Frailty State Utility and Minimally Important Difference

Dr Mark Thompson and Professor Renuka Visvanathan are based with the NHMRC Centre for Research Excellence in Frailty and Health Ageing. University of Adelaide, Adelaide, South Australia. Professor Julie Ratcliffe is based at Flinders University, Adelaide. South Australia.

Frailty is common among older adults and is a state of decreased functional reserve resulting from a cumulative decline in multiple physiological systems. Frailty is associated with a range of poor health outcomes. However, it is also a dynamic and potentially modifiable condition. In our Age and Ageing paper, we report on frailty state ‘utility’ and Minimally Important Difference (MID) in a cohort of South Australian older adults. Frailty was measured in this study using both the frailty phenotype and frailty index approaches.

Utility is a form of measurement of quality of life, and is a value that represents the strength of an individual’s preference for specific health states, such as frailty. Utilities range between 1 (perfect health) and 0 (dead) and may be used in evaluating the comparative effectiveness of health interventions. MID is the smallest change in a treatment outcome which an individual would perceive as being important. MID may be useful in providing a patient perspective that informs clinical decision making regarding the effectiveness of frailty interventions. To date, very little has been published on either utility or MID regarding frailty.

The results of our study showed that frailty and pre-frailty classification were significantly associated with lower health state utility for both frailty measures, compared to their non-frail counterparts. The utility estimates we have reported for different levels of frailty classification provide important data for model-based economic evaluation of frailty interventions for older people in community settings. For MID, a single characteristic was enough to be considered a minimally important difference for the frailty phenotype by participants; while for frailty index, this was represented in a range of 7-11% of proportion of deficits present. Our MID estimates represent a ‘plausible range’ of difference for continuous frailty scores. These are valuable as they offer a perspective on meaningful difference as rated by older adults themselves.

These findings are relevant to the design of frailty RCTs, health economic evaluations of frailty interventions, and to clinicians evaluating patient responsiveness to frailty interventions.

Read the Age and Ageing paper Frailty state utility and minimally important difference: findings from the North West Adelaide Health Study.