Table 2.

Power comparisons of Cox regression, Time-dependent Cox regression and Landmark methods on simulated data under the alternative hypothesis that cumulative drug dose is significantly related to disease outcome

β 0	Cumulative dose	Cox regression	Time-dependent Cox regression	Landmark analyses
				τ = 5	τ = 7
log(0.015)	Low	214	58	119	74
	Moderate	1000	109	348	181
	High	1000	627	186	396
log(5∗0.015)	Low	723	47	289	103
	Moderate	1000	296	792	471
	High	1000	977	452	827
log(10∗0.015)	Low	957	52	288	94
	Moderate	1000	322	836	523
	High	1000	989	538	888

β ₀ values represent disease incidence rate; τ= Landmark time