Volume 118, Issue 17 p. 4331-4338
Original Article
Free Access

Augmenting advance care planning in poor prognosis cancer with a video decision aid

A preintervention-postintervention study

Angelo E. Volandes MD, MPH

Corresponding Author

Angelo E. Volandes MD, MPH

General Medicine Unit, Department of Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts

General Medicine Unit, Massachusetts General Hospital, 50 Staniford Street, 9th floor, Boston, MA 02114; Fax: (617) 724-3544Search for more papers by this author
Tomer T. Levin MBBS

Tomer T. Levin MBBS

Department of Psychiatry, Memorial Sloan-Kettering Comprehensive Cancer Center and Department of Psychiatry, Weill Cornell Medical College, New York, New York

Search for more papers by this author
Susan Slovin MD

Susan Slovin MD

Department of Medicine, Memorial Sloan-Kettering Comprehensive Cancer Center and Deparment of Medicine, Weill Cornell Medical College, New York, New York

Search for more papers by this author
Richard D. Carvajal MD

Richard D. Carvajal MD

Department of Medicine, Memorial Sloan-Kettering Comprehensive Cancer Center and Deparment of Medicine, Weill Cornell Medical College, New York, New York

Search for more papers by this author
Eileen M. O'Reilly MD

Eileen M. O'Reilly MD

Department of Medicine, Memorial Sloan-Kettering Comprehensive Cancer Center and Deparment of Medicine, Weill Cornell Medical College, New York, New York

Search for more papers by this author
Mary Louise Keohan MD

Mary Louise Keohan MD

Department of Medicine, Memorial Sloan-Kettering Comprehensive Cancer Center and Deparment of Medicine, Weill Cornell Medical College, New York, New York

Search for more papers by this author
Maria Theodoulou MD

Maria Theodoulou MD

Department of Medicine, Memorial Sloan-Kettering Comprehensive Cancer Center and Deparment of Medicine, Weill Cornell Medical College, New York, New York

Search for more papers by this author
Maura Dickler MD

Maura Dickler MD

Department of Medicine, Memorial Sloan-Kettering Comprehensive Cancer Center and Deparment of Medicine, Weill Cornell Medical College, New York, New York

Search for more papers by this author
John F. Gerecitano MD

John F. Gerecitano MD

Department of Medicine, Memorial Sloan-Kettering Comprehensive Cancer Center and Deparment of Medicine, Weill Cornell Medical College, New York, New York

Search for more papers by this author
Michael Morris MD

Michael Morris MD

Department of Medicine, Memorial Sloan-Kettering Comprehensive Cancer Center and Deparment of Medicine, Weill Cornell Medical College, New York, New York

Search for more papers by this author
Andrew S. Epstein MD

Andrew S. Epstein MD

Department of Medicine, Memorial Sloan-Kettering Comprehensive Cancer Center and Deparment of Medicine, Weill Cornell Medical College, New York, New York

Search for more papers by this author
Anastazia Naka-Blackstone BA

Anastazia Naka-Blackstone BA

Department of Psychiatry, Memorial Sloan-Kettering Comprehensive Cancer Center and Department of Psychiatry, Weill Cornell Medical College, New York, New York

Search for more papers by this author
Elizabeth S. Walker-Corkery MPH

Elizabeth S. Walker-Corkery MPH

General Medicine Unit, Department of Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts

Search for more papers by this author
Yuchiao Chang PhD

Yuchiao Chang PhD

General Medicine Unit, Department of Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts

Search for more papers by this author
Ariela Noy MD

Ariela Noy MD

Department of Medicine, Memorial Sloan-Kettering Comprehensive Cancer Center and Deparment of Medicine, Weill Cornell Medical College, New York, New York

Search for more papers by this author
First published: 17 January 2012
Citations: 55

The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the National Cancer Institute, the Agency for Healthcare Research and Quality, or the National Institutes of Health.

Abstract

BACKGROUND:

The authors tested whether an educational video on the goals of care in advanced cancer (life-prolonging care, basic care, or comfort care) helped patients understand these goals and had an impact on their preferences for resuscitation.

METHODS:

A survey of 80 patients with advanced cancer was conducted before and after they viewed an educational video. The outcomes of interest included changes in goals of care preference and knowledge and consistency of preferences with code status.

