The Temperature Feature of ChatGPT: Modifying Creativity for Clinical Research.

More clinicians and researchers are exploring uses for large language model chatbots, such as ChatGPT, for research, dissemination, and educational purposes. Therefore, it becomes increasingly relevant to consider the full potential of this tool, including the special features that are currently available through the application programming interface. One of these features is a variable called temperature, which changes the degree to which randomness is involved in the model's generated output. This is of particular interest to clinicians and researchers. By lowering this variable, one can generate more consistent outputs; by increasing it, one can receive more creative responses. For clinicians and researchers who are exploring these tools for a variety of tasks, the ability to tailor outputs to be less creative may be beneficial for work that demands consistency. Additionally, access to more creative text generation may enable scientific authors to describe their research in more general language and potentially connect with a broader public through social media. In this viewpoint, we present the temperature feature, discuss potential uses, and provide some examples.

Bolstering Advance Care Planning Measurement Using Natural Language Processing.

Despite its growth as a clinical activity and research topic, the complex dynamic nature of advance care planning (ACP) has posed serious challenges for researchers hoping to quantitatively measure it. Methods for measurement have traditionally depended on lengthy manual chart abstractions or static documents (e.g., advance directive forms) even though completion of such documents is only one aspect of ACP. Natural language processing (NLP), in the form of an assisted electronic health record (EHR) review, is a technological advancement that may help researchers better measure ACP activity. In this article, we aim to show how NLP-assisted EHR review supports more accurate and robust measurement of ACP. We do so by presenting three example applications that illustrate how using NLP for this purpose supports (1) measurement in research, (2) detailed insights into ACP in quality improvement, and (3) identification of current limitations of ACP in clinical settings.

Assessing performance of Geriatric Surgery Verification Program preoperative communication standards in spine surgery.

BACKGROUND: To assess performance of the American College of Surgeons Geriatric Surgery Verification (GSV) Program preoperative communication standards in older patients undergoing high risk spine surgery. METHODS: We performed an external validation of a natural language processing (NLP) method for identifying documentation meeting GSV communication standards. We then applied this method to a retrospective cohort of patients aged 65 and older who underwent spinal fusion procedures between January 2018-December 2020 in a large healthcare system in Massachusetts. Our primary outcome of interest was fulfillment of GSV communication domains: overall health goals, treatment goals, and patient-centered outcomes. Factors associated with the fulfillment of at least one domain were assessed using Poisson regression to adjust for confounding. RESULTS: External validation of the NLP method had a sensitivity of 88.6% and specificity of 99.0%. Our study population included 1294 patients, of whom only 0.8% (n = 10) patients contained documentation of all three GSV domains, and 33.7% (n = 436) had documentation fulfilling at least one GSV domain. The GSV domain with lowest frequency of documentation was overall health goals, with only 35 (2.7%) of patients meeting this requirement. Adjusted analysis suggested that patients with a Charlson comorbidity score of one or more had higher fulfillment of GSV criteria (CCI 1-3: prevalence rate ratio (PRR) 1.8, 95% confidence interval (CI) 1.5-2.1; CCI >3: PRR 1.5, 95% CI 1.2-1.9). CONCLUSION: A paucity of geriatric patients undergoing spine surgery had preoperative documentation consistent with GSV standards. Given that spine surgery is one of the highest risk surgeries in older adults and GSV standards are relevant to all surgical specialties, wider promulgation of these standards is warranted.

Role matters in understanding 'quality' in palliative care: a qualitative analysis of patient, caregiver and practitioner perspectives.

OBJECTIVES: To compare the discussions from two panels on the concept of palliative care quality for patients with advanced cancer, exploring the priorities reflected in each group's perspectives. DESIGN: We convened two RAND-UCLA appropriateness panel discussions on palliative care quality in advanced cancer. Discussions were audio-recorded and transcribed verbatim. Panel transcripts were analysed thematically using a matrix approach to examine perceptions and experiences of quality. SETTING: Discussions were framed within the context of advanced cancer care and palliative care. PARTICIPANTS: The patient-caregiver panel had 9 patients with current or a history of cancer and caregivers, and the practitioner panel had 10 expert practitioners representing fields of oncology, primary care, social work, palliative care, nursing, pain management and ethics. RESULTS: Our analysis identified three thematic categories for understanding quality common across both groups and nine subthemes within those categories. At the highest level, quality was conceived as: (1) the patient and caregiver experience of care, (2) technical competence and (3) the structure of health system. Among the subthemes, four were present in only one of the two group's discussions: 'purpose and action' was specific to the patient-caregiver panel, whereas 'adhering to best medical practice', 'mitigating unintended consequences and side effects', and 'health system resources and costs' were specific to the practitioner panel. CONCLUSIONS: While both panels aligned on the three key domains of quality, the particular dimensions through which they perceived quality varied in relation to their experience and role as a professional provider of care versus recipient of healthcare services. These differences suggest the importance of adopting a collaborative approach to quality measurement and improvement so that the values of all interested parties are represented in improvement efforts.

