Table 3.

Sample size estimates for varying absolute difference in proportions, given the follow-up window and mortality outcome

Computed Total N (2-Group Trial)			Absolute Difference in Proportions							SS Needed in PROPPR
Follow-up window	Mortality Outcome	Reference Proportion	0.02	0.04	0.05	0.07	0.08	0.1	0.12	SS Needed in PROPPR
30-day	Overall	0.22	19232	4950	3212	1682	1302	854	604	5826
30-day	Any HS	0.11	11466	3068	2026	1096	862	580	422	3008
24-hour	Overall	0.13	13074	3458	2272	1218	954	638	462	2968
24-hour	Any HS	0.09	9772	2656	1766	966	764	520	382	1558
6-hour	Overall	0.1	10630	2864	1898	1032	814	550	402	3264
6-hour	Any HS	0.09	9772	2656	1766	966	764	520	382	2570
4-hour	Overall	0.09	9772	2656	1766	966	764	520	382	2252
4-hour	Any HS	0.08	8892	2440	1630	900	714	488	360	2084
3-hour	Overall	0.06	7068	1996	1348	760	608	422	316	1246
3-hour	Any HS	0.05	6122	1764	1202	686	552	388	292	1290
2-hour	Overall	0.05	6122	1764	1202	686	552	388	292	2722
2-hour	Any HS	0.047	5834	1694	1158	664	536	378	286	2502

Ref proportion from PROPPR (1:1:1 group)

Fixed scenario elements: asymptotic normal distribution, normal approximation method, alpha (adjusted for 3 looks) = 0.044, nominal power=0.9, 2 sided hypothesis test, 0 null proportion difference