Ravi B Parikh1,2,3,4, Christopher R Manz5, Maria N Nelson6, Chalanda N Evans6,7, Susan H Regli8, Nina O'Connor6,9,8, Lynn M Schuchter6,9,8, Lawrence N Shulman6,9,8, Mitesh S Patel6,9,10,8,7,11, Joanna Paladino12, Judy A Shea6. 1. Perelman School of Medicine, University of Pennsylvania, 423 Guardian Drive, Blockley 1102, Philadelphia, PA, 19104, USA. Ravi.Parikh@pennmedicine.upenn.edu. 2. Abramson Cancer Center, University of Pennsylvania, Philadelphia, PA, USA. Ravi.Parikh@pennmedicine.upenn.edu. 3. Corporal Michael J. Crescenz Veterans Affairs Medical Center, Philadelphia, PA, USA. Ravi.Parikh@pennmedicine.upenn.edu. 4. University of Pennsylvania Health System, Philadelphia, PA, USA. Ravi.Parikh@pennmedicine.upenn.edu. 5. Dana Farber Cancer Institute, Boston, MA, USA. 6. Perelman School of Medicine, University of Pennsylvania, 423 Guardian Drive, Blockley 1102, Philadelphia, PA, 19104, USA. 7. Penn Medicine Nudge Unit, Philadelphia, PA, USA. 8. University of Pennsylvania Health System, Philadelphia, PA, USA. 9. Abramson Cancer Center, University of Pennsylvania, Philadelphia, PA, USA. 10. Corporal Michael J. Crescenz Veterans Affairs Medical Center, Philadelphia, PA, USA. 11. Wharton School of the University of Pennsylvania, Philadelphia, PA, USA. 12. Ariadne Labs, Brigham and Women's Hospital & Harvard Chan School of Public Health, Boston, MA, USA.
Abstract
PURPOSE: Oncologists may overestimate prognosis for patients with cancer, leading to delayed or missed conversations about patients' goals and subsequent low-quality end-of-life care. Machine learning algorithms may accurately predict mortality risk in cancer, but it is unclear how oncology clinicians would use such algorithms in practice. METHODS: The purpose of this qualitative study was to assess oncology clinicians' perceptions on the utility and barriers of machine learning prognostic algorithms to prompt advance care planning. Participants included medical oncology physicians and advanced practice providers (APPs) practicing in tertiary and community practices within a large academic healthcare system. Transcripts were coded and analyzed inductively using NVivo software. RESULTS: The study included 29 oncology clinicians (19 physicians, 10 APPs) across 6 practice sites (1 tertiary, 5 community) in the USA. Fourteen participants had previously had exposure to an automated machine learning-based prognostic algorithm as part of a pragmatic randomized trial. Clinicians believed that there was utility for algorithms in validating their own intuition about prognosis and prompting conversations about patient goals and preferences. However, this enthusiasm was tempered by concerns about algorithm accuracy, over-reliance on algorithm predictions, and the ethical implications around disclosure of an algorithm prediction. There was significant variation in tolerance for false positive vs. false negative predictions. CONCLUSION: While oncologists believe there are applications for advanced prognostic algorithms in routine care of patients with cancer, they are concerned about algorithm accuracy, confirmation and automation biases, and ethical issues of prognostic disclosure.
PURPOSE: Oncologists may overestimate prognosis for patients with cancer, leading to delayed or missed conversations about patients' goals and subsequent low-quality end-of-life care. Machine learning algorithms may accurately predict mortality risk in cancer, but it is unclear how oncology clinicians would use such algorithms in practice. METHODS: The purpose of this qualitative study was to assess oncology clinicians' perceptions on the utility and barriers of machine learning prognostic algorithms to prompt advance care planning. Participants included medical oncology physicians and advanced practice providers (APPs) practicing in tertiary and community practices within a large academic healthcare system. Transcripts were coded and analyzed inductively using NVivo software. RESULTS: The study included 29 oncology clinicians (19 physicians, 10 APPs) across 6 practice sites (1 tertiary, 5 community) in the USA. Fourteen participants had previously had exposure to an automated machine learning-based prognostic algorithm as part of a pragmatic randomized trial. Clinicians believed that there was utility for algorithms in validating their own intuition about prognosis and prompting conversations about patient goals and preferences. However, this enthusiasm was tempered by concerns about algorithm accuracy, over-reliance on algorithm predictions, and the ethical implications around disclosure of an algorithm prediction. There was significant variation in tolerance for false positive vs. false negative predictions. CONCLUSION: While oncologists believe there are applications for advanced prognostic algorithms in routine care of patients with cancer, they are concerned about algorithm accuracy, confirmation and automation biases, and ethical issues of prognostic disclosure.
Authors: Benjamin Chastek; Carolyn Harley; Joel Kallich; Lee Newcomer; Carly J Paoli; April H Teitelbaum Journal: J Oncol Pract Date: 2012-07-03 Impact factor: 3.840
Authors: Craig C Earle; Bridget A Neville; Mary Beth Landrum; John Z Ayanian; Susan D Block; Jane C Weeks Journal: J Clin Oncol Date: 2004-01-15 Impact factor: 44.544
Authors: Craig C Earle; Mary Beth Landrum; Jeffrey M Souza; Bridget A Neville; Jane C Weeks; John Z Ayanian Journal: J Clin Oncol Date: 2008-08-10 Impact factor: 44.544
Authors: Alexi A Wright; Baohui Zhang; Alaka Ray; Jennifer W Mack; Elizabeth Trice; Tracy Balboni; Susan L Mitchell; Vicki A Jackson; Susan D Block; Paul K Maciejewski; Holly G Prigerson Journal: JAMA Date: 2008-10-08 Impact factor: 56.272
Authors: Timothy Gilligan; Nessa Coyle; Richard M Frankel; Donna L Berry; Kari Bohlke; Ronald M Epstein; Esme Finlay; Vicki A Jackson; Christopher S Lathan; Charles L Loprinzi; Lynne H Nguyen; Carole Seigel; Walter F Baile Journal: J Clin Oncol Date: 2017-09-11 Impact factor: 44.544