RESULTS:

Before viewing the video, 10 patients (13%) preferred life-prolonging care, 24 patients (30%) preferred basic care, 29 patients (36%) preferred comfort care, and 17 patients (21%) were unsure. Preferences did not change after the video, when 9 patients (11%) chose life-prolonging care, 28 patients (35%) chose basic care, 29 patients (36%) chose comfort care, and, 14 patients (18%) were unsure (P = .28). Compared with baseline, after the video presentation, more patients did not want cardiopulmonary resuscitation (CPR) (71% vs 62%; P = .03) or ventilation (80% vs 67%; P = .008). Knowledge about goals of care and likelihood of resuscitation increased after the video (P < .001). Of the patients who did not want CPR or ventilation after the video augmentation, only 4 patients (5%) had a documented do-not-resuscitate order in their medical record (kappa statistic, −0.01; 95% confidence interval, −0.06 to 0.04). Acceptability of the video was high.

CONCLUSIONS:

Patients with advanced cancer did not change care preferences after viewing the video, but fewer wanted CPR or ventilation. Documented code status was inconsistent with patient preferences. Patients were more knowledgeable after the video, reported that the video was acceptable, and said they would recommend it to others. The current results indicated that this type of video may enable patients to visualize “goals of care,” enriching patient understanding of worsening health states and better informing decision making. Cancer 2012. © 2012 American Cancer Society.

INTRODUCTION

Advance care planning (ACP) involves planning with patients for future medical care for situations in which the patient may not be able to imagine future health states and medical care.1, 2 ACP usually includes a conversation regarding future health states and elucidation of the goals of care, which may lead to documentation of patient preferences in the medical record or the completion of an advance directive (eg, a living will or designating a health care proxy).1, 2 The flaws of current methods of ACP are well documented,3, 4 including concerns regarding the fundamental premise of ACP, which is that patients can accurately imagine future disease states that are often troubling.5

The current ACP model entails clinicians verbally communicating information about a condition and then asking patients what they would want in the given scenario.6 Central to this process is the ability of a patient to imagine being a part of that scenario and use their own value system to develop preferences for care. Patients are encouraged to document whether they would want mechanical ventilation or other life-prolonging interventions, but most patients have little knowledge of what these interventions entail.5 Thus, there is growing recognition that a new model of ACP is needed to more effectively elicit a patient's goals of care, particularly with regard to future health states. One potential remedy is the use of decision aids.

Over the last decade, decision support tools have been developed to help patients make more informed decisions by clarifying benefits and risks for a wide variety of medical interventions.7-10 More than 500 decision support tools have been developed, including 12 videos.10 An increasingly used health care information technology, video, can improve decision-making by providing visual information that captures complex medical and emotional scenarios.11, 12 Video can help patients understand their medical options and clarify their preferences by reinforcing the goals-of-care discussion with the use of realistic images, something that is lacking in verbal descriptions.7-10, 13-15 However, to the best of our knowledge, few video support tools have focused on end-of-life communication in advanced cancer.10

In our previous work, we demonstrated that a video of the goals of care in advanced cancer improved understanding and decision-making in a small group of patients with advanced brain cancer.16 This work was limited by studying only 1 type of cancer, and it did not compare patients' stated preferences with their documented code status. We were interested in extending this work to 1) observe the influence of the video on patients with a diversity of advanced cancers, 2) explore the effect of reinforcing a verbal description with the video on overall knowledge and cardiopulmonary resuscitation (CPR)/ventilation preferences, and 3) compare patients' stated preferences after watching the video with their documented code status.

We hypothesized that supplementing verbal discussions with visual images would change the choices for overall care of patients with advanced cancer in the event they were critically ill. In addition, we hypothesized that reinforcing discussions with visual images would improve overall knowledge regarding the decision-making process and would lead to preferences against the use of CPR and ventilation. Finally, we also were interested in comparing patients' stated preferences regarding CPR/ventilation after watching the video with their documented code status in the medical record. We completed a study before and after the intervention to study whether a goals-of-care video would shape the choices made by patients with advanced cancer about their preferences for future medical care.

MATERIALS AND METHODS

Participants

Study participants were recruited from the ambulatory oncology practices affiliated with Memorial Sloan-Kettering Comprehensive Cancer Center, a quaternary academic medical center specializing in the care of patients with cancer. All English-speaking patients with advanced cancer who were returning to visit an oncologist were eligible to participate. Patients were eligible if they had terminal, progressive cancer with a poor prognosis and limited/poor response to usual treatment algorithms and whose treatment intent was palliative. The terms terminal, progressive, poor prognosis, palliative intent, and limited/poor response were judged by the patient's oncologist (and not by the research assistant or consenting professional).

