• 
• 

Three Part Question

In adult patients who present [with critical illness in the emergency department] what are the [factors that influence] the institution of [ceilings of treatment].

Clinical Scenario

An 84 year old man presents to your Emergency Department with septic shock. He has a long list of medications, but you learn from his medical notes that he is normally independent at home. You need to decide what level of intervention is in this patient’s best interests. What factors should you consider in order to institute an appropriate ceiling of treatment for this patient?

Search Strategy

MEDLINE – 1948 to Present with Daily Update
[Terminal Care/ OR End of Life.mp. OR ceilings adj2 care.mp. OR Palliative Care/ OR Treatment adj Limiting.mp. OR Withdrawing treatment.mp. OR Withholding Treatment/ OR Resuscitation/ OR Cardiopulmonary Resuscitation/ OR Resuscitation orders/ OR Advance Directives/ OR Advance Care Planning/] AND [Decision Making/ OR Attitude/ OR Attitude of Health Personnel/ OR Establishing.mp. OR (Practice Guideline as Topic/] AND [Emergency Medicine/ OR Accident and Emergency.mp. OR Emergency Physician.mp. OR Emergency Service, Hospital]

EMBASE – 1974 to 2016 November 29
[End adj2 Life.mp OR Ceilings adj2 Care.mp. OR Palliative Therapy/ OR Treatment adj Limiting.mp. OR Withdrawing adj Treatment.mp. OR Withholding adj Treatment.mp. OR Resuscitation/ OR Advance Directive.mp. OR Living Will/ OR Advance Care Planning.mp.] AND [Decision Making/ OR Health Personnel Attitude/ OR Attitude.mp. OR Establishing.mp. OR Practice Guideline/] AND [Emergency Medicine/ OR Accident and Emergency.mp. OR Emergency Physician/ OR Emergency Health Service/]

[Terminal Care/ OR End of Life.mp. OR ceilings adj2 care.mp. OR Palliative Care/ OR Treatment adj Limiting.mp. OR Withdrawing treatment.mp. OR Withholding Treatment/ OR Resuscitation/ OR Cardiopulmonary Resuscitation/ OR Resuscitation orders/ OR Advance Directives/ OR Advance Care Planning/] AND [Decision Making/ OR Attitude/ OR Attitude of Health Personnel/ OR Establishing.mp. OR (Practice Guideline as Topic/] AND [Emergency Medicine/ OR Accident and Emergency.mp. OR Emergency Physician.mp. OR Emergency Service, Hospital]

[End adj2 Life.mp OR Ceilings adj2 Care.mp. OR Palliative Therapy/ OR Treatment adj Limiting.mp. OR Withdrawing adj Treatment.mp. OR Withholding adj Treatment.mp. OR Resuscitation/ OR Advance Directive.mp. OR Living Will/ OR Advance Care Planning.mp.] AND [Decision Making/ OR Health Personnel Attitude/ OR Attitude.mp. OR Establishing.mp. OR Practice Guideline/] AND [Emergency Medicine/ OR Accident and Emergency.mp. OR Emergency Physician/ OR Emergency Health Service/]

Search Outcome

Using MEDLINE the search strategy identified 332 articles, of which 7 were found to be of sufficient quality and relevance to the topic to be included.

