Special Feature

Life After the ICU: Improving Decision Making About Chronic Critical Illness in the Acute Care Setting

By Betty Tran, MD, MS, Assistant Professor of Medicine, Pulmonary and Critical Care Medicine, Rush University Medical Center, Chicago, is Associate Editor for Critical Care Alert.

Dr. Tran reports no financial relationships relevant to this field of study.

In their 1985 article “The Chronically Critically Ill: To Save or Let Die?,” Girard and Raffin created the term “chronically critically ill” to describe patients admitted to an ICU who survived their acute insults but remained dependent on intensive care therapies.1 It is estimated that more than 100,000 such patients exist in the United States at any point in time, and this number is projected to increase over the coming years. Respiratory failure necessitating prolonged dependence on mechanical ventilation is invariably a key component of chronic critical illness (CCI). A consensus conference recommended a formal definition of prolonged mechanical ventilation (PMV) as ≥ 21 consecutive days on the ventilator for ≥ 6 hours/day, although many studies have also included patients who have undergone tracheotomy placement after ≥ 4 days of mechanical ventilation for an expected prolonged ventilator weaning course.2 Current evidence, however, suggests that CCI is a larger syndrome comprised of additional comorbidities such as profound neuromuscular weakness, brain dysfunction, endocrinopathy, malnutrition, increased vulnerability to infection, skin breakdown due to anasarca, incontinence, prolonged immobility, and symptom distress.3

Although prognosis for patients with CCI is dismal, this group uses a disproportionate amount of health care resources, and studies have revealed that both surrogate decision makers and physicians are overly optimistic about long-term outcomes. This special feature aims to present a current overview of CCI from economic and ethical perspectives and discuss strategies for improving discussion about CCI in the acute ICU setting.

Chronic Critical Illness: The Costly Few

Although only 5-10% of patients transition from acute to CCI with PMV, these patients generate 13% of all hospital costs in the United States, a figure that exceeds $20 billion per year.4 In one study, the provision of PMV costs $82,411 per quality adjusted life-year saved when compared to comfort care.4 Notably, these estimates do not include average 1-year costs accrued after discharge from an acute care setting related to hospital readmissions, interfacility transportation, and outpatient care, as well as the significant direct costs to families caring for patients with CCI in terms of lost productivity and work days.

The high costs for long acute care stays have created financial incentives for early transfer of patients with CCI to post-acute care facilities, such as long-term acute care hospitals (LTACs) where the mean length of stay is ≥ 25 days, and for further ventilator weaning, physical rehabilitation, and other intensive care therapies. This has resulted in a rapid increase in the number of LTACs (8.8% per year) and LTAC beds (5.9% per year) from 1997 to 2006.5 In contrast to acute care hospitals, however, LTACs have the ability to select their patient population for admission, are not governed by laws requiring treatment regardless of ability to pay, and often operate at high profit margins, all factors that can contribute to disparities in terms of access. Furthermore, the overall cost “savings” associated with LTACs may be offset by higher Medicare payments and the trend of earlier transfer to LTACs of increasingly sicker patients who may ultimately require subsequent readmission to acute care hospitals.3,6 Notably, the expansion of the LTAC model of care has yet to be supported by evidence that such venues improve patient outcomes.

Data from retrospective cohort studies have also revealed racial and insurance differences with regard to LTAC utilization. Patient-level hospitalization data from the Medicare Provider Analysis and Review files reveal a higher age-adjusted incidence of LTAC transfers for male and black individuals, the latter of which had more than twice the transfer rates as white individuals.5 Whether this observation reflects the preference among black patients for more aggressive care at the end of life, a higher incidence of CCI among black individuals, or greater accessibility given a high concentration of LTACs in urban areas is yet to be determined. In contrast, a separate study using data from the Pennsylvania Health Care Cost Containment Council found no race effect in patients aged ≥ 65 years of age (nearly all of whom have Medicare as their payor), but did find that for patients < 65 years of age, black race was significantly associated with a decreased odds of transfer to an LTAC, an observation that was attributed to insurance status (black patients were more likely to have Medicaid or be uninsured) and hospital-level effects (hospitals that tend to discharge to LTACs had a higher proportion of white patients).7 Given that LTACs have the freedom to select their patients based on criteria such as their ability to pay and may have special relationships with the referring acute care hospital, these findings are not surprising, although the disparities this can create are disturbing.

