Stephanie Harman

Improving palliative care with deep learning

Improving the quality of end-of-life care for hospitalized patients is a priority for healthcare organizations. Studies have shown that physicians tend to over-estimate prognoses, which in combination with treatment inertia results in a mismatch between patients wishes and actual care at the end of life. We describe a method to address this problem using Deep Learning and Electronic Health Record (EHR) data, which is currently being piloted, with Institutional Review Board approval, at an academic medical center. The EHR data of admitted patients are automatically evaluated by an algorithm, which brings patients who are likely to benefit from palliative care services to the attention of the Palliative Care team. The algorithm is a Deep Neural Network trained on the EHR data from previous years, to predict all-cause 3–12 month mortality of patients as a proxy for patients that could benefit from palliative care. Our predictions enable the Palliative Care team to take a proactive approach in reaching out to such patients, rather than relying on referrals from treating physicians, or conduct time consuming chart reviews of all patients. We also present a novel interpretation technique which we use to provide explanations of the models predictions.

To Stent or Not to Stent: An Evidence-Based Approach to Palliative Procedures at the End of Life

Patients near the end of life often undergo invasive procedures, such as biliary stenting for obstructive jaundice, with the intent of relieving symptoms. We describe a case in which the medical team and a patient and family are considering a second palliative biliary stent despite the patients limited life expectancy. We review available evidence to inform the decision, focusing on the specific question of whether the benefits of palliative biliary stents in patients with advanced cancer outweigh the risks. We then apply the evidence to the issue of how the primary and/or palliative care team and the interventionist communicate with patients and their families about the risks and benefits of palliative procedures. Review of the evidence found several prospective case series without control groups that measured patient-centered outcomes. Studies had high attrition rates, results for improvements in symptoms and quality of life were mixed, and rates of complications and short-term mortality were high. In conclusion, the limited evidence does not support that the benefits of palliative biliary stents in this population outweigh the risks. We propose that primary care teams consider and discuss the larger picture of the goals of care with patients and families when considering offering these procedures, as well as benefits and potential harms, and consider involving palliative care services early, before consultation with an interventionist.

The State of Medical Student Performance Evaluations: Improved Transparency or Continued Obfuscation?

Purpose The medical student performance evaluation (MSPE), a letter summarizing academic performance, is included in each medical student’s residency application. The extent to which medical schools follow Association of American Medical Colleges (AAMC) recommendations for comparative and transparent data is not known. This study’s purpose was to describe the content, interpretability, and transparency of MSPEs. Method This cross-sectional study examined one randomly selected MSPE from every Liaison Committee on Medical Education–accredited U.S. medical school from which at least one student applied to the Stanford University internal medical residency program during the 2013–2014 application cycle. The authors described the number, distribution, and range of key words and clerkship grades used in the MSPEs and the proportions of schools with missing or incomplete data. Results The sample included MSPEs from 117 (89%) of 131 medical schools. Sixty schools (51%) provided complete information about clerkship grade and key word distributions. Ninety-six (82%) provided comparative data for clerkship grades, and 71 (61%) provided complete key word data. Key words describing overall performance were extremely heterogeneous, with a total of 72 used and great variation in the assignment of the top designation (median: 24% of students; range: 1%–60%). There was also great variation in the proportion of students awarded the top internal medicine clerkship grade (median: 29%; range: 2%–90%). Conclusions The MSPE is a critical component of residency applications, yet data contained within MSPEs are incomplete and variable. Approximately half of U.S. medical schools do not follow AAMC guidelines for MSPEs.

Patients Who Lack Capacity and Lack Surrogates: Can They Enroll in Hospice?

Patients who lack capacity and lack surrogates are among the most vulnerable patients we care for in palliative care. In the case we present here, we have considered how to make end-of-life decisions for a patient who lacks both capacity and surrogates, who has a terminal illness, and who is not a candidate for disease-modifying treatments. We first define and characterize this population of patients through a review of the literature and then explore some decision-making quandaries that are encountered at the end of life. Finally, we make recommendations on how best to proceed with decision making for this vulnerable population.

A resident-created hospitalist curriculum for internal medicine housestaff.

The growth of hospital medicine has led to new challenges, and recent graduates may feel unprepared to meet the expanding clinical duties expected of hospitalists. At our institution, we created a resident-inspired hospitalist curriculum to address the training needs for the next generation of hospitalists. Our program provided 3 tiers of training: (1) clinical excellence through improved training in underemphasized areas of hospital medicine, (2) academic development through required research, quality improvement, and medical student teaching, and (3) career mentorship. In this article, we describe the genesis of our program, our final product, and the challenges of creating a curriculum while being internal medicine residents. Journal of Hospital Medicine 2016;11:646-649.