Using EMBASE the search strategy identified 1734 articles, of which 1 was found to be of sufficient quality and relevance to the topic to be included.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Wrenn, K. Brody, S. L. 1992 United States	37 consecutive adult patients for whom a DNR order was issued during the author’s shifts in the ED of a hospital in Atlanta over a 10 month period (1987-1988).	Prospective single-centre case series with collection and analysis of patient data.	Analysis of characteristics for patients assigned a DNR order including: patient age, acute and chronic problems, prior DNR orders and capacity.	• Elderly (overall mean age 70) o Subset (32% of patients) were younger (mean age 54) and critically ill with malignancy or AIDS. • Chronic disease & comorbidity (uniformly present) o Dementia (46%), decubitus ulcers (41%), malignancy (24%), stroke (19%), chronic renal failure (11%), AIDS (8%), chronic heart disease (8%), chronic neurologic disease (5%), malnutrition (5%), alcoholism (3%). • Acute disease o Sepsis (59%), coma (49%), hypotension (49%), respiratory failure (16%), dehydration (11%), hypothermia (8%), congestive heart failure, stroke, status epilepticus, acute MI, severe anaemia, hyponatraemia (3%) each. • 14% of patients had capacity • 14% of patients had prior DNR orders	Small sample size. Only represents those patients for whom the authors initiated a DNR order, introducing a large potential for bias. This study only looks at DNR orders, a subset of treatment limiting decisions. Does not provide comparison group.
Le Conte, Philippe Baron, Denis Trewick, David Touzé, Marie Dominique Longo, Céline Vial, Irshaad Ya 2004 France	All non-trauma patients (n= 119) for whom a decision to withhold (WH) or withdraw (WD) life-sustaining treatment was taken by senior staff in a French ED between January and September 1998.	Prospective single-centre survey. When making a WH/WD decision, physicians were required to justify it by choosing from 17 pre-defined criteria. Patient characteristics were also recorded.	Predefined criteria used to justify treatment limiting decision	• Mean of 6±2 criteria chosen per case. ‘Expected quality of life unacceptably poor’ was never chosen alone. o Principal acute medical disorder (83%) o Irreversibility of acute disorder in the first 24h (60%) o Level of care considered maximal (59%) o Severity of illness using scoring systems (40%) o Vegetative state (40%) o Post-morbid expected quality of life unacceptably poor (39%) o Underlying disease expected to be fatal within 6 months (37%) o Age (24%) o Underlying chronic debilitating disease (22%) o Choice of patient (8%)	Single centre study, exclusion of trauma patients. Pre-defined list of criteria used to justify treatment limiting decisions limits the range of identifiable factors.
			Analysis of characteristics for patients in whom a WH/WD decision was made.	• Elderly (mean age 75 years) • Chronic disease (77%) o Cause of acute presentation in 56% of patients • Acute disease o Neurological (38%), cardiovascular (24%), respiratory (17%), digestive (14%), cancer (14%) • Severity of illness o Prognosis of underlying disease  Fatal within 5 years (35%), fatal within 1 year (40%) o Organ system failure score of >3 (14%) o Mean SAPS score of 14 (indicating expected 30% mortality rate) • Functional limitation o None to moderate (42%), severe (53%) • 73% of patients judged unable to enter the decision making process
Sedillot, N. Holzapfel, L. Jacquet-Francillon, T. Tafaro, N. Eskandanian, A. Eyraud, S. Metton, P. P 2008 France	All adult patients admitted to a French general hospital ED in a one year period between 2004-2005 who did not already have a treatment limiting decision in place (98 patients – 1.5% of admission).	Prospective observational study. Pattern of treatment limitation chosen from a five step protocol ranging from no limitation to active withdrawal of life support.	Characteristics of patients in whom a WD/WH decision was made	• Elderly (mean age 82 years). • Chronic disease (95%) o Dementia (39%), cardiac insufficiency (34%), cancer (32)%, neurological disease (21%), COPD (6%), chronic renal failure (5%), psychiatric disorders (4%). • Acute organ failure was observed at admission in 82%. Physicians preferred choose a pattern of treatment limitation (83%) rather than treatment withdrawal as they were not seen as ethically equivalent.	Single centre study. Does not attempt to identify reasons for WD/WH life support.
