Serious Illness Care

Healthcare, “A systematic intervention to improve serious illness communication in primary care: Effect on expenses at the end of life,” June 2020

Partners: Ariadne Labs, Brigham and Women’s Hospital, Harvard T. H. Chan School of Public Health, Harvard Medical School, Partners Healthcare, Massachusetts General Hospital, Dana Farber Cancer Institute

Study Summary: Previous research has shown that serious illness conversations between clinicians and patients can improve care and patient well-being. New research by Ariadne’s Serious Illness Care Program finds that these conversations can also be associated with lower health care expenses. Researchers analyzed the impact of a quality improvement intervention for better communication on total medical expenses in a high-risk care management program. Included were patients who died between January 2014 and September 2016 and who were selected for serious illness conversations.

What Did We Learn?  The evaluation of 124 patients in 14 primary care clinics showed no differences in total medical expenses between intervention and comparison clinic patients. When comparing patients in intervention clinics who did and did not have conversations, researchers observed lower average monthly expenses over the last six and three months of life for patients who had conversations.

Conclusions: Programs that are designed to drive more, earlier, and better serious illness communication hold the potential to reduce costs.

Journal of Palliative Medicine, “Training Clinicians in Serious Illness Communication Using a Structured Guide: Evaluation of a Training Program in Three Health Systems,” February 2020.

Study Summary: Failure to initiate discussions about patients’ values and goals in serious illness remains a common problem. From 2016 to 2018, faculty at Ariadne Labs conducted train-the-trainer courses at three institutions to equip trainers to teach serious illness communication to clinicians. Ariadne Labs trained 22 trainers (19 of whom were palliative care specialists) in three systems. They, in turn, trained 297 clinicians, including physicians, nurses, social workers, and chaplains primarily in the fields of palliative care and primary care. Ariadne researchers then analyzed responses to a post-training evaluation questionnaire completed by clinicians who attended the training.

What Did We Learn? 

• For all clinician learners, the self-rating of their skill increased from an average of 3.1 (skilled) before the training to 3.9 (very skilled) after the training.
• 96% of clinicians mostly or fully agreed that the course objectives were met; 97% or more rated the training as mostly or extremely effective, and 96% or more reported that they were mostly or extremely likely to recommend the training to their colleagues. 
• Qualitative data indicate that after a single training session, clinicians reported benefits of having concrete language and a framework, positive attitudinal shifts, increased focus on patient-centered communication techniques, enhanced self-efficacy, and intention of trying the new tool in practice.

Conclusions: 

The train-the-trainer model resulted in significant self-reported improvements in clinicians’ competencies. Although some issues were noted that require further evaluation, the streamlined and adaptable nature of this training suggests that this may be a scalable model to meet the goal of serious illness communication training for large numbers of health care professionals.

Partners: Dana-Farber Cancer Institute; Harvard Medical School; Harvard T.H. Chan School of Public Health; University Medical Center, Groningen, the Netherlands

Study Results: JAMA Internal Medicine, “Effect of the Serious Illness Care Program in Outpatient Oncology: A Cluster Randomized Clinical Trial,” March 2019; JAMA Oncology, “Evaluating an Intervention to Improve Communication Between Oncology Clinicians and Patients With Life-Limiting Cancer: A Cluster Randomized Clinical Trial of the Serious Illness Care Program,” March 2019

Previous studies have shown that patients who discuss end-of-life care with their clinicians experience better quality of life, less distress, and a higher likelihood of receiving care consistent with their preferences. By contrast, absent, late, and poor-quality clinician-patient communication during serious illness contributes to avoidable suffering and overly aggressive, lower-quality care that may not align with patient goals. While palliative-care clinicians are trained to conduct end-of-life conversations, there is a need for other clinicians to have such conversations. Developed by Ariadne Labs, the Serious Illness Care Program is an intervention that has been systematically developed and tested to ensure patients with life-threatening illnesses have more, better, and earlier conversations with their clinicians about what matters most.

The program’s centerpiece is the Serious Illness Conversation Guide. Developed by palliative care experts, the guide offers clinicians questions to ask patients about their goals, values, and wishes. The guide is one element of a multi-component program that includes training and system-level support for clinicians to have these important conversations with patients. A four-year, randomized clinical trial of the Serious Illness Care Program was conducted with 278 patients with advanced cancer and 91 oncology clinicians at the Dana Farber Cancer Institute. The trial evaluated the feasibility and acceptability of the Serious Illness Care Program and its effect on patient outcomes. The primary outcomes were goal-concordant care and peacefulness at the end of life; secondary outcomes included communication documentation, therapeutic alliance anxiety, depression, and survival.

