Your browser doesn't support javascript.
loading
Show: 20 | 50 | 100
Results 1 - 20 de 2.947
Filter
1.
Acta Oncol ; 63: 518-525, 2024 Jun 30.
Article in English | MEDLINE | ID: mdl-38946287

ABSTRACT

AIM: The study aims to investigate the impact of the COVID-19 pandemic on cancer patients' perceptions of the quality of their oncological treatment and care. BACKGROUND: The COVID-19 pandemic disrupted healthcare delivery and oncological resources were repurposed, potentially leading to prolonged treatment and reduced access to innovative therapies and clinical trials. Still, little is known about how patients perceived the quality of their treatment. METHODS: A cross-sectional study was conducted in the spring of 2020 among cancer patients at the Department of Oncology, Aarhus University Hospital and Rigshospitalet, Denmark. Patients were invited to complete an online questionnaire on clinical, socioeconomic, emotional, behavioural, and quality-related aspects of oncological cancer care. Patients who experienced reduced treatment quality and those who reported no or slight reductions were compared using multiple logistic regression, exploring the associations with patient characteristics, behaviours, and fear of cancer progression or recurrence. RESULTS: A total of 2,040/5,372 patients experienced changes in their treatment plans during the pandemic, and 1,570/5,372 patients experienced reduced treatment quality, with 236 reporting a high degree of reduction. Patients with breast, head and neck, and upper gastrointestinal cancers were more likely to experience reduced treatment quality. Altered interactions with healthcare providers, along with isolation, lack of social support, and heightened fear of cancer progression, were significant risk factors for experiencing reduced cancer care quality. INTERPRETATION: We identified subgroups of cancer patients needing targeted communication and care during health crises affecting cancer treatment. The findings underscore the importance of safeguarding the needs of vulnerable patient populations in future healthcare emergencies.


Subject(s)
COVID-19 , Neoplasms , Patient Reported Outcome Measures , Quality of Health Care , Humans , COVID-19/epidemiology , Cross-Sectional Studies , Female , Male , Neoplasms/therapy , Neoplasms/psychology , Neoplasms/epidemiology , Middle Aged , Aged , Denmark/epidemiology , Quality of Health Care/standards , Adult , SARS-CoV-2 , Surveys and Questionnaires , Medical Oncology/standards , Pandemics
3.
Support Care Cancer ; 32(7): 476, 2024 Jul 02.
Article in English | MEDLINE | ID: mdl-38954101

ABSTRACT

CONTEXT: Home palliative care service increases the chance of dying at home, particularly for patients with advanced cancer, but late referrals to home palliative care services still exist. Indicators for evaluating programs that can facilitate the integration of oncology and home palliative care have not been defined. OBJECTIVES: This study developed quality indicators for the integration of oncology and home palliative care in Japan. METHODS: We conducted a systematic literature review (Databases included CENTRAL, MEDLINE, EMBASE, and Emcare) and a modified Delphi study to develop the quality indicators. Panelists rated a potential list of indicators using a 9-point scale over three rounds according to two criteria: appropriateness and feasibility. The criterion for the adoption of candidate indicators was set at a total mean score of 7 or more. Final quality indicators with no disagreement were included. RESULTS: Of the 973 publications in our initial search, 12 studies were included. The preliminary list of quality indicators by systematic literature review comprised 50 items. In total, 37 panelists participated in the modified Delphi study. Ultimately, 18 indicators were identified from the following domains: structure in cancer hospitals, structure in home palliative care services, the process of home palliative care service delivery, less aggressive end-of-life care, patient's psychological comfort, caregiver's psychological comfort, and patient's satisfaction with home palliative care service. CONCLUSION: Comprehensive quality indicators for the integration of oncology and home palliative care were identified. These indicators may facilitate interdisciplinary collaboration between professional healthcare providers in both cancer hospitals and home palliative care services.


