Table 2.
Secondary efficacy variables: cross-titration/stabilization and relapse-prevention phases: final analysis
| Visit | Cross-titration/stabilization (N = 629) |
Relapse prevention | |
|---|---|---|---|
| Iloperidone (N = 151) |
Placebo (N = 150) |
||
| PANSS total score, change from baseline to last visit | |||
| N | 607 | 150 | 149 |
| Baseline mean (SD) | 76.8 (11.54) | 55.6 (10.76) | 55.2 (10.08) |
| Change mean (SE) | −14.0 (14.05) | 1.1 (1.12)* (Adj) | 12.4 (1.15) (Adj) |
| CGI-S, last visit | |||
| N | 609 | 112 | 105 |
| Median | 3.0 | 3.0 | 4.0 |
| SDS total score, change from baseline to last visit | |||
| N | 586 | 142 | 128 |
| Baseline mean (SD) | 15.6 (7.42) | 8.8 (6.47) | 8.0 (6.31) |
| Change mean (SE) | −4.6 (7.62) | −0.2 (0.54)** (Adj) | 1.8 (0.61) (Adj) |
OC data from the safety set are presented for the PANSS in the cross-titration/stabilization phase; LOCF data are presented for the PANSS and SDS for the relapse-prevention phase. The CGI-S reflects OC data, as this measure does not include a baseline assessment. Baseline refers to the respective study phases shown at the top of each column that displays data. Cross-titration/stabilization baseline is defined as the last available assessment prior to the start of medication in the cross-titration phase. Relapse prevention baseline is defined as the last available assessment prior to the first dose of double-blind study medication. Change from cross-titration/stabilization baseline is calculated as post value minus cross-titration/stabilization baseline value. A negative change indicates improvement. Change from relapse prevention baseline is calculated as post value minus relapse prevention baseline value. A negative change indicates improvement. P values are based on an ANCOVA model with treatment and site as main effects and relapse prevention baseline as a covariate. Adjusted change is the least squared mean change obtained from the ANCOVA model
Adj adjusted, ANCOVA analysis of covariance, CGI-S Clinical Global Impression of Severity, LOCF last observation carried forward, OC observed cases, PANSS Positive and Negative Syndrome Scale, SD standard deviation, SDS Sheehan Disability Scale, SE standard error
* p < 0.0001, ** p = 0.0062, log-rank test