Examining systemic steroid use in older inflammatory bowel disease patients using hurdle models: a cohort study

Table 1 Baseline Participant Characteristics, Overall and by Drug Therapy

	Full cohort (n = 1682)		Ever systemic steroid users (n = 726)^a		Ever nonbiologic users (n = 85)^a
Sociodemographics^b	n	%	n	%	n	%
Age Mean (SD)^c		79 (8)		78 (7)^e		75 (7)^e
Female	1136	68	494	68	50	59
Caucasian	1474	88	654	90^d	80	94
Region
Northeast	444	26	182	25	19	22
Southeast	459	27	208	29	24	28
Midwest	417	25	189	26	28	33
Rocky Mountain	34	2	12	2	<11^g	<11^g
Southwest	163	10	96	10	<11^g	<11^g
Pacific (incl HI, AK, PR)^h	165	10	68	9	<11^g	<11^g
Urban/Suburban	1266	76	516	72^e	61	72
Medicaid coverage	547	33	219	30	21	25
Clinical^b	n	%	n	%	n	%
Polypharmacy (>5 drugs)	1290	81	602	85^e	69	83
Charlson Index Mean (SD)		4 (3)		4 (3)		4 (2)
Primary managing provider type
Primary Care Provider	769	46	320	44.4	38	44.7
Gastroenterologist	19	1	<11^g	<11^g	<11^g	<11^g
Other specialists	868	52	393	55	44	52
antiTNF	37	2	26	4^e	18	21^e
Aminosalicylates	374	22	215	30^e	38	45^e
Locally administered steroids	12	1	<11^g	<11^d,g	<11^g	<11^g
Antidiarrheals	209	12	111	15^e	14	17
>1 endoscopy	452	27	201	28	25	29
IBD surgery	106	6	41	6	<11^g	<11^g
Hospitalizations Mean (SD)		2 (2)		2 (2)^e		2 (2)
ED visits Mean (SD)		1 (2)		1 (2)^e		1 (2)

^aBivariate analyses is the comparison of Steroid Use to NonUse; Nonbiologic Use to Nonuse; ^bBaseline characteristics were ascertained during 12 months prior to study inclusion; ^cSD = standard deviation ^d p < .05, ^e p < .01, ^f p < .001; ^gCell sizes are too small to include variable in regression model & requires cell suppression
^hHI = Hawaii, AK = Alaska, PR = Puerto Rico

ISSN: 2050-6511