Subject(s)
Delphi Technique , Home Care Services , Neoplasms , Palliative Care , Quality Indicators, Health Care , Humans , Palliative Care/standards , Palliative Care/organization & administration , Palliative Care/methods , Home Care Services/standards , Home Care Services/organization & administration , Japan , Neoplasms/therapy , Medical Oncology/organization & administration , Medical Oncology/standards
4.
JCO Clin Cancer Inform ; 8: e2400039, 2024 Jun.
Article in English | MEDLINE | ID: mdl-38950323

ABSTRACT

Randomized trials provide high-quality, internally consistent data on selected clinical questions, but lack generalizability for the aging population who are most often diagnosed with cancer and have comorbid conditions that may affect the interpretation of treatment benefit. The need for high-quality, relevant, and timely data is greater than ever. Promising solutions lie in the collection and analysis of real-world data (RWD), which can potentially provide timely insights about the patient's course during and after initial treatment and the outcomes of important subgroups such as the elderly, rural populations, children, and patients with greater social health needs. However, to inform practice and policy, real-world evidence must be created from trustworthy and comprehensive sources of RWD; these may include pragmatic clinical trials, registries, prospective observational studies, electronic health records (EHRs), administrative claims, and digital technologies. There are unique challenges in oncology since key parameters (eg, cancer stage, biomarker status, genomic assays, imaging response, side effects, quality of life) are not recorded, siloed in inaccessible documents, or available only as free text or unstructured reports in the EHR. Advances in analytics, such as artificial intelligence, may greatly enhance the ability to obtain more granular information from EHRs and support integrated diagnostics; however, they will need to be validated purpose by purpose. We recommend a commitment to standardizing data across sources and building infrastructures that can produce fit-for-purpose RWD that will provide timely understanding of the effectiveness of individual interventions.


Subject(s)
Electronic Health Records , Medical Oncology , Neoplasms , Humans , Medical Oncology/methods , Medical Oncology/standards , Neoplasms/therapy , Neoplasms/diagnosis , Registries
5.
Am Soc Clin Oncol Educ Book ; 44(3): e432794, 2024 Jun.
Article in English | MEDLINE | ID: mdl-38924707

ABSTRACT

The completion of multiple national pediatric precision oncology platform trials and the incorporation of standardized molecular profiling into the diagnostic care of pediatric and young adult patients with sarcomas have proven the feasibility and potential of the approach. In this work, we explore the current state of the art of precision oncology for pediatric and young adults with sarcoma. We highlight important lessons learned and the challenges that should be addressed in the next generation of trials. The chapter outlines current efforts to improve standardization of molecular assays, harmonization of data collection, and novel molecular tools such as cell-free DNA analyses. Finally, we discuss the impacts and psychosocial outcomes experienced by patients and communication strategies for providers.


Subject(s)
Precision Medicine , Sarcoma , Humans , Sarcoma/therapy , Precision Medicine/methods , Child , Young Adult , Adolescent , Medical Oncology/methods , Medical Oncology/standards , Adult , Molecular Targeted Therapy , Biomarkers, Tumor
6.
JCO Clin Cancer Inform ; 8: e2300249, 2024 Jun.
Article in English | MEDLINE | ID: mdl-38935887

ABSTRACT

PURPOSE: The expanding presence of the electronic health record (EHR) underscores the necessity for improved interoperability. To test the interoperability within the field of oncology research, our team at Vanderbilt University Medical Center (VUMC) enabled our Epic-based EHR to be compatible with the Minimal Common Oncology Data Elements (mCODE), which is a Fast Healthcare Interoperability Resources (FHIR)-based consensus data standard created to facilitate the transmission of EHRs for patients with cancer. METHODS: Our approach used an extract, transform, load tool for converting EHR data from the VUMC Epic Clarity database into mCODE-compatible profiles. We established a sandbox environment on Microsoft Azure for data migration, deployed a FHIR server to handle application programming interface (API) requests, and mapped VUMC data to align with mCODE structures. In addition, we constructed a web application to demonstrate the practical use of mCODE profiles in health care. RESULTS: We developed an end-to-end pipeline that converted EHR data into mCODE-compliant profiles, as well as a web application that visualizes genomic data and provides cancer risk assessments. Despite the complexities of aligning traditional EHR databases with mCODE standards and the limitations of FHIR APIs in supporting advanced statistical methodologies, this project successfully demonstrates the practical integration of mCODE standards into existing health care infrastructures. CONCLUSION: This study provides a proof of concept for the interoperability of mCODE within a major health care institution's EHR system, highlighting both the potential and the current limitations of FHIR APIs in supporting complex data analysis for oncology research.


