1
|
Zachariah FJ, Rossi LA, Roberts LM, Bosserman LD. Prospective Comparison of Medical Oncologists and a Machine Learning Model to Predict 3-Month Mortality in Patients With Metastatic Solid Tumors. JAMA Netw Open 2022; 5:e2214514. [PMID: 35639380 PMCID: PMC9157269 DOI: 10.1001/jamanetworkopen.2022.14514] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 11/17/2021] [Accepted: 03/24/2022] [Indexed: 12/29/2022] Open
Abstract
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.
Collapse
Affiliation(s)
- Finly J. Zachariah
- Department of Supportive Care Medicine, City of Hope National Medical Center, Duarte, California
| | - Lorenzo A. Rossi
- Department of Applied AI and Data Science, City of Hope National Medical Center, Duarte, California
| | - Laura M. Roberts
- Department of Clinical Informatics, City of Hope National Medical Center, Duarte, California
| | - Linda D. Bosserman
- Department of Medical Oncology, City of Hope National Medical Center, Duarte, California
| |
Collapse
|
2
|
White N, Oostendorp LJ, Vickerstaff V, Gerlach C, Engels Y, Maessen M, Tomlinson C, Wens J, Leysen B, Biasco G, Zambrano S, Eychmüller S, Avgerinou C, Chattat R, Ottoboni G, Veldhoven C, Stone P. An online international comparison of palliative care identification in primary care using the Surprise Question. Palliat Med 2022; 36:142-151. [PMID: 34596445 PMCID: PMC8796152 DOI: 10.1177/02692163211048340] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 11/15/2022]
Abstract
BACKGROUND The Surprise Question ('Would I be surprised if this patient died within 12 months?') identifies patients in the last year of life. It is unclear if 'surprised' means the same for each clinician, and whether their responses are internally consistent. AIM To determine the consistency with which the Surprise Question is used. DESIGN A cross-sectional online study of participants located in Belgium, Germany, Italy, The Netherlands, Switzerland and UK. Participants completed 20 hypothetical patient summaries ('vignettes'). Primary outcome measure: continuous estimate of probability of death within 12 months (0% [certain survival]-100% [certain death]). A threshold (probability estimate above which Surprise Question responses were consistently 'no') and an inconsistency range (range of probability estimates where respondents vacillated between responses) were calculated. Univariable and multivariable linear regression explored differences in consistency. Trial registration: NCT03697213. SETTING/PARTICIPANTS Registered General Practitioners (GPs). Of the 307 GPs who started the study, 250 completed 15 or more vignettes. RESULTS Participants had a consistency threshold of 49.8% (SD 22.7) and inconsistency range of 17% (SD 22.4). Italy had a significantly higher threshold than other countries (p = 0.002). There was also a difference in threshold levels depending on age of clinician, for every yearly increase, participants had a higher threshold. There was no difference in inconsistency between countries (p = 0.53). CONCLUSIONS There is variation between clinicians regarding the use of the Surprise Question. Over half of GPs were not internally consistent in their responses to the Surprise Question. Future research with standardised terms and real patients is warranted.