Oncologists were contacted by e-mail during each week of the study to identify patients who met the above eligibility criteria. Once a potential participant was identified, the oncologist would introduce the study to the patient after the patient's visit. Patients who agreed to participate were then referred to the study research assistant, who further explained the study and obtained informed consent. If family members were present with the patient, then they were allowed to remain during the interview and view the video, but they were asked not to make decisions for the patient.

Potential participants were screened with the Folstein et al Mini-Mental State examination (MMSE),17 which has a possible score range from 0 to 30, with scores <25 indicative of cognitive impairment. Patients were ineligible if they scored <25 on the MMSE or if they were deemed to be in a psychological state not appropriate for end-of-life discussions by the treating clinician. Interviews were conducted primarily by A.N.-B. from July 1, 2009 to June 30, 2010. The hospital's institutional review board approved the study, and all participants provided written informed consent.

Design

A structured verbal questionnaire was designed after a review of the ACP and oncology literature and in consultation with experts in oncology, palliative care, medical ethics and decision making. An overview of the study design is provided in Figure 1.

Details are in the caption following the image

This is an overview of the study design. CPR indicates cardiopulmonary resuscitation; ICU, intensive care unit.

After informed consent was obtained, participants had a baseline assessment to gather sociodemographic data, knowledge about goals-of-care choices, and individual choices for CPR and ventilation. The sociodemographic data included age, race, sex, religion, level of education, marital status, and self-reported completion of an advance directive. The knowledge measurement included 6 questions, including 5 true/false queries and 1 multiple-choice item, that were intended to measure the participants' understanding of the various choices of health care in the advanced stages of cancer (Table 1). Overall scores ranged from 0 to 6, and higher scores represented more knowledge of the choices of medical care, and this scale was used in our previous work.16 Participants were queried regarding whether or not they wished to have CPR attempted or to be placed on a breathing machine.

Table 1. The Knowledge-Assessment Questionnaire Used at Baseline and After Viewing the Videoa
Knowledge-Assessment Questionnaire
True or false
 1. Cardiopulmonary resuscitation or CPR is a medical procedure that is done on patients whose heart stops beating in an attempt to restart their heart. (True)
 2. Most cancer patients that get CPR in the hospital survive and get to leave the hospital. (False)
 3. Most cancer patients who survive CPR and are placed on a breathing machine have very few complications from these procedures. (False)
 4. Comfort care is a type of medical care that can only be provided for cancer patients living in hospice. (False)
 5. Once you tell your doctor what kind of medical care you want if your cancer becomes very advanced, you cannot change your wishes in the future. (False)
Multiple choice
 6. How many cancer patients who get CPR in the hospital survive and get to leave the hospital?
  a. Almost all (more than 90%)
  b. About half (about 50%)
  c. Very few (less than 10%)b
  • a One point is given for each correct answer, and the score ranges from 0 to 6.
  • b The correct answer is c.

Next, all participants were read a verbal description of the 3 choices of health care available in advanced cancer: life-prolonging care, basic medical care, and comfort care. The choices of medical care were created by reviewing the ACP literature and then critiqued by 10 oncologists in a systematic review process. Explanations for each of the choices for medical care were provided as follows: Life-prolonging care attempts to sustain life regardless of cost and features all potential medical interventions, including CPR, breathing machines, and medical care in the intensive care unit (ICU); basic medical care proposes to sustain physical and cognitive functions and features interventions, such as hospitalization and intravenous medicines, but does not include CPR, breathing machines, or medical care in the ICU; and comfort care intends to maximize comfort and relieve symptoms and usually involves medicines to ease symptoms but would not normally include being admitted to the hospital.

Participants were then asked to select which level of care (life-prolonging, basic, or comfort care) they would prefer if they became critically ill. Participants who were unable to choose a category were recorded as “not sure.”

