Back to 2011 Program

Objective(s): Identify differences in utilization of Palliative Care Medicine (PCM) consultations among various ICUs and its effect on invasive procedures, hospital length of stay (LOS) and ICU stay.
Design: Retrospective database and chart review. Univariate analysis between type of ICU and key variables. Logistic regression between key variables and two dichotomized versions of invasive procedures (any, multiple).
Setting: Tertiary university affiliated institution.
Patients: All patients >75 years, admitted to medical, surgical or neuro ICU for >48hrs and <30 days with palliative care consultation.
Interventions: None.
Main Outcome Measure(s): Utilization and timing of PCM, number of invasive procedures, LOS.
Results: 142 patients were included from Jan 2009 - Dec 2010. Univariate analysis showed a significantly higher number of invasive procedures in the surgical ICU patients (p=0.01). Patients whose palliative care referral was made after 5 days from ICU admission had significantly longer ICU stay (12.5 vs. 7.0 days, p<0.001). In logistic regression, admission to surgical ICU was a predictor of at least one invasive procedure (p=0.01) but there was no significant difference between medical and surgical ICUs for multiple invasive procedures (p=0.051). Late PCM utilization was a predictor of multiple invasive procedures (p=0.01). In both cases, APACHE score was not a significant predictor (p=0.70).
Conclusions: Timing of palliative care involvement in critically ill elderly patients has a significant effect on the number of invasive procedures, and ICU length of stay. Surgical ICU is associated with more invasive procedures and a slightly delayed PCM intervention.

ICU	Age	Apache	Mortality	Days to death	Days to PCM	Early PCM consult	Invasive procedures	LOS days	ICU days
Medical (N=75)	83.2	20.6	81.3%	21.6	6.7	50.7%	1.1	18.2	9.4
Surgical (N=24)	83.2	20.7	87.5%	20.4	9.6	33.3%	1.9*	18.0	9.8
Neuro (N=43)	82.9	19.9	83.7%	21.5	8.3	53.5%	1.1	16.1	10.6