Table 4.

Multivariable Models for Describing Factors Associated With a Short Hospice Stay (≤3 Days)

	Full Model		Trimmed Model		Slim Model
	OR (95% CI)	P Value	OR (95% CI)	P Value	OR (95% CI)	P Value
Age at consent: <65 vs ≥65 y	1.12 (1.01–1.24)	.04	1.13 (1.01–1.25)	.03	1.12 (1.01–1.24)	.03
Sex: female vs male	0.98 (0.88–1.08)	.68	0.97 (0.87–1.07)	.50
Race: White vs Black or other race	1.07 (0.94–1.22)	.29	1.07 (0.94–1.22)	.30
Facility: referred to vs originally treated at JHH	1.13 (0.96–1.32)	.14
Referred to palliative care before disease progression: yes vs no	0.84 (0.73–0.97)	.02	0.84 (0.76–0.93)	<.001	0.83 (0.75–0.92)	<.001
Goals of care: documented vs not documented	0.98 (0.84–1.14)	.80
Remained at JHH after progression: yes vs no	0.80 (0.68–0.94)	.008	0.73 (0.66–0.81)	<.001	0.72 (0.65–0.80)	<.001
Hospice care: discussed after progression vs not discussed	1.01 (0.90–1.14)	.85

All variables were included in a single logistic regression model, shown as one model per column of results.

Abbreviations: JHH, Johns Hopkins Hospital; OR, odds ratio.