Applications of Artificial Intelligence in Health Care Delivery.

Health care costs now comprise nearly one-fifth of the United States' gross domestic product, with the last 25 years marked by rising administrative costs, a lack of labor productivity growth, and rising patient and physician dissatisfaction. Policy experts have responded with a series of reforms that have - ironically - increased patient and physician administrative burden with little meaningful effect on cost and quality. Artificial intelligence (AI), a topic of great consternation, can serve as the "wheat thresher" for health care delivery, empowering and freeing both patients and physicians by decreasing administrative burden and improving labor productivity. In this Viewpoint, we discuss three principal areas where AI poses an unprecedented opportunity to reduce cost, improve care, and markedly enhance the patient and physician experience: (1) automation of administrative process, (2) augmentation of clinical practice, and (3) automation of elements of clinical practice.

Improving Health Care Quality Measurement to Combat Clinician Burnout.

This Viewpoint explores a critical source of administrative burden and physician frustration­the explosive growth of health care quality metrics­and how improving these metrics could combat clinician burnout.

Using phenotypic data from the Electronic Health Record (EHR) to predict discharge.

BACKGROUND: Timely discharge to post-acute care (PAC) settings, such as skilled nursing facilities, requires early identification of eligible patients. We sought to develop and internally validate a model which predicts a patient's likelihood of requiring PAC based on information obtained in the first 24 h of hospitalization. METHODS: This was a retrospective observational cohort study. We collected clinical data and commonly used nursing assessments from the electronic health record (EHR) for all adult inpatient admissions at our academic tertiary care center from September 1, 2017 to August 1, 2018. We performed a multivariable logistic regression to develop the model from the derivation cohort of the available records. We then evaluated the capability of the model to predict discharge destination on an internal validation cohort. RESULTS: Age (adjusted odds ratio [AOR], 1.04 [per year]; 95% Confidence Interval [CI], 1.03 to 1.04), admission to the intensive care unit (AOR, 1.51; 95% CI, 1.27 to 1.79), admission from the emergency department (AOR, 1.53; 95% CI, 1.31 to 1.78), more home medication prescriptions (AOR, 1.06 [per medication count increase]; 95% CI 1.05 to 1.07), and higher Morse fall risk scores at admission (AOR, 1.03 [per unit increase]; 95% CI 1.02 to 1.03) were independently associated with higher likelihood of being discharged to PAC facility. The c-statistic of the model derived from the primary analysis was 0.875, and the model predicted the correct discharge destination in 81.2% of the validation cases. CONCLUSIONS: A model that utilizes baseline clinical factors and risk assessments has excellent model performance in predicting discharge to a PAC facility.

What Should "Shopping" Look Like in Actual Practice?

In health care, lack of transparency about the cost of health care services to patients during clinical encounters has contributed to increased costs and high out-of-pocket expenses. Federal policy has responded to the need for more transparency and spurred discussion about ethics and the clinician's role in being transparent with patients at the point of service. This article investigates and encourages state, private market, and federal policy efforts to address what health care costs patients. This article also applies the ethical framework of principlism to cases and considers what a "shoppable service" model would demand of clinicians in practice.

U.S. medical school admissions and enrollment practices: status of LGBTQ inclusivity.

CONTEXT: The failure to collect information on lesbian, gay, bisexual, transgender, and queer (LGBTQ) identity in healthcare and medical education is a part of a systemic problem that limits academic medical institutions' ability to address LGBTQ health disparities. OBJECTIVES: To determine whether accurate sexual and gender minority (SGM) demographic data is being consistently collected for all US medical schools during admissions and enrollment, and whether differences exist between collection practices at osteopathic and allopathic schools. METHODS: Secure, confidential electronic were sent via email in July 2019 to 180 osteopathic (n=42) and allopathic (n=138) medical schools identified through the American Association of Colleges of Osteopathic Medicine Student Guide to Osteopathic Medical Colleges database and the American Association of Medical Colleges Medical School Admissions Requirements database. The nine question survey remained open through October 2019 and queried for; (1) the ability of students to self report SGM status during admissions and enrollment; and (2) availability of SGM specific resources and support services for students. Chi square analysis and the test for equality of proportions were performed. RESULTS: Seventy five of 180 (41.7%) programs responded to the survey; 74 provided at least partial data. Of the 75 respondent schools, 55 (73.3%) allowed applicants to self report a gender identity other than male or female, with 49 (87.5%) of those being allopathic schools compared with 6 (31.6%) osteopathic schools. Similarly, 15 (20.0%) allowed applicants to report sexual orientation, with 14 (25.5%) of those being allopathic schools compared with one (5.3%) osteopathic school. Fifty four of 74 (73.0%) programs allowed matriculants to self report a gender identity other than male or female; 11 of 74 (14.7%) allowed matriculants to report sexual orientation. CONCLUSIONS: Demographics collection practices among American medical education programs that responded to our survey indicated that they undervalued sexual orientation and gender identity, with osteopathic programs being less likely than allopathic programs to report inclusive best practices in several areas. American medical education programs, and their supervising bodies, must update their practices with respect to the collection of sexual orientation and gender identity demographics as part of a holistic effort to address SGM health disparities.