What did we find?

• As a result of the intervention, the proportion of patients with moderate to severe anxiety and depression was reduced by half, and the anxiety improvements were sustained for at least 24 weeks.
• Patients and clinicians had more, earlier, and better serious illness conversations. Such conversations occurred 2.4 months earlier in the intervention group than the control arm. Conversations centered on what mattered most to patients, with 90 percent discussing goals and values in the intervention group (compared to fewer than half in the control group).
• The intervention resulted in more accessible and more complete documentation of these conversations in the electronic medical record
• The study was unable to demonstrate whether the conversations resulted in care that aligned with patient goals, or brought about greater peacefulness at the end of life. The intervention did not impact survival rates.

Conclusions: Although the intervention did not affect the primary outcomes of goal-concordant care and peacefulness, improvements in anxiety and depression suggest that these values-based discussions can have immediate psychological benefits for patients with serious illness. The significant increase in conversations suggests that the intervention could be successfully integrated into a typical oncology practice.

Partners: Branta Foundation, Charina Endowment Fund, Margaret T. Morris Foundation, Richard A. Cantor Fund, the John A. Hartford Foundation and Partners HealthCare.

Study preliminary findings: Journal of Clinical Oncology, Abstract: Delivering more, earlier, and better goals-of-care conversations to seriously ill oncology patients, October 2015

We enrolled oncology clinicians from the Dana-Farber Cancer Institute (DFCI) to participate in a cluster-randomized controlled trial to test the multi-component Serious Illness Care Program in the adult oncology setting. The clinicians were asked to identify patients eligible for the Serious Illness Conversation using the “surprise question” – would you be surprised if this patient died within the next year? If the answer was “no,” the patient was contacted to be enrolled in the study. Results of the study are still to be analyzed for final publication. Our preliminary analysis was shared at the American Society of Clinical Oncology’s 2015 symposium.

Conclusion: Our preliminary analysis indicates a brief communication skills training program for oncology clinicians, accompanied by simple tools and workflow changes, resulted in more, better and earlier serious illness conversations and reductions in anxiety and depression for patients with advanced cancer.

Partners: Dana-Farber Cancer Institute, Duke University School of Medicine, Medical University of South Carolina, Brigham and Women’s Hospital, Medical University of South Carolina, Cambia Health Foundation, Richard A. Cantor Fund for Communications Research in Palliative Care and American Cancer Society.

Study results: Palliative and Supportive Care, From Barriers to Assets: Rethinking factors impacting advance care planning for African Americans, April 2018.

Advanced care planning (ACP) can provide clarity on critical end-of-life care issues by clarifying care preferences beforehand. African Americans do not participate in ACP as much as white Americans, which may impact their quality of life and cause challenges when they suffer serious illness. As part of a broader study to improve the quality of clinician-led ACP, researchers performed a qualitative study with a small group of health disparities experts, community members and seriously ill African American patients and caregivers to understand factors that limit or enable ACP.

This study is the first to examine and compare these barriers and facilitators from multiple perspectives.

What did we learn?

• Seven factors emerged to deter or enable ACP among African Americans: religion and spirituality, trust and mistrust, family relationships and experiences, patient-clinician relationships, prognostic communication, care preferences and preparation and control.
• However, patients viewed some factors differently than clinicians. The experts interviewed for the study framed religion and spirituality as potential barriers to care, but patients and caregivers stated that faith did not conflict with their desire to participate in ACP. Experts stated that African Americans were not interested in prognostic information because of mistrust. But patients strongly desired direct, honest communication and seriously ill patients and caregivers expected this information from their physicians, in whom they described a high degree of trust. Experts suggested that multiple family decision-makers presented a barrier to care, but patient-caregiver participants emphasized the value of independent ACP to minimize burden on and conflict among family and caregivers, as well as to ensure that their preferences informed their care.
• Overall, some factors traditionally viewed as barriers to ACP by experts were seen as assets by patients. The expert participants interviewed for this study expressed ideas reflected in previous literature on the barriers that prevent African Americans from pursuing ACP. But their beliefs about these barriers were not always in agreement with those of community members, patients and caregivers.

