David Oliver: What has the pandemic taught us about using frailty scales?

For decades the geriatric medicine clinical and research community has been discussing the concept, definition, causation, and clinical relevance of frailty,12 eventually influencing mainstream health policy and thinking outside our specialties.34 England’s pandemic response has pushed frailty onto a bigger stage: notably, the use of the clinical frailty scale (CFS) to triage, target, and potentially ration scarce intensive and high dependency care.5

While I’m pleased to see frailty and structured assessments promoted so keenly, this hasn’t been without controversies. In particular, the National Institute for Health and Care Excellence’s rapid guidelines on covid-19 critical care in adults, issued at the start of the pandemic in March 2020, had to be revised and updated when advocacy groups raised concerns about using the CFS to withhold care from some patients.6

Clearly, frailty does have great relevance to health service use, healthcare outcomes, and design of services. For instance, in over 75s registered with NHS practices, severe frailty as defined by an electronic frailty index is associated with far higher risk of hospital admission, death, or care home admission in the following 12 months.7 People with frailty have less functional reserve and are far more likely (with or without covid-19) to present with immobility, falls, confusion, or generalised failure to thrive, or to get stranded in hospital or experience acute loss of function.89 Those in care homes, or who have dementia or are receiving home care or post-acute rehabilitation, will often be frailer and older.10 These groups are all at high risk from covid-19.

I think that the explicit use of a CFS cut-off was a well intentioned attempt to put some structure and transparent decisions around the rationing, or at least targeting, of critical care for the people with covid-19 who were most likely to benefit when services were likely to be overwhelmed. And since the early pandemic decisions, several systematic reviews and cohort studies have indeed shown close correlations between very high scores and a poor chance of survival or benefit.1112131415 Better, surely, an explicit than a covert or opaque decision making process, and one that evolves as empirical evidence emerges?

But, apart from the CFS not being designed or validated for the subsequently excluded groups, there was significant disquiet among the public, patients, media, and lobbying groups about what could be seen as the crude, depersonalising use of scales to determine whether someone might be given a chance of living.161718 Don’t all such assessments need to be based on personalised, individual clinical judgments, with a scale used merely to guide decision making? Beyond this, instruments such as the electronic frailty index, CFS, or Fried frailty index19 are not designed to predict outcomes or the ability to benefit from critical or subcritical care in a SARS virus pandemic.

So, what have we learnt over the past few months about using frailty scores in clinical practice? An editorial in the British Journal of Anaesthesia20 concluded that tools such as the CFS could never be used in isolation from individual clinical judgment or more recent, dynamic information about the patient’s current physiology and acute comorbidities.

NICE itself has emphasised that patients’ wishes, preferences, and best interests are a key factor in all of this. Let’s keep it that way, but let’s also not be squeamish in discussing these hard choices with the public, our patients, and the press.