Cost-effective critical care: What does it look like? Scott D. Halpern, M.D., Ph.D. Associate Professor of Medicine, Epidemiology, and Medical Ethics & Health Policy Director, Fostering Improvement in End-of-Life Decision Science (FIELDS) program Deputy Director, Center for Health Incentives and Behavioral Economics (CHIBE)
Improving cost effectiveness requires: Improving outcomes Reducing Costs
Options for improving CC cost-effectiveness 1. Improve (long-term) outcomes 2. Improve value of services used for admitted patients 3. Admit (only) patients who truly benefit from the ICU 4. Improve efficiency reduce LOS without worse outcomes 5. Reduce ICU staffing costs without threatening outcomes 6. Develop cheaper (non-icu-based) care pathways
Options for improving CC cost-effectiveness 1. Improve (long-term) outcomes 2. Improve value of services used for admitted patients 3. Admit (only) patients who truly benefit from the ICU 4. Improve efficiency reduce LOS without worse outcomes 5. Reduce ICU staffing costs without threatening outcomes 6. Develop cheaper (non-icu-based) care pathways
Options for improving CC cost-effectiveness 1. Improve long-term outcomes 2. Improve value of services used for admitted patients 3. Admit (only) patients who truly benefit from the ICU 4. Improve efficiency reduce LOS without worse outcomes 5. Reduce ICU staffing costs without threatening outcomes 6. Develop cheaper (non-icu-based) care pathways
Why so much variability in catheter use? Gershengorn, et al. Anesthesiology 2014
Top 5 List in Critical Care Medicine released January 11, 2014 www.choosingwisely.org Problem: ~85% of critical care costs are fixed (beds, docs, nurses, etc)
Options for improving CC cost-effectiveness 1. Improve long-term outcomes 2. Improve value of services used for admitted patients 3. Admit (only) patients who truly benefit from the ICU 4. Improve efficiency reduce LOS without worse outcomes 5. Reduce ICU staffing costs without threatening outcomes 6. Develop cheaper (non-icu-based) care pathways
DKA patients don t seem to benefit from ICU No relationship with hospital mortality or LOS Gershengorn NB, et al. Crit Care Med 2012
Heart failure patients don t seem to benefit from ICU No relationship with hospital mortality Safavi KC, et al. Circulation 2013
Patients with PE don t seem to benefit from ICU No relationship with hospital mortality, cost, or readmission Admon AJ, et al. Chest 2014
ICU utilization associated with procedures Chang DW, Schapiro MF. JAMA-IM 2016
ICU utilization associated with higher costs Chang DW, Schapiro MF. JAMA-IM 2016
ICU utilization not associated with mortality Chang DW, Schapiro MF. JAMA-IM 2016
Perhaps pneumonia patients benefit? Valley TS, et al. JAMA 2015
Leveraging within-hospital variability in ICU admission to define net ICU benefit R01 HL136719 Anesi GL et al. (under review)
Options for improving CC cost-effectiveness 1. Improve long-term outcomes 2. Improve value of services used for admitted patients 3. Admit (only) patients who truly benefit from the ICU 4. Improve efficiency reduce LOS without worse outcomes 5. Reduce ICU staffing costs without threatening outcomes 6. Develop cheaper (non-icu-based) care pathways
ICU strain improves efficiency among patients who survive ICU stay * Extreme increases in all 3 strain measures yielded 6.3 hour reduction in ICU LOS and 1% increase in ICU readmissions among discharged patients (both p < 0.02) No differences in subsequent mortality or probability of being discharged home Wagner J, et al. Ann Intern Med 2013; 159: 447-455
ICU strain also improves efficiency among patients who will die 10,000 patients dying in 161 U.S. ICUs 2001-2008 Hua, M et al. Intensive Care Medicine 2016
Steroids in sepsis (probably) don t save lives But they do shorten LOS without increasing adverse outcomes What s the debate all about?
Options for improving CC cost-effectiveness 1. Improve long-term outcomes 2. Improve value of services used for admitted patients 3. Admit (only) patients who truly benefit from the ICU 4. Improve efficiency reduce LOS without worse outcomes 5. Reduce ICU staffing costs without threatening outcomes 6. Develop cheaper (non-icu-based) care pathways
Nighttime intensivists add costs, don t save lives Kerlin MP, et al. Am J Resp Crit Care Med (in press)
Nighttime intensivists don t improve efficiency Rate ratio: 0.98 (0.88 1.09) Rate ratio: 0.98 (0.84 1.13) No effects on ICU/hospital mortality or probability of discharge home Kerlin MP, et al. N Engl J Med 2013; 368: 2201-9
Future directions Hypothesis: optimal ways to improve ICU cost-effectiveness: 1) Better manage ambulatorycare-sensitive conditions 2) identify Identify those patients too too well well to to benefit (in from ER) ICU and (e.g., design in ED) alternate and design triage alternate and care triage pathways and care pathways 3) Combat ward strain -- to reduce ICU readmissions and rapid responses, and decrease ICU length of stay
Penn Acute Care Health Services Research Group Meeta Kerlin, MD, MSCE Elizabeth Cooney, MPH Kate Courtright, MD Dylan Small, PhD Michael Josephs Joanna Hart, MD, MSHP Anna Buehler Sydney Brown, MD, PhD Meghan Lane-Fall, MD, MSHP Rachel Kohn, MD Kit Delgado, MD, MS Sarah Ratcliffe, PhD Mark Mikkelsen, MD, MSCE Michael Detsky, MD, MSHP Michael Harhay, PhD (c) Nicole Gabler, PhD Kuldeep Yadav Gary Weissman, MD Mark Neumann, MD, MSc George Anesi, MD, MBE Alexis Zebrowski, PhD (c) shalpern@exchange.upenn.edu