PRECIOUS: PREvention of Complications to Improve OUtcome in elderly patients with acute Stroke—statistical analysis plan of a randomised, open, phase III, clinical trial with blinded outcome assessment

Table 2 Primary outcome. Analyses are adjusted except where stated

	Analysis	Paracetamol	Control	DIM or OR (95% CI	Metoclopramide	Control	DIM or OR (95% CI	Ceftriaxone	Control	DIM or OR (95% CI
mRS, median	aOLR	Median [IQR]	Median [IQR]	aOR (95% CI)	Median [IQR]	Median [IQR]	aOR (95% CI)	Median [IQR]	Median [IQR]	aOR (95% CI)
Sensitivity analyses
mRS, unadjusted	OLR	Median [IQR]	Median [IQR]	OR (95% CI)	Median [IQR]	Median [IQR]	OR (95% CI)	Median [IQR]	Median [IQR]	OR (95% CI)
mRS, imputed	aOLR	Median [IQR]	Median [IQR]	aOR (95% CI)	Median [IQR]	Median [IQR]	aOR (95% CI)	Median [IQR]	Median [IQR]	aOR (95% CI)
mRS, mean	aMLR	Mean (SD)	Mean (SD)	aDIM (95% CI)	Mean (SD)	Mean (SD)	aDIM (95% CI)	Mean (SD)	Mean (SD)	aDIM (95% CI)
mRS > 2	aBLR	n (%)	n (%)	aOR (95% CI)	n (%)	n (%)	aOR (95% CI)	n (%)	n (%)	aOR (95% CI)
Death	aCPHR	n (%)	n (%)	aHR (95% CI)	n (%)	n (%)	aHR (95% CI)	n (%)	n (%)	aHR (95% CI)

Data are n (%), median [IQR], and mean (SD). aDIM adjusted difference in means, aHR adjusted hazards ratio, aOR adjusted odds ratio. Comparison by adjusted ordinal logistic regression (aOLR), multiple linear regression (aMLR), cox proportional hazards regression (CPHR), or adjusted binary logistic regression (aBLR)

ISSN: 1745-6215