Background

Using shared decision-making (SDM) to define therapeutic goals and shape individualized treatment plans for relapsed/refractory diffuse large B-cell lymphoma (R/R DLBCL) can profoundly improve patient-reported outcomes. However, unique barriers within oncology systems can impede patient-centered care coordination and delivery. In this quality improvement (QI) initiative, we assessed barriers to patient-centered R/R DLBCL care in 2 community oncology systems and conducted team-based audit-feedback (AF) sessions within each system to facilitate improved care coordination.

Methods

Between 12/2019 and 1/2020, 33 hematology/oncology healthcare professionals (HCPs) completed team-based surveys designed to assess barriers to quality R/R DLBCL care in 2 community oncology systems (Table 1). In addition, we retrospectively audited electronic medical records (EMRs) of 75 patients with R/R DLBCL to compare documented practice patterns with self-reported survey results.

To address identified gaps, 31 HCPs practicing within the 2 systems participated in AF sessions; together, clinical teams developed action plans guided by survey insights and EMR findings. Additional surveys completed before and after the AF sessions measured changes in participants' beliefs and confidence in care delivery. A planned prospective follow-up EMR audit of 75 patients will assess changes in documentation and practice behavior.

Results

Team Surveys and EMR Audit: Using a 5-point Likert scale (0 = extremely unlikely; 5 = extremely likely), HCPs indicated a high likelihood of using prognostic scores (mean score, 3.8) and cell of origin (mean score, 3.9) to inform DLBCL treatment decisions. However, despite documentation of individual prognostic factors, only 30% of EMRs included the calculated International Prognostic Index (IPI) risk score and only 8% included cell of origin. No EMRs included an age-adjusted IPI, stage-modified IPI, or NCCN-IPI score (Figure 1).

Despite only 6% of HCPs identifying engaging patients in SDM as a major practice challenge, SDM resources were consistently underutilized (Figure 2). Only 30% of HCPs estimated using SDM in more than half of their patients; 23% of HCPs reported using SDM with none of their patients with R/R DLBCL. Moreover, while 63% of HCPs reported routinely asking the patient what role he/she wishes to play (active, passive, collaborative), fewer HCPs integrated other SDM tools: using visual aids to communicate treatment benefits/risks (59%), referring patients to online education resources (53%), including the patient's spouse/family members in decisions (47%), and discussing financial toxicity (38%).

Small-Group AF Sessions: When asked to identify a single aspect of R/R DLBCL care in greatest need for improvement in their systems, HCPs most commonly selected care coordination (34%) and adverse event recognition/management (20%), followed by individualizing treatment (16%) and prognostic scoring (14%). As part of their action plans, HCPs prioritized 3 practice behaviors to address with their clinical teams: individualizing treatment decisions based on patient- and disease-related factors (40%), improving communication during care transitions (40%), and providing adequate patient education about treatment options and potential side effects (20%).

As a result of participating in the AF sessions, HCPs reported a meaningful shift in beliefs about collaborative care: 84% of HCPs agreed or strongly agreed that collaboration across the extended oncology care team is essential for achieving DLBCL treatment goals, an increase from 64% prior to the AF sessions. Further, HCPs reported increased confidence in their ability to perform each of 6 aspects of evidence-based, collaborative, patient-centered care (Figure 3).

Conclusions

Key system-based barriers to providing individualized R/R DLBCL care include adequate documentation of prognostic factors, care coordination, and effective SDM. After participating in this QI initiative, HCPs demonstrated improved commitment to team-based collaboration and increased confidence in delivering patient-centered care. Remaining practice gaps and challenges can inform future QI programs.

Study Sponsor Statement

The study reported in this abstract was funded by an independent educational grant from Genentech. The grantor had no role in the study design, execution, analysis, or reporting.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.