Conclusions:

Previous research has shown a culture in which African Americans do not seek out ACP and has offered potentials reasons why this might be. However, this paper shows that existing research might not accurately represent the diversity of preferences in the African American community or the experience of African Americans who are seriously ill. To more effectively encourage ACP and engage with this community, clinicians and caregivers may need to reframe factors like religion/spirituality and family as assets. Most importantly, African American patients said they desired respectful communication and a rapport with the medical professionals they work with, which is an important takeaway for increasing ACP. Future research aims to explore this area to identify most impactful methods of communication.

Partners: Partners HealthCare, Charina Foundation, Brigham and Women’s Physician Group and Branta Foundation

Study Results: Health Affairs, 2017; Journal of the American Medical Association Internal Medicine, 2016; second paper in Journal of the American Medical Association Internal Medicine, 2016.

Given the workforce shortage of palliative care clinicians, there is a need for clinicians in other specialities to have the skills, confidence, and systems to allow them to have conversations about values and preferences with their seriously ill patients. The Serious Illness Integrated Care Management Program (iCMP) study demonstrated that the Serious Illness Care Program can be successfully implemented in the primary care setting.

For this pragmatic implementation trial, we adapted the Serious Illness Care Program to be implemented in primary care clinics affiliated with the Brigham and Women’s Hospital in Boston, focusing on patients enrolled in the Integrated Care Management Program (iCMP). This program matches chronically-ill, medically-complex patients with a nurse care coordinator who works with the patients and their primary care physicians and social workers to develop a customized health care plan. The intervention arm consisted of six clinics trained in the Serious Illness Care Program, while eight clinics served as the comparison sites.

What did we find?

• More and better conversations: For patients who died during the study period, more patients in the intervention group (62 percent) had at least one serious illness conversation compared to the control group (43 percent). Conversations for patients in the intervention group were more comprehensive, covering more elements of patient goals and values (an average of 2.8 elements) compared to the control group (1.8 elements).
• Conversations happened earlier than in previous studies. This study was the first to look at timing of conversations in primary care. While there wasn’t a difference in timing between the intervention and control groups, we did see that the conversations in primary care took place earlier than demonstrated in prior oncology studies.
• Documentation was easier to access: For those patients whose conversation was documented, patients in the intervention (44 percent) were more likely to have their conversations recorded in the advance care planning module of their electronic medical record compared to the control group (3 percent).
• Training feasibility: The training portion of the program received high satisfaction ratings from the clinicians (4.7/5)
• Timing fits within workflow: We found that the time required for clinicians to have a serious illness conversation – 22 minutes – could fit within their workflow.
• Teamwork counts: We also learned that the majority of the conversations that occurred involved more than one clinician. A team approach was seen to be an effective way to engage in the conversation and contributed to increased access for this population.

Conclusion: Overall, this primary palliative care intervention was feasible and endorsed by clinicians, and it improved discussions about patients’ goals and values. By providing a comprehensive approach, including patient identification, teamwork and coaching, we have seen how beneficial this can be for both patients and providers. This intervention contributes to a new model for improving care for seriously ill patients in the primary care setting.

Partners: Brigham and Women’s Faulkner Hospital / DCI Dialysis Unit

Background: Clinical Journal of the American Society of Nephrology, 2016

In surveys, 90 percent of patients with chronic kidney disease indicate that they wish to have serious illness conversations, yet only 10 percent report having this discussion with their nephrologist. Furthermore, although the five-year survival rate for patients on dialysis is only 20 percent, communication around patients’ values, goals and care preferences is not standardized in nephrology.

In collaboration with Brigham and Women’s Faulkner Hospital, Ariadne Labs led a trial to learn how to integrate serious illness care conversations into dialysis practice. We have completed a background paper outlining the opportunities and challenges in conducting discussions with dialysis patients. Results from the study are pending analysis and publication.

Partners: Rx Foundation, Spaulding Hospital for Continuing Care, Brigham and Women’s Hospital

Study Results: Journal of Palliative Medicine, 2017; Critical Care Medicine, 2017.

Patients with chronic critical illness have survived an acute critical illness but have not recovered, often requiring prolonged mechanical ventilation, suffering repeated infections, weakness and delirium. They are often transferred from the acute care hospital to another type of hospital, called a long-term acute care hospital, where they stay for weeks to months, in the hopes of returning home at or near their pre-illness functioning state. However, approximately 50 percent of this population dies within one year, and little is known about this growing population’s experience in the long-term care setting. The study team sought to understand the expectations and goals of these patients and their surrogate decision makers through the use of a modified Serious Illness Conversation Guide.

In 2015, a modified Conversation Guide was used to conduct 50 interviews, 30 with patients and 20 with surrogates, at Spaulding Hospital for Continuing Care in Cambridge, MA.

