以 relative risk 驗證 individual person level responsiveness

Hobart 以 relative risk (RR) 驗證 FIM 是否比 BI 能呈現較多的個案具備顯著ADL進步
用在我的數據也就是驗證 PASS 是否比 SFPASS 能呈現較多的個案具備顯著 balance 進步

結果如下：

Table. Individual patient level responsiveness of the Postural Assessment Scale for Stroke Patients (PASS) and short form PASS (SFPASS)

Group	PASS	SFPASS
	% (n)	% (n)	Relative risk (95% CI, P)
Significant improvement	51.2 (133)	41.5 (108)	1.23 (1.02 – 1.48, 0.029)
Non-significant improvement	40.0 (104)	55.4 (144)	0.72 (0.60 – 0.87, 0.001)
Others	8.8 (23)	3.1 (8)	2.88 (1.31 – 6.31, 0.008)

RR 的算式是 [(PASS 呈現個案具備顯著進步之 probability) / (SFPASS 呈現個案具備顯著進步之 probability)] 也就是 [(133/260)/(108/260)] = 1.23。解釋上可謂：PASS呈現個案層級顯著進步的能力是SFPASS的 1.23 倍

SPSS 計算 RR & p value 有些複雜。還好網路上： http://www.medcalc.org/calc/relative_risk.php 可協助計算這些數值。

語意上， odds ratio （勝算比）是否更好！？概念上亦然！？