Table 2.

Multiple regression* of clinical characteristics on Vitamin D Level in 376 patients with gastroparesis (delayed gastric emptying)

Clinical Characteristic	Comparison	Cumulative Odds Ratio of decreasing Vitamin D level	95% CI	P-value
Age	Per year	0.97	0.96, 0.99	0.001
Race	White vs non-white	0.36	0.17, 0.76	0.007
Body mass index	Per kg/m2	1.04	1.01, 1.07	0.02
Season	Summer vs other seasons	0.47	0.28, 0.80	0.005
Diabetes	Yes vs no	0.47	0.23, 0.98	0.04
HbA1c	Per %	1.28	1.08, 1.51	0.004
Nausea	Per score	1.30	1.12, 1.50	0.001
GES at 4 h	Per %	1.01	1.00, 1.02	0.01
Change in normogastria	Per %	0.95	0.92, 0.98	< 0.001
Change in bradygastria	Per %	0.96	0.94, 0.99	0.009
Change in tachygastria	Per %	0.96	0.93, 1.00	0.04

^*Multiple ordinal logistic regression model using backward stepwise selection (P > 0.05 for removal) regressing Vitamin D deficiency categorization (0 = normal, 1 = insufficient, 2 = deficient) on 39 candidate variables including age, race, sex, ethnicity, summer, BMI, HbA1c, diabetes, GE at 1 h., GE at 2 h., GE at 4 h., nausea, retching, vomiting, stomach fullness, not able to finish meal, feeling excessively full, loss of appetite, bloating, stomach visibly larger and calcium, water load amount, following symptoms at baseline and change from baseline: fullness, hunger, nausea, bloating and abdominal discomfort, following GMA regions at baseline and change from baseline: bradygastria, normogastria, tachygastria, duodenal-respiratory range