as desired. Participants are provided the option of using headphones, and, when possible, study staff exit the room for privacy until the patient has completed viewing the videos.

Outcomes were decided through a process that included iterative discussions with the entire investigative team, including patient investigators, as well as feedback from the study sponsor peer review. On days 1 through 7 and 14 after enrollment, the study portal sends participants automated email or text messages containing links to secure, online follow-up surveys. These brief surveys collect information on patient satisfaction and outlook [40], pain management strategies and functional outcomes, behavioral and functional characteristics [41,43], medication use behaviors [44], quality of life [41], health literacy [45], and health numeracy [46]. Patients randomized to the video narrative–enhanced PRT arm also receive messages encouraging them to continue to view the eight video narratives through an individualized portal. A total of 3 months after enrollment, the portal will send participants automated email or text messages containing a link to a secure final follow-up survey of the above attributes, as well as recall of opioid risk level and/or viewing of video narratives for patients randomized to the corresponding intervention arms.

Text and email reminders are used to promote continuous engagement and follow-up among study participants. Small cash incentives are used to maximize retention and promote follow-up. All subjects are eligible to receive up to US $84 for completing enrollment and follow-up surveys. Because of lower-than-expected response rates to the day-14 follow-up at the beginning of the study, the total eligible incentive was increased on January 15, 2018, from US $50 to US $84 to improve response rates and minimize study attrition. For the survey at 3 months, if the subjects do not respond to the electronic reminders, research staff will call the participants to remind them to complete survey; if necessary, they will conduct the survey by telephone.

An important limitation is the lack of standardization of the definition of aberrant drug use that was used as an exclusion criterion. During enrollment, the emergency department provider determined whether they believed each potential participant demonstrated aberrant drug use. Whether they used objective or subjective evidence was not predetermined. This exclusion criterion, as well as the exclusion of patients with prior use, was made in response to sponsor reviewer concerns that these tools would be of little use for patients who are “doctor shopping” or who already have an opioid use disorder.

The ORT has previously been tested in the emergency department but not fully validated. It was designed for primary care settings, among which acute pain for back and kidney stone pain are common complaints. It remains an easy-to-implement tool that could be stratified in a way that the visual and risk tools could be tested.

Patients were provided with the possibility of repeat electronic exposure to the narrative intervention but not their specific risk score display. However, the risk thermometer was provided as a hard copy to patients on discharge, with the possibility of repeat exposure. Because risk recall was a primary outcome, we did not wish patients to look up their score in real time as an answer to the risk recall question (ie, social desirability). Home viewing of the narratives was deemed an important outcome and part of the intervention as they were generalized and could be shared and viewed; however, this required some additional time that could not be achieved at the point of care during a busy emergency department visit.