Great Expectations: Surrogate and Physician Predictions in Chronic Critical Illness

Recent studies have revealed significant weaknesses in the process of decision making regarding life support and prognostication in CCI. One-year outcomes for patients with PMV are grim: 56% are alive, only 27% report a good quality of life, and a mere 9% are at home and independently functioning.8 The average patient with PMV spent 74% of all days alive either in a hospital, a post-acute care facility, or at home with paid home care.8 Patients who reported a “good” quality of life based on ratings in areas of mobility, self-care, usual activities, pain/discomfort, and anxiety/depression tended to be younger, have fewer comorbid conditions, be admitted for trauma, and have private insurance.8

Given these dismal figures, would patients or their surrogates opt for aggressive care? If not, why is the incidence of CCI and LTAC utilization increasing? Unfortunately, data show that we are doing a poor job in the acute care hospital setting discussing CCI with patients and surrogates. Based on surrogate and physician interviews at the time of tracheotomy and 1-year follow-up, only 26% of surrogates reported that physicians discussed the patient’s prognosis for survival, functional limitations, quality of life, and expected caregiver needs.9 The majority of surrogates expected the patient to be alive (93%), have no major functional limitations (71%), and have a good quality of life (83%) at 1 year.9 This contrasted significantly with physicians’ expectations. Although 44% of physicians had expectations for patient survival at 1 year, only 6% expected patients to be free of major functional limitations and only 4% believed patients would have a good quality of life.9 The degree of concordance among physicians’ and surrogates’ expectations as well as the accuracy of their outcome predictions were low overall. These single-center findings are also supported by similar data at other centers, and are concerning as they imply that surrogate decisions regarding life-prolonging treatments are oftentimes not appropriately informed.10

Strategies for Improving Communication in CCI

Several strategies for improving communication during acute critical illness have been studied. An “intensive communication intervention” at a single center consisting of multidisciplinary meetings with patients and families within 72 hours of ICU admission resulted in a significantly decreased median length of ICU stay, mainly in the group of patients with the highest acuity, by allowing them earlier access to palliative care without an overall increase in ICU mortality.11 This proactive communication intervention covered four main objectives during the formal meeting: 1) a review of medical facts and treatment options, 2) a discussion of patients’ perspectives on death, chronic dependence, loss of function, and acceptability of critical care, 3) an agreement on a care plan, and 4) an agreement on criteria by which success or failure of the plan was to be judged.11 An intervention with similar objectives but conducted by separate ethics teams at multiple centers was also associated with reductions in hospital and ICU stays, mainly in patients who did not survive to discharge, also without differences in overall mortality.12 Although many centers are now relying on specialized palliative care teams to hold such discussions, studies of communication training reveal that clinicians can be taught the necessary skills involved in giving bad news and discussing transitions to palliative care with sustained improvements.13

In addition to strategies to improve face-to-face communication, other studies have shown benefits such as decreased post-traumatic stress disorder, depression, anxiety, improved comprehension and subsequent satisfaction among family members with distribution of printed material with or without a proactive approach to family conferences.14,15 One such brochure had information on the basic operation of the ICU, including names and essential contact information, a diagram of the ICU room, names of all the devices in the room, and a glossary of commonly used ICU terms.15 In another study, a 14-page decision aid on PMV given to surrogates resulted in lower surrogate-physician discordance, greater comprehension, and improved quality of communication without differences in overall ICU mortality.16 This decision aid aimed to provide medical information, elicit surrogate understanding of patient preferences and role in decision making, and guide surrogate deliberation and decision making with specific examples.16 Although we do not know the effects of these interventions on long-term resource utilization or their widespread applicability to populations of varying ethnicities, cultural, and religious beliefs, these findings are encouraging.

