Palliative Care and Advance Care Planning Intervention Fidelity Monitoring: Methods and Lessons Learned From PCORI-Funded Large-Scale, Pragmatic Clinical Trials.

Over the past decade, the Patient-Centered Outcomes Research Institute (PCORI) funded multiple large-scale, comparative effectiveness clinical trials evaluating palliative care and advance care planning interventions. These are complex multicomponent interventions that need robust but flexible fidelity monitoring. Fidelity is necessary to maintain both internal and external validity within palliative care intervention research and to ultimately evaluate the real-world impact of high-quality interventions. Different trials not only took varying approaches to fidelity monitoring but also uncovered both unique and common challenges and facilitators. This article summarizes 8 of these trials and highlights approaches, adaptations, barriers, and facilitators for intervention fidelity monitoring. Identifying and delivering core elements while simultaneously allowing adaptations of noncore elements is a vital part of fidelity monitoring. Dissemination of such experiences can inform both future palliative care research as well as ongoing implementation of palliative care and advance care planning interventions across diverse clinical practices. Adoption of rigorous intervention fidelity methods is critical to advancing the science and reproducibility of palliative care interventions.

Implementing Patient-Centered Outcomes Research Institute Stakeholder Engagement Principles in Models of Palliative Care Delivery and Advance Care Planning Research.

BACKGROUND: Given the many challenges of conducting research that addresses the palliative and end-of-life care needs of patients with serious illnesses, stakeholder engagement starting from the moment of study conceptualization and design is critical to ensure successful participant recruitment, data collection, intervention delivery, data analysis, and dissemination. METHODS: Guided by a conceptual model published by the Patient-Centered Outcomes Research Institute (PCORI) entitled, "Measuring What Matters for Advancing the Science and Practice of Engagement"14 and with the support of a PCORI Engagement Officer, representatives from 9 PCORI-funded study teams formed a working group to survey team members and review, outline, and describe key lessons learned and best practices for promoting stakeholder engagement in palliative care research. RESULTS: Almost all study teams engaged with patients/caregivers, clinicians, researchers, and health care system experts as stakeholder partners. About half the teams also included payers and training institutions as part of their stakeholder advisors as well as a range of content experts. Study teams relied on a variety of support structures and resources, and they employed 10 distinct methods for maintaining engagement. All engagement methods were generally considered to be effective by teams who used the method, though there was some variability in team-rated engagement quality of each method. Nine barriers to stakeholder engagement were identified across the 9 studies as well as 9 strategies (or facilitators) to overcome these barriers. We share examples of how stakeholder engagement impacted studies in all phases, including the preparatory phase, study initiation phase, execution phase, and data analysis/dissemination phase. CONCLUSIONS: Teams utilized a variety of resources and support structures as well as capitalized on multiple engagement methods for fostering stakeholder engagement, resulting in a high level of collaboration and integration.

Conducting Comparative Effectiveness, Multisite Palliative Care and Advance Care Planning Trials: Lessons Learned and Future Directions From PCORI-Funded Studies.

The Patient-Centered Outcomes Research Institute (PCORI) funded multiple large-scale comparative effectiveness clinical trials evaluating palliative care (PC) and advance care planning (ACP) healthcare delivery models. This article provides an overview of the most common barriers our investigative teams encountered while implementing these trials and the strategies we utilized to overcome these challenges, with particular attention to identifying research partners for multisite trials; addressing contracting and regulatory issues; creating a team governance structure; training and engaging study staff across sites; recruiting, consenting, and enrolling study participants; collecting PC and ACP data and study outcomes; and managing multisite collaborations. The goal of this article is to provide guidance on how to best plan for and conduct rigorous trials evaluating PC and ACP healthcare delivery interventions moving forward.

Automated Extraction of Patient-Centered Outcomes After Breast Cancer Treatment: An Open-Source Large Language Model-Based Toolkit.