Subject(s)
Academic Medical Centers , Electronic Health Records , Genomics , Medical Oncology , Humans , Pilot Projects , Medical Oncology/methods , Medical Oncology/standards , Genomics/methods , Neoplasms/genetics , Common Data Elements , Software , Health Information Interoperability
7.
J Natl Compr Canc Netw ; 22(2 D)2024 06.
Article in English | MEDLINE | ID: mdl-38862008

ABSTRACT

Colorectal cancer (CRC) is the fourth most frequently diagnosed cancer and the second leading cause of cancer death in the United States. Management of disseminated metastatic CRC involves various active drugs, either in combination or as single agents. The choice of therapy is based on consideration of the goals of therapy, the type and timing of prior therapy, the mutational profile of the tumor, and the differing toxicity profiles of the constituent drugs. This manuscript summarizes the data supporting the systemic therapy options recommended for metastatic CRC in the NCCN Guidelines for Colon Cancer.


Subject(s)
Colonic Neoplasms , Humans , Colonic Neoplasms/diagnosis , Colonic Neoplasms/therapy , Colonic Neoplasms/pathology , Colonic Neoplasms/drug therapy , Medical Oncology/standards , Medical Oncology/methods , Antineoplastic Combined Chemotherapy Protocols/therapeutic use , Antineoplastic Combined Chemotherapy Protocols/adverse effects , United States
8.
JCO Glob Oncol ; 10: e2300405, 2024 Jun.
Article in English | MEDLINE | ID: mdl-38870438

ABSTRACT

Executing global clinical trials for cancer is a long, expensive, and complex undertaking. While selecting countries global studies, sponsors must consider several aspects including patient pool, quality of trained investigators, competing trials, availability of infrastructure, and financial investment versus returns. With a large, often treatment-naïve, and diverse patient pool, relatively low cost, good quality health care facilities in urban areas, and a robust and well-trained workforce, India offers several advantages for conducting oncology clinical trials. However, there remains challenges, including a shifting regulatory environment in recent decades. With the implementation of the New Drugs and Clinical Trial Rules in 2019, India's regulatory atmosphere seems to have stabilized. In this article, we present a review of the evolving clinical trial landscape in India, highlight the current regulatory scenario, and discuss the advantages and challenges of selecting India as a potential location for conducting global oncology clinical trials.


Subject(s)
Clinical Trials as Topic , Neoplasms , India , Humans , Neoplasms/therapy , Medical Oncology/standards
9.
Am Soc Clin Oncol Educ Book ; 44(3): e432102, 2024 Jun.
Article in English | MEDLINE | ID: mdl-38870439

ABSTRACT

Quality cancer care is efficient, accessible, coordinated, and evidence-based. Recognizing the necessary key components, development of pathways and guidelines to incorporate these key domains, and finally respectful adaptation to cultural differences can ensure that cancer care globally is of the highest quality. This quality care should be judged not only on how it optimizes health outcomes for patients, but also its impact on the care providers and the global community.


Subject(s)
Medical Oncology , Neoplasms , Quality of Health Care , Humans , Delivery of Health Care/standards , Global Health , Medical Oncology/standards , Neoplasms/therapy
10.
Oncologist ; 29(7): 547-550, 2024 Jul 05.
Article in English | MEDLINE | ID: mdl-38824414