What Patients Facing Cancer and Caregivers Want From Communication in Times of Crisis: A Qualitative Study in the Early Months of the COVID-19 Pandemic.

BACKGROUND: Interpersonal communication is a cornerstone of patient-centered care. We aimed to identify what patients with cancer and caregivers may want from communication during a public health crisis. METHODS: We interviewed 15 patients (8 Veteran, 7 non-Veteran) and caregivers from regionally, racially, and ethnically diverse backgrounds across the US about serious illness care and quality of care during the COVID-19 pandemic Using an iterative, inductive and deductive process, 2 coders analyzed content associated with the code "Communication," which appeared 71 times, and identified 5 themes. RESULTS: Participants identified as White (10), Latino/a (3), Asian (1), and Black (1). (1) Help patients and caregivers prepare for care during crisis by communicating medical information directly and proactively. (2) Explain how a crisis might influence medical recommendations and impact on recovery from illness. (3) Use key messengers to improve communication between primary teams, patients, and caregivers. (4) Include caregivers and families in communication when they cannot be physically present. (5) Foster bidirectional communication with patients and families to engage them in shared decision-making during a vulnerable time. CONCLUSION: Communication is critical during a public health crisis yet overwhelmed clinicians may not be able to communicate effectively. Communicating with caregivers and family, transparent and timely communication, ensuring diverse providers are on the same page, and effective listening are known gaps even before the COVID-19 pandemic. Clinicians may need quick interventions, like education about goals of care, to remind them about what seriously ill patients and their caregivers want from communication and offer patient-centered care during crises.

Video Intervention and Goals-of-Care Documentation in Hospitalized Older Adults: The VIDEO-PCE Randomized Clinical Trial.

Importance: Despite the benefits of goals-of-care (GOC) communication, many hospitalized individuals never communicate their goals or preferences to clinicians. Objective: To assess whether a GOC video intervention delivered by palliative care educators (PCEs) increased the rate of GOC documentation. Design, Setting, and Participants: This pragmatic, stepped-wedge cluster randomized clinical trial included patients aged 65 years or older admitted to 1 of 14 units at 2 urban hospitals in New York and Boston from July 1, 2021, to October 31, 2022. Intervention: The intervention involved PCEs (social workers and nurses trained in GOC communication) facilitating GOC conversations with patients and/or their decision-makers using a library of brief, certified video decision aids available in 29 languages. Patients in the control period received usual care. Main Outcome and Measures: The primary outcome was GOC documentation, which included any documentation of a goals conversation, limitation of life-sustaining treatment, palliative care, hospice, or time-limited trials and was obtained by natural language processing. Results: A total of 10 802 patients (mean [SD] age, 78 [8] years; 51.6% male) were admitted to 1 of 14 hospital units. Goals-of-care documentation during the intervention phase occurred among 3744 of 6023 patients (62.2%) compared with 2396 of 4779 patients (50.1%) in the usual care phase (P < .001). Proportions of documented GOC discussions for Black or African American individuals (865 of 1376 [62.9%] vs 596 of 1125 [53.0%]), Hispanic or Latino individuals (311 of 548 [56.8%] vs 218 of 451 [48.3%]), non-English speakers (586 of 1059 [55.3%] vs 405 of 863 [46.9%]), and people living with Alzheimer disease and related dementias (520 of 681 [76.4%] vs 355 of 570 [62.3%]) were greater during the intervention phase compared with the usual care phase. Conclusions and Relevance: In this stepped-wedge cluster randomized clinical trial of older adults, a GOC video intervention delivered by PCEs resulted in higher rates of GOC documentation compared with usual care, including among Black or African American individuals, Hispanic or Latino individuals, non-English speakers, and people living with Alzheimer disease and related dementias. The findings suggest that this form of patient-centered care delivery may be a beneficial decision support tool. Trial Registration: ClinicalTrials.gov Identifier: NCT04857060.

