Health Care Cost Reductions with Machine Learning-Directed Evaluations during Radiation Therapy - An Economic Analysis of a Randomized Controlled Study.

BACKGROUND: Machine learning (ML) may cost-effectively direct health care by identifying patients most likely to benefit from preventative interventions to avoid negative and expensive outcomes. System for High-Intensity Evaluation During Radiation Therapy (SHIELD-RT; NCT04277650) was a single-institution, randomized controlled study in which electronic health record-based ML accurately identified patients at high risk for acute care (emergency visit or hospitalization) during radiotherapy (RT) and targeted them for supplemental clinical evaluations. This ML-directed intervention resulted in decreased acute care utilization. Given the limited prospective data showing the ability of ML to direct interventions cost-efficiently, an economic analysis was performed. METHODS: A post hoc economic analysis was conducted of SHIELD-RT that included RT courses from January 7, 2019, to June 30, 2019. ML-identified high-risk courses (≥10% risk of acute care during RT) were randomized to receive standard of care weekly clinical evaluations with ad hoc supplemental evaluations per clinician discretion versus mandatory twice-weekly evaluations. The primary outcome was difference in mean total medical costs during and 15 days after RT. Acute care costs were obtained via institutional cost accounting. Physician and intervention costs were estimated via Medicare and Medicaid data. Negative binomial regression was used to estimate cost outcomes after adjustment for patient and disease factors. RESULTS: A total of 311 high-risk RT courses among 305 patients were randomized to the standard (n=157) or the intervention (n=154) group. Unadjusted mean intervention group supplemental visit costs were $155 per course (95% confidence interval, $142 to $168). The intervention group had fewer acute care visits per course (standard, 0.47; intervention, 0.31; P=0.04). Total mean adjusted costs were $3110 per course for the standard group and $1494 for the intervention group (difference in means, $1616 [95% confidence interval, $1450 to $1783]; P=0.03). CONCLUSIONS: In this economic analysis of a randomized controlled, health care ML study, mandatory supplemental evaluations for ML-identified high-risk patients were associated with both reduced total medical costs and improved clinical outcomes. Further study is needed to determine whether economic results are generalizable. (Funded in part by The Duke Endowment, The Conquer Cancer Foundation, the Duke Department of Radiation Oncology, and the National Cancer Institute of the National Institutes of Health [R01CA277782]; ClinicalTrials.gov number, NCT04277650.).

A novel multi-modality imaging phantom for validating interstitial needle guidance for high dose rate gynecological brachytherapy.

PURPOSE: To design, manufacture, and validate a female pelvic phantom for multi-modality imaging (CT, MRI, US) to benchmark a commercial needle tracking system with application in HDR gynecological (GYN) interstitial procedures. MATERIALS AND METHODS: A GYN needle-tracking phantom was designed using CAD software to model an average uterus from a previous patient study, a vaginal canal from speculum dimensions, and a rectum to accommodate a transrectal ultrasound (TRUS) probe. A target volume (CTVHR ) was designed as an extension from the cervix-uterus complex. Negative space molds were created from modeled anatomy and 3D printed. Silicone was used to cast the anatomy molds. A 3D printed box was constructed to house the manufactured anatomy for structural integrity and to accommodate the insertion of a speculum, tandem, needles, and TRUS probe. The phantom was CT-imaged to identify potential imperfections that might impact US visualization. Free-hand TRUS was used to guide interstitial needles into the phantom. The commercial tracking system was used to generate a 3D US volume. After insertion, the phantom was imaged with CT and MR and the uterus and CTVHR dimensions were verified against the CAD model. RESULTS/CONCLUSIONS: The manufactured phantom allows for accurate visualization with multiple imaging modalities and is conducive to applicator and needle insertion. The phantom dimensions from the CAD model were verified with those from each imaging modality. The phantom is low cost and can be reproducibly manufactured with the 3D printing and molding processes. Our initial experiments demonstrate the ability to integrate the phantom with a commercial tracking system for future needle tracking validation studies.

Preventing medication history errors in high-risk patients: Impact of California Senate Bill 1254.