ABSTRACT

Missing visual elements (MVE) in Kaplan-Meier (KM) curves can misrepresent data, preclude curve reconstruction, and hamper transparency. This study evaluated KM plots of phase III oncology trials. MVE were defined as an incomplete y-axis range or missing number at risk table in a KM curve. Surrogate endpoint KM curves were additionally evaluated for complete interpretability, defined by (1) reporting the number of censored patients and (2) correspondence of the disease assessment interval with the number at risk interval. Among 641 trials enrolling 518 235 patients, 116 trials (18%) had MVE in KM curves. Industry sponsorship, larger trials, and more recently published trials were correlated with lower odds of MVE. Only 3% of trials (15 of 574) published surrogate endpoint KM plots with complete interpretability. Improvements in the quality of KM curves of phase III oncology trials, particularly for surrogate endpoints, are needed for greater interpretability, reproducibility, and transparency in oncology research.


Subject(s)
Clinical Trials, Phase III as Topic , Kaplan-Meier Estimate , Humans , Clinical Trials, Phase III as Topic/standards , Neoplasms/therapy , Medical Oncology/standards , Medical Oncology/methods
11.
Adv Gerontol ; 37(1-2): 87-94, 2024.
Article in Russian | MEDLINE | ID: mdl-38944778

ABSTRACT

A review of the use of comprehensive geriatric assessment and its components in oncology is introduced. The evidence base for the increasing incidence of cancer in the elderly worldwide also presented. The management of older adults with cancer is challenging. Comprehensive Geriatric Assessment (CGA) has been shown by many authors to be a strong predictor of adverse events in geriatric oncology patients. CGA is recommended in oncology practice for many reasons: to identify health problems not usually detected in routine oncologic screening, to perform non-oncologic interventions, and to modify the cancer treatment plan. Comprehensive geriatric assessment is the gold standard in geriatric oncology for identifying patients at high risk for adverse outcomes and optimizing cancer treatment and overall management. Nevertheless, it can be stated that the final point in the search for evidence-based and effective frailty assessment tools in the practice of geriatric oncology has not yet been reached. It is concluded that the development of new scales and index scores, as well as the application of the CGA model in general, can provide adequate care for elderly cancer patients.


Subject(s)
Frailty , Geriatric Assessment , Medical Oncology , Neoplasms , Humans , Geriatric Assessment/methods , Aged , Neoplasms/therapy , Neoplasms/epidemiology , Neoplasms/diagnosis , Medical Oncology/methods , Medical Oncology/standards , Frailty/diagnosis , Frailty/epidemiology , Frail Elderly , Geriatrics/methods
12.
J Natl Compr Canc Netw ; 22(2D)2024 Jun.
Article in English | MEDLINE | ID: mdl-38862005

ABSTRACT

Mastocytosis is a heterogeneous group of disorders comprising cutaneous mastocytosis, systemic mastocytosis, and mast cell sarcoma. It is associated with a variety of symptoms related to the release of mast cell mediators and mast cell tissue infiltration. Referral to specialized centers with expertise in the management of mastocytosis and multidisciplinary collaboration with subspecialists (eg, allergists for the management of anaphylaxis and drug hypersensitivities, anesthesiologists for invasive procedures or surgery, high-risk obstetrician for pregnancy) is recommended. The NCCN Guidelines for Systemic Mastocytosis provide evidence- and consensus-based recommendations for the diagnosis and comprehensive care of patients with systemic mastocytosis. The multidisciplinary panel of experts convenes at least once a year to review requested changes to the guidelines from both internal and external entities as well as to discuss data on existing and new therapies. These NCCN Guidelines Insights focus on some of the recent updates to the guidelines.


Subject(s)
Mastocytosis, Systemic , Humans , Mastocytosis, Systemic/diagnosis , Mastocytosis, Systemic/therapy , Disease Management , Medical Oncology/standards , Medical Oncology/methods
13.
Am Soc Clin Oncol Educ Book ; 44(3): e432104, 2024 Jun.
Article in English | MEDLINE | ID: mdl-38862140