PURPOSE: Patient-centered outcomes (PCOs) are pivotal in cancer treatment, as they directly reflect patients' quality of life. Although multiple studies suggest that factors affecting breast cancer-related morbidity and survival are influenced by treatment side effects and adherence to long-term treatment, such data are generally only available on a smaller scale or from a single center. The primary challenge with collecting these data is that the outcomes are captured as free text in clinical narratives written by clinicians. MATERIALS AND METHODS: Given the complexity of PCO documentation in these narratives, computerized methods are necessary to unlock the wealth of information buried in unstructured text notes that often document PCOs. Inspired by the success of large language models (LLMs), we examined the adaptability of three LLMs, GPT-2, BioGPT, and PMC-LLaMA, on PCO tasks across three institutions, Mayo Clinic, Emory University Hospital, and Stanford University. We developed an open-source framework for fine-tuning LLM that can directly extract the five different categories of PCO from the clinic notes. RESULTS: We found that these LLMs without fine-tuning (zero-shot) struggle with challenging PCO extraction tasks, displaying almost random performance, even with some task-specific examples (few-shot learning). The performance of our fine-tuned, task-specific models is notably superior compared with their non-fine-tuned LLM models. Moreover, the fine-tuned GPT-2 model has demonstrated a significantly better performance than the other two larger LLMs. CONCLUSION: Our discovery indicates that although LLMs serve as effective general-purpose models for tasks across various domains, they require fine-tuning when applied to the clinician domain. Our proposed approach has the potential to lead more efficient, adaptable models for PCO information extraction, reducing reliance on extensive computational resources while still delivering superior performance for specific tasks.

Implementation Strategies for Quality Improvement in Palliative Care: A Scoping Review.

BACKGROUND: Quality improvement (QI) programs based on person-centred outcome measures (PCOMs) play an important role in promoting optimal palliative care. However, routine use of PCOMs has been slow and difficult to implement, including within QI programs. OBJECTIVE: This study aimed to identify implementation strategies that support the implementation of PCOMs as routine practice in hospital-based palliative care, as well as the implementation theories, models and frameworks (TMFs) guiding the design of these implementation strategies. METHODS: A scoping review was conducted in accordance with the Joanna Briggs Institute (JBI) Scoping Review framework. Four databases (Medline, CINAHL, Scopus and PubMed) were systematically searched for literature published between 1 January 1990 and 8 March 2024. RESULTS: One hundred and fifteen unique implementation strategies, identified from 11 included studies, were mapped onto the 73 Expert Recommendations for Implementing Change (ERIC) discrete implementation strategies, covering 52% of the ERIC strategies. The most commonly used categories were train and educate stakeholders, and support clinicians, followed by develop stakeholder interrelationships and use evaluation and iterative strategies. Three key themes emerged: what to do; how to do it; and who to do it with. Only four studies employed TMFs to guide the design of the implementation strategies in this review. CONCLUSIONS: To promote the implementation of PCOM-based QI programs, strategies should be developed based on identified/potential barriers and facilitators by using rigorous TMFs. The components of the implementation strategies must be reported transparently and consistently to enable replication and measurement in future research and practice. PATIENT AND PUBLIC CONTRIBUTION: This scoping review does not directly involve patients or the general public in its design or execution. However, it is part of an implementation study aimed at integrating the Palliative Care Outcome Collaboration (PCOC) model into routine clinical practice at a cancer hospital in China. Before the formal implementation, palliative care professionals from this hospital highlighted the need for a comprehensive analysis of existing evidence to support the effective adoption of the PCOC model in their specific clinical setting.

The Tennessee Heart Health Network effectiveness study: A stepped wedge cluster randomized controlled trial to assess the effectiveness of statewide quality improvement cooperative participation on cardiovascular outcomes.

BACKGROUND: Cardiovascular disease (CVD) is the primary cause of premature morbidity and mortality in the United States and Tennessee ranks among the highest in CVD events. While patient-centered outcomes research (PCOR) evidence-based approaches that reach beyond the traditional doctor-patient visit hold promise to improve CVD care and prevent serious complications, most primary care providers lack time, knowledge, and infrastructure to implement these proven approaches. Statewide primary care quality improvement (QI) collaboratives hold potential to help address primary care needs, however, little is known regarding their effectiveness in improving uptake of PCOR evidence-based population health approaches and improving CVD outcomes. This study describes the design and implementation of a stepped-wedge cluster randomized controlled trial to assess the effectiveness of participation in a statewide quality improvement cooperative (The Tennessee Heart Health Network [TN-HHN]) on cardiovascular outcomes. METHODS/DESIGN: The TN-HHN Effectiveness Study randomized 77 practices to 4 waves (i.e., clusters), with each wave beginning three months after the start of the prior wave and lasting for 18 months. All practice clusters received one of three Network interventions, and outcomes are measured for each three months both in the control phase and the intervention phase. Primary outcomes include Center for Medicare and Medicaid Services measures for aspirin use, blood pressure control, cholesterol control, and smoking cessation (ABCS). CONCLUSIONS: This trial, upon its conclusion, will allow us to assess the effect of participation in a statewide quality improvement cooperative on cardiovascular outcomes as well as key contributors to successful practice transformation.