PURPOSE: California Senate Bill (SB) 1254 (effective January 1, 2019) requires pharmacy staff at acute hospitals with more than 100 beds to obtain a medication profile for high-risk patients upon hospital admission. This multicenter study sought to evaluate the statewide impact of California SB 1254 by capturing the errors intercepted and harm prevented as a result of the passage of the bill. METHODS: This was a multicenter, prospective, observational study conducted at 11 hospitals in California for 6 consecutive weeks between January 2020 and March 2020. Participating sites captured medication history errors identified among high-risk patients using organization-specific criteria. Errors were categorized by type and ranked for severity of potential or actual harm based on the modified National Coordinating Council for Medication Error Reporting and Prevention (NCC MERP) categories. RESULTS: Study sites had an average daily census of 180 to 800 patients. Approximately 94% (n = 2,554) of medication histories conducted disclosed at least 1 error. Approximately 54% (n = 1,474) of histories disclosed at least 1 serious or potentially life-threatening error. Approximately 6 errors were identified and prevented per patient (95% CI, 5.62-6.01 errors per patient), and 1 in 4 errors (25%) was categorized as potentially serious or life-threatening. CONCLUSION: Among high-risk patients, pharmacy-led medication histories significantly reduced medication errors. If not intercepted, these errors would have likely resulted in substantial morbidity and mortality. Future research should evaluate opportunities to standardize high-risk criteria to support patient prioritization and allocation of resources.

Healthcare provider evaluation of machine learning-directed care: reactions to deployment on a randomised controlled study.

OBJECTIVES: Clinical artificial intelligence and machine learning (ML) face barriers related to implementation and trust. There have been few prospective opportunities to evaluate these concerns. System for High Intensity EvaLuation During Radiotherapy (NCT03775265) was a randomised controlled study demonstrating that ML accurately directed clinical evaluations to reduce acute care during cancer radiotherapy. We characterised subsequent perceptions and barriers to implementation. METHODS: An anonymous 7-question Likert-type scale survey with optional free text was administered to multidisciplinary staff focused on workflow, agreement with ML and patient experience. RESULTS: 59/71 (83%) responded. 81% disagreed/strongly disagreed their workflow was disrupted. 67% agreed/strongly agreed patients undergoing intervention were high risk. 75% agreed/strongly agreed they would implement the ML approach routinely if the study was positive. Free-text feedback focused on patient education and ML predictions. CONCLUSIONS: Randomised data and firsthand experience support positive reception of clinical ML. Providers highlighted future priorities, including patient counselling and workflow optimisation.

Implementation of machine learning in the clinic: challenges and lessons in prospective deployment from the System for High Intensity EvaLuation During Radiation Therapy (SHIELD-RT) randomized controlled study.

BACKGROUND: Artificial intelligence (AI) and machine learning (ML) have resulted in significant enthusiasm for their promise in healthcare. Despite this, prospective randomized controlled trials and successful clinical implementation remain limited. One clinical application of ML is mitigation of the increased risk for acute care during outpatient cancer therapy. We previously reported the results of the System for High Intensity EvaLuation During Radiation Therapy (SHIELD-RT) study (NCT04277650), which was a prospective, randomized quality improvement study demonstrating that ML based on electronic health record (EHR) data can direct supplemental clinical evaluations and reduce the rate of acute care during cancer radiotherapy with and without chemotherapy. The objective of this study is to report the workflow and operational challenges encountered during ML implementation on the SHIELD-RT study. RESULTS: Data extraction and manual review steps in the workflow represented significant time commitments for implementation of clinical ML on a prospective, randomized study. Barriers include limited data availability through the standard clinical workflow and commercial products, the need to aggregate data from multiple sources, and logistical challenges from altering the standard clinical workflow to deliver adaptive care. CONCLUSIONS: The SHIELD-RT study was an early randomized controlled study which enabled assessment of barriers to clinical ML implementation, specifically those which leverage the EHR. These challenges build on a growing body of literature and may provide lessons for future healthcare ML adoption. TRIAL REGISTRATION: NCT04277650. Registered 20 February 2020. Retrospectively registered quality improvement study.

Early-stage Breast Cancer: Tailored External Beam Fractionation Approaches for Treatment of the Whole or Partial Breast.

