Prospective Comparison of Medical Oncologists and a Machine Learning Model to Predict 3-Month Mortality in Patients With Metastatic Solid Tumors.

Importance: To date, oncologist and model prognostic performance have been assessed independently and mostly retrospectively; however, how model prognostic performance compares with oncologist prognostic performance prospectively remains unknown. Objective: To compare oncologist performance with a model in predicting 3-month mortality for patients with metastatic solid tumors in an outpatient setting. Design, Setting, and Participants: This prognostic study evaluated prospective predictions for a cohort of patients with metastatic solid tumors seen in outpatient oncology clinics at a National Cancer Institute-designated cancer center and associated satellites between December 6, 2019, and August 6, 2021. Oncologists (57 physicians and 17 advanced practice clinicians) answered a 3-month surprise question (3MSQ) within clinical pathways. A model was trained with electronic health record data from January 1, 2013, to April 24, 2019, to identify patients at high risk of 3-month mortality and deployed silently in October 2019. Analysis was limited to oncologist prognostications with a model prediction within the preceding 30 days. Exposures: Three-month surprise question and gradient-boosting binary classifier. Main Outcomes and Measures: The primary outcome was performance comparison between oncologists and the model to predict 3-month mortality. The primary performance metric was the positive predictive value (PPV) at the sensitivity achieved by the medical oncologists with their 3MSQ answers. Results: A total of 74 oncologists answered 3099 3MSQs for 2041 patients with advanced cancer (median age, 62.6 [range, 18-96] years; 1271 women [62.3%]). In this cohort with a 15% prevalence of 3-month mortality and 30% sensitivity for both oncologists and the model, the PPV of oncologists was 34.8% (95% CI, 30.1%-39.5%) and the PPV of the model was 60.0% (95% CI, 53.6%-66.3%). Area under the receiver operating characteristic curve for the model was 81.2% (95% CI, 79.1%-83.3%). The model significantly outperformed the oncologists in short-term mortality. Conclusions and Relevance: In this prognostic study, the model outperformed oncologists overall and within the breast and gastrointestinal cancer cohorts in predicting 3-month mortality for patients with advanced cancer. These findings suggest that further studies may be useful to examine how model predictions could improve oncologists' prognostic confidence and patient-centered goal-concordant care at the end of life.

Adoption of Patient-Generated Health Data in Oncology: A Report From the NCCN EHR Oncology Advisory Group.

BACKGROUND: Collecting, monitoring, and responding to patient-generated health data (PGHD) are associated with improved quality of life and patient satisfaction, and possibly with improved patient survival in oncology. However, the current state of adoption, types of PGHD collected, and degree of integration into electronic health records (EHRs) is unknown. METHODS: The NCCN EHR Oncology Advisory Group formed a Patient-Reported Outcomes (PRO) Workgroup to perform an assessment and provide recommendations for cancer centers, researchers, and EHR vendors to advance the collection and use of PGHD in oncology. The issues were evaluated via a survey of NCCN Member Institutions. Questions were designed to assess the current state of PGHD collection, including how, what, and where PGHD are collected. Additionally, detailed questions about governance and data integration into EHRs were asked. RESULTS: Of 28 Member Institutions surveyed, 23 responded. The collection and use of PGHD is widespread among NCCN Members Institutions (96%). Most centers (90%) embed at least some PGHD into the EHR, although challenges remain, as evidenced by 88% of respondents reporting the use of instruments not integrated. Forty-seven percent of respondents are leveraging PGHD for process automation and adherence to best evidence. Content type and integration touchpoints vary among the members, as well as governance maturity. CONCLUSIONS: The reported variability regarding PGHD suggests that it may not yet have reached its full potential for oncology care delivery. As the adoption of PGHD in oncology continues to expand, opportunities exist to enhance their utility. Among the recommendations for cancer centers is establishment of a governance process that includes patients. Researchers should consider determining which PGHD instruments confer the highest value. It is recommended that EHR vendors collaborate with cancer centers to develop solutions for the collection, interpretation, visualization, and use of PGHD.

Cancer-Related Fatigue, Version 2.2015.

Cancer-related fatigue is defined as a distressing, persistent, subjective sense of physical, emotional, and/or cognitive tiredness or exhaustion related to cancer or cancer treatment that is not proportional to recent activity and interferes with usual functioning. It is one of the most common side effects in patients with cancer. Fatigue has been shown to be a consequence of active treatment, but it may also persist into posttreatment periods. Furthermore, difficulties in end-of-life care can be compounded by fatigue. The NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) for Cancer-Related Fatigue provide guidance on screening for fatigue and recommendations for interventions based on the stage of treatment. Interventions may include education and counseling, general strategies for the management of fatigue, and specific nonpharmacologic and pharmacologic interventions. Fatigue is a frequently underreported complication in patients with cancer and, when reported, is responsible for reduced quality of life. Therefore, routine screening to identify fatigue is an important component in improving the quality of life for patients living with cancer.