Response to Dr Caballero-Mora's comments

Reply to letter submitted 15 May April 2017 regarding Age and Ageing paper https://doi.org/10.1093/ageing/afx005. Dear Sir, In response to Dr Caballero-Mora's comments, we wish to make the following number of points. First, STOPPFrail criteria were not developed to identify frailty but rather assist in accurately recognising patients in whom a high proportion of preventative medications should be stopped. The four preliminary criteria (end-stage irreversible pathology plus a poor 1-year survival prognosis plus severe functional and/or cognitive impairment plus symptom control as the first priority) were agreed upon, by an expert Delphi panel, to be the most important parameters in identifying these patients. Many older adults with a poor 1-year survival prognosis are not clinically frail. Conversely, some older adults who are clinically frail do not have a poor one year survival prognosis, however, for many patients frailty and poor survival prognosis do co-exist. Therefore, we developed our criteria to identify, in our and our panel's opinion, the sicker frailer cohort of patients with a life limiting condition in whom most preventive treatments are least likely to be of benefit. We did not intend that STOPPFrail would be applicable in all frail older people, thus a frailty index was not appropriate to identify our target population. As Dr Caballero-Mora correctly points out, frailty in some circumstances can be reversible, but in patients where there is an end-stage diagnosis e.g. dementia or disseminated progressive malignancy with associated severe functional impairment, frailty is implicit and prognosis is likely to be poor. The literature to date supports that co-morbidities and functional impairment are the most consistent predictors of mortality and our criteria reflect this [1]. The articles Dr Caballero-Mora makes reference to further support the view that ‘frailty is sometime difficult to define’ as different criteria are used in each frailty tool. Of 79 frailty instruments reviewed by Azzopardi RV et al. [2], 24 include physical components alone, with all other instruments consisting of two or more components from the following domains: medical, physical, functional, cognitive, psychological and/or social components. To confound matters further the above tools have been validated across different healthcare settings and for different age groups. Lastly, of the frailty instruments reviewed some components are self-reported, others are objective measures and in those remaining a combination of both are employed. Finally, we intend that STOPPFrail criteria will assist physicians’ clinical judgement in deciding which patients are appropriate for discontinuation of certain medications, but not replace clinical judgement. In this context, STOPPFrail is not intended to be a rigid set of deprescribing rules. Similarly, frailty indices are used to assist identification and diagnosis of frailty. Prescribing and frailty tools are not intended to supersede or dictate clinical judgement but rather guide physicians in clinical decision-making. We believe that STOPPFrail accurately identifies the sicker frailer cohort of patients in whom medication should, in most cases, be deprescribed and we anticipate that STOPPFrail will be of the use as both a clinical and research tool in the future. None declared. Yours sincerely, etc.