Participants next viewed a 6-minute video portraying the 3 choices of health care in advanced cancer. The video was displayed on a laptop computer. The definitions of the 3 levels of health care were the same as those used in the verbal description, but they also included visual images of the choices. For example, life-prolonging care scenes included an ICU with a patient on a ventilator being tended to by nursing staff. Additional images included a simulated code with physicians depicting CPR and intubation and physicians administering various intravenous medicines. Visual scenes to portray basic medical care included a patient receiving oral medicines, scenes from an inpatient ward service, and a patient receiving oxygen. The video portrayal of comfort care used images of a patient on home hospice care receiving medicines for pain, a patient at home receiving supplemental oxygen with a nasal cannula, and a medical assistant helping a patient with self-care.

The creation of the video followed a systematic approach, starting with a review of the ACP literature in advanced cancer. The video's overall design was reviewed for appropriateness and accuracy by 10 physicians specializing in cancer, by 3 ICU physicians, by 3 experts in palliative medicine, by 3 experts in bioethics and by 2 decision-making experts using an iterative process. The film was created without using stage directions or special effects to impart a realistic film style known as cinema verite.18 The overall designing and creation of the film was done by the principal investigator (A.E.V.) using standardized filming criteria.19

After viewing the video, all participants were asked again to choose which category of care they would prefer if they became critically ill with advanced cancer (life-prolonging, basic, or comfort care). Participants who were unable to choose a level of medical care were recorded as “not sure.”

Once again, participants also were asked the knowledge questions and whether they would want CPR or ventilation attempted. Finally, we measured the perceived value of the video by asking patients to rate on a 4-point scale whether they were comfortable viewing the video, whether they would recommend the video to others, and whether they found the video helpful in their understanding of their choices. All data were collected in a private, quiet room with the aid of a structured script. For all patients, whether or not they had a do-not-resuscitate (DNR) order documented in their medical record before viewing the video was determined by chart abstraction.

Analysis

Participant characteristics and outcomes were described using proportions for categorical variables and means and standard deviations for continuous variables. The primary outcome measure was change in the level of care chosen. Additional measures included change in knowledge compared with baseline and change in CPR/ventilator preferences compared with baseline. We also were interested in participants' ratings of the visual aid in terms of comfort, utility, and helpfulness. Finally, we compared stated preferences for CPR/ventilation after viewing the video with the code status documented in the medical record before patients viewed the video.

Changes in the levels of care elected by participants before and after viewing the video were analyzed with the McNemar test for situations involving more than a binary outcome.20 Changes in knowledge scores from baseline to postintervention were compared between the 2 groups using 2-sample t tests. Preferences for preintervention CPR/ventilation compared with baseline were compared using exact chi-square tests.

Finally, kappa statistics were used to summarize the agreement between a stated postintervention preference regarding CPR/ventilation and documentation of DNR orders in the medical record before viewing the video. A 2-sided P value < .05 was considered statistically significant for all analyses. Data were analyzed using SAS software (version 9.2; SAS Institute Inc., Cary, NC).

RESULTS

In total, 179 eligible patients were screened for participation in the study, and 86 of those patients (48%) agreed to participate. The most often cited reason for nonparticipation was lack of interest. Four participants withdrew because of lack of time to complete the survey. Two participants were disqualified, 1 because of a recruitment error and another at the request of a spouse, leaving a total of 80 participants. The characteristics of this group are listed in Table 2. Most the patients were men (72.5%), white (91.3%), highly educated (83.7% had some college or higher), and had an advance directive (75%). Prostate cancer was the most common type of cancer (45% of participants) followed by melanoma (18.8%). The average knowledge score at baseline was 2.8 (range, 0-6).

Table 2. Baseline Characteristics of the Study Participants, N = 80
Characteristic No. of Patients (%)
Age, y
 Mean ± SD 65 ± 12
 Range 36-89
Sex
 Women 22 (27.5)
Race
 White 73 (91.3)
 Black 5 (6.3)
 Other 2 (2.4)
Education
 ≤High school 13 (16.3)
 Some college/technical school or higher 67 (83.7)
Religious affiliation
 Christian
  Non-Catholic 14 (17.6)
  Catholic 34 (42.5)
 Jewish 13 (16.3)
 Other 19 (23.6)
Marital status
 Never married 5 (6.3)
 Married/with partner 66 (82.5)
 Divorced 2 (2.5)
 Widowed 5 (6.3)
 Missing 2 (2.4)
Advance directive
 Yesa 60 (75)
Type of cancer
 Prostate 36 (45)
 Melanoma 15 (18.8)
 Pancreatic 8 (10)
 Breast 6 (7.5)
 Lymphoma 4 (5)
 Leiomyosarcoma 2 (2.5)
 Other 9 (11.3)
Desire CPR at baseline
 No 49 (61.3)
Desire ventilation at baseline
 No 54 (67.5)
Knowledge score at baseline
 Mean ± SDb 2.8 ± 0.9
  • Abbreviations: CPR, cardiopulmonary resuscitation; SD, standard deviation.
  • a Advance directive included those patients who had a designated health care proxy, who completed a living will, or both.
  • b Knowledge scores ranged from 0 to 6, and higher scores indicated more knowledge.

