ImportanceAcute back pain is a common reason for primary care visits and often results in low-value spinal imaging.ObjectiveTo evaluate the effect of a standardized patient-delivered intervention on rates of low-value spinal imaging among primary care patients with acute low back pain.Design, Setting, and ParticipantsIn this randomized clinical trial, physicians or advanced practice clinicians were recruited from March 22 to August 5, 2021, from 10 adult primary care or urgent care clinics in Sacramento, California. The intervention period was from May 1, 2021, to March 30, 2022, with follow-up from October 28, 2021, to June 30, 2023. Analyses were performed from April 1 to June 25, 2024.InterventionClinicians were randomized 1:1 to intervention or control. Intervention clinicians received 3 simulated office visits, each with a standardized patient instructor (SPI) portraying a patient with acute uncomplicated back pain. At each visit, SPIs provided clinician feedback guided by a 3-step model: (1) set the stage for deferred imaging by building trust, (2) convey empathy, and (3) communicate optimism while advocating watchful waiting without imaging. Control clinicians received no intervention.Main Outcomes and MeasuresThe primary outcome was lumbar spinal imaging completion within 90 days of acute low back pain visits, with study clinicians assessed up to 18 months of follow-up. Secondary outcomes were cervical spine imaging completion after acute neck pain visits, any imaging completion after an adult visit, patient experience ratings of clinicians (scale range, 0-100), and use of targeted communication skills during an audio-recorded standardized patient evaluation visit at median follow-up of 16.8 months (range, 14.1-18.0 months).ResultsThe analysis included 53 clinicians; mean (SD) age was 46.7 (1.0) years, and 35 (66.0%) reported female gender. A total of 25 were in the intervention group and 28 in the control group. After adjustment for prerandomization rates, patients with acute low back pain who saw intervention and control clinicians during follow-up had similar rates of lumbar imaging (194 of 1234 clinic visits [15.7%] vs 226 of 1306 clinic visits [17.3%]; adjusted ratio of postintervention vs preintervention odds ratios [AORR], 1.00; 95% CI, 0.72-1.40). Adjusted follow-up rates of imaging for acute neck pain (AORR, 1.16; 95% CI, 0.83-1.63) and overall imaging (AORR, 1.07; 95% CI, 0.97-1.19) were not significantly different among patients of intervention and control clinicians. Intervention and control clinicians had similar mean (SD) patient experience ratings during follow-up (88.6 [28.7] vs 88.8 [28.3]; adjusted mean difference-in-differences, −1.0; 95% CI, −3.0 to 0.9). During audio-recorded standardized patient visits, intervention clinicians had significantly better ratings than controls on eliciting the patient’s perspective (adjusted standardized difference [ASD], 0.62; 95% CI, 0.05-1.19) and conveying empathy (ASD, 1.16; 95% CI, 0.55-1.77).Conclusions and RelevanceIn this randomized clinical trial of an educational intervention using simulated office visits to encourage a watchful waiting approach for acute low back pain, the intervention had no significant effect on low-value spinal imaging rates or patient experience ratings.Trial RegistrationClinicalTrials.gov Identifier: NCT04255199