Forms or Free-Text? Measuring Advance Care Planning Activity Using Electronic Health Records.

Advance care planning (ACP) discussions seek to guide future serious illness care. These discussions may be recorded in the electronic health record by documentation in clinical notes, structured forms and directives, and physician orders. Yet, most studies of ACP prevalence have only examined structured electronic health record elements and ignored data existing in notes. We sought to investigate the relative comprehensiveness and accuracy of ACP documentation from structured and unstructured electronic health record data sources. We evaluated structured and unstructured ACP documentation present in the electronic health records of 435 patients with cancer drawn from three separate healthcare systems. We extracted structured ACP documentation by manually annotating written documents and forms scanned into the electronic health record. We coded unstructured ACP documentation using a rule-based natural language processing software that identified ACP keywords within clinical notes and was subsequently reviewed for accuracy. The unstructured approach identified more instances of ACP documentation (238, 54.7% of patients) than the structured ACP approach (187, 42.9% of patients). Additionally, 16.6% of all patients with structured ACP documentation only had documents that were judged as misclassified, incomplete, blank, unavailable, or a duplicate of a previously entered erroneous document. ACP documents scanned into electronic health records represent a limited view of ACP activity. Research and measures of clinical practice with ACP should incorporate information from unstructured data.

Depression care quality among patients with solid tumor cancers detected to have depression in Veterans Health Administration primary care clinics.

Patients with cancer, especially advanced cancer, experience depression at high rates. We aimed to evaluate the quality of depression care received by patients with solid tumor cancer and advanced solid tumor cancer in Veterans Affairs (VA) primary care clinics. This is a retrospective cohort study of patients seen in 82 VA primary care clinics who newly screened positive for depression on the Patient Health Questionnaire (PHQ-2). Outcomes included timely follow-up within 84 or 180 days (3+ mental health specialty, 3+ psychotherapy, or 3+ primary care visits with depression diagnosis codes) and minimum treatment within 1 year (60+ days antidepressants prescribed, 4+ mental health specialty visits, or 3+ psychotherapy visits). 608,042 individuals were seen in VA primary care clinics during this period; 49,839 patients (8.2%) had solid tumor cancer and 9,278 (1.5%) had advanced or poor-prognosis solid tumor cancer. For 686 observations of patients with cancer and new depression, rates of appropriate follow-up were 22.3% within 84 days and 38.2% within 180 days. For 73 observations of patients with advanced or poor-prognosis cancer and new depression, rates of appropriate follow-up were 21.9% within 84 days and 34.3% within 180 days. Rates of minimum treatment within 1 year were 68.4% and 64.4% for patients with cancer and patients with advanced or poor-prognosis cancer, respectively. Quality of timely depression management is low in patients with solid tumor cancers. Even in health systems with well-integrated mental health services, care gaps remain for patients with cancer and depression. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Mitigating bias in AI at the point of care.

Promoting equity in AI in health care requires addressing biases at cli nical implementation.

Integrating Patient and Expert Perspectives to Conceptualize High-Quality Palliative Cancer Care for Symptoms in the US Veterans Health Administration: A Qualitative Study.

Quality measurement is typically the domain of clinical experts and health system leaders; patient/caregiver perspectives are rarely solicited. We aimed to describe and integrate clinician and patient/caregiver conceptualizations of high-quality palliative symptom care for patients receiving care for advanced cancer within the US Veterans Health Administration in the context of existing quality measures. We conducted a secondary qualitative analysis of transcripts from prioritization discussions of process quality measures relevant to cancer palliative care. These discussions occurred during 2 modified RAND-UCLA appropriateness panels: a panel of 10 palliative care clinical expert stakeholders (7 physicians, 2 nurses, 1 social worker) and a panel of 9 patients/caregivers with cancer experience. Discussions were recorded, transcribed, and independently double-coded using an a priori logical framework. Content analysis was used to identify subthemes within codes and axial coding was used to identify crosscutting themes. Patients/caregivers and clinical experts contributed important perspectives to 3 crosscutting themes. First, proactive elicitation of symptoms is critical. Patients/caregivers especially emphasized importance of comprehensive and proactive screening and assessment, especially for pain and mental health. Second, screening and assessment alone is not enough; information elicited from patients must inform care. Measuring screening/assessment and management care processes separately has important limitations. Lastly, high-quality symptom management can be broadly defined if it is patient-centered; high-quality care takes an individualized approach and might include non-medical or non-pharmacological symptom management. Integrating the perspectives of clinical experts and patients/caregivers is critical for health systems to consider as they design and implement quality measures for palliative cancer care.