After they heard a verbal description of the 3 levels of medical care in advanced cancer, 10 participants (13%) preferred life-prolonging care, 24 (30%) chose basic care, 29 (36%) preferred comfort care, and 17 (21%) were unsure of their preferences. Participant preferences did not change significantly after viewing the video: Nine participants (11%) preferred life-prolonging care, 28 (35%) chose basic care, 29 (36%) chose comfort care, and, 14 (18%) were unsure of their preferences (P = .28) (Fig. 2). Participants significantly improved their knowledge scores after viewing the video compared with baseline (mean ± standard deviation change: 1.6 ± 0.95; P < .001).

Details are in the caption following the image

These charts illustrate the distribution of patient preferences for goals of care after hearing a verbal description (pre-video) and after viewing the video (post-video).

Preferences regarding CPR and ventilation also changed significantly after the video augmentation. At baseline, 49 participants (61%) did not want CPR; and, after the video augmentation, 57 (71%) preferred not to have CPR (P = .03) (Fig. 3). With regard to ventilation, 54 participants (68%) did not want ventilation at baseline, and 64 (80%) did not want ventilation after hearing the verbal description and watching the video (P = .008) (Fig. 4).

Details are in the caption following the image

These charts illustrate the distribution of preferences regarding cardiopulmonary resuscitation (CPR) after hearing a verbal description (pre-video) and after viewing the video (post-video) (P = .03).

Details are in the caption following the image

These charts illustrate the distribution of preferences regarding ventilation after hearing a verbal description (pre-video) and after viewing the video (post-video) (P = .008).

The video was highly acceptable to the participants: Sixty-four participants (80%) indicated that they were either “very” or “somewhat” comfortable watching the video, 61 (76%) indicated that they would “definitely” or “probably” recommend the video to other cancer patients, and 51 (64%) indicated that the video was either “very” or “somewhat” helpful.

Discrepancy existed between participants' stated preferences against CPR or ventilation after the video augmentation and their documented code status before the survey. Of the participants who did not want CPR or ventilation after the video augmentation, only 4 (5%) had a documented DNR order in the medical record (kappa statistic, −0.01; 95% confidence interval, −0.06 to 0.04).

DISCUSSION

Participants with advanced cancer did not significantly change their preferences for life-prolonging care, basic medical care, or comfort care after watching an educational video on the goals of care in advanced cancer compared with a verbal description of these goals of care. However, participants did change their preferences for CPR and ventilation after watching the video augmentation, and they were more knowledgeable about ACP. It is noteworthy that the participants who stated a preference against CPR or ventilation after the video were highly unlikely to have a DNR order in the medical record. Overwhelmingly, participants indicated that they were comfortable watching the video and would recommend it to others. Video decision aids supplementing ACP discussions may play a significant role in helping patients with advanced cancer make more informed end-of-life decisions.

An important factor in ACP is the patient's ability to realistically imagine and comprehend future goals of care. This usually requires envisioning a health state that may be difficult to imagine for some patients, such as being attached to a ventilator in the last moments of life. Watching a video may provide details that are not necessarily communicated by a provider's verbal discussion. At baseline, a significant number of patients were uncertain about their preferences for CPR or ventilation. After viewing the video and compared with baseline, many participants had more knowledge about their decision and were more likely to prefer not to have CPR or ventilation. Comparing knowledge and preferences after the video augmentation simulates how video decision-support tools are intended to reinforce and supplement patient-physician discussions. The findings that many participants would recommend the video to others and believed that the video was helpful lend support for the informational value of visual media.

Often, ACP conversations rely solely on oral communication to provide information regarding future health states. The supplementing of ACP discussions with visual images may have significant clinical implications, because patients sometimes incorrectly imagine future health states.21 Innovative models of ACP must make sure that the media used to educate patients about their options correctly depict future health states and interventions. A more thorough comprehension of future health states with visual images allows patients to observe and better understand their options.

