Table 5.

Logistic Regression Analyses of Enrollment Characteristics as Predictors of Severe Vomiting^* in Idiopathic and Diabetic Gastroparetics (n=159)

Enrollment characteristic	Unadjusted Analyses			Adjusted Analyses
	OR	CI	P†	OR	CI	P‡
Etiology (diabetic vs idiopathic)	2.12	(0.87, 5.21)	0.10	1.02	(0.19, 5.59)	0.98
Age, years	0.98	(0.94, 1.01)	0.16	0.97	(0.93, 1.01)	0.16
Race (white vs non-white)	0.22	(0.08, 0.64)	0.006	0.20	(0.05, 0.80)	0.02
PAGI-SYM, satiety sub-score	1.95	(1.18, 3.21)	0.009	2.17	(1.19, 3.95)	0.01
PAGI-SYM, bloating sub-score	1.19	(0.89, 1.60)	0.24	0.62	(0.41, 0.95)	0.03
PAGI-SYM, GERD sub-score	1.60	(1.15, 2.22)	0.005	1.77	(1.16, 2.72)	0.009
PAGI-QOL score	0.63	(0.42, 0.95)	0.03	0.71	(0.42, 1.20)	0.20
HbA1c, %	1.28	(1.04, 1.59)	0.02	1.29	(0.85, 1.95)	0.24

^*Severe vomiting defined as ‘severe’ or ‘very severe’ vomiting score on the PAGI-SYM instrument

^†Unadjusted odds ratios, 95% confidence limits, P values determined from logistic regression models of severe vomiting on each predictor

^‡Adjusted odds ratios, 95% confidence limits, P values were determined from a multiple logistic regression analyses of severe nausea using all baseline predictors indicated. This model was determined from Akaike Information criteria (AIC) with forward selection using a candidate set of baseline variables: gender, age at enrollment, etiology, race, SF-36 physical score, SF-36 mental score, PAGI-QOL total score, solid GES 2 hour retention percent, solid GES 4 hour retention percent, HbA1c, ESR, gastroparesis severity, and the following PAGI-SYM items: satiety sub-score, bloating sub-score, upper abdominal pain sub-score, and GERD sub-score. Etiology was forced into the model.