Historically, radiotherapy fractionation for early-stage breast cancer primarily consisted of 1.8-2 Gy per fraction given once daily to a total dose of 45-66 Gy over 5-7 weeks for whole breast treatment. Partial breast treatment employed larger dose per fraction (3.4-3.85 Gy) in 10 fractions given twice daily over 1 week. Radiobiologically, breast cancer is increasingly appreciated as a low alpha-beta ratio malignancy that is best treated with larger dose per fraction. Over the past 10 years, there have been increasing data from multiple large randomized clinical trials that support the use of shorter treatment courses: first hypofractionated regimens consisting of 15-20 treatments, and more recently, ultra-hypofractionated regimens as short as 5 treatments. Simultaneously, data from modern partial breast irradiation (PBI) trials support once daily treatment regimens ranging from 1-5 treatments. Shorter treatment courses represent less treatment burden on patients, reduced financial impact, and potentially improved access to care for patients with transportation and/or socioeconomic barriers. Here we review the evolution of whole and partial breast treatment regimens for early-stage breast cancer.

PrecisionFDA Truth Challenge V2: Calling variants from short and long reads in difficult-to-map regions.

The precisionFDA Truth Challenge V2 aimed to assess the state of the art of variant calling in challenging genomic regions. Starting with FASTQs, 20 challenge participants applied their variant-calling pipelines and submitted 64 variant call sets for one or more sequencing technologies (Illumina, PacBio HiFi, and Oxford Nanopore Technologies). Submissions were evaluated following best practices for benchmarking small variants with updated Genome in a Bottle benchmark sets and genome stratifications. Challenge submissions included numerous innovative methods, with graph-based and machine learning methods scoring best for short-read and long-read datasets, respectively. With machine learning approaches, combining multiple sequencing technologies performed particularly well. Recent developments in sequencing and variant calling have enabled benchmarking variants in challenging genomic regions, paving the way for the identification of previously unknown clinically relevant variants.

Reduction in medical emergency team activation among postoperative surgical patients at risk for undiagnosed obstructive sleep apnea.

STUDY OBJECTIVES: Obstructive sleep apnea (OSA) is an under-recognized condition that results in morbidity and mortality. Postoperative complications, including medical emergency team activation (META), are disproportionally increased among surgical patients at risk for OSA. A systematic approach is needed to improve provider recognition and treatment, but protocols that demonstrate improvement in META are lacking. As part of a multidisciplinary quality improvement project, DOISNORE50 (DIS), a sleep apnea questionnaire and proactive safety measure, was algorithmically applied to all perioperative patients. METHODS: Consecutive sleep screening was conducted among perioperative patients. Of the 49,567 surgical navigation center patients, 11,932 had previous diagnosis of OSA. Of the 37,572 (96%) patients screened with DIS, 25,171 (66.9%) were Low Risk (DIS < 4), 9,211 (24.5%) were At Risk (DIS ≥ 4), and 3,190 (8.5%) were High Risk (DIS ≥ 6) for OSA, respectively. High Risk patients received same-day sleep consultation. On the day of surgery, patients with Known OSA, At Risk, and High Risk for OSA received an "OSA Precaution Band." An electronic chart reminder alerted admission providers to order postoperative continuous positive airway pressure (CPAP) machine and sleep consult for patients High Risk for OSA. RESULTS: Implementation of a comprehensive program was associated with increased sleep consultation, sleep testing, and inpatient CPAP use (P < .001). For every 1,000 surgical patients screened, 30 fewer META, including rapid responses, reintubation, code blues, and code strokes, were observed. However, inpatient sleep consultation and inpatient CPAP use were not independently associated with reduced META. In the subgroup of patients hospitalized longer than 3 days, inpatient CPAP use was independently associated with reduced META. CONCLUSIONS: In this single-center, institution-wide, multidisciplinary-approach, quality improvement project, a comprehensive OSA screening process and treatment algorithm with appropriate postoperative inpatient CPAP therapy and inpatient sleep consultations was associated with increased CPAP use and reduced META. Further prospective studies are needed to assess cost, feasibility, and generalizability of these findings. CITATION: Namen AM, Forest D, Saha AK, et al. Reduction in medical emergency team activation among postoperative surgical patients at risk for undiagnosed obstructive sleep apnea. J Clin Sleep Med. 2022;18(8):1953-1965.

Role of stereotactic body radiotherapy in gynecologic radiation oncology.