Leveraging sensor-based functional outcomes to enhance understanding of the patient experience: challenges and opportunities.

INTRODUCTION: Sensor-based digital health technology (DHT) has emerged as a promising means to assess patient functioning within and outside clinical trials. Sensor-based functional outcomes (SBFOs) provide valuable insights that complement other measures of how a patient feels or functions to enhance understanding of the patient experience to inform medical product development. AREAS COVERED: This perspective paper provides recommendations for defining SBFOs, discusses the core evidence required to support SBFOs to inform decision-making, and considers future directions for the field. EXPERT COMMENTARY: The clinical outcome assessment (COA) development process provides an important starting point for developing patient-centered SBFOs; however, given the infancy of the field, SBFO development may benefit from a hybrid approach to evidence generation by merging exploratory data analysis with patient engagement in measure development. Effective SBFO development requires combining unique expertise in patient engagement, measurement and regulatory science, and digital health and analytics. Challenges specific to SBFO development include identifying concepts of interest, ensuring measurement of meaningful aspects of health, and identifying thresholds for meaningful change. SBFOs are complementary to other COAs and, as part of an integrated evidence strategy, offer great promise in fostering a holistic understanding of patient experience and treatment benefits, particularly in real-world settings.

Patient outcome quality indicators for older persons in acute care: original development data using interRAI AC-CGA.

BACKGROUND: A range of strategies are available that can improve the outcomes of older persons particularly in relation to basic activities of daily living during and after an acute care (AC) episode. This paper outlines the original development of outcome-oriented quality indicators (QIs) in relation to common geriatric syndromes and function for the care of the frail aged hospitalized in acute general medical wards. METHODS: Design QIs were developed using evidence from literature, expert opinion, field study data and a formal voting process. A systematic literature review of literature identified existing QIs (there were no outcome QIs) and evidence of interventions that improve older persons' outcomes in AC. Preliminary indicators were developed by two expert panels following consideration of the evidence. After analysis of the data from field testing (indicator prevalence, variability across sites), panel meetings refined the QIs prior to a formal voting process. SETTING: Data was collected in nine Australian general medical wards. PARTICIPANTS: Patients aged 70 years and over, consented within 24 h of admission to the AC ward. MEASUREMENTS: The interRAI Acute Care - Comprehensive Geriatric Assessment (interRAI AC-CGA) was administered at admission and discharge; a daily risk assessment in hospital; 28-day phone follow-up and chart audit. RESULTS: Ten outcome QIs were established which focused on common geriatric syndromes and function for the care of the frail aged hospitalized in acute general medical wards. CONCLUSION: Ten outcome QIs were developed. These QIs can be used to identify areas where specific action will lead to improvements in the quality of care delivered to older persons in hospital.

Patient-centered outcomes and outcome measurements for people aged 65 years and older-a scoping review.

INTRODUCTION: The aging population is a challenge for the healthcare system that must identify strategies that meet their needs. Practicing patient-centered care has been shown beneficial for this patient-group. The effect of patient-centered care is called patient-centered outcomes and can be appraised using outcomes measurements. OBJECTIVES: The main aim was to review and map existing knowledge related to patient-centered outcomes and patient-centered outcomes measurements for older people, as well as identify key-concepts and knowledge-gaps. The research questions were: How can patient-centered outcomes for older people be measured, and which patient-centered outcomes matters the most for the older people? STUDY DESIGN: Scoping review. METHODS: Search for relevant publications in electronical databases, grey literature databases and websites from year 2000 to 2021. Two reviewers independently screened titles and abstracts, followed by full text review and extraction of data using a data extraction framework. RESULTS: Eighteen studies were included, of which six with involvement of patients and/or experts in the process on determine the outcomes. Outcomes that matter the most to older people was interpreted as: access to- and experience of care, autonomy and control, cognition, daily living, emotional health, falls, general health, medications, overall survival, pain, participation in decision making, physical function, physical health, place of death, social role function, symptom burden, and time spent in hospital. The most frequently mentioned/used outcomes measurements tools were the Adult Social Care Outcomes Toolkit (ASCOT), EQ-5D, Gait Speed, Katz- ADL index, Patient Health Questionnaire (PHQ9), SF/RAND-36 and 4-Item Screening Zarit Burden Interview. CONCLUSIONS: Few studies have investigated the older people's opinion of what matters the most to them, which forms a knowledge-gap in the field. Future research should focus on providing older people a stronger voice in what they think matters the most to them.