Shared Decision Making in the Geriatric Surgery Verification Program: Assessing Baseline Performance.

CONTEXT: As part of the launch of the Geriatric Surgery Verification program in 2019, the American College of Surgeons issued care standards for older patients, including requirements for preoperative documentation of patients' goals. Hospital performance on these standards prior to the Geriatric Surgery Verification program is unknown. OBJECTIVES: To assess baseline performance of the Geriatric Surgery Verification (GSV) standard for documentation of preoperative goals for older patients, and to determine factors associated with standard adherence. METHODS: Using natural language processing, this study examines the electronic health records of patients aged 65 years or older who underwent coronary artery bypass grafts (CABG) or colectomies in 2017 or 2018 at three hospitals. The primary outcome was adherence to at least one of the three components of GSV Standard 5.1, which requires preoperative documentation of overall health goals, treatment goals, and patient-centered outcomes. RESULTS: A total of 2630 operations and 2563 patients were included. At least one component of the standard was met in 307 (11.7%) operations and all three components were met in 5 (0.2%). Higher likelihood of meeting the standard was demonstrated for patients who were female (odds ratio [OR] 1.30; 95% CI 1.00-1.68), undergoing colectomy (OR 2.82; 95% CI 2.15-3.72), or with more comorbidities (Charlson scores >3 [OR 1.55; 95% CI 1.14-2.09]). CONCLUSION: Before GSV program implementation, clinicians for two major operations almost never met the GSV standard for preoperative discussion of patient goals. Interdisciplinary teams will need to adjust clinical practice to meet best-practice communication standards for older patients.

Development of a keyword library for capturing PRO-CTCAE-focused "symptom talk" in oncology conversations.

Objectives: As computational methods for detecting symptoms can help us better attend to patient suffering, the objectives of this study were to develop and evaluate the performance of a natural language processing keyword library for detecting symptom talk, and to describe symptom communication within our dataset to generate insights for future model building. Materials and Methods: This was a secondary analysis of 121 transcribed outpatient oncology conversations from the Communication in Oncologist-Patient Encounters trial. Through an iterative process of identifying symptom expressions via inductive and deductive techniques, we generated a library of keywords relevant to the Patient-Reported Outcome version of the Common Terminology Criteria for Adverse Events (PRO-CTCAE) framework from 90 conversations, and tested the library on 31 additional transcripts. To contextualize symptom expressions and the nature of misclassifications, we qualitatively analyzed 450 mislabeled and properly labeled symptom-positive turns. Results: The final library, comprising 1320 terms, identified symptom talk among conversation turns with an F1 of 0.82 against a PRO-CTCAE-focused gold standard, and an F1 of 0.61 against a broad gold standard. Qualitative observations suggest that physical symptoms are more easily detected than psychological symptoms (eg, anxiety), and ambiguity persists throughout symptom communication. Discussion: This rudimentary keyword library captures most PRO-CTCAE-focused symptom talk, but the ambiguity of symptom speech limits the utility of rule-based methods alone, and limits to generalizability must be considered. Conclusion: Our findings highlight opportunities for more advanced computational models to detect symptom expressions from transcribed clinical conversations. Future improvements in speech-to-text could enable real-time detection at scale.

Delirium and Healthcare Utilization in Patients Undergoing Hematopoietic Stem Cell Transplantation.