Stereotactic body radiotherapy (SBRT, also referred to as stereotactic ablative radiotherapy (SABR)) has been used in the treatment of primary and metastatic solid tumors, and increasingly so in gynecologic oncology. This review article aims to summarize the current literature describing the utility of SBRT in the primary, recurrent, and limited metastatic settings for gynecologic malignancies. The use of SBRT in both retrospective and prospective reports has been associated with adequate control of the treated site, particularly in the setting of oligometastatic disease. It is not, however, recommended as an alternative to brachytherapy for intact disease unless all efforts to use brachytherapy are exhausted. While phase I and II trials have established the relative safety and potential toxicities of SBRT, there remains a dearth of phase III randomized evidence, including the use of immunotherapy, in order to better establish the role of this technique as a method of improving more global outcomes for our patients with gynecologic cancers.

High-Alert Medication Stratification Tool-Revised Phase II: A Multisite Study Examining the Validity, Interrater Reliability, and Ease of Use of the High-Alert Medication Stratification Tool-Revised.

OBJECTIVES: The primary objective of this multisite study, High-Alert Medication Stratification Tool-Revised (HAMST-R) phase II, was to assess the content validity of HAMST-R. Secondary outcomes included interrater reliability and ease of use. METHODS: HAMST-R was designed as an objective tool to evaluate high-alert medications (HAMs) at a single site during HAMST-R Phase I. Medication safety experts from 7 health systems across the United States volunteered to participate in this phase II study. Participants completed a demographic survey, oversaw evaluation of 47 HAMs and 35 non-HAMs using HAMST-R, and submitted scores for each medication evaluated. In addition, participants rated each question of HAMST-R on its relevance to assess a medication's safety risk, measured as scale-content validity index. Positive and negative predictive values were evaluated in a post hoc analysis. Interrater reliability was evaluated using the Kendall coefficient of concordance (K), and ease of use was assessed using a mixed-methods approach. RESULTS: Scale-content validity index was 0.80, indicating that the tool was valid. Positive predictive value was 90.5% (95% confidence interval, 87.2%-93.0%), and negative predictive value was 98.2% (95% confidence interval, 95.4%-99.3%). A score of 4 or more differentiated between HAMs and non-HAMs, confirming phase I findings. K was 0.56, indicating moderate agreement. Participants confirmed that the tool was easy to use and plan to incorporate the tool into HAM policies and procedures, formulary review, and safety strategy implementation. CONCLUSIONS: HAMST-R is a valid, objective, and easy to use method that institutions may implement to evaluate a medication's potential safety risk.

System for High-Intensity Evaluation During Radiation Therapy (SHIELD-RT): A Prospective Randomized Study of Machine Learning-Directed Clinical Evaluations During Radiation and Chemoradiation.

PURPOSE: Patients undergoing outpatient radiotherapy (RT) or chemoradiation (CRT) frequently require acute care (emergency department evaluation or hospitalization). Machine learning (ML) may guide interventions to reduce this risk. There are limited prospective studies investigating the clinical impact of ML in health care. The objective of this study was to determine whether ML can identify high-risk patients and direct mandatory twice-weekly clinical evaluation to reduce acute care visits during treatment. PATIENTS AND METHODS: During this single-institution randomized quality improvement study (ClinicalTrials.gov identifier: NCT04277650), 963 outpatient adult courses of RT and CRT started from January 7 to June 30, 2019, were evaluated by an ML algorithm. Among these, 311 courses identified by ML as high risk (> 10% risk of acute care during treatment) were randomized to standard once-weekly clinical evaluation (n = 157) or mandatory twice-weekly evaluation (n = 154). Both arms allowed additional evaluations on the basis of clinician discretion. The primary end point was the rate of acute care visits during RT. Model performance was evaluated using receiver operating characteristic area under the curve (AUC) and decile calibration plots. RESULTS: Twice-weekly evaluation reduced rates of acute care during treatment from 22.3% to 12.3% (difference, -10.0%; 95% CI, -18.3 to -1.6; relative risk, 0.556; 95% CI, 0.332 to 0.924; P = .02). Low-risk patients had a 2.7% acute care rate. Model discrimination was good in high- and low-risk patients undergoing standard once-weekly evaluation (AUC, 0.851). CONCLUSION: In this prospective randomized study, ML accurately triaged patients undergoing RT and CRT, directing clinical management with reduced acute care rates versus standard of care. This prospective study demonstrates the potential benefit of ML in health care and offers opportunities to enhance care quality and reduce health care costs.