Collapse
Affiliation(s)
- Nicola White
- Marie Curie Palliative Care Research Department, Division of Psychiatry, University College London, London, UK
| | - Linda Jm Oostendorp
- Marie Curie Palliative Care Research Department, Division of Psychiatry, University College London, London, UK
| | - Victoria Vickerstaff
- Marie Curie Palliative Care Research Department, Division of Psychiatry, University College London, London, UK.,Primary Care and Population Health, Institute of Epidemiology and Health Care, University College London, London, UK
| | - Christina Gerlach
- Palliative Care Unit, Department of Oncology, Hematology and BMT, and Pneumology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.,Interdisciplinary Palliative Care Unit, Department of Hematology, Oncology, and Pneumology, University Medical Center, Mainz, Germany
| | - Yvonne Engels
- Anesthesiology, Pain and Palliative Medicine, Radboud University Medical Centre, Nijmegen, The Netherlands
| | - Maud Maessen
- University Center for Palliative Care, Inselspital University Hospital Bern, University of Bern, Bern, Switzerland.,Institute of Social and Preventive Medicine, University of Bern, Bern, Switzerland
| | - Christopher Tomlinson
- Department of Metabolism, Digestion and Reproduction, Faculty of Medicine, Imperial College London, London, UK
| | - Johan Wens
- Department Family Medicine and Population Health (FamPop), Faculty of Medicine and Health Sciences, University of Antwerp, Antwerp, Belgium
| | - Bert Leysen
- Department of Metabolism, Digestion and Reproduction, Faculty of Medicine, Imperial College London, London, UK
| | - Guido Biasco
- Department of Experimental, Diagnostic and Specialty Medicine, University of Bologna & Academy of the Sciences of Palliative Medicine, Bologna, Italy
| | - Sofia Zambrano
- University Center for Palliative Care, Inselspital University Hospital Bern, University of Bern, Bern, Switzerland
| | - Steffen Eychmüller
- University Center for Palliative Care, Inselspital University Hospital Bern, University of Bern, Bern, Switzerland
| | - Christina Avgerinou
- Primary Care and Population Health, Institute of Epidemiology and Health Care, University College London, London, UK
| | - Rabih Chattat
- Department of Psychology, University of Bologna, Bologna, Italy
| | | | - Carel Veldhoven
- Anesthesiology, Pain and Palliative Medicine, Radboud University Medical Centre, Nijmegen, The Netherlands
| | - Patrick Stone
- Marie Curie Palliative Care Research Department, Division of Psychiatry, University College London, London, UK
| |
Collapse
|
3
|
Le K, Lee J, Desai S, Ho A, van Heukelom H. The Surprise Question and Serious Illness Conversations: A pilot study. Nurs Ethics 2021; 28:1010-1025. [PMID: 33686904 DOI: 10.1177/0969733020983392] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/15/2022]
Abstract
BACKGROUND Serious Illness Conversations aim to discuss patient goals. However, on acute medicine units, seriously ill patients may undergo distressing interventions until death. OBJECTIVES To investigate the feasibility of using the Surprise Question, "Would you be surprised if this patient died within the next year?" to identify patients who would benefit from early Serious Illness Conversations and study any changes in the interdisciplinary team's beliefs, confidence, and engagement as a result of asking the Surprise Question. DESIGN A prospective cohort pilot study with two Plan-Do-Study-Act cycles. PARTICIPANTS/CONTEXT Fifty-eight healthcare professionals working on Acute Medicine Units participated in pre- and post-intervention questionnaires. The intervention involved asking participants the Surprise Question for each patient. Patient charts were reviewed for Serious Illness Conversation documentation. ETHICAL CONSIDERATIONS Ethical approval was granted by the institutions involved. FINDINGS Equivocal overall changes in the beliefs, confidence, and engagement of healthcare professionals were observed. Six out of 23 patients were indicated as needing a Serious Illness Conversation; chart review provided some evidence that these patients had more Serious Illness Conversation documentation compared with the 17 patients not flagged for a Serious Illness Conversation. Issues were identified in equating the Surprise Question to a Serious Illness Conversation. DISCUSSION Appropriate support for seriously ill patients is both a nursing professional and ethical duty. Flagging patients for conversations may act as a filtering process, allowing healthcare professionals to focus on conversations with patients who need them most. There are ethical and practical issues as to what constitutes a "serious illness" and if answering "no" to the Surprise Question always equates to a conversation. CONCLUSION The barriers of time constraints and lack of training call for institutional change in order to prioritise the moral obligation of Serious Illness Conversations.
Collapse
Affiliation(s)
| | - Jenny Lee
- 102794Providence Health Care, Canada
| | - Sameer Desai
- Centre for Health Evaluation and Outcome Sciences, Canada
| | - Anita Ho
- 8166University of British Columbia, Canada; University of California San Francisco, USA; Centre for Health Evaluation and Outcome Sciences, Canada
| | | |
Collapse
|
4
|
White N, Oostendorp L, Vickerstaff V, Gerlach C, Engels Y, Maessen M, Tomlinson C, Wens J, Leysen B, Biasco G, Zambrano S, Eychmüller S, Avgerinou C, Chattat R, Ottoboni G, Veldhoven C, Stone P. Correction to: An online international comparison of thresholds for triggering a negative response to the “Surprise Question”: a study protocol. Palliat Care 2020; 19:15. [PMID: 31987055 PMCID: PMC6986139 DOI: 10.1186/s12904-020-0518-2] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/10/2022] Open
|