Table 1. Data layout under four recurrent event models with patient 1 having three events (at times 8, 12 and 26) and patient 2 having two events (at times 10, 18).

Model	Patient i.d.	Time interval	Eventa	Stratumb
Conditional A	1	(0,8]	1	1
	1	(8,12]	1	2
	1	(12,26]	1	3
	1	(26,31]	0	4
	2	(0,10]	1	1
	2	(10,18]	1	2
Conditional Bc	1	(0,8]	1	1
	1	(0,4]	1	2
	1	(0,14]	1	3
	1	(0,5]	0	4
	2	(0,10]	1	1
	2	(0,8]	1	2
Marginal model	1	(0,8]	1	1
	1	(0,12]	1	2
	1	(0,26]	1	3
	1	(0,31]	0	4
	2	(0,10]	1	1
	2	(0,18]	1	2
	2	(0,18]	0	3
	2	(0,18]	0	4
Independent increment	1	(0,8]	1	1
	1	(8,12]	1	1
	1	(12,26]	1	1
	1	(26,31]	0	1
	2	(0,10]	1	1
	2	(10,18]	1	1

^a1=had event of interest, 0=censored.

^bRelates to the number of events and is used in the fitting of the model as the strata variable.

^cGap time model.