The overall goal of this protocol is to pilot test a clinician training intervention that uses standardized patients (trained actors playing patient roles) as instructors who impart communication skills to primary care clinicians. This project will conduct a pilot clinical trial of the intervention developed by the primary investigator in order to evaluate intervention feasibility and generate data to plan a subsequent fully-powered, multisite trial. Primary care clinicians will be randomized to receive either the intervention or control; 48 patients (2 per clinician) will then be recorded during clinic visits with study clinicians and will provide data on post-visit perceptions and health outcomes. Study hypotheses are that visits with clinicians who receive the intervention (versus control) will be associated with more frequent use of targeted communication skills, lower probability of high-risk opioid prescribing, higher patient-reported agreement with treatment plan, and lower pain interference 2 months later.

The only intervention in the randomized controlled trial (RCT) will be clinician education. This RCT will not directly manipulate or influence patient care. Therefore, this study is minimal risk. The only foreseeable patient risks relate to data confidentiality and privacy.

The investigators will recruit approximately 48 primary care clinicians for the pilot RCT. The study goal is to enroll 24 residents and 48 patients (2 patient visits per resident). However, based on previous experience with similar recruitment methods in these clinics, only about 50% of enrolled clinicians will ultimately have visits with study patients. Thus, it will be necessary to over-enroll clinicians because the investigators expect a large proportion of residents will have to be dropped from study due to not seeing any study patients. After enrollment, clinicians will complete a brief questionnaire including demographics and self efficacy regarding communicating with patients about chronic pain and opioids.

Randomization assignment will take place after all clinicians have provided informed consent and enrolled, to prevent randomization status from influencing enrollment decisions. Randomization assignment will be done by study personnel after all clinicians have been recruited. Randomization is at the clinician level. Because of the nature of the intervention, it is not possible to blind subjects or investigators to randomization assignment. However, patients will be unaware of clinicians' randomization assignment.

Intervention clinicians will complete 2 standardized patient visits during regular clinic time. The first visit will include viewing an 8-10 minute video summarizing the key communication skills, a 10-12 minute roll-play session to practice using these skills, and 8-10 minutes of constructive feedback. The second video will have only roll-play and feedback. Control clinicians will receive a written summary of the 2016 Centers for Disease Control and Prevention (CDC) opioid prescribing guidelines which include recommendations for best practices for use of opioids to treat chronic non-cancer pain. To the extent possible, the investigators will use CDC-produced materials for the control group. CDC guidelines will serve as an attention control. Intervention clinicians will complete a brief questionnaire evaluating the intervention. Some time later, enrolled clinicians will see 2 patients who've been screened and enrolled by the research team and who've agreed to allow a previously scheduled visit with pcp to be audio recorded. Data will include the audio recording transcripts, a pre-visit and post-visit patient questionnaire and a clinician post-visit questionnaire.

2 months after each patient's visit, a research assistant will call patients and obtain 2-month follow up data (e.g. Brief Pain Inventory).

Inclusion Criteria:

Internal Medicine or Family Medicine residents at UC Davis:

• completed ≥1 year of training
• see primary care patients in the Ambulatory Care Center (ACC) building. Clinicians will be recruited through clinic huddles, emails, and presentations at meetings.

UC Davis Patients:

• 18-80 years old
• taking opioids (≥1 opioid dose per day) prescribed by their primary care physician for >90 days to treat chronic musculoskeletal pain
• have an appointment scheduled with a participating clinician at which they report they are likely to discuss pain management

Exclusion Criteria:

• active cancer
• hospice
• do not speak English
• prisoners
• pregnant women
• unable to consent

Physicians will be randomized after physician recruitment is complete. This ensures that arm assignment does not impact physicians' decision to participate. Arm assignment will be done using a random number generator. Due to the nature of the intervention, blinding participants (physicians) is not feasible. However, patients will not be aware of their physician's arm assignment.