Table 4.

Treatment persistence, by the number of treatment episodes and period (using a gap of >30 days as discontinuation of medications)

	Period 1 (10/1/1999–3/31/2002)
Drug class	1 episode		2 episodes			3 episodes
	Mean ± SD	N	(1) Mean ± SD	(2) Mean ± SD	N	(1) Mean ± SD	(2) Mean ± SD	(3) Mean ± SD	N
Atypical agents	197 ± 143	14,352	68 ± 57	151 ± 97	2,505	50 ± 38	51 ± 35	114 ± 111	480
Typical agents	139 ± 133	3,126	62 ± 53	113 ± 112	707	54 ± 38	42 ± 27	88 ± 88	156
	Period 2 (10/1/2002–3/31/2005)
Drug class	1 episode		2 episodes			3 episodes
	Mean ± SD	N	(1) Mean ± SD	(2) Mean ± SD	N	(1) Mean ± SD	(2) Mean ± SD	(3) Mean ± SD	N
Atypical agents	174 ± 141	18,523	67 ± 57	139 ± 126	3,305	51 ± 37	47 ± 32	114 ± 107	712
Typical agents	141 ± 130	2,698	61 ± 61	133 ± 127	546	57 ± 45	46 ± 30	68 ± 64	109

The conventional approach in calculating treatment persistence included the prescription episode among patients with only one prescription episode, the first prescription episode among those with two prescription episodes, and the first prescription episode among those with three prescription episodes (shaded columns).