Frailty and Geriatric Assessment Instruments

Frailty instruments take two fundamentally different approaches and produce different answers about the same person. Phenotype models such as the Fried Frailty Phenotype define frailty as a syndrome with specific components including weakness, slowness, exhaustion, low activity and weight loss. Cumulative deficit models count how many things are wrong out of a long list. Judgement-based scales such as the Clinical Frailty Scale ask an experienced clinician to place the person on a described spectrum. Brief screens such as FRAIL and the Edmonton Frail Scale sit between these, and the SARC-F screens specifically for sarcopenia rather than frailty as a whole.

This section also holds instruments about the people around the patient. The Caregiver Strain Index and the Caregiver Reaction Assessment measure a burden that predicts admission and placement independently of the patient's own scores, and that no patient-facing instrument captures.

In practice these instruments are used to decide intensity of care rather than to diagnose. A frailty score informs decisions about surgical risk, treatment aggressiveness, rehabilitation intensity and discharge planning. That makes the reproducibility of the score more important than its precision, which is part of why the Clinical Frailty Scale has spread so widely despite being a judgement call rather than a measurement.

The main practical complication is that frailty overlaps heavily with disability, comorbidity and cognitive impairment without being identical to any of them. Instruments that mix these together produce a number that is easy to record and hard to interpret, because two people with the same score can be frail for entirely different reasons and need entirely different plans.

Check which frailty model the instrument uses, and do not compare across models. A phenotype instrument and a deficit-count instrument will classify overlapping but different groups of people as frail, and pooling them in an analysis produces a category that means nothing.

Check whether the instrument requires physical measurement. Phenotype models typically need grip strength and gait speed, which requires equipment, space and a participant able to attempt the tasks. In a remote study this is a real constraint, and a questionnaire-only substitute is a different instrument.

Check the reference population for cut-offs, particularly for sarcopenia screening, where thresholds differ by sex and by regional working group definitions. Finally, in this population sensory and cognitive impairment affect completion rates for every self-report instrument, so plan for assisted completion and record when it was used.

What is the difference between frailty and disability?

Disability describes what a person cannot currently do. Frailty describes reduced physiological reserve, meaning reduced ability to withstand a stressor such as surgery or infection. A person can be frail without being disabled and disabled without being frail, and instruments that conflate the two are hard to act on.

Which frailty instrument should I use?

It depends on setting and available time. The Clinical Frailty Scale is quick and needs clinical judgement rather than equipment. Phenotype models need physical measurements and give a more mechanistic answer. Brief screens such as FRAIL suit primary care and large studies. Choose one and use it consistently, because scores do not transfer between models.

Can frailty be assessed remotely?

Partly. Judgement-based and questionnaire-based instruments can be adapted. Phenotype models that require grip strength and gait speed cannot be completed remotely without substituting the physical components, which changes what is being measured.

Why measure caregiver strain?

Because it predicts outcomes that patient-facing instruments do not, including admission and placement, and because it is often the factor that determines whether a care plan is sustainable. It is measured on the carer, not the patient, and needs its own consent pathway.