Le Conte, Philippe Riochet, David Batard, Eric Volteau, Christelle Giraudeau, Bruno Arnaudet, Idriss 2010 France	All patients who died in 174 EDs in France and Belgium over two 2-month periods in 2004/2005 were enrolled (n= 2512), 1907 of whom had a treatment WD/WH decision made.	Prospective cross-sectional survey. Physicians were required to justify WD/WH decisions from 9 predefined criteria. Patient characteristics were recorded, including whether a WD/WH decision was made. A logistic regression model for treatment limiting decisions was created.	Predefined criteria used to justify treatment limiting decision	• Mean of 3±3 criteria chosen per case. Neither ‘Expected quality of life unacceptably poor’ nor ‘age’ were ever chosen alone. o Principal acute presenting medical disorder (77%) o Irreversibility of acute disorder in the first 24h (54%) o Age (39%) o Previous functional limitation (38%) o Underlying chronic disease (35%) o Absence of improvement following active treatment (26%) o Recovery but expected quality of life unacceptably poor (25%) o Underlying disease expected to be fatal within 6 months (20%) o Level of care considered to be maximal (17%).	37% of patients were transported to the ED by mobile intensive care units staffed by physicians. May limit transferability of findings to other systems. Seasonal variation of study population. Pre-defined list of criteria used to justify treatment limiting decisions limits the range of identifiable factors.
			Patient factors associated with level of care limitation	• Old age (71-81 OR 1.6, 81-88 OR 2.51, >88 OR 3.27). • Chronic disease o Immunodeficiency OR 1.9, liver disease OR 2.18, metastatic cancer OR 2.34. • Principal acute presenting disorder o Brain haemorrhage OR 2.62, neurologic OR 1.91, respiratory OR 1.61, cardiovascular OR 0.63, traumatic 0.34. • Severe functional limitation o Knauss C (OR 3.54) and Knauss D ( OR 5.84) • 92% of patients unable to enter decision making process.
			Ethical arguments for limiting life support	Futility of care (57.6%), age (35.6%), physical pain (18.9%), psychological pain (11.1%), DNR order issued by patient or transmitted by relative (6.3%).
Rodriguez-Molinero, A. Lopez-Dieguez, M. Tabuenca, A. I. de la Cruz, J. J. Banegas, J. R. 2010 Spain	101 randomly selected elderly patients (>80 or 65-79 with ≥ 2 comorbid conditions) admitted to the ED, and their respective physicians. Selected patients did not need intensive care treatment at the time of the study, which took place in 2003 over 5 months in 4 Spanish hospitals.	Cross-sectional study. Reasons underpinning WH/WD decisions were recorded via open-ended questionnaires. Patient cognitive/functional status was recorded.	Physician-reported factors taken into account during decision making	• Prior functional status (69%), • Age (42%), • Patient’s pathologic background (30%), • Current disease (18%), • Prior mental status (10%), • Subjective assessment of quality of life (4%), • Family or patient preferences (2%), • Social status (2%) • Moral considerations (1%).	Based on hypothetical events. Small sample size. Only ‘elderly’ population included. Large proportion of participating physicians were trainees.
			Patient factors associated with decisions to administer intensive treatment (CPR, ICU/CCU referral)	• Functional status as perceived by physician o (CPR administration OR 1.97, ICU/CCU referral OR 4.09/4.32) • Cognitive function as perceived by physician o (ICU referral OR 15.38) • Age o (ICU/CCU referral OR 0.86/0.76).
de Decker, L. Beauchet, O. Gouraud-Tanguy, A. Berrut, G. Annweiler, C. Le Conte, P. 2012 France	Data from LeConte et al. (2010) used. Exclusion criteria were age <65. N=2095	Prospective cross-sectional survey. Post-hoc analysis of medical notes to calculate the Charlson Comorbidity Index (CCI)’s association with treatment WH/WD decisions. A logistic regression model for treatment limiting decisions was created.	Patient factors associated with a treatment limiting decision.	• Comorbidities o CCI ≥5 (OR 25.56), • Age o ≥ 85 (OR 20.00) • Haematological disease o (OR 6.92). • Factors found to be protective from treatment limiting decisions o Living in an institution (OR 0.15), having respiratory disease (OR 0.17), neurologic causes of organ failure (OR 0.2)	All included patients died in the ED introducing the potential for bias. CCI was a post-hoc calculation.
Fassier, T. Valour, E. Colin, C. Danet, F. 2016 France	15 ED physicians and 9 ICU physicians interviewed over a one year period in 2010 in France. Data was collected across 8 units in 2 hospitals	Qualitative study. Non-participant observations identified physicians making end-of-life decisions. They were subsequently interviewed and thematic analysis was carried out.	Patient factors	• Age o Physiologic age preserved/altered o Old age thresholds (<70, 70 to 80-85, >85) • Information on patient’s EOL preference o Presence/absence of information or advance directives • Family-related factors o Presence/absence of family, conflicts	May have limited transferability of results outside France due to legal/ethical/healthcare system differences, potentially leaving influential factors unexplored in interviews.
			Physician factors	• Experience and training in EOL decision making • Physician’s positive/negative age-related stereotypes of elderly • Physician’s familial experience of ageing
			Time/resource factors	• Heterogeneous terminology and unclear acronyms in patient notes. • Unit-related factors o ED vs step-down vs short-stay units • Timing-related factors o Time available, handoffs, night/weekend shifts, leadership changes

