Diabetes Conversation Tool Pleases Docs, but Misses the Mark on Clinical Outcomes

— Agenda-setting cards were designed to help patients ID concerns to discuss with their clinicians

Use of agenda-setting conversation cards alongside usual primary care improved clinician satisfaction but not care fit or clinical outcomes for patients with type 2 diabetes, a cluster randomized trial showed.

To determine how well treatment fits into patients' daily lives, priorities, and capacity, the researchers used the Illness Intrusiveness Ratings Scale (range 13-91, with higher scores indicating greater intrusiveness); mean scores were 32.1 in the intervention group and 37.1 in the control group (usual primary care alone) at 6 months.

The average score at baseline was 35.6, which is "comparable to [scores] reported by patients living with cancer or after undergoing organ transplant," wrote Kasia Lipska, MD, of the Yale School of Medicine in New Haven, Connecticut, and colleagues in JAMA Network Openopens in a new tab or window.

In addition, mean HbA1c levels were 8.8% in the intervention group and 9.1% in the control group.

Other outcomes that did not differ significantly between the intervention and control groups included:

• Treatment Burden Questionnaire scores (range 0-150, with higher scores indicating greater burden): 41.2 vs 46.8
• 17-Item Diabetes Distress Scale scores (<2.0 indicates little or no distress, 2.0-2.9 moderate distress, and ≥3.0 high distress): 2.0 vs 2.3

However, clinicians in the intervention group were significantly more likely to report feeling satisfied with the way the visit went, with 62.2% strongly agreeing that they were satisfied compared with 19.1% in the control group, and 59.5% strongly agreeing that they were feeling confident in responding to patient concerns versus 30.9%.

Meanwhile, patients' levels of satisfaction were overall very high and did not vary between groups.

The intervention "may serve as a low-burden, scalable tool for supporting person-centered conversations, even if their impact on specific outcomes may be modest without further supports or tools for clinicians to help them respond to patient concerns," Lipska and team noted.

The agenda-setting kit involved a set of 14 conversation cards regarding life with diabetes and its treatment, designed to help patients identify concerns, barriers, and successes to discuss with their clinicians. Patients were asked to select up to three cards they wanted to prioritize before the visit. Clinicians were told to conduct the encounter as usual but to review and respond to the cards first.

"It was intentionally designed as a brief, pragmatic intervention to improve the quality of patient-clinician conversations during routine care, rather than as a comprehensive self-management program," the authors explained. The idea behind the intervention was that poor diabetes outcomes "arise when the workload of diabetes exceeds a patient's capacity to manage it."

The novelty of the intervention "lies in its focus on care fit -- the extent to which treatment is compatible with a patient's daily life, priorities, and capacity -- rather than simply adding another diabetes self-management intervention," pointed out Megha K. Shah, MD, MSc, and Ambar Kulshreshtha, MD, PhD, both of Emory University School of Medicine in Atlanta, in an accompanying commentaryopens in a new tab or window.

The duo praised the intervention for not increasing encounter length, but emphasized that the lack of measurable patient benefits suggests that "identifying a problem does not necessarily provide the resources required to solve it."

"This finding may be the most important lesson from the trial -- and an opportunity rather than simply a limitation," wrote Shah and Kulshreshtha. "Primary care clinicians can ask about food insecurity, medication costs, transportation, competing caregiving responsibilities, work schedules, treatment adverse effects, or emotional distress, but they cannot fully address all these during the brief visit."

"The next step is building the clinical and community infrastructure capable of acting on those priorities to deliver true person-centered diabetes care," they suggested.

The QBSafe cluster trialopens in a new tab or window randomized clinicians 1:1 to the intervention and control groups and ran from February 2023 to July 2024 at a federally qualified health center in New Haven. Participating clinicians included primary care physicians, nurse practitioners, and physician assistants who prescribed medication to treat adults with type 2 diabetes.

Patients had a baseline HbA1c level over 8% (mean 10%) and were generally younger, with a high burden of complications and social risk factors. Average patient age was 56.2 years, 60.8% were women, 35.7% were white, 19.6% were Black, and 72.7% were Hispanic or Latino. Most (61.5%) reported Spanish was their primary language, and 91.3% reported family income less than 150% of the federal poverty level.

At 6 months, there were no differences between groups in the total number of diabetes medications used or in net additions/discontinuations.

Post-encounter adjustments in existing medications -- such as changes in dose, frequency, or switching medications within the same class -- were higher in the intervention group (33.8% vs 23.2%). Additionally, medication adherence measured by the proportion of days covered was significantly higher in the intervention group (77.2% vs 64.6%).

A slightly higher proportion of control group patients had an emergency department visit (24% vs 18%) or were hospitalized (18% vs 14%) during the trial.

Due to the single-center nature of this study, generalizability may be limited, the researchers acknowledged.