Patient experiences and perspectives of health service access for carpal tunnel syndrome in Aotearoa New Zealand: a normalisation process theory-informed qualitative study.

BACKGROUND: Early access to care for carpal tunnel syndrome (CTS) can avoid higher rates of surgery and permanent harm yet is often delayed, particularly for populations more likely to underutilise care. OBJECTIVE: We sought to explore patient experiences and perspectives of health service access for CTS to inform an equity-focussed co-design of a health service for improving early care access. METHODS: In this Normalisation Process Theory (NPT)-informed qualitative study we conducted semistructured in-depth interviews with 19 adults with experience of CTS. Recruitment prioritised New Zealand Maori, Pasifika, low-income, and rural populations. Data were analysed using deductive then inductive thematic analysis. RESULTS: We identified five major themes: (1) the 'Significant Impact of CTS' of the sense-making and relational work to understand the condition, deciding when to get care, compelling clinicians to provide care, and garnering help from others; (2) 'Waiting and Paying for Care'- the enacting, relational, and appraising work to avoid long wait times unless paying privately, particularly where quality of care was low, employment relations poor, or injury compensation processes faltered; (3) circumstances of 'Occupation and CTS Onset' whereby the burden of proof to relate onset of CT symptoms to occupation created excessive relational and enacting work; (4) the 'Information Scarcity' of good information about CTS and the high relational and appraising work associated with using online resources; (5) 'Negotiating Telehealth Perspectives' where telehealth was valued if it meant earlier access for all despite the challenges it held for many. CONCLUSION: Quality, culturally and linguistically responsive information and communication from clinicians and health services will improve equitable early access to CTS care including realising the potential of telehealth modes of care. Policy changes that reduce individual burden of proof in injury compensation claims processes, enable time off work to attend health appointments, and increase public funding for surgical resources would improve early access to CTS care particularly for Maori and Pacific populations and those in small and rural workplaces. NPT is valuable for understanding where opportunities lie to reduce inequitable delays to accessing care including the impact of racism, particularly for populations more likely to underutilise care.

Recreating the village: the patient experience with a hybrid model of Group Perinatal Care (GPPC) in an academic family health team.

BACKGROUND: Group prenatal care (GPC) has been shown to have a positive impact on social support, patient knowledge and preparedness for birth. We developed an interprofessional hybrid model of care whereby the group perinatal care (GPPC) component was co-facilitated by midwives (MW) and family medicine residents (FMR) and alternating individual visits were provided by family physicians (FP's) within our academic family health team (FHT) In this qualitative study, we sought to explore the impact of this program and how it supports patients through pregnancy and the early newborn period. METHODS: Qualitative study that was conducted using semi-structured telephone interviews with 18 participants who had completed GPPC in the Mount Sinai Academic Family Health Team in Toronto, Canada and delivered between November 2016 and October 2018. Interviews were audio-recorded and transcribed verbatim. Thematic analysis was conducted by team members using grounded theory. RESULTS: Four over-arching themes emerged from the data: (i) Participants highly valued information they received from multiple trusted sources, (ii) Participants felt well cared for by the collaborative and coordinated interprofessional team, (iii) The design of GPPC enabled a shared experience, allowing for increased support of the pregnant person, and (iv) GPPC facilitated a supportive transition into the community which positively impacted participants' emotional well- being. CONCLUSIONS: The four constructs of social support (emotional, informational, instrumental and appraisal) were central to the value that participants found in GPPC. This support from the team of healthcare providers, peers and partners had a positive impact on participants' mental health and helped them face the challenges of their transition to parenthood.

Patient experience with bronchoscopy: topical versus monitored anesthesia.

