Table 2.

A mixed model for repeated measures showing PUQE scores on days 3, 4, 5, and 15, using the ITT-E population

Nature of statistic	N	Diclegis LS Mean	Placebo LS Mean	Estimated treatment effect (Diclegis minus placebo)	95% confidence interval	P value
Number of Observations Used	920
Number of Subjects Included in Analysis	252
Overall Tests of Fixed Effects
TRT						0.0004
DAY						<.0001
TRT*DAY						0.3374
BASELINE						<.0001
Treatment Effects by Day
Day 3		–3.03	–2.15	–0.88	–1.43 to–0.33	0.0019
Day 4		–3.45	–2.51	–0.95	–1.48 to–0.41	0.0006
Day 5		–3.81	–3.05	–0.76	–1.29 to–0.24	0.0045
Day 15		–5.00	–4.60	–0.40	–0.86 to 0.06	0.0867

LS Mean Least Squares Mean

The change from baseline in PUQE score was the dependent variable, and treatment, study day, and treatment by study day interaction were included as categorical fixed effects, and the baseline score was included as a continuous fixed effect. The correlation between measures on different days within the same subject was modeled with an unstructured covariance matrix. Least squares means for the change from baseline at each treatment and timepoint were calculated from the model, along with the estimated treatment effect (Diclegis minus placebo) at each timepoint and its 95% confidence interval and P value