Background
Rationale
Methods
Protocol and registration
Eligibility criteria
Eligibility criterion | Rationale |
---|---|
Randomized, controlled trial or prospective observational study, published in peer-reviewed journal | Randomized controlled trials and prospective observational study experimental designs are least likely to lead to biased results |
Evaluates a structured communication tool (decision aid, structured meeting, educational strategy) compared to control group | • Interest in comparing wide variety of interventions, in multiple formats (verbal, paper, video, computer, etc.) • A control group is required to assess whether the intervention is better than usual care as routinely practiced (recognizing that usual care may vary based on setting) |
Communication tool must address end-of-life decision-making | Review interest is in interventions that assist patients with decision-making, as opposed to those that address breaking bad news, patient comfort alone |
Adult patients (age >18 years) | End-of-life decision-making process in frail adults is likely to be qualitatively different from that in children |
English language | Communication tools published in other languages, with no English translation available, may not be generalizable to english-language settings |
Information sources and search strategy
Study selection
Data collection process and data items
Risk of bias in individual studies
Synthesis of results
Investigation of heterogeneity and subgroup analyses
Publication bias
Rating of quality of evidence
Results
Study selection
Study characteristics
Study settings and populations
Study ID | Sample size | Population description | Country | Design | Target of intervention | Intervention | Comparator | Measured outcome |
---|---|---|---|---|---|---|---|---|
Andereck 2014 [14] | Intervention 174, control 210 | Patients admitted to the medical or surgical ICU of a tertiary care hospital for at least 5 consecutive days | USA | RCT | Family/SDM | The intervention group received a proactive ethics consultation. The ethics consultation assessed patient capacity and preferences, and assisted SDMs in medical decision-making, including DNR. The ethicist continued to follow the patient until discharge | Usual care | • Health care resource utilization • Satisfaction with end-of-life care |
Lautrette 2007 [21] | Intervention 63, control 63 | Adult patients admitted to medical or surgical ICUs judged to be likely to die within a few days, with an identified SDM | France | RCT | Family/SDM | An intensive end-of-life communication intervention aimed at eliciting the patient’s values, acknowledging the family member’s voice and emotions, and to allow questions. Following the meeting, families were given a brochure on bereavement | Usual care | • Quality of communication • Preference on life-sustaining treatment options • Advance directive discussions • Health care resource utilization |
Schneiderman 2000 [23] | Intervention 35, control 35 | ICU patients in whom value-based treatment conflicts arose (e.g., disagreements over CPR status, withdrawal of life support, etc.) | USA | RCT | Family/SDM | Offering of an ethics consultation from the hospital ethics service | Usual care | • Health care resource utilization • Preference on life-sustaining treatment options • Acceptability of intervention |
Schneiderman 2003 [24] | Intervention 278, control 273 | Critically ill adult patients admitted to medical or surgical ICUs | USA | RCT | Family/SDM | The intervention group received a proactive ethics consultation, which addressed current ethical issues, reviewed patient wishes and values, and provided recommendations for next steps regarding communication and decision-making | Usual | • Health care resource utilization • Acceptability of intervention |
Ahrens 2003 [32] | Intervention 43, control 108 | Patients admitted to an academic tertiary care medical ICU | USA | Cohort | Family/SDM | Families/SDMs were provided with an intensive communication strategy, including daily medical updates by the attending physician, provision of treatment options, including non-curative/palliative options, and support by a clinical nurse specialist | Usual care | • Health care resource utilization |
Campbell 2003 [29] | Intervention 20, control 18 | Patients admitted to the medical ICU with either global cerebral ischemia or multisystem organ failure, with a retrospective control cohort and prospective interventional cohort | USA | Cohort | Family/SDM | Early involvement of palliative care service in communicating prognosis to the family, identifying advance directives and preference, and assisting with discussion and implementation of treatment options and palliative care | Usual care | • Preference on life-sustaining treatment options • Health care resource utilization |
