Comparing primary care communication interventions to improve patient engagement

A cluster randomized trial evaluating interventions to improve patient-clinician communication found no evidence that the considered interventions had varying effects on patient engagement, despite strong implementation efforts. The study, published in the Journal of the American Medical Association (JAMA) Health Forum examined three intervention arms—a high-touch, high-tech and shared decision-making approach—across 21 primary care clinics in three U.S. health systems. Some secondary outcomes suggested a positive effect of in-person coaching, but ultimately the study’s findings indicate the need for implementation fidelity and alternative outcome measures for patient engagement.

“Our multiple study found no evidence of intervention effects on the primary outcome,” the authors of the study wrote. “Implementing successful pilot interventions into diverse clinical practices is challenging, particularly when minimizing disruptions to workflows…To better align outcome measures with interventions, alternative patient engagement metrics should be considered.”

According to the study, clinicians are often unfamiliar with patients’ reason for visits prior to meeting face to face. It’s also said that patients can struggle to recall the details of decisions made during their visits. The study is intended to enhance patient-clinician communication by building on a prior pilot study with the same goal. The pilot study created two interventions aimed at enhancing patient-clinician communication that was highly accessible to users.

“We chose the primary care setting for three reasons: it is essential for health care, most shared decision-making research is in specialty care and the pilot study in primary care showed potential utility,” the authors explained. “This study aimed to replicate and expand those findings in diverse settings.”

The intervention arms assessed in the study are multilevel, targeting patients, care teams and health systems:

• AskShareKnow (ASK): Posters placed in examination rooms encouraging shared decision-making.

• High-tech: Mobile application-based coaching promoting shared decision-making.
• High-touch: In-person coaching of clinicians with standardized training to promote shared decision-making.

The primary outcome was patient engagement as measured by CollaboRATE, a validated survey assessing shared decision-making. CollaboRATE asks patients to rate, on a scale of zero to nine, how much of an effort was made to:

1. Help you understand your health issues.
2. Listen to the things that matter most to you about your health issues.
3. Include what matters most to you in choosing what to do next.

Secondary outcomes included the Net Promoter Score (NPS), which gauges likelihood of recommending the clinician, and patient confidence in managing their health. Statistical adjustments were made for demographic and contextual factors, including clinician characteristics, type of visit (telehealth or in-person) and health system variability.

The current trial was conducted in three health systems, UC San Diego Health, Reliant Medical Group and Sutter Health. Researchers conducted the randomized controlled trial with 4,852 patients (median age of 54 years) and 114 primary care clinicians. Clinic visits occurred between August 2019 and December 2021. Prior to intervention implementation, baseline surveys collected patient-reported experiences from January to July 2019. Patients were randomly assigned to one of the three arms—ASK, high-touch or high-tech—and were surveyed post-visit and again three months later.

Survey results

Among the 4,852 enrolled patients, participation rates were modest, with just 7% agreeing to take part in the study. Survey completion rates were stronger, with 73% of participants returning surveys within seven days post-visit. The study’s intervention arms had similar demographic characteristics across patients and clinicians.

Regarding CollaboRATE, the study’s primary outcome, there were no statistically significant differences between the intervention arms. At baseline, 63.9% of patients in the ASK group, 70.2% in high-tech and 67.9% in high-touch reported top CollaboRATE scores for shared decision-making. Postintervention, these rates remained consistent across the groups, with ASK achieving 68.2%, high-tech 67.4% and high-touch 67.0%. Adjusted analyses identified several significant covariates: older patients, those with higher confidence in managing their own health and those attending in-person visits were more likely to give top scores, while telehealth visits and certain system-level factors were associated with lower scores.

NPS results demonstrated higher proportions of “promoters” in the high-touch (86.8%) and ASK (87.0%) arms compared to high-tech (83.0%), although the differences were not statistically robust. Patients’ confidence in managing their health declined slightly across all groups three months post-intervention, with no meaningful differences between the arms.

Despite the absence of measurable effects in the primary outcome, the study highlights lessons that are critical for the implementation of communication interventions in real-world clinical settings. Researchers noted that the transition from a physical booklet to an electronic format may have diluted cognitive engagement with the intervention.

Secondary findings suggest that the more intensive high-touch intervention might offer advantages for the enhancement of NPS and patient confidence, although these findings lacked statistical significance. Researchers also emphasized the importance of adjusting for confounding variables, including patient demographics, the type of visit and differences in local health systems. Telehealth visits, for example, consistently returned lower engagement scores, suggesting challenges unique to virtual care environments.

Although the interventions failed to demonstrate significant effects on shared decision-making, the process of prompting patients to articulate their priorities remains valuable and continues to be used at one of the health systems studied. Researchers recommend integrating these interventions as quality improvement initiatives rather than research trials, waiving informed consent to boost participation and employing metrics like NPS to monitor impact. By refining engagement metrics and targeting implementation efforts, health systems may achieve better outcomes in the future.