TABLE 3.

Comparing PT versus PH approaches

# Events N for PT assumption (80% power, α = 0.05 one-sided, ΔPT = 2, r = 1)											# Events N for PH assumption (80% power, α = 0.05 one-sided, r = 1)
		β											N
		4	3	2	1.5	1	0.75	0.5	0.25	0.1		0.85	938
	0.10	-	-	-	-	-	-	-	10	52		0.8	498
	0.25	-	-	-	-	-	-	14	52	322		0.75	300
	0.50	-	-	-	-	14	24	52	208	1288		0.7	196
	0.75	-	-	-	14	30	54	118	466	2898		0.65	134
|λ|	1.00	-	-	14	24	52	94	208	826	5150	ΔHR	0.6	96
	1.50	10	16	32	56	120	210	466	1858	11586		0.55	70
	2.00	16	28	56	96	212	372	828	3300	20596		0.5	52
	2.50	26	44	88	150	332	582	1294	5158	-		0.45	40
	3.00	38	64	128	218	478	838	1864	7426	-		0.4	30

N is rounded up to an even number to avoid fractional values for n₀ and n₁.

A “-” indicates that N is either too small or too large to be considered practically meaningful.

This table shows N for a one-sided hypothesis. Similar calculations can be done for a two-sided hypothesis.

This table is constructed for Δ_PT = 2. For Δ_PT > 2, N will decrease and vice versa.