A Call for LGBTQ Content in Graduate Medical Education Program Requirements.

A well-developed body of literature demonstrates that lesbian, gay, bisexual, transgender, and queer (LGBTQ) individuals experience poorer health outcomes and report worse health care experiences than straight/cisgender individuals. Many reforms since 2010 have addressed the LGBTQ-related education of future health care professionals at the undergraduate medical education (UME) level; however, reforms at the graduate medical education (GME) level are lagging, and new literature suggests that didactic education at the UME level is not enough to prepare future physicians to properly and compassionately care for LGBTQ patients. Recently, the Accreditation Council for Graduate Medical Education (ACGME) implemented a major revision of its Common Program Requirements that requires residents to demonstrate, as a competence, respect and responsiveness to diverse populations. Given these revisions and the ongoing failure of many GME training programs to adequately prepare future physicians to care for LGBTQ patients, the authors argue that now is the time for the ACGME to develop and implement LGBTQ health-related residency requirements. In addition, the authors outline a path by which the academic medical community may develop and implement these requirements.

Changes in health insurance coverage, access to care, and health services utilization by sexual minority status in the United States, 2013-2018.

OBJECTIVE: To examine the changes in health insurance coverage, access to care, and health services utilization among nonelderly sexual minority and heterosexual adults between pooled years 2013-2014 and 2017-2018. DATA SOURCES: Data on 3223 sexual minorities (lesbians, gay men, bisexual individuals, and other nonheterosexual populations) and 86 181 heterosexuals aged 18-64 years were obtained from the 2013, 2014, 2017, and 2018 National Health Interview Surveys. STUDY DESIGN: Unadjusted and regression-adjusted estimates compared changes in health insurance status, access to care, and health services utilization for nonelderly adults by sexual minority status. Regression-adjusted changes were obtained from logistic regression models controlling for demographic and socioeconomic characteristics. PRINCIPAL FINDINGS: Uninsurance declined for both sexual minority adults (5 percentage points, P < .05) and heterosexual adults (2.5 percentage points, P < .001) between 2013-2014 and 2017-2018. Reductions in uninsurance for sexual minority and heterosexual adults were associated with increases in Medicaid coverage. Sexual minority and heterosexual adults were also less likely to report unmet medical care in 2017-2018 compared with 2013-2014. Low-income adults (regardless of sexual minority status) experienced relatively large increases in Medicaid coverage and substantial improvements in access to care over the study period. The gains in coverage and access to care across the study period were generally similar for heterosexual and sexual minority adults. CONCLUSIONS: Sexual minority and heterosexual adults have experienced improvements in health insurance coverage and access to care in recent years. Ongoing health equity research and public health initiatives should continue to monitor health care access and the potential benefits of recent health insurance expansions by sexual orientation and sexual minority status when possible.

Understanding the Information Needs and Context of Trauma Handoffs to Design Automated Sensing Clinical Documentation Technologies: Qualitative Mixed-Method Study of Military and Civilian Cases.