Delirium, a common neuropsychiatric syndrome among hospitalized patients, has been associated with significant morbidity and mortality in patients undergoing hematopoietic stem cell transplantation (HSCT). Although delirium is often reversible with prompt diagnosis and appropriate management, timely screening of hospitalized patients, including HSCT recipients at risk for delirium, is lacking. The association between delirium symptoms and healthcare utilization among HSCT recipients is also limited. We conducted a retrospective analysis of 502 hospitalized patients admitted for allogeneic or autologous HSCT at 2 tertiary care hospitals between April 2016 and April 2021. We used Natural Language Processing (NLP) to identify patients with delirium symptoms, as defined by an NLP-assisted chart review of the electronic health record (EHR). We used multivariable regression models to examine the associations between delirium symptoms, clinical outcomes, and healthcare utilization, adjusting for patient-, disease-, and transplantation-related factors. Overall, 44.4% (124 of 279) of patients undergoing allogeneic HSCT and 39.0% (87 of 223) of those undergoing autologous HSCT were identified as having delirium symptoms during their index hospitalization. Two-thirds (139 of 211) of the patients with delirium symptoms were prescribed treatment with antipsychotic medications. Among allogeneic HSCT recipients, delirium symptoms were associated with longer hospital length of stay (ß = 7.960; P < .001), fewer days alive and out of the hospital (ß = -23.669; P < .001), and more intensive care unit admissions (odds ratio, 2.854; P = .002). In autologous HSCT recipients, delirium symptoms were associated with longer hospital length of stay (ß = 2.204; P < .001). NLP-assisted EHR review is a feasible approach to identifying hospitalized patients, including HSCT recipients at risk for delirium. Because delirium symptoms are negatively associated with health care utilization during and after HSCT, our findings underscore the need to efficiently identify patients hospitalized for HSCT who are at risk of delirium to improve their outcomes. © 2023 American Society for Transplantation and Cellular Therapy. Published by Elsevier Inc.

Top Ten Tips Palliative Care Clinicians Should Know About Leveraging the Electronic Health Record for Data Collection and Quality Improvement.

As palliative care (PC) programs rapidly grow and expand across settings, the need to measure, improve, and standardize high-quality PC has also grown. The electronic health record (EHR) is a key component of these efforts as a central hub of care delivery and a repository of patient and system data. Deliberate efforts to leverage the EHR for PC quality improvement (QI) can help PC programs and health systems improve care for patients with serious illnesses. This article, written by clinicians with experience in QI, informatics, and clinical program development, provides practical tips and guidance on EHR strategies and tools for QI and quality measurement.

The Use of Natural Language Processing to Assess Social Support in Patients With Advanced Cancer.

BACKGROUND: Data examining associations among social support, survival, and healthcare utilization are lacking in patients with advanced cancer. METHODS: We conducted a cross-sectional secondary analysis using data from a prospective longitudinal cohort study of 966 hospitalized patients with advanced cancer at Massachusetts General Hospital from 2014 through 2017. We used NLP to identify extent of patients' social support (limited versus adequate as defined by NLP-aided review of the Electronic Health Record (EHR)). Two independent coders achieved a Kappa of 0.90 (95% CI: 0.84-1.00) using NLP. Using multivariable regression models, we examined associations of social support with: 1) OS; 2) death or readmission within 90 days of hospital discharge; 3) time to readmission within 90 days; and 4) hospital length of stay (LOS). RESULTS: Patients' median age was 65 (range: 21-92) years, and a plurality had gastrointestinal (GI) cancer (34.3%) followed by lung cancer (19.5%). 6.2% (60/966) of patients had limited social support. In multivariable analyses, limited social support was not significantly associated with OS (HR = 1.13, P = 0.390), death or readmission (OR = 1.18, P = 0.578), time to readmission (HR = 0.92, P = 0.698), or LOS (ß = -0.22, P = 0.726). We identified a potential interaction suggesting cancer type (GI cancer versus other) may be an effect modifier of the relationship between social support and OS (interaction term P = 0.053). In separate unadjusted analyses, limited social support was associated with lower OS (HR = 2.10, P = 0.008) in patients with GI cancer but not other cancer types (HR = 1.00, P = 0.991). CONCLUSION: We used NLP to assess the extent of social support in patients with advanced cancer. We did not identify significant associations of social support with OS or healthcare utilization but found cancer type may be an effect modifier of the relationship between social support and OS. These findings underscore the potential utility of NLP for evaluating social support in patients with advanced cancer.

Surgeon-reported Factors Influencing Adoption of Quality Standards for Goal-concordant Care in Patients With Advanced Cancer: A Qualitative Study.