ABSTRACT

Cancer care in low- and middle-income countries (LMICs) faces numerous challenges, such as limited resources, infrastructure constraints, and a shortage of specialized training. To address these challenges, innovative quality improvement (QI) approaches are required. The Quality Oncology Practice Initiative (QOPI) by the ASCO provides a framework for improving care quality through evidence-based standards. This article explores how QOPI has been adapted to the local context of the Uganda Cancer Institute (UCI) and highlights the importance of aligning international best practices with local health care realities to bridge disparities in care standards. The adaptation of the QOPI program at UCI commenced with a collaborative meeting with the ASCO-QOPI team in 2020. A tailored implementation plan was developed focusing on incorporating ASCO's QOPI measures and additional metrics relevant to the Ugandan context, engaging multidisciplinary teams, and optimizing resource use by leveraging existing resources for data collection and analysis. The execution of the plan relied heavily on staff training, participatory data collection, and continuous quality improvement processes that utilized data-driven methodologies. A retrospective analysis of QOPI data of UCI from 2020 to 2023 shows significant improvements in oncology care quality, highlighted by an upward trend in QOPI assessment scores across various metrics. These reflect the journey of UCI toward aligning its oncology care practices with international standards despite facing significant challenges. UCI's experience demonstrated the feasibility and impact of implementing international QI programs in LMICs. The success demonstrates that significant improvements in cancer care quality can be achieved in resource-constrained settings through adaptability, stakeholder engagement, and strategic resource optimization. UCI's journey is a model for other LMICs seeking to raise their cancer care standards, demonstrating that QI is necessary and attainable worldwide.


Subject(s)
Medical Oncology , Neoplasms , Quality Improvement , Humans , Uganda , Medical Oncology/standards , Neoplasms/therapy , Quality of Health Care , Delivery of Health Care/standards
15.
BMJ Open Qual ; 13(2)2024 Jun 19.
Article in English | MEDLINE | ID: mdl-38901878

ABSTRACT

BACKGROUND: Evaluation of quality of care in oncology is key in ensuring patients receive adequate treatment. American Society of Clinical Oncology's (ASCO) Quality Oncology Practice Initiative (QOPI) Certification Program (QCP) is an international initiative that evaluates quality of care in outpatient oncology practices. METHODS: We retrospectively reviewed free-text electronic medical records from patients with breast cancer (BR), colorectal cancer (CRC) or non-small cell lung cancer (NSCLC). In a baseline measurement, high scores were obtained for the nine disease-specific measures of QCP Track (2021 version had 26 measures); thus, they were not further analysed. We evaluated two sets of measures: the remaining 17 QCP Track measures, as well as these plus other 17 measures selected by us (combined measures). Review of data from 58 patients (26 BR; 18 CRC; 14 NSCLC) seen in June 2021 revealed low overall quality scores (OQS)-below ASCO's 75% threshold-for QCP Track measures (46%) and combined measures (58%). We developed a plan to improve OQS and monitored the impact of the intervention by abstracting data at subsequent time points. RESULTS: We evaluated potential causes for the low OQS and developed a plan to improve it over time by educating oncologists at our hospital on the importance of improving collection of measures and highlighting the goal of applying for QOPI certification. We conducted seven plan-do-study-act cycles and evaluated the scores at seven subsequent data abstraction time points from November 2021 to December 2022, reviewing 404 patients (199 BR; 114 CRC; 91 NSCLC). All measures were improved. Four months after the intervention, OQS surpassed the quality threshold and was maintained for 10 months until the end of the study (range, 78-87% for QCP Track measures; 78-86% for combined measures). CONCLUSIONS: We developed an easy-to-implement intervention that achieved a fast improvement in OQS, enabling our Medical Oncology Department to aim for QOPI certification.