BACKGROUND: Current methods of communication between the point of injury and receiving medical facilities rely on verbal communication, supported by brief notes and the memory of the field medic. This communication can be made more complete and reliable with technologies that automatically document the actions of field medics. However, designing state-of-the-art technology for military field personnel and civilian first responders is challenging due to the barriers researchers face in accessing the environment and understanding situated actions and cognitive models employed in the field. OBJECTIVE: To identify design insights for an automated sensing clinical documentation (ASCD) system, we sought to understand what information is transferred in trauma cases between prehospital and hospital personnel, and what contextual factors influence the collection, management, and handover of information in trauma cases, in both military and civilian cases. METHODS: Using a multi-method approach including video review and focus groups, we developed an understanding of the information needs of trauma handoffs and the context of field documentation to inform the design of an automated sensing documentation system that uses wearables, cameras, and environmental sensors to passively infer clinical activity and automatically produce documentation. RESULTS: Comparing military and civilian trauma documentation and handoff, we found similarities in the types of data collected and the prioritization of information. We found that military environments involved many more contextual factors that have implications for design, such as the physical environment (eg, heat, lack of lighting, lack of power) and the potential for active combat and triage, creating additional complexity. CONCLUSIONS: An ineffectiveness of communication is evident in both the civilian and military worlds. We used multiple methods of inquiry to study the information needs of trauma care and handoff, and the context of medical work in the field. Our findings informed the design and evaluation of an automated documentation tool. The data illustrated the need for more accurate recordkeeping, specifically temporal aspects, during transportation, and characterized the environment in which field testing of the developed tool will take place. The employment of a systems perspective in this project produced design insights that our team would not have identified otherwise. These insights created exciting and interesting challenges for the technical team to resolve.

Development and Implementation of a Nonclinical Professional Development Rotation.

Residency programs often struggle with strategies to formally teach leadership and communications skills. To provide a catalyst for professional development, Vanderbilt University Medical Center implemented a curriculum aimed at addressing this gap. Starting in 2014, the authors implemented a 2-week professional development rotation for first-year anesthesiology residents. Experts provided a series of didactic and experiential sessions focused on various professional development topics. Outcomes were assessed using pre- and postrotation surveys. Sixty-nine residents completed the rotation over a 4-year period, and 82% (54 of 66) strongly agreed that nonclinical professional development should be a component of training.

Patient satisfaction survey scores are not an appropriate metric to differentiate performance among anesthesiologists.

STUDY OBJECTIVE: With the focus of patient-centered care in healthcare organizations, patient satisfaction plays an increasingly important role in healthcare quality measurement. We sought to determine whether an automated patient satisfaction survey could be effectively used to identify outlying anesthesiologists. DESIGN: Retrospective Observational Study. SETTING: Vanderbilt University Medical Center (VUMC). MEASUREMENTS: Patient satisfaction data were obtained between October 24, 2016 and November 1, 2017. A multivariable ordered probit regression was conducted to evaluate the relationship between the mean scores of responses to Likert-scale questions on SurveyVitals' Anesthesia Patient Satisfaction Questionnaire 2. Fixed effects included demographics, clinical variables, providers and surgeons. Hypothesis tests to compare each individual anesthesiologist with the median-performing anesthesiologist were conducted. MAIN RESULTS: We analyzed 10,528 surveys, with a 49.5% overall response rate. Younger patient (odds ratio (OR) 1.011 [per year of age]; 95% confidence interval (CI) 1.008 to 1.014; p < 0.001), regional anesthesia (versus general anesthesia) (OR 1.695; 95% CI 1.186 to 2.422; p = 0.004) and daytime surgery (versus nighttime surgery) (OR 1.795; 95% CI 1.091 to 2.959; p = 0.035) were associated with higher satisfaction scores. Compared with the median-ranked anesthesiologist, we found the adjusted odds ratio for an increase in satisfaction score ranged from 0.346 (95% CI 0.158 to 0.762) to 1.649 (95% CI 0.687 to 3.956) for the lowest and highest scoring providers, respectively. Only 10.10% of anesthesiologists at our institution had an odds ratio for satisfaction with a 95% CI not inclusive of 1. CONCLUSIONS: Patient satisfaction is impacted by multiple factors. There was very little information in patient satisfaction scores to discriminate the providers, after adjusting for confounding. While patient satisfaction scores may facilitate identification of extreme outliers among anesthesiologists, there is no evidence that this metric is useful for the routine evaluation of individual provider performance.

Price Transparency: Empowering Patient Choice and Promoting Provider Competition.

In light of recent health policy efforts to promote price transparency, this perspective reviews the challenges and benefits of price transparency. These price transparency efforts include the recent executive order and associated rulemaking directing providers to disclose negotiated and out-of-pocket costs for "shoppable" healthcare services. First, we explore the previous efforts of states and health plans targeted at price transparency, reviewing lessons for future implementation. Second, we address the value of price transparency in light of various policy concerns and objections. Finally, we jointly hypothesize potential effects of and opportunities presented by price transparency for patients, physicians, and other healthcare industry stakeholders.

