Send to

Choose Destination
PLoS One. 2018 Feb 28;13(2):e0190569. doi: 10.1371/journal.pone.0190569. eCollection 2018.

Reversals and limitations on high-intensity, life-sustaining treatments.

Author information

Stanford University School of Medicine, Stanford, California, United States of America.
Center for Outcomes Research & Evaluation, Yale Medicine, New Haven, CT, United States of America.
Stanford Quantitative Sciences Unit, Stanford, California, United States of America.
Stanford Center for Clinical Informatics, Stanford, California, United States of America.
Department of Bioengineering, Stanford University School of Medicine, Stanford, California, United States of America.
VA Palo Alto Health Care System, Palo Alto, California, United States of America.
Department of Medicine, Stanford University School of Medicine, Stanford, California, United States of America.
Center for Biomedical Informatics Research and Division of Hospital Medicine, Department of Medicine, Stanford University School of Medicine, Stanford, CA, United States of America.



Critically ill patients often receive high-intensity life sustaining treatments (LST) in the intensive care unit (ICU), although they can be ineffective and eventually undesired. Determining the risk factors associated with reversals in LST goals can improve patient and provider appreciation for the natural history and epidemiology of critical care and inform decision making around the (continued) use of LSTs.


This is a single institution retrospective cohort study of patients receiving life sustaining treatment in an academic tertiary hospital from 2009 to 2013. Deidentified patient electronic medical record data was collected via the clinical data warehouse to study the outcomes of treatment limiting Comfort Care and do-not-resuscitate (DNR) orders. Extended multivariable Cox regression models were used to estimate the association of patient and clinical factors with subsequent treatment limiting orders.


10,157 patients received life-sustaining treatment while initially Full Code (allowing all resuscitative measures). Of these, 770 (8.0%) transitioned to Comfort Care (with discontinuation of any life-sustaining treatments) while 1,669 (16%) patients received new DNR orders that reflect preferences to limit further life-sustaining treatment options. Patients who were older (Hazard Ratio(HR) 1.37 [95% CI 1.28-1.47] per decade), with cerebrovascular disease (HR 2.18 [95% CI 1.69-2.81]), treated by the Medical ICU (HR 1.92 [95% CI 1.49-2.49]) and Hematology-Oncology (HR 1.87 [95% CI 1.27-2.74]) services, receiving vasoactive infusions (HR 1.76 [95% CI 1.28, 2.43]) or continuous renal replacement (HR 1.83 [95% CI 1.34, 2.48]) were more likely to transition to Comfort Care. Any new DNR orders were more likely for patients who were older (HR 1.43 [95% CI 1.38-1.48] per decade), female (HR 1.30 [95% CI 1.17-1.44]), with cerebrovascular disease (HR 1.45 [95% CI 1.25-1.67]) or metastatic solid cancers (HR 1.92 [95% CI 1.48-2.49]), or treated by Medical ICU (HR 1.63 [95% CI 1.42-1.86]), Hematology-Oncology (HR 1.63 [95% CI 1.33-1.98]) and Cardiac Care Unit-Heart Failure (HR 1.41 [95% CI 1.15-1.72]).


Decisions to reverse or limit treatment goals occurs after more than 1 in 13 trials of LST, and is associated with older female patients, receiving non-ventilator forms of LST, cerebrovascular disease, and treatment by certain medical specialty services.

[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for Public Library of Science Icon for PubMed Central
Loading ...
Support Center