The delivery of end-of-life medical care to patients with advanced cancer that is consistent with their stated preferences is a critical component of high-quality medical care. Our current results suggest that the current modality of ACP discussions for complex decision-making at the end of life often is inadequate and may leave patients with inaccurate impressions. Tools like video decision aids can empower and inform patients with advanced cancer as they deliberate about complicated issues at a vulnerable time in their medical trajectories.

In the current study, we examined the expressed wishes of patients and compared them with documented code status, and we observed a profound lack of correlation. Video decision aids may be a feasible and effective approach toward ascertaining the goals of care; however, our findings also demonstrate that ascertainment of goals of care may not be enough, because steps also must be taken to translate those wishes into a medical order (eg, DNR orders). Lack of correlation between patients' desires for comfort-oriented care at the end of life and their documented orders in the medical record should be considered a medical error no different from missing documentation of allergies, because this lack of correlation may lead to unwanted and tragic outcomes.

Our study has some limitations. Visual images may be edited to steer patients toward a particular decision. We attempted to avoid bias in the creation of the video by including a variety of experts in the review of the film and using the cinema verite style of documentary film making, which favors realistic scenery over staged directions.18 Nevertheless, using a visual medium like film can introduce esthetic biases.

Our trial did not study additional film clips that changed the characteristics of the individuals filmed, such as sex or race. In addition, our before-and-after study design did not address the influence of the film in isolation. We also did not include longer versions of the oral description of the choices of medical care. Rather, we hoped to simulate realistic ACP conversations that occur in practice, which focus on brevity and the broad levels of medical care.

In addition, our participants were recruited from 1 quaternary medical center in New York City that specializes in the care of patients with cancer and included primarily highly educated men with prostate cancer, a unique population reflective of the large numbers of patients with prostate cancer referred to Memorial Sloan-Kettering Comprehensive Cancer Center. Our patients did not include 2 of the leading causes of death from cancer: lung cancer and colon cancer. Attitudes toward the use of video and preferences of patients with advanced cancer may differ by cancer type, geographic location, education, and type of institution. Finally, the majority of participants already had an advance directive, and they probably had considered questions regarding their overall goals of care before seeking help at a highly specialized institution that focuses on cancer care. A greater impact may have been observed had the intervention been presented to patients who had not yet formulated an advance directive.

Our use of visual images to depict the goals of medical care may enhance patients' understanding of their options beyond that provided by oral ACP discussions. Future studies using the visual medium for ACP include trials that study whether patients actually document their choices in advance directives after watching the video as well as following the stability of patients' preferences longitudinally, because many of the patients exposed to the video may make very different decisions in the future as a result of the video. Additional studies also may include use of the video earlier in the disease process, because ACP discussions are ongoing and not a 1-time discussion.

The visual medium offers a powerful aid for providers to ground ACP discussions. Issues surrounding the creation of these films, such as how to design them and who should approve them, must be carefully examined before instituting the routine use of visual decision aids to other diseases. Deliberation among patients, families, physicians, and film-makers is an initial step toward addressing these important factors.

In conclusion, involving cancer patients in ACP discussions surrounding care at the end of life respects their autonomy but requires that providers offer patients the tools with which to make fully informed decisions. The use of video decision aids to supplement verbal descriptions of the goals of care is one powerful means to better inform patients. The use of video decision aids is palatable to patients, increases their understanding of their options, and has the potential to improve the quality of end-of-life care.

Acknowledgements

We thank Patricia Agre, RN, EdD; Susan Holland, MA; Matthew Fury, MD; David Solit, MD; Dana Rathkopf, MD; Ethan Basch, MD; Heather Landau, MD; and David Kissane, MD of the Memorial Sloan-Kettering Cancer Center for their contributions to patient recruitment.

    FUNDING SOURCES

    The project described was supported by award R21CA139121 from the National Cancer Institute. Dr. Volandes also was supported by grant K08HS018780 from the Agency for Healthcare Research and Quality.

    CONFLICT OF INTEREST DISCLOSURES

    Dr. Volandes is on the Board of the Nous Foundation, Inc., a nonprofit foundation that disseminates educational videos. Dr. Volandes has no financial interest in and receives no payment from the Foundation.