Comment(s)

8 papers aiming to identify factors that influence ceiling of treatment decisions in the ED were included in this review. There was considerable variation in the terminology used to define the patient group including ceilings of treatment, treatment limiting and ceilings of care. The heterogeneity in study type and results perhaps reflecting this variation of both definition and methodology but also the nature of this challenging clinical issue. Observational studies identified factors that can be broadly categorised into patient-related factors (of which the most recurring were age, chronic disease and severe functional disability) and disease-related factors (of which the most recurring were severity of acute disorder, incurable disease and absence of improvement following a period of active treatment). Qualitative methods elucidated a number of factors which reflect the complexity of end of life decisions, including physician, timing and resource related factors.

Clinical Bottom Line

It is difficult to pinpoint the factors that influence the institution of ceilings of care in the ED, but they can be loosely subdivided into patient, disease, physician, unit and timing factors. How these factors are combined, their weighting and influence on the decision to institute ceilings of care is variable. The decision to institute a ceiling of care is complex and the clinician should be cognizant of these factors and their associated biases. Although challenging, the formulation of a set criteria of patient/disease related factors could act as a guide for physicians making end of life decisions in the ED.

References

1. Wrenn, K. Brody, S. L. Do-not-resuscitate orders in the emergency department Am J Med. 1992 Feb;92(2):129-33.
2. Le Conte, Philippe Baron, Denis Trewick, David Touzé, Marie Dominique Longo, Céline Vial, Irshaad Yatim, Danielle Potel, Gille Withholding and withdrawing life-support therapy in an Emergency Department: prospective survey Intensive Care Medicine 2004;2216-2221
3. Sedillot, N. Holzapfel, L. Jacquet-Francillon, T. Tafaro, N. Eskandanian, A. Eyraud, S. Metton, P. Prost, S. Serre, P. Souton, L. A five-step protocol for withholding and withdrawing of life support in an emergency department: an observational study Eur J Emerg Med 2008;145-9
4. Le Conte, Philippe Riochet, David Batard, Eric Volteau, Christelle Giraudeau, Bruno Arnaudet, Idriss Labastire, Laetitia Levraut, Jacques Thys, Frédéric Lauque, Dominique Piva, Claude Schmidt, Jeannot Death in emergency departments: a multicenter cross-sectional survey with analysis of withholding and withdrawing life support Intensive Care Medicine 2010; 765-772
5. Rodriguez-Molinero, A. Lopez-Dieguez, M. Tabuenca, A. I. de la Cruz, J. J. Banegas, J. R. Physicians' impression on the elders' functionality influences decision making for emergency care Am J Emerg Med 2010; 757-65
6. de Decker, L. Beauchet, O. Gouraud-Tanguy, A. Berrut, G. Annweiler, C. Le Conte, P. Treatment-limiting decisions, comorbidities, and mortality in the emergency departments: a cross-sectional elderly population-based study J Nutr Health Aging 2012; 914-8
7. Wong, R. E. Weiland, T. J. Jelinek, G. A. Emergency clinicians' attitudes and decisions in patient scenarios involving advance directives Emerg Med J 2012; 720-4
8. Fassier, T. Valour, E. Colin, C. Danet, F. Who Am I to Decide Whether This Person Is to Die Today? Physicians' Life-or-Death Decisions for Elderly Critically Ill Patients at the Emergency Department-ICU Interface: A Qualitative Study Ann Emerg Med 2016; 28-39