What is the risk of adverse clinical outcomes in middle-aged and older adults with Type 2 Diabetes and Frailty?
Problem
The prevalence of type 2 diabetes (T2D) is growing worldwide, and the average age of people with T2D is rising. Frailty, a state of increased vulnerability to adverse health outcomes, is an important factor in T2D. Clinical guidelines recommend less stringent glycated haemoglobin (HbA1c) targets for older frail people with T2D. The implications for relatively younger people meeting criteria for frailty are less clear. We aim to assess the association between frailty and clinical outcomes in T2D and explore if frailty modifies the association between HbA1c and mortality and hypoglycaemia.
Approach
We identified people with T2D form the UK Biobank cohort. We assessed baseline frailty using the Frailty Phenotype model, comprising five characteristics (low grip strength, weight loss, slow walking speed, exhaustion, and low physical activity). People with 1-2 of these features are considered pre-frail. People with 3 or more considered frail. Clinical outcomes were identified from linkage to national mortality registers and hospital episode statistics over a median 75 months follow-up. Outcomes included all-cause mortality, cardiovascular mortality, cancer mortality, major adverse cardiovascular event (MACE), hypoglycaemia, and fall/fracture. Cox-proportional hazards models assessed the association between frailty and each outcome, adjusted for age, sex, body mass index, socioeconomic status, smoking and alcohol. We then assessed the association between HbA1c and both all-cause mortality and hypoglycaemia at different levels of frailty.
Findings
Among 20,569 people with T2D (mean age 60.2, 36.8% female) 2,506 (12.7%) met the criteria for frailty and 11,026 (55.7%) were pre-frail. Frailty was associated with older age, female sex, higher socioeconomic deprivation, higher body mass index, current smoking, and no alcohol drinking. Compared with participants with no frailty characteristics, frailty was associated with all-cause mortality (hazard ratio 2.23, 95% confidence intervals 1.92-2.60), cardiovascular mortality (3.65, 2.16-5.12), cancer mortality (1.75, 1.31-2.35), MACE (1.75, 1.31-2.35), hypoglycaemia (3.21, 2.29-4.49) and fall/fracture (2.22, 1.85-2.67). Pre-frailty was also associated with each of these outcomes, with smaller effect sizes. HbA1c showed a J-shaped relationship with all-cause mortality, with increased risk below 45 mmol/mol and above 55mmol/mol. This relationship was consistent at all levels of frailty (p for interaction = 0.196). Higher HbA1c was associated with greater risk of hypoglycaemia at all levels of frailty.
Consequences
Frailty is identifiable and common among middle-aged as well as older adults with T2D. It is associated with a range of adverse outcomes, with particularly strong associations with cardiovascular mortality and hypoglycaemia. Low and high HbA1c are associated with mortality, with similar hazard ratios at all levels of frailty. Consequently, the absolute risks of low and high HbA1c are necessarily larger in people who are frail.