What are the implications of frailty and multimorbidity in middle-aged and older people with type 2 diabetes?

Talk Code: 
Peter Hanlon
Peter Hanlon, Bhautesh D Jani, Elaine Butterly, Barbara Nicholl, Jim Lewsey, David A McAllister, Frances S Mair
Author institutions: 
University of Glasgow


Frailty and multimorbidity are common in type 2 diabetes (T2D), including people <65 years. Guidelines recommend adjustment of treatment targets in people with frailty or multimorbidity, however guidelines do not differentiate these two related states. It is unclear how recommendations to adjust treatment targets in people with frailty or multimorbidity should be applied to different ages. It is also not known if the relationship between HbA1c and outcomes is similar in people with and without frailty. We assess implications of frailty/multimorbidity in middle/older-aged people with T2D.


Analysis of UK Biobank participants (n=20,566) with T2D aged 40-72 years comparing two frailty measures (frailty phenotype and frailty index) and two multimorbidity measures (Charlson comorbidity index and a simple count of 40 long-term conditions (LTCs)). Outomes: mortality (all-cause, cardiovascular- and cancer-related mortality), Major Adverse Cardiovascular Event (MACE), hospitalization with hypoglycaemia or fall/fracture.


Measure choice influenced the population identified: 42% of participants were identified as frail/multimorbid by at least one measure; only 2.2% were identified by all four measures. Both frailty and multimorbidity, by all measures, were prevalent throughout the age range studied. Each measure was associated with mortality, MACE, hypoglycaemia and falls. The absolute 5-year mortality risk was higher in older versus younger participants with a given level of frailty (e.g. 1.9%, and 9.9% in men aged 45 and 65, respectively, using frailty phenotype) or multimorbidity (e.g. 1.3%, and 7.8% in men with 4 LTCs aged 45 and 65, respectively). Using frailty phenotype, the relationship between higher HbA1c and mortality was stronger in frail compared with pre-frail or robust participants.


Both frailty and multimorbidity, regardless of measure used, identify people at greater risk of mortality, cardiovascular events, and hypoglycaemia. However, among younger people the absolute mortality risk remained low even among the most frail groups. While these findings support calls to embed identification of both multimorbidity and frailty within routine diabetic reviews, care should be taken to assess risk and appropriate treatment targets at an individual level. This assessment should reflect patients’ age, measure used to identify frailty or multimorbidity, as well as other other risk factors. Clinicians should actively identify frailty and multimorbidity in people with type 2 diabetes and tailor management (particularly glycaemic control) to the individual risk and needs of the patient.

Submitted by: 
Peter Hanlon
Funding acknowledgement: 
Medical Research Council