Early Experience With the California End of Life Option Act: Balancing Institutional Participation and Physician Conscientious Objection

Effective since June 9, 2016, the California End of Life Option Act allows physicians to provide prescriptions for a lethal dose of medication to patients with a terminal illness, decision-making capacity, and a prognosis of less than 6 months to live. The patient can ingest the medication and end his or her life.1 The act outlines specific processes for physicians, patients, and health care systems to ensure that patients to whom the law applies receive all needed information regarding their options for end-of-life care and are neither impaired nor coerced into making such a profound decision. Although similar to laws in Oregon and Washington, the California law will sunset in 10 years if the state takes no further action. In the 2016 elections, Colorado became the sixth state to pass such legislation. At Stanford, where we work, as of April 1, 2017, 13 patients have been prescribed lethal medications under the End of Life Option Act and 6 are known to have died by ingesting these medications since the law’s passage. Physicians at Stanford have several options for this prescription, including secobarbital by itself, a compounded liquid mixture of phenobarbital/morphine/ chloral hydrate, or the ingestion of diazepam, propranolol, morphine, and digoxin. Stanford Health Care is participating in the End of Life Option Act. Participation by individual physicians is voluntary, however; they may opt out of prescribing or consulting for physician-assisted death. Institutions may also choose to opt out of the act, and not allow any of their employees to serve as prescribers, as the Department of Veteran’s Affairs and Dignity Health, a Catholic health system, have done. In Colorado, Catholic health systems have opted out, and nearly one-third of the hospitals will not offer physician-assisted death, creating a major challenge to access.2 Some hospice organizations in California have opted out as well, meaning that their medical directors cannot write a lethal prescription as hospice employees nor be present with the patient at the time of death (from the lethal prescription). Although physicians should clearly not be forced to write lethal prescriptions, such conscientious objection affects access to physician-assisted death. There is a potential for injustice because similar patients with similar diseases and life expectancies may find wide variation in the availability of prescriptions for lethal medications. For a participating institution, such as Stanford, the potential for injustice raised the question of the institutional obligation to patients, should their primary physician opt out. In general, our policy about conscientious objection allows a physician to opt out of directly providing a service or medication that he or she objects to, but requires indirect involvement and an institutional commitment to finding another physician who is willing to provide the care. For example, if a physician in the emergency department objects to providing Plan B to a rape victim, the hospital must make every effort to respect the physician’s ethical stance, without affecting the patient’s care. The situation is different when physicians exercise clinical judgment, such as when a surgeon decides that a patient is not a surgical candidate. In this situation, there is no institutional obligation to find another surgeon for the patient. A major operational challenge has been finding a physician willing to prescribe for another physician’s patient and the process for that physician to write the prescription. The number of physicians willing to prescribe for others’ patients has been small. The social workers and ethics consultants charged with identifying alternative prescribers found that they were burdened by taking on this responsibility, particularly for a patient from another division or department. Physicians cited concerns for being known as prescribers for physicianassisted death. Moreover, after agreeing to see the patient, they often experienced an undue urgency given the patient’s prognosis and the fact that their primary physician had opted out. After reviewing the first few cases, we addressed this urgency in subsequent cases. An alternate physician should have sufficient time and clinic visits to establish a relationship with the patient and feel confident that he or she can fulfill the necessary components of the physician-assisted death process. Although the law describes the required process elements in detail, it does not set a maximum or minimum required number of visits.1 Our experiences have also raised questions about what constitutes conscientious objection and when the institution will take on an obligation that is not required by the law.3-6 A physician may place additional requirements on the process beyond the elements in the law, such as a longer waiting period between visits or insisting on psychiatric evaluation for all patients requesting physician-assisted death.7 These additional requirements do not constitute a formal objection. Nonetheless, they may preclude some patients from receiving a prescription. Experience in Oregon suggests that reVIEWPOINT

Psychiatric and Palliative Care in the Intensive Care Unit

Palliative care is specialized medical care focused on patients with serious illness and their families. In the intensive care unit (ICU), palliative care encompasses core skills to support patients and their families throughout their ICU course and post-ICU stays. Psychiatric symptoms are common among patients approaching the end of life and require particular attention in the setting of sedating medications, typically used when patients require ventilators and other life-sustaining treatments. For patients with preexisting severe mental illness who have a concurrent serious medical illness, a palliative psychiatric approach can address complex symptom management and support ethical and value-based shared decision making.