What did we learn?

Five themes emerged from interviews with chronically critically ill patients and their surrogates:

• Poor patient quality of life. The majority of respondents described the patient’s quality of life as poor, noting hunger, thirst, difficulty communicating, boredom, and limited mobility.
• Surrogate stress and anxiety. Surrogates reported concerns about what the patient’s life would look like if they lived, whether the patient would resent the decisions the surrogate made, and finances.
• Optimistic health expectations. The majority of respondents reported optimism for the patient’s future health, characterized by a return home at or near the patient’s previous  level of functioning.
• Poor planning for medical setbacks. In the majority of conversations with respondents, they reported not having a discussion with the clinicians at the long-term acute care institution about care if the patient were to become sicker.
• Disruptive care transitions. Nearly half of surrogates expressed concern over the transition between acute care and long-term acute care institutions, notably the disruption in provider continuity and how difficult it was to adapt to the differences between the two types of facilities. Surrogates also noted receiving conflicting messages from the clinicians, suggesting a lack of coordination.

Conclusion: Patients reported an overly optimistic expectation for returning home and unmet palliative care needs, suggesting the need for integration of palliative care within the long-term acute care hospital. Both patients and surrogates found this type of conversation acceptable, and the primary hospitalists reported the information to be useful and that it would affect clinical care decisions. This suggests there is an opening to use the Conversation Guide to improve the conversations clinicians have with their chronically critically ill patients and surrogates, and to continue the discussions even when initial decisions may have already been made.

Partners: Brigham and Women’s Hospital

Study Results: Journal of Palliative Medicine, 2016; Annals of Surgery, 2016; Journal of Palliative Medicine, 2014.

In a high-stakes emergency room environment, surgeons have limited time to engage in important conversations with seriously ill patients about their values and goals. In collaboration with Brigham and Women’s Hospital, Ariadne Labs led a study to examine the intersection of acute surgery and end-of-life care. The goal was to develop a communication tool that surgeons can use with patients who have advanced disease and require surgery for sudden, life-threatening problems in order to ensure that patients can be actively involved in decision-making about their surgery.

In the second phase, Ariadne Labs led a symposium of national experts in palliative care and emergency medicine to identify the key questions a surgeon should explore with a critically ill patient prior to surgery.

What did we learn?

Interviews with emergency surgeons showed that they felt it was their responsibility to have conversations with seriously ill patients and their families to avoid performing operations that would ultimately not benefit the patient. However, the surgeons noted several barriers to having these conversations, often leading to non-beneficial surgery: the difficulty in estimating prognosis, inadequate data about the quality of life a patient could expect after a procedure, patients and surrogates being unprepared to discuss end-of-life issues, differing perceptions of the role of palliative care, and time constraints.

Participants in the expert symposium identified nine key components of a framework for structured conversations with these patients:

• formulating prognosis,
• creating a personal connection,
• disclosing information regarding the acute problem in the context of the underlying illness,
• establishing a shared understanding of the patient’s condition,
• allowing silence and dealing with emotion,
• describing surgical and palliative treatment options,
• eliciting patient’s goals and priorities,
• making a treatment recommendation, and
• affirming ongoing support for the patient and family.

Conclusion: The group of experts identified the need for educational opportunities for surgeons to strengthen clinical communication skills for the serious illness setting, and the need to study the impact of such initiatives in improving patient outcomes.

Partners: South Carolina Hospital Association, Medical University of South Carolina in Charleston, SC

Within urban African-American communities, numerous barriers to advance care planning exist, some cultural, some structural, which contribute to marked disparities in end-of-life care. While some community health facilities offer advance care planning discussions, there are few studies about preferences regarding content and timing of advance care planning discussions in African American communities.

In collaboration with the South Carolina Hospital Association and researchers from the Medical University of South Carolina, Ariadne Labs led a two-phase, mixed-methods study to understand the acceptability of the Serious Illness Conversation Guide in African-American communities and to modify and test the Conversation Guide to improve generalizability for future versions of the Conversation Guide.

What did we learn?

The study found that a conversation guide can help overcome barriers to advance care planning for a community that has historically been negatively impacted by health care experiences. The guide encourages direct communication, and the questions at the heart of the conversation are about hearing from the patient about their goals, values, and priorities. Focus group participants found the language used in the Guide easy to understand. As a result of the study, an additional question was added to the Guide – “What gives you strength as you think about the future with your illness?”

Future work will seek to extend these findings and to gain further insight into rural and other underserved populations.