A Scoping Review of Intravenous Opioid Pain Protocols and Their Dose-Time Intervals in the Management of Acute Postoperative Pain in the Postanesthesia Care Unit.

PURPOSE: This review aims to explore intravenous opioid pain protocols and their dose-time intervals in managing acute postoperative pain in adults in the postanesthesia care unit (PACU). DESIGN: A scoping review using a systematic search strategy. METHODS: Sixteen articles were identified from MEDLINE, CINAHL, PubMed, Embase, and Cochrane specific to the aims. FINDINGS: The literature demonstrated several variations on dose-time intervals used for opioid pain protocol administration globally. Furthermore, opioid analgesic pain protocols in the PACU appear to be effective in postoperative pain management. However, the literature did not identify optimal time intervals related to dose administration within these protocols. CONCLUSIONS: Literature gaps were identified regarding the significance of dose-time intervals when using opioid analgesic pain protocols in the PACU.

Evaluating for disparities in place of death for head and neck cancer patients in the United States utilizing the CDC WONDER database.

OBJECTIVE: Evaluate trends in place of death for patients with head and neck cancers (HNC) in the U.S. from 1999 to 2017 based on the CDC WONDER (Wide-ranging Online Data for Epidemiologic Research) database. MATERIALS/METHODS: Using patient-level data from 2015 and aggregate data from 1999 to 2017, multivariable logistic regression analyses (MLR) were performed to evaluate for disparities in place of death. RESULTS: We obtained aggregate data for 101,963 people who died of HNC between 1999 and 2017 (25.9% oral cavity, 24.6% oropharynx/pharynx, 0.4% nasopharynx, and 49.1% larynx/hypopharynx). Most were Caucasian (92.7%) and male (87.0%). Deaths at home or hospice increased over the study period (R2 = 0.96, p < 0.05) from 29.2% in 1999 to 61.2% in 2017. On MLR of patient-level data from 2015, those who were single (ref), ages 85+ (OR 0.78; 95% CI: 0.68, 0.90), African American (OR 0.73; 95% CI: 0.65, 0.82), or Asian/Pacific Islanders (OR 0.66; 95% CI: 0.54, 0.81) were less likely to die at home or hospice. On MLR of the aggregate data (1999-2017), those who were female (OR 0.87; 95% CI: 0.83, 0.91) or ages 75-84 (OR 0.79; 95% CI: 0.76, 0.82) were also less likely to die at home or hospice. In both analyses, those who died from larynx/hypopharynx cancers were less likely to die at home or hospice. CONCLUSIONS: HNC-related deaths at home or hospice increased between 1999 and 2017. Those who were single, female, African American, Asian/Pacific Islander, older (ages 75+), or those with larynx/hypopharynx cancers were less likely to die at home or hospice.

The Effects of Health Disparities on Perceptions About Lung Cancer Screening (LCS): Survey Results of a Patient Sample.

Lung cancer screening (LCS) is currently advocated in a subset of current or former smokers with a thirty pack-year smoking history or higher. Studies report that few patients meeting the criteria for screening are undergoing LCS. We conducted a survey to assess if barriers to LCS (race, ethnicity, and socioeconomic status) affect the perceptions about LCS that could influence screening uptake. We did not detect different perceptions based on race, ethnicity, or socioeconomic status; however, our survey found that fewer barriers and more benefits to LCS may be perceived in patients who undergo other types of health screening and more benefits for those with internet capable devices.

Local Tumor Control and Patient Outcome Using Stereotactic Body Radiation Therapy for Hepatocellular Carcinoma: iRECIST as a Potential Substitute for Traditional Criteria.