BACKGROUND: This study aimed to compare patient experiences during bronchoscopy procedures using either topical anesthesia (TA) or monitored anesthesia care (MA). The goal was to identify circumstances where patients could achieve similar levels of tolerance and satisfaction using only TA, especially in resource-limited settings. METHODS: This study included consecutive patients who underwent bronchoscopy with either TA or MA. Data collected included demographics, indications for bronchoscopy, procedure time, and complications during the procedure. A quality assurance survey was administered to assess patient experience and satisfaction with both procedures. A pre-specified subgroup analysis was performed based on procedure invasiveness and time. RESULTS: This study enrolled 350 (TA 251; MA 99) patients, with an average age of 65 years. Main indications for bronchoscopy included tumor diagnosis (38%), esophageal cancer staging (18%), and pulmonary infection (17%). The average duration of the procedures was 20 min, with MA being associated with a significantly longer procedure time than TA (31 min vs. 16 min; P < 0.001). The overall satisfaction rating with bronchoscopy was significantly higher in the MA group (visual analogue scale, 8.9 vs. 8.2; P = 0.001). Subgroup analyses showed that when less invasive or shorter procedures were performed, TA patients reported tolerance and satisfaction levels comparable to MA patients. CONCLUSIONS: Bronchoscopy with MA offered patients a better experience and greater satisfaction; however, in settings with limited resources, TA alone may provide similar levels of patient tolerance and satisfaction during less invasive or shorter procedures.

Development and Validation of a Measure to Assess Patient Experiences With Video Care Encounters.

Importance: As video-based care expands in many clinical settings, assessing patient experiences with this care modality will help optimize health care quality, safety, and communication. Objective: To develop and assess the psychometric properties of the video visit user experience (VVUE) measure, a patient-reported measure of experiences with video-based technology. Design, Setting, and Participants: In this survey study, veterans completed a web-based, mail, or telephone survey about their use of Veterans Healthcare Administration (VHA) virtual care between September 2021 and January 2022. The survey was completed by patients who reported having a VHA video visit on their own device or a VHA-issued device and linked to VHA utilization data for the 6 months following the survey. Data analysis was performed from March 2022 to February 2023. Main Outcomes and Measures: The survey included 19 items about experiences with video visits that were rated using a 4-point Likert-type scale (strongly disagree to strongly agree). First, an exploratory factor analysis was conducted to determine the factor structure and parsimonious set of items, using the McDonald Omega test to assess internal consistency reliability. Then, a confirmatory factor analysis was conducted to test structural validity, and bivariate correlations between VVUE and VHA health care engagement were calculated to test concurrent validity. Finally, predictive validity was assessed using logistic regression to determine whether VVUE was associated with future VHA video visit use. Results: Among 1887 respondents included in the analyses, 83.2% (95% CI, 81.5%-84.8%) were male, 41.0% (95% CI, 38.8%-43.1%) were aged 65 years or older, and the majority had multiple chronic medical and mental health conditions. The exploratory factor analysis identified a 10-item single-factor VVUE measure (including questions about satisfaction, user-centeredness, technical quality, usefulness, and appropriateness), explaining 96% of the total variance, with acceptable internal consistency reliability (ω = 0.95). The confirmatory factor analysis results confirmed a single-factor solution (standardized root mean squared residual = 0.04). VVUE was positively associated with VHA health care engagement (ρ = 0.47; P < .001). Predictive validity models demonstrated that higher VVUE measure scores were associated with future use of video visits, where each 1-point increase on VVUE was associated with greater likelihood of having a video visit in subsequent 6 months (adjusted odds ratio, 1.04; 95% CI, 1.02-1.06). Conclusions and Relevance: The findings of this study of veterans using video visits suggest that a brief measure is valid to capture veterans' experiences receiving VHA virtual care.

Improving Communication with Family Members during a Global Pandemic: Effect on Patient Experience.

OBJECTIVES: Communication with patients and their families/caregivers to facilitate informed decision making is an integral part of patient/family-centered care. Due to the high coronavirus disease 2019 (COVID-19) infection rates and limited personal protective equipment, healthcare systems were forced to restrict patient visitors, limit patient-provider interactions, and implement other changes in treatment protocols that disturbed traditional communications and risked eroding patient/family-centered care and adversely affected patient satisfaction. This article focuses on changes in patient experience in two dedicated COVID-19 units of an academic medical center located in the US South as a result of the enhanced communication process implemented specifically during the COVID-19 pandemic. METHODS: This retrospective quality improvement project used data from Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) surveys, fielded between January 1, 2021 and August 31, 2021, to understand the role of a proactive communication initiative in patient satisfaction. RESULTS: Our results show that HCAHPS scores for hospital unit hospital unit 4 (HSP4) in all categories increased over time, with the greatest improvements seen in the responsiveness of staff and care transition; however, HCAHPS scores for hospital unit HSP3 remained stable, with a small increase in responsiveness of staff. CONCLUSIONS: Our findings suggest that communication is a critical factor in patient satisfaction, demonstrating the efficacy of a swift and innovative initiative to improve communication with family/caregivers, which may have been linked to better patient experiences. Developing communication strategies is crucial for enhancing patient satisfaction.

