Examining HbA1c in frail older adults: a linked data cohort study

Objective: Diabetes and frailty pose a significant issue for society. Previous research has shown a mixed relationship between diabetes and frailty in older people, indicating a complex interrelationship. The aim of this study was to examine the association between routinely collected HbA1c testing data and frailty in a large cohort from Western Sydney. Methods: Data from the Western Sydney Frailty Registry Study were linked to the Blacktown and Mount Druitt hospital HbA1c testing datasets. These datasets include information on patient outcomes, pathological metrics, hospitalisation indicators, and a range of morbidity and demographic information. This was then analysed in Stata using multivariable regression. Results: Diabetes was more common in frail patients than in pre-frail or non-frail patients. Those with both diabetes and frailty did not have an increased risk of death or rehospitalisation in age and sex corrected analyses (OR: 0.66, 95% CI: 0.36–1.21 and 0.90, 95% CI: 0.37–2.17, respectively) compared to those with frailty alone. There was a reduced risk of death associated with higher HbA1c for frail people (OR: 0.58, 95% CI: 0.37–0.90). Conclusions: There is a complex interrelationship between diabetes and frailty. For frail people with diabetes, there are important lessons for management. These include the possibility that HbA1c is not a useful metric and that overtreatment with hypoglycaemic agents must be considered.