Research Letter

摘要

SIR—We thank Dr Chalcroft for his comments relating tonour paper [3]. The categories of the Reported Edmonton FrailnScale (REFS) to determine frail and non-frail patients werenreported under the sub-heading of ‘Data Analysis’ in thenMethods section of the paper. The cut-off scores to determinenfrailty remained the same as those used in the originalnEdmonton Frail Scale [4] with not frail (0–5) and apparentlynvulnerable (6–7) making up the ‘non-frail’ group and thenmildly frail (8–9), moderately frail (10–11) and severely frailn(12+) making up the ‘frail’ group. The groups were combinednin order to maximise the power of the data to find differences,nparticularly when undertaking survival analysis to determinenkey outcomes (stroke, haemorrhage, death).