OBJECTIVE. The purpose of this study was to investigate whether, compared with traditional criteria, the modified Response Evaluation Criteria in Solid Tumors version 1.1 for immune-based therapeutics (iRECIST) improves prediction of local tumor control and survival in patients with hepatocellular carcinoma (HCC) treated with stereotactic body radiotherapy (SBRT). MATERIALS AND METHODS. Fifty-one HCC lesions (mean size, 3.1 cm) treated with SBRT in 41 patients (mean age, 67 years) were retrospectively included. Each patient underwent CT or MRI before SBRT and at least once after SBRT. Best overall response was categorized using Response Evaluation Criteria in Solid Tumors version 1.1 (RECIST 1.1), iRECIST, World Health Organization (WHO) criteria, modified Response Evaluation Criteria in Solid Tumors (mRECIST), and European Association for the Study of the Liver (EASL) criteria. Lesions were then classified as local tumor control (i.e., stable disease, partial response, or complete response) or local treatment failure (i.e., progressive disease) by each tumor response criteria. Proportions of local tumor control were compared using the McNemar exact test. The 1-year overall survival was estimated using the Kaplan-Meier method. RESULTS. The median follow-up after SBRT was 21.0 months. The local tumor control rate was 94.1% (48/51) by iRECIST, 88.2% (45/51) by RECIST 1.1, 72.5% (37/51) by WHO criteria, 80.4% (41/51) by mRECIST, and 72.5% (37/51) by EASL criteria. The local tumor control rate was significantly higher according to iRECIST compared with WHO (p = 0.0010) and EASL (p = 0.0225) criteria. The 1-year survival rate for patients with local tumor control according to iRECIST (86.4%) was higher (although not statistically significant) compared with the 1-year survival rate for patients with local tumor control according to the other response criteria. CONCLUSION. iRECIST may provide more robust interpretation of HCC response after SBRT, yielding improved prediction of local tumor control and 1-year survival rates compared with traditional criteria.

The role of external beam radiotherapy in the treatment of hepatocellular cancer.

Hepatocellular carcinoma (HCC) is increasing in incidence and mortality. Although the prognosis remains poor, long-term survival has improved from 3% in 1970 to an 18% 5-year survival rate today. This is likely because of the introduction of well tolerated, oral antiviral therapies for hepatitis C. Curative options for patients with HCC are often limited by underlying liver dysfunction/cirrhosis and medical comorbidities. Less than one-third of patients are candidates for surgery, which is the current gold standard for cure. Nonsurgical treatments include embolotherapies, percutaneous ablation, and ablative radiation. Technological advances in radiation delivery in the past several decades now allow for safe and effective ablative doses to the liver. Conformal techniques allow for both dose escalation to target volumes and normal tissue sparing. Multiple retrospective and prospective studies have demonstrated that hypofractionated image-guided radiation therapy, used as monotherapy or in combination with other liver-directed therapies, can provide excellent local control that is cost effective. Therefore, as the HCC treatment paradigm continues to evolve, ablative radiation treatment has moved from a palliative treatment to both a "bridge to transplant" and a definitive treatment.

Managing Patients With Oligometastatic Non-Small-Cell Lung Cancer.

Metastatic lung cancer has long been considered incurable, with the goal of treatment being palliation. However, a clinically meaningful number of these patients with limited metastases (approximately 25%) are living long term after definitive treatment to all sites of active disease. These patients with so-called oligometastatic disease likely represent a distinct clinical group who may possess a more indolent biology compared with their more widely metastatic counterparts. Hellman and Weichselbaum proposed the existence of the oligometastatic state, on the basis of the spectrum theory of cancer spread. The literature suggests that an oligometastatic state exists in patients with non-small-cell lung cancer (NSCLC). This observation in the setting of rapidly evolving systemic therapies, including immune checkpoint inhibitors and an increasing number of targeted therapies, represents a unique clinical opportunity. Metastasis-directed therapies to address sites of disease include surgery (metastasectomy) and/or radiation therapy. Available evidence suggests that treating patients with limited or oligometastases may improve outcomes in a meaningful way; however, the majority of the randomized data includes patients with intracranial metastatic disease, and there are limited robust, randomized data available in the setting of NSCLC with only extracranial sites of metastatic disease. Ongoing randomized trials, including NRG-LU002 and the UK Conventional Care Versus Radioablation (Stereotactic Body Radiotherapy) for Extracranial Oligometastases trial, are aimed at evaluating this question further. One of the current limitations of aggressive treatment of oligometastatic NSCLC is the inability to accurately identify these patients before therapy, yet molecular markers, including microRNA profiles, are being investigated as a promising way to identify these patients.