Background/Purpose: Hydroxychloroquine (HCQ) is a key therapy for lupus (or SLE). Yet, challenged to weigh benefits vs. harms, ~80% of patients self-discontinue HCQ. Shared decision-making (SDM) tools, like HCQ-SAFE© (Garg, 2023), can support patients and clinicians in complex treatment decisions. In a clinical trial, HCQ-SAFE SDM intervention (Fig. 1A) was used by 12 clinicians during 200 visits in 8 clinics of 2 health systems. Still adoption of SDM tools can be challenging in non-research settings, particularly busy clinics and low-resource settings. We aimed to: 1) examine implementation outcome metrics mapped to the RE-AIM framework for using the HCQ-SAFE SDM intervention; 2) develop strategies based on clinician experiences using the Normalization Process Theory (NPT) to revise HCQ-SAFE implementation workflow for easier integration in clinics.

Methods: The HCQ-SAFE SDM intervention (Fig. 1A) was implemented in 200 unique visits by 6 MDs, 3 PharmDs, and 3 RNs (NCT05922722). Clinicians completed a questionnaire to report intervention details (version, topics, language), time, and rated feasibility & likelihood to use on a 1-9 Likert scale (9=best). Clinicians also participated in a cognitive interview to share experiences, barriers, & facilitators to HCQ-SAFE use. Quantitative data from 200 questionnaires estimated RE-AIM implementation metrics. Reach: % eligible visits completing HCQ-SAFE. Effectiveness: change in decisional conflict scores (overall/by role). Adoption: % clinicians completing intervention. Implementation: Fidelity (% visits covering ≥1 benefit/harm, language) & Feasibility (median score ≥7, time ≤5 min).

Maintenance: monthly completion rate & Net Promoter Score (NPS ≥80%, motivators – detractors). Qualitative data from interviews were analyzed using a coding scheme informed by 4 NPT domains to identify strategies to refine workflows for easier integration of HCQ-SAFE in clinics.

Results: In 200 visits, MDs (50%), PharmDs (28%), and RNs (22%) completed the HCQ-SAFE intervention (Table 1). Data analysis showed excellent completion rates (100%) across MDs, PharmDs, RNs, with equally effective resolution of decisional conflicts regardless of clinician role (98% PharmD/RN vs. 99% MD; Table 1). Adoption was slightly lower among MDs (75%) vs. PharmDs/RNs (100%). HCQ-SAFE was rated highly feasible (median score 9, median time 5 mins) and received 100% NPS, indicating strong clinician willingness for continued use. Content analysis of clinician experiences using NPT yielded 6 strategies for HCQ-SAFE implementation (Table 2; Fig. 1B). Key recommendations included: tailoring training by role, simplifying eligibility to include all SLE patients, using PharmDs/RNs to lead reviews. Clinicians preferred this equally effective staff-led model for easier integration and facilitate team-based care. The revised clinical workflow (Fig. 1B) gained unanimous endorsement from participating clinicians.

Conclusion: The impact of this study is 2-fold. First, it establishes HCQ-SAFE is a highly feasible and effective SDM intervention for clinical use. Second, it delivers an end-user-informed practical workflow to ensure sustainable integration in clinics to promote medication adherence via SDM.

Table 1. A. Key Visits Characteristics; B. Implementation Outcome Metrics Mapped to RE-AIM

Table 2. Qualitative Analysis of Clinician Interviews Informed by Normalization Process Theory Domains to Deliver Strategies to Develop Practical Workflow for HCQ-SAFE Integration in Clinics

Figure 1A. Swim Lane diagram showing the study workflow used to implement the HCQ-SAFE shared decision-making intervention across 200 unique patient visits in 8 clinics of 2 health systems; Figure 1B. An end-user-informed practical workflow for routine use of HCQ-SAFE shared decision-making intervention in clinics

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/bridging-the-gap-a-mixed-methods-study-to-enhance-integration-of-hcq-safe-a-shared-decision-making-tool-for-hydroxychloroquine-use-in-routine-lupus-care/