Improving Data Infrastructure for Person-Centered Outcomes Research on Intellectual and Developmental Disabilities.

Individuals with intellectual and developmental disabilities (IDD) continue to experience disparities in health and well-being despite improved provisions of person-centered care. Patient-centered outcomes research (PCOR) translates evidence into practice for meaningful outcomes. This piece describes findings from an environmental scan and stakeholder outreach to identify and prioritize opportunities to enhance IDD PCOR data infrastructure. These opportunities include developing a standardized research definition; advancing data standards for service systems; improving capture of IDD at point of care; developing standardized outcome measures; and encouraging Medicaid data use for IDD research. Within this piece, we discuss the implications of addressing data gaps for enhanced research. While the identified activities provide a path towards advancing IDD PCOR data infrastructure, collaborative efforts between government, researchers, and others are paramount.

Why surveys are 'very hard': exploring challenges and insights for collection of authentic patient experience information with speakers of Australian First Nations languages.

INTRODUCTION: Health services collect patient experience data to monitor, evaluate and improve services and subsequently health outcomes. Obtaining authentic patient experience information to inform improvements relies on the quality of data collection processes and the responsiveness of these processes to the cultural and linguistic needs of diverse populations. This study explores the challenges and considerations in collecting authentic patient experience information through survey methods with Australians who primarily speak First Nations languages. METHODS: First Nations language experts, interpreters, health staff and researchers with expertise in intercultural communication engaged in an iterative process of critical review of two survey tools using qualitative methods. These included a collaborative process of repeated translation and back translation of survey items and collaborative analysis of video-recorded trial administration of surveys with languages experts (who were also receiving dialysis treatment) and survey administrators. All research activities were audio- or video-recorded, and data from all sources were translated, transcribed and inductively analysed to identify key elements influencing acceptability and relevance of both survey process and items as well as translatability. RESULTS: Serious challenges in achieving equivalence of meaning between English and translated versions of survey items were pervasive. Translatability of original survey items was extensively compromised by the use of metaphors specific to the cultural context within which surveys were developed, English words that are familiar but used with different meaning, English terms with no equivalent in First Nations languages and grammatical discordance between languages. Discordance between survey methods and First Nations cultural protocols and preferences for seeking and sharing information was also important: the lack of opportunity to share the 'full story', discomfort with direct questions and communication protocols that preclude negative or critical responses constrained the authenticity of the information obtained through survey methods. These limitations have serious implications for the quality of information collected and result in frustration and distress for those engaging with the survey. CONCLUSION: Profound implications for the acceptability of a survey tool as well as data quality arise from differences between First Nations cultural and communication contexts and the cultural context within which survey methods have evolved. When data collection processes are not linguistically and culturally congruent there is a risk that patient experience data are inaccurate, miss what is important to First Nations patients and have limited utility for informing relevant healthcare improvement. Engagement of First Nations cultural and language experts is essential in all stages of development, implementation and evaluation of culturally safe and effective approaches to support speakers of First Nations languages to share their experiences of health care and influence change.

Patient experiences of resection versus responsive neurostimulation for drug-resistant epilepsy.

This study explored illness experiences and decision-making among patients with epilepsy who underwent two different types of surgical interventions: resection versus implantation of the NeuroPace Responsive Neurostimulation System (RNS). We recruited 31 participants from a level four epilepsy center in an academic medical institution. We observed 22 patient clinic visits (resection: n = 10, RNS: n = 12) and conducted 18 in-depth patient interviews (resection: n = seven, RNS: n = 11); most visits and interviews included patient caregivers. Using an applied ethnographic approach, we identified three major themes in the experiences of resection versus RNS patients. First, for patients in both cohorts, the therapeutic journey was circuitous in ways that defied standardized first-, second-, and third- line of care models. Second, in conceptualizing risk, resection patients emphasized the permanent loss of "taking out" brain tissue whereas RNS patients highlighted the reversibility of "putting in" a device. Lastly, in considering benefit, resection patients perceived their surgery as potentially curative while RNS patients understood implantation as primarily palliative with possible additional diagnostic benefit from chronic electrocorticography. Insight into the perspectives of patients and caregivers may help identify key topics for counseling and exploration by clinicians.