Cox 2012 [15] | Intervention 10, control 17 | SDMs for adult medical and surgical ICU patients on mechanical ventilation for equal to or greater than 10 days, expected to survive for greater than 72 hours without pre-existing tracheostomy | USA | Cohort | Family/SDM | The prolonged mechanical ventilation decision aid reviewed medical information, elicited the SM understanding of the patient’s preferences, clarified the role of the SDM, and provided guidance in decision-making | Usual care | • Quality of communication • Comfort and confidence (decision conflict) • Health care knowledge and literacy • Health care resource utilization • Preference on life-sustaining treatment options |
Daly 2010 [16] | Intervention 354, control 135 | Incapable patients with 72 hours of mechanical ventilation, with an identified SDM, admitted to surgical, medical, or neuroscience ICUs at two university-affiliated medical centers | USA, Canada | Cohort | Family/SDM | An intensive communication system, including a family meeting with a medical update, identification of goals of care, a treatment plan, and milestones for determining if the treatment was effective, conducted within 5 days of ICU admission and weekly thereafter. | Usual care | • Preference on life-sustaining treatment options • Health care resource utilization • Quality of communication |
Dowdy 1998 [17] | Intervention 31, control 31 | Sequential patients treated with mechanical ventilation for more than 96 hours, between June 1992 and October 1994 | USA | Cohort | Family/SDM | Proactive ethics consultation, and daily as required, addressing advance directives, patient capacity, SDM knowledge of patient advance directive, anticipated conflicts, and limits of treatment | Usual care | • Preference on life-sustaining treatment options • Health care resource utilization • Quality of communication |
Hatler 2012 [18] | Intervention 98, control 105 | Patients admitted to a territory neurosurgical ICU who received mechanical ventilation for >96 hours, remained in ICU for 7 days or longer, and were not awaiting transfer out of ICU during that time | USA | Cohort | Family/SDM and HCPs | A surrogacy information and decision-making tool was filled out by the admitting nurse, documenting patient’s decision-making capacity, the identity of the SDM/POA, and prior advance directive. The nurse gave the patient or SDM an information sheet about surrogate decision-making and advance directives. | Usual care | • Health care resource utilization |
Holloran 1995 [28] | Intervention 6, control 24 | Patients admitted to a large, tertiary care ICU for any reason. | USA | Cohort | HCPs | “Decisions near the End of Life” program, a small-group workshop using cases to facilitate discussion of issues such as withholding or withdrawing treatment, eliciting patient and family wishes, patient competency, and conflict with families | Pre-intervention hospital cohort | • Health care resource utilization • Preference on life-sustaining treatment options |
Knaus 1990 [25] | Intervention 705, control 760 | All adult patients admitted to ICU, excluding those with uncomplicated myocardial infarction or those admitted with acute burns | France | Cohort | HCPs | HCPs were provided with a calculated estimate of hospital mortality daily on rounds until the patient died, or until 7 days, whichever came first | Usual care | • Preference on life-sustaining treatment options |
Lamba 2012 [27] | Intervention 104, control 79 | Patients admitted to a surgical ICU between March 2003 and May 2005 for liver transplantation | USA | Cohort | Family/SDM | Each patient had a palliative care assessment delineating prognosis, advance directives, family support, surrogate decision maker, and pain, within 24 hours of admission. The patient’s family received psychosocial and/or bereavement support. An interdisciplinary family meeting was held at 72 hours to address patient outcomes, treatment options, and goals of care, and family support was provided by a multidisciplinary team. | Usual care | • Quality of communication • Preference on life-sustaining treatment options • Advance directive discussions • Health care resource utilization |
Lilly 2000 [26] | Intervention 396, control 134 | Consecutive admitted to the ICU of a tertiary care teaching hospital | USA | Cohort | Family/SDM | An intensive communication strategy, including a meeting with the attending physician within 72 hours for patients expected to stay >4 days, with predicted mortality >25 %, or change in functional status, unlikely to return to home | Usual care | • Advance directive discussions • Quality of communication • Health care resource utilization |
