Table 3.

Mean Change from Baseline to First Relief (Standard Error) of TNSS and Individual Symptom Scores for ICX72 and Placebo Groups

Symptom	ICX72 (N = 20)			Placebo (N = 22)*
	Baseline	Change: baseline –first relief	% Rel change	Baseline	Change: baseline –first Relief	% Rel change
TNSS	23.7 (2.2)	3.4 (1.5)	14%	26.0 (2.2)	2.3 (7.7)	9%
Nasal congestion	5.1 (0.5)	1.3 (0.4)	25%	5.3 (0.4)	0 (2.2)	0
Sinus pressure	5.4 (0.5)	1.7 (0.5)	32%	3.7 (0.4)	−0.1 (1.9)	−1%
Sinus pain	4.5 (0.5)	1.2 (0.6)	27%	4.8 (0.5)	−0.2 (2.7)	−5%
Headache	4.9 (0.5)	2.2 (0.6)	44%	4.8 (0.5)	−0.1 (2.0)	−2%
Sneezing	1.6 (0.4)	1.2 (0.3)	75%	2.4 (0.4)	0.5 (1.8)	19%
Rhinorrhea	2.6 (0.5)	−2.3 (0.6)	−89%	2.6 (0.4)	−1.9 (4.3)	−74%
PND	4.7 (0.5)	0.4 (0.7)	7%	5.6 (0.5)	−0.6 (5.4)	−10%

Abbreviation: N, number of patients.

For each symptom, baseline differences between groups were not statistically different. Mean changes from baseline to first relief were significantly different between ICX72 and placebo for nasal congestion, sinus pressure, and headache (P < .05). Percent relative change of each symptom was larger for ICX72, compared with placebo, except rhinorrhea.

^*Four of 22 placebo patients had missing data at first relief for each symptom. When a symptom at first relief was missing, less relief was assumed than was determined from all reported symptom changes at first relief, after adjusting for baseline. The minimum values at first relief of missing data were symptom-specific. Mean symptom changes in the placebo group were obtained from Tobit analysis, which accounts for the upward bias of changes calculated from observed data only. The probability of unreported symptoms reflecting less relief than the minimum relief of all other patients was combined with the probabilities of the observed symptom values at first relief, to estimate mean changes at first relief of each symptom for the placebo group.