Subject(s)
Electronic Health Records , Quality Improvement , Humans , Electronic Health Records/statistics & numerical data , Electronic Health Records/standards , Retrospective Studies , Female , Spain , Male , Middle Aged , Quality of Health Care/standards , Quality of Health Care/statistics & numerical data , Aged , Data Collection/methods , Data Collection/standards , Medical Oncology/standards , Medical Oncology/methods , Medical Oncology/statistics & numerical data , Colorectal Neoplasms/therapy , Adult , Breast Neoplasms/therapy , Carcinoma, Non-Small-Cell Lung/therapy
16.
Ann Oncol ; 35(7): 588-606, 2024 Jul.
Article in English | MEDLINE | ID: mdl-38834388

ABSTRACT

BACKGROUND: Advancements in the field of precision medicine have prompted the European Society for Medical Oncology (ESMO) Precision Medicine Working Group to update the recommendations for the use of tumour next-generation sequencing (NGS) for patients with advanced cancers in routine practice. METHODS: The group discussed the clinical impact of tumour NGS in guiding treatment decision using the ESMO Scale for Clinical Actionability of molecular Targets (ESCAT) considering cost-effectiveness and accessibility. RESULTS: As for 2020 recommendations, ESMO recommends running tumour NGS in advanced non-squamous non-small-cell lung cancer, prostate cancer, colorectal cancer, cholangiocarcinoma, and ovarian cancer. Moreover, it is recommended to carry out tumour NGS in clinical research centres and under specific circumstances discussed with patients. In this updated report, the consensus within the group has led to an expansion of the recommendations to encompass patients with advanced breast cancer and rare tumours such as gastrointestinal stromal tumours, sarcoma, thyroid cancer, and cancer of unknown primary. Finally, ESMO recommends carrying out tumour NGS to detect tumour-agnostic alterations in patients with metastatic cancers where access to matched therapies is available. CONCLUSION: Tumour NGS is increasingly expanding its scope and application within oncology with the aim of enhancing the efficacy of precision medicine for patients with cancer.


Subject(s)
High-Throughput Nucleotide Sequencing , Neoplasms , Precision Medicine , Humans , High-Throughput Nucleotide Sequencing/methods , High-Throughput Nucleotide Sequencing/standards , Precision Medicine/methods , Precision Medicine/standards , Neoplasms/genetics , Neoplasms/pathology , Neoplasms/therapy , Medical Oncology/methods , Medical Oncology/standards , Europe
18.
J Clin Oncol ; 42(19): 2336-2357, 2024 Jul 01.
Article in English | MEDLINE | ID: mdl-38748941

ABSTRACT

PURPOSE: To provide evidence-based guidance to oncology clinicians, patients, nonprofessional caregivers, and palliative care clinicians to update the 2016 ASCO guideline on the integration of palliative care into standard oncology for all patients diagnosed with cancer. METHODS: ASCO convened an Expert Panel of medical, radiation, hematology-oncology, oncology nursing, palliative care, social work, ethics, advocacy, and psycho-oncology experts. The Panel conducted a literature search, including systematic reviews, meta-analyses, and randomized controlled trials published from 2015-2023. Outcomes of interest included quality of life (QOL), patient satisfaction, physical and psychological symptoms, survival, and caregiver burden. Expert Panel members used available evidence and informal consensus to develop evidence-based guideline recommendations. RESULTS: The literature search identified 52 relevant studies to inform the evidence base for this guideline. RECOMMENDATIONS: Evidence-based recommendations address the integration of palliative care in oncology. Oncology clinicians should refer patients with advanced solid tumors and hematologic malignancies to specialized interdisciplinary palliative care teams that provide outpatient and inpatient care beginning early in the course of the disease, alongside active treatment of their cancer. For patients with cancer with unaddressed physical, psychosocial, or spiritual distress, cancer care programs should provide dedicated specialist palliative care services complementing existing or emerging supportive care interventions. Oncology clinicians from across the interdisciplinary cancer care team may refer the caregivers (eg, family, chosen family, and friends) of patients with cancer to palliative care teams for additional support. The Expert Panel suggests early palliative care involvement, especially for patients with uncontrolled symptoms and QOL concerns. Clinicians caring for patients with solid tumors on phase I cancer trials may also refer them to specialist palliative care.Additional information is available at www.asco.org/supportive-care-guidelines.


