Presentation (CTI)

OBJECTIVES: The impact of clinical interventions for diabetes mellitus (DM) may differ according to frailty status. Prior studies suggest that frail patients with DM may derive some benefit from intensive glucose lowering compared with non-frail patients. This study examined the association between HbA1c levels and mortality among frail and non-frail older adults with DM.
METHODS: We conducted a retrospective cohort study using electronic medical records (EMR) from a large health maintenance organization, including all patients with DM aged ≥75 years old. HbA1c category was defined by the most recent value recorded in the year before cohort entry, defined as the earlier of January 1, 2015, or age 75. Patients were followed until the earlier of death or December 31, 2024. Frailty was assessed using a validated EMR-based index. Multivariable Cox proportional-hazards models were used to evaluate the association between glycemic control and mortality across frailty categories.
RESULTS: The cohort included 15,407 older adults with DM, of whom 70% were classified as frail. Frail patients were older, had shorter diabetes duration, and had more comorbidities, while baseline HbA1c values were like those of non-frail patients; 31% had HbA1c levels of 7-7.9%, and 17% had levels ≥8%. Compared with HbA1c <7%, adjusted mortality risk was lower among patients with HbA1c 7-7.9% and higher among those with HbA1c ≥8%, regardless of frailty status; however, the effect size was greater among non-frail (aHR 0.69, 95% CI 0.49-0.97 and aHR 1.67, 95% CI 1.22-2.29, respectively) compared to frail patients (aHR 0.92, 95% CI 0.86-0.98 and aHR 1.10, 95% CI 1.05-1.20, respectively).
CONCLUSIONS: In this large cohort of older DM adults, the association between HbA1c levels and mortality differed by frailty status. These findings support individualized glycemic targets in older adults with DM, with frailty considered when weighing the potential benefits and risks of more intensive glucose control.