Patient- and clinician-directed implementation strategies to improve serious illness communication for high-risk patients with cancer: A cluster-randomized pragmatic trial.

Abstract

1514 Background: Early serious illness conversations (SICs) elicit patients’ values, goals, and care preferences and have been shown to improve outcomes and reduce end-of-life healthcare utilization for patients with cancer. However, most patients with cancer die without a documented SIC. Given prior evidence that strategies informed by behavioral economics (“nudges”) increase SIC rates, our objective was to test the independent and additive effects of clinician- and patient-directed nudges to increase SIC completion. Methods: We conducted a 2 × 2 factorial, cluster-randomized pragmatic trial (NCT04867850) to test the effects of nudges to clinicians, patients, or both, compared to usual care, on SIC completion. Usual care was an active control consisting of clinician-directed text messages sent before routine clinic sessions, identifying patients at high risk of 6-month mortality as predicted by a validated machine-learning prognostic algorithm. The clinician nudge additionally included weekly peer comparisons on clinician-level SIC completion rates. The patient nudge consisted of a patient-facing message sent electronically before an index clinic visit, asking a 3-question survey designed to prime patients for an SIC with their oncology team. Participants included medical/gynecologic oncologists and advanced practice providers (APPs) within a large academic health system and their high-risk patients. We independently randomized oncologist/APP clusters and patients to receive nudges vs. usual care. The primary outcome was a documented SIC in the electronic health record within 6 months of enrollment. Using a Cox proportional hazards model with cluster robust standard errors, we performed a time-to-event analysis and tested for heterogeneity of effect across prespecified subgroups. Results: From September 2021 to March 2022, the study accrued 4,450 patients (median age 67, 52.9% female, 17.3% Black, 2.7% Hispanic) seen by 166 clinicians across 4 hospitals and 6 community sites, randomized to clinician nudge (n=1,179), patient nudge (n=997), both (n=1,270), or active control (n=1,004). Overall patient-level rates of 6-month SIC completion were: 11.5% (clinician nudge), 11.5% (patient nudge), 14.1% (both), and 11.2% (active control). Compared to the active control, participants in the combination nudge arm were more likely to engage in SICs (hazard ratio [HR] 1.55, 95% confidence interval [CI] 1.00-2.40), whereas those in the clinician (HR 0.95, 95% CI 0.64-1.41) and patient (HR 0.99, 95% CI 0.73-1.33) nudge arms were not. There was no effect heterogeneity across age and race subgroups. Conclusions: Clinician- and patient-directed nudges may be synergistic in promoting serious illness communication at scale and equitably in routine cancer care. Effects on end-of-life care among decedents are forthcoming. Clinical trial information: NCT04867850 .