Predicting Mortality in Prolonged Mechanical Ventilation

Another contributing factor to suboptimal conversations surrounding continuation of life support is physician uncertainty about the long-term outcomes of individual patients with CCI. Although the prognosis for CCI overall is known to be poor, predicting which patients will fare better after discharge from an acute care hospital is difficult. Recently, a mortality prediction model (ProVent) was developed using patient data from five academic tertiary care centers to help identify patients requiring PMV who are at high risk for 1-year mortality with good results. The ProVent score assigns points to four different variables (age, platelet count, need for vasopressors, hemodialysis) obtained on day 21 of mechanical ventilation with the range of possible scores being 0-5. Higher scores are associated with increasing mortality. This ProVent categorical probability model had an area under the receiver operator curve (AUC) of 0.77.17 For comparison, the Acute Physiology and Chronic Health Evaluation III score in common use has an AUC of 0.63.17 An additional benefit of the ProVent model is that it is based on objective data rather than subjective assessments. The investigators are quick to affirm that such a model is not meant to replace clinical judgment, but hope that in conjunction with medical judgment, it may increase clinicians’ abilities to discuss likely outcomes and enhance patient-centered care.


Advances in critical care have enabled more patients to survive their acute ICU admissions, but have also created a growing population of patients with CCI with a myriad of comorbidities. The care of these patients is fraught with ethical and economical issues, and evidence-driven care in this area has not kept pace with trends in clinical practice. However, while we hone our ability to prognosticate effectively, data on long-term outcomes in this population are available, and we do our patients and their families a disservice if this information is not communicated along with a thorough discussion of the values, preferences, and goals of the patient and surrogate with regard to continuation of life-sustaining treatment in the setting of CCI.


  1. Girard K, Raffin TA. The chronically critically ill: To save or let die? Respir Care 1985;30:339-347.
  2. MacIntyre NR, et al. Management of patients requiring prolonged mechanical ventilation: Report of a NAMDRC consensus conference. Chest 2005;128:3937-3954.
  3. Nelson JE, et al. Chronic critical illness. Am J Respir Crit Care Med 2010;182:446-454.
  4. Cox CE, Carson SS. Medical and economic implications of prolonged mechanical ventilation and expedited post-acute care. Semin Respir Crit Care Med 2012;33:357-361.
  5. Kahn JM, et al. Long-term acute care hospital utilization after critical illness. JAMA 2010;303:2253-2259.
  6. Kahn JM, et al. Effectiveness of long-term acute care hospitalization in elderly patients with chronic critical illness. Med Care 2013;51:4-10.
  7. Lane-Fall MB, et al. Insurance and racial differences in long-term acute care utilization after critical illness. Crit Care Med 2012;40: 1143-1149.
  8. Unroe M, et al. One-year trajectories of care and resource utilization for recipients of prolonged mechanical ventilation. Ann Intern Med 2010;153:167-175.
  9. Cox CE, et al. Expectations and outcomes of prolonged mechanical ventilation. Crit Care Med 2009;37:2888-2894.
  10. Nelson JE, et al. Communication about chronic critical illness. Arch Intern Med 2007;167:2509-2515.
  11. Lilly CM, et al. An intensive communication intervention for the critically ill. Am J Med 2000;109:469-475.
  12. Schneiderman LJ, et al. Effect of ethics consultations on nonbeneficial life-sustaining treatments in the intensive care setting: A randomized controlled trial. JAMA 2003;290:1166-1172.
  13. Back AL, et al. Efficacy of communication skills training for giving bad news and discussing transitions to palliative care. Arch Intern Med 2007;167:453-460.
  14. Lautrette A, et al. A communication strategy and brochure for relatives of patients dying in the ICU. N Engl J Med 2007;356: 469-478.
  15. Azoulay E, et al. Impact of a family information leaflet on effectiveness of information provided to family members of intensive care unit patients: A multicenter, prospective, randomized, controlled trial. Am J Resp Crit Care Med 2002;165:438-442.
  16. Cox CE, et al. Development and pilot testing of a decision aid for surrogates of patients with prolonged mechanical ventilation. Crit Care Med 2012;40:2327-2334.
  17. Carson SS, et al. A multicenter mortality prediction model for patients receiving prolonged mechanical ventilation. Crit Care Med 2012;40:1171-1176.