OBJECTIVE: This study explored surgical oncologists' perspectives on factors influencing adoption of quality standards in patients with advanced cancer. BACKGROUND: The American College of Surgeons Geriatric Surgery Verification Program includes communication standards designed to facilitate goal-concordant care, yet little is known about how surgeons believe these standards align with clinical practice. METHODS: Semistructured video-based interviews were conducted from November 2020 to January 2021 with academic surgical oncologists purposively sampled based on demographics, region, palliative care certification, and years in practice. Interviews addressed: (1) adherence to standards documenting care preferences for life-sustaining treatment, surrogate decision-maker, and goals of surgery; and (2) factors influencing their adoption into practice. Interviews were audio-recorded, transcribed, qualitatively analyzed, and conducted until thematic saturation was reached. RESULTS: Twenty-six surgeons participated (57.7% male, 8.5 mean years in practice, 19.2% palliative care board-certified). Surgeons reported low adherence to documenting care preferences and surrogate decision-maker and high adherence to discussing, but not documenting, goals of surgery. Participants held conflicting views about the relevance of care preferences to preoperative conversations and surrogate decision-maker documentation by the surgeon and questioned the direct connection between documentation of quality standards and higher value patient care. Key themes regarding factors influencing adoption of quality standards included organizational culture, workflow, and multidisciplinary collaboration. CONCLUSIONS: Although surgeons routinely discuss goals of surgery, documentation is inconsistent; care preferences and surrogate decision-makers are rarely discussed or documented. Adherence to these standards would be facilitated by multidisciplinary collaboration, institutional standardization, and evidence linking standards to higher value care.

Prevalence, Management, and Outcomes Related to Preoperative Medical Orders for Life Sustaining Treatment (MOLST) in an Adult Surgical Population: Preoperative MOLST and Code Status Discussions.

OBJECTIVE: To determine prevalence of documented preoperative code status discussions and postoperative outcomes (specifically mortality, readmission, and discharge disposition) of patients with completed MOLST forms before surgery. SUMMARY OF BACKGROUND DATA: A MOLST form documents patient care preference regarding treatment limitations. When considering surgery in these patients, preoperative discussion is necessary to ensure concordance of care. Little is known about prevalence of these discussions and postoperative outcomes. METHODS: A retrospective cohort study was conducted consisting of all patients having surgery during a 1-year period at a tertiary care academic center in Boston, Massachusetts. RESULTS: Among 21,787 surgical patients meeting inclusion criteria, 402 (1.8%) patients had preoperative MOLST. Within the MOLST, 224 (55.7%) patients had chosen to limit cardiopulmonary resuscitation and 214 (53.2%) had chosen to limit intubation and mechanical ventilation. Code status discussion was documented presurgery in 169 (42.0%) patients with MOLST. Surgery was elective or nonurgent for 362 (90%), and median length of stay (Q1, Q3) was 5.1 days (1.9, 9.9). The minority of patients with preoperative MOLST were discharged home [169 (42%), and 103 (25.6%) patients were readmitted within 30 days. Patients with preoperative MOLST had a 30-day mortality of 9.2% (37 patients) and cumulative 90-day mortality of 14.9% (60 patients). CONCLUSIONS: Fewer than half of surgical patients with preoperative MOLST have documented code status discussions before surgery. Given their high risk of postoperative mortality and the diversity of preferences found in MOLST, thoughtful discussion before surgery is critical to ensure concordant perioperative care.

Study protocol for Video Images about Decisions to Improve Ethical Outcomes with Palliative Care Educators (VIDEO-PCE): a pragmatic stepped wedge cluster randomised trial of older patients admitted to the hospital

INTRODUCTION: Despite the known benefit to patients and families, discussions about goals, values and preferences for medical care in advancing serious illness often do not occur. Many system and clinician factors, such as patient and clinician reticence and shortage of specialty palliative care teams, contribute to this lack of communication. To address this gap, we designed an intervention to promote goals-of-care conversations and palliative care referrals in the hospital setting by using trained palliative care educators and video decision aids. This paper presents the rationale, design and methods for a trial aimed at addressing barriers to goals-of-care conversations for hospitalised adults aged 65 and older and those with Alzheimer's disease and related Dementias, regardless of age. METHODS AND ANALYSIS: The Video Image about Decisions to Improve Ethical Outcomes with Palliative Care Educators is a pragmatic stepped wedge, cluster randomised controlled trial, which aims to improve and extend goals-of-care conversations in the hospital setting with palliative care educators trained in serious illness communication and video decision aids. The primary outcome is the proportion of patients with goals-of-care documentation in the electronic health record. We estimate that over 9000 patients will be included. ETHICS AND DISSEMINATION: The Institutional Review Board (IRB) at Boston Medical Center will serve as the single IRB of record for all regulatory and ethical aspects of this trial. BMC Protocol Number: H-41482. Findings will be presented at national meetings and in publications. This trial is registered at ClinicalTrials.gov. TRIAL REGISTRATION NUMBER: NCT04857060; ClinicalTrials.gov.