McCannon 2012 [30] | Intervention 27, control 23 | Patients admitted to the medical ICU age >50 years, currently incapable, likely to survive >24 hours, with an identified adult SDM. | USA | Cohort | Family/SDM | A 3-minute video decision support-tool was shown which reviewed CPR methods and outcomes, and the care of a sedated, mechanically ventilated patient, within 72 hours of ICU admission | Usual care | • Health care knowledge and literacy • Preference on life-sustaining treatment options • Acceptability of intervention |
Norton 2007 [19] | Intervention 126, control 65 | Adult patients admitted to a medical ICU with a hospital stay of 10 days, age >80 years, or two or more life-threatening comorbidities | USA | Cohort | Family/SDM | The intervention group had a proactive palliative care consultation, which facilitated decision-making and family member support, and followed the patient until discharge | Usual | • Health care resource utilization |
Quenot 2012 [31] | Intervention 823, control 678 | All patients who died in the ICU, or in hospital after discharge to another department, during two periods, one before and one after a 2005 French law on end-of-life and patient rights. | France | Cohort | Family/SDM | An intensive communication strategy, including daily meetings with the attending team, modalities for withdrawing and withholding treatment, a special ‘ethics’ section in the chart, and debriefing sessions | Pre-intervention hospital cohort | • Preference on life-sustaining treatment options • Health care resource utilization • Quality of communication |
Shelton 2010 [20] | Intervention 114, control 113 | Patients admitted to the surgical ICU, anticipated by the attending physician to remain for at least 7 days, or were expected to die within that time, during two periods | USA | Cohort | Family/SDM | During the intervention period, a family support coordinator assessed the family’s information needs, interpreted and explained relevant medical information, assisted the family in decision-making, and identified the need for referrals to spiritual care and to enhance the health care team’s understanding of the family’s needs. | Usual care | • Satisfaction with end-of-life care • Quality of communication with HCPs • Health care resource utilization |
Curtis 2011 [22] | Intervention 514, control 565 | Medical and surgical ICUs with sufficient ICU deaths to meet study sample size requirements (6 intervention hospitals, 6 control hospitals) Patients included those who died in ICU or within 30 hours of transfer to another hospital location. | USA | Cluster RCT | HCPs | A multifaceted intervention including education about palliative care, identification and training of ICU clinician local champions for palliative care, nurse and physician ICU directors to address barriers to improving end-of-life care, feedback of quality data including family satisfaction, and implementation of system supports such as palliative care order forms. | Usual care | • Satisfaction with end-of-life care • Preference on life-sustaining treatment options • Quality of communication • Health care resource utilization |
Interventions
Characteristics of excluded studies
Risk of bias within studies
Study ID | Newcastle-Ottawa scale for cohort studies - selection | Newcastle-Ottawa scale for cohort studies - comparability | Newcastle-Ottawa scale for cohort studies - outcome | Overall Newcastle-Ottawa scale risk of bias |
---|---|---|---|---|
Ahrens 2003 [32] | ★★★★ | ☆☆ | ★★★ | Poor |
Campbell 2003 [29] | ★★★★ | ☆☆ | ★★★ | Poor |
Cox 2012 [15] | ★★★★ | ☆☆ | ★★☆ | Poor |
Daly 2010 [16] | ★★★★ | ★★ | ★★★ | Good |
Dowdy 1998 [17] | ★★★★ | ★☆ | ★★★ | Good |
Hatler 2012 [18] | ★★★★ | ☆☆ | ★★★ | Poor |
Holloran 1995 [28] | ★★★☆ | ☆☆ | ★★★ | Poor |
Knaus 1990 [25] | ★★★☆ | ★☆ | ★★★ | Good |
Lamba 2012 [27] | ★★★★ | ☆☆ | ★★★ | Poor |
Lilly 2000 [26] | ★★★★ | ★☆ | ★★★ | Good |
McCannon 2012 [30] | ★★★★ | ☆☆ | ★★★ | Poor |
Norton 2007 [19] | ★★★☆ | ☆☆ | ★★★ | Poor |
Quenot 2012 [31] | ★★★★ | ☆☆ | ★★★ | Poor |
Shelton 2010 [20] | ★★★★ | ☆☆ | ★★★ | Poor |
Synthesis of results
Primary outcomes
Secondary outcomes
Discussion
Strengths
Limitations
Conclusions
Key messages
-
Many studies of structured communication tools to improve end-of-life decision-making in adult ICU patients have been published
-
The evidence that such interventions increase the documentation of goals-of-care discussions and code status, or withdrawal and withholding of life-sustaining treatments is low to very low in quality
-
Future studies of simple interventions in targeted ICU populations are needed, and should report upon patient-level, family-level, and system-level outcomes