Subject(s)
Neoplasms , Palliative Care , Humans , Palliative Care/standards , Neoplasms/therapy , Quality of Life , Medical Oncology/standards
19.
Int J Clin Oncol ; 29(7): 873-888, 2024 Jul.
Article in English | MEDLINE | ID: mdl-38753042

ABSTRACT

BACKGROUND: The Japan Society of Clinical Oncology Clinical Practice Guidelines for Antiemesis 2023 was extensively revised to reflect the latest advances in antineoplastic agents, antiemetics, and antineoplastic regimens. This update provides new evidence on the efficacy of antiemetic regimens. METHODS: Guided by the Minds Clinical Practice Guideline Development Manual of 2017, a rigorous approach was used to update the guidelines; a thorough literature search was conducted from January 1, 1990, to December 31, 2020. RESULTS: Comprehensive process resulted in the creation of 13 background questions (BQs), 12 clinical questions (CQs), and three future research questions (FQs). Moreover, the emetic risk classification was also updated. CONCLUSIONS: The primary goal of the present guidelines is to provide comprehensive information and facilitate informed decision-making, regarding antiemetic therapy, for both patients and healthcare providers.


Subject(s)
Antiemetics , Medical Oncology , Vomiting , Humans , Japan , Medical Oncology/standards , Antiemetics/therapeutic use , Vomiting/prevention & control , Vomiting/drug therapy , Neoplasms/drug therapy , Antineoplastic Agents/therapeutic use , Societies, Medical , Nausea/prevention & control , Nausea/drug therapy
20.
BMC Palliat Care ; 23(1): 124, 2024 May 20.
Article in English | MEDLINE | ID: mdl-38769564

ABSTRACT

BACKGROUND: Ex-ante identification of the last year in life facilitates a proactive palliative approach. Machine learning models trained on electronic health records (EHR) demonstrate promising performance in cancer prognostication. However, gaps in literature include incomplete reporting of model performance, inadequate alignment of model formulation with implementation use-case, and insufficient explainability hindering trust and adoption in clinical settings. Hence, we aim to develop an explainable machine learning EHR-based model that prompts palliative care processes by predicting for 365-day mortality risk among patients with advanced cancer within an outpatient setting. METHODS: Our cohort consisted of 5,926 adults diagnosed with Stage 3 or 4 solid organ cancer between July 1, 2017, and June 30, 2020 and receiving ambulatory cancer care within a tertiary center. The classification problem was modelled using Extreme Gradient Boosting (XGBoost) and aligned to our envisioned use-case: "Given a prediction point that corresponds to an outpatient cancer encounter, predict for mortality within 365-days from prediction point, using EHR data up to 365-days prior." The model was trained with 75% of the dataset (n = 39,416 outpatient encounters) and validated on a 25% hold-out dataset (n = 13,122 outpatient encounters). To explain model outputs, we used Shapley Additive Explanations (SHAP) values. Clinical characteristics, laboratory tests and treatment data were used to train the model. Performance was evaluated using area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC), while model calibration was assessed using the Brier score. RESULTS: In total, 17,149 of the 52,538 prediction points (32.6%) had a mortality event within the 365-day prediction window. The model demonstrated an AUROC of 0.861 (95% CI 0.856-0.867) and AUPRC of 0.771. The Brier score was 0.147, indicating slight overestimations of mortality risk. Explanatory diagrams utilizing SHAP values allowed visualization of feature impacts on predictions at both the global and individual levels. CONCLUSION: Our machine learning model demonstrated good discrimination and precision-recall in predicting 365-day mortality risk among individuals with advanced cancer. It has the potential to provide personalized mortality predictions and facilitate earlier integration of palliative care.


Subject(s)
Electronic Health Records , Machine Learning , Palliative Care , Humans , Machine Learning/standards , Electronic Health Records/statistics & numerical data , Palliative Care/methods , Palliative Care/standards , Palliative Care/statistics & numerical data , Male , Female , Middle Aged , Aged , Risk Assessment/methods , Neoplasms/mortality , Neoplasms/therapy , Cohort Studies , Adult , Medical Oncology/methods , Medical Oncology/standards , Aged, 80 and over , Mortality/trends
SELECTION OF CITATIONS
SEARCH DETAIL
...