Shared Decision-Making During Inpatient Rounds: Opportunities for Improvement in Patient Engagement and Communication

BACKGROUND Shared decision-making (SDM) improves patient engagement and may improve outpatient health outcomes. Little is known about inpatient SDM. OBJECTIVE To assess overall quality, provider behaviors, and contextual predictors of SDM during inpatient rounds on medicine and pediatrics hospitalist services. DESIGN A 12-week, cross-sectional, single-blinded observational study of team SDM behaviors during rounds, followed by semistructured patient interviews. SETTING Two large quaternary care academic medical centers. PARTICIPANTS Thirty-five inpatient teams (18 medicine, 17 pediatrics) and 254 unique patient encounters (117 medicine, 137 pediatrics). INTERVENTION Observational study. MEASUREMENTS We used a 9-item Rochester Participatory Decision-Making Scale (RPAD) measured team-level SDM behaviors. Same-day interviews using a modified RPAD assessed patient perceptions of SDM. RESULTS Characteristics associated with increased SDM in the multivariate analysis included the following: service, patient gender, timing of rounds during patient’s hospital stay, and amount of time rounding per patient (P < .05). The most frequently observed behaviors across all services included explaining the clinical issue and matching medical language to the patient’s level of understanding. The least frequently observed behaviors included checking understanding of the patient’s point of view, examining barriers to follow-through, and asking if the patient has any questions. Patients and guardians had substantially higher ratings for SDM quality compared to peer observers (7.2 vs 4.4 out of 9). CONCLUSIONS Important opportunities exist to improve inpatient SDM. Team size, number of learners, patient census, and type of decision being made did not affect SDM, suggesting that even large, busy services can perform SDM if properly trained.

A high value care curriculum for interns: a description of curricular design, implementation and housestaff feedback

Purpose Most residency programmes do not have a formal high value care curriculum. Our goal was to design and implement a multidisciplinary high value care curriculum specifically targeted at interns. Design Our curriculum was designed with multidisciplinary input from attendings, fellows and residents at Stanford. Curricular topics were inspired by the American Board of Internal Medicine’s Choosing Wisely campaign, Alliance for Academic Internal Medicine, American College of Physicians and Society of Hospital Medicine. Our topics were as follows: introduction to value-based care; telemetry utilisation; lab ordering; optimal approach to thrombophilia work-ups and fresh frozen plasma use; optimal approach to palliative care referrals; antibiotic stewardship; and optimal approach to imaging for low back pain. Our curriculum was implemented at the Stanford Internal Medicine residency programme over the course of two academic years (2014 and 2015), during which 100 interns participated in our high value care curriculum. After each high value care session, interns were offered the opportunity to complete surveys regarding feedback on the curriculum, self-reported improvements in knowledge, skills and attitudinal module objectives, and quiz-based knowledge assessments. Results The overall survey response rate was 67.1%. Overall, the material was rated as highly useful on a 5-point Likert scale (mean 4.4, SD 0.6). On average, interns reported a significant improvement in their self-rated knowledge, skills and attitudes after the six seminars (mean improvement 1.6 points, SD 0.4 (95% CI 1.5 to 1.7), p<0.001). Conclusions We successfully implemented a novel high value care curriculum that specifically targets intern physicians.

TRIAD VIII: Nationwide Multicenter Evaluation to Determine Whether Patient Video Testimonials Can Safely Help Ensure Appropriate Critical Versus End-of-Life Care.

Objective End-of-life interventions should be predicated on consensus understanding of patient wishes. Written documents are not always understood; adding a video testimonial/message (VM) might improve clarity. Goals of this study were to (1) determine baseline rates of consensus in assigning code status and resuscitation decisions in critically ill scenarios and (2) determine whether adding a VM increases consensus. Methods We randomly assigned 2 web-based survey links to 1366 faculty and resident physicians at institutions with graduate medical education programs in emergency medicine, family practice, and internal medicine. Each survey asked for code status interpretation of stand-alone Physician Orders for Life-Sustaining Treatment (POLST) and living will (LW) documents in 9 scenarios. Respondents assigned code status and resuscitation decisions to each scenario. For 1 of 2 surveys, a VM was included to help clarify patient wishes. Results Response rate was 54%, and most were male emergency physicians who lacked formal advanced planning document interpretation training. Consensus was not achievable for stand-alone POLST or LW documents (68%–78% noted “DNR”). Two of 9 scenarios attained consensus for code status (97%–98% responses) and treatment decisions (96%–99%). Adding a VM significantly changed code status responses by 9% to 62% (P ⩽ 0.026) in 7 of 9 scenarios with 4 achieving consensus. Resuscitation responses changed by 7% to 57% (P ⩽ 0.005) with 4 of 9 achieving consensus with VMs. Conclusions For most scenarios, consensus was not attained for code status and resuscitation decisions with stand-alone LW and POLST documents. Adding VMs produced significant impacts toward achieving interpretive consensus.