Prospective Investigation of the Operating Room Time-Out Process.

BACKGROUND: Although the surgical pause or time-out is a required part of most hospitals' standard operating procedures, little is known about the quality of execution of the time-out in routine clinical practice. An interactive electronic time-out was implemented to increase surgical team compliance with the time-out procedure and to improve communication among team members in the operating room. We sought to identify nonroutine events that occur during the time-out procedure in the operating room, including distractions and interruptions, deviations from protocol, and the problem-solving strategies used by operating room team members to mitigate them. METHODS: Direct observations of surgical time-outs were performed on 166 nonemergent surgeries in 2016. For each time-out, the observers recorded compliance with each step, any nonroutine events that may have occurred, and whether any operating room team members were distracted. RESULTS: The time-out procedure was performed before the first incision in 100% of cases. An announcement was made to indicate the start of the time-out procedure in 163 of 166 observed surgeries. Most observed time-outs were completed in <1 minute. Most time-outs were completed without interruption (92.8%). The most common reason for an interruption was to verify patient information. Ten time-out procedures were stopped due to a safety concern. At least 1 member of the operating room team was actively distracted in 10.2% of the time-out procedures observed. CONCLUSIONS: Compliance with preincision time-outs is high at our institution, and nonroutine events are a rare occurrence. It is common for ≥1 member of the operating room team to be actively distracted during time-out procedures, even though most time-outs are completed in under 1 minute. Despite distractions, there were no wrong-site or wrong-person surgeries reported at our hospital during the study period. We conclude that the simple act of performing a preprocedure checklist may be completed quickly, but that distractions are common.

Implementation of an Enhanced Recovery Protocol (ERP) is associated with an increase in the perioperative use of non-opioid multimodal analgesia for non-ERP patients.

STUDY OBJECTIVE: Incorporation of multimodal, non-opioid analgesic medications into patient care pathways has become a common theme of enhanced recovery pathways (ERPs), which have been shown to improve patient outcomes such as complication rates and length of stay. With surgical care episodes, patients also incur a significant risk of persistent postoperative opioid use, whether the surgery is classified as major or minor surgery. One method that has been shown to reduce perioperative opioid administration is a structured non-opioid multimodal analgesic strategy, widely utilized in ERPs. Despite well-defined benefits, the time to translate evidence-based approaches into clinical practice can be prolonged. This study examines the effect of implementation of an Enhanced Recovery Protocol (ERP) on the adoption of intraoperative multimodal analgesia outside of the auspices of an ERP care pathway, describing factors influencing the clinical implementation of non-opioid multimodal analgesia (NOMA) in routine practice. DESIGN: Retrospective cohort analysis. SETTING: We identified all surgical cases between January 2013 and December 2016 at Vanderbilt University Medical Center (VUMC). INTERVENTIONS: None. MEASUREMENTS: Using both segmented and logistic regression approaches, we compared non-ERP surgical cases before and after the initial ERP education and implementation in April 2014. Outcomes included provider, patient, and procedural factors associated with utilization of non-opioid multimodal analgesia (NOMA) in the immediate perioperative period. MAIN RESULTS: We studied 73,560 non-ERP cases. Cases utilizing any element of NOMA increased from 17.06% to 35.21% (X2 = 2358, df = 1, p < 0.01) before and after the initial ERP pathway implementation. Patient factors influencing this increased adoption of multimodal analgesia included lower American Society of Anesthesiologists Physical Status Class, younger age, and Caucasian race. Cases with in-room providers who were residents or trainees (as opposed to nurse anesthetists) or providers who had a greater number of prior ERP pathway cases were more likely to use multimodal. Procedure-specific factors favoring multimodal included use of laparoscopy. The gynecologic, neurosurgical, and orthopedic cases were more likely to utilize multimodal analgesics. CONCLUSIONS: From 2013 to 2016, NOMA usage in non-ERP patients increased significantly and in association with departmental education and concomitant implementation of an ERP pathway. Factors associated with increased uptake of multimodal analgesia included the presence of trainees, providers with a higher number of previous ERP pathway cases, patients who were younger, healthier, female, Caucasian race, and having specific types of surgery.