{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T23:22:36Z","timestamp":1785367356955,"version":"3.55.0"},"reference-count":40,"publisher":"SAGE Publications","issue":"6","license":[{"start":{"date-parts":[[2025,7,4]],"date-time":"2025-07-04T00:00:00Z","timestamp":1751587200000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2025,7,4]],"date-time":"2025-07-04T00:00:00Z","timestamp":1751587200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"funder":[{"DOI":"10.13039\/100000054","name":"National Cancer Institute","doi-asserted-by":"publisher","award":["K08CA263541"],"award-info":[{"award-number":["K08CA263541"]}],"id":[{"id":"10.13039\/100000054","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["Med Decis Making"],"published-print":{"date-parts":[[2025,8]]},"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Machine learning (ML) algorithms may improve the prognosis for serious illnesses such as cancer, identifying patients who may benefit from earlier palliative care (PC) or advance care planning (ACP). We evaluated the impact of various presentation strategies of a hypothetical ML algorithm on clinician prognostic accuracy and decision making.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This was a randomized clinical vignette survey study among medical oncologists who treat metastatic non-small-cell lung cancer (mNSCLC). Between March and June 2023, clinicians were shown 3 vignettes of patients presenting with mNSCLC. The vignettes varied by prognostic risk, as defined from the Lung Cancer Prognostic Index (LCPI). Clinicians estimated life expectancy in months and made recommendations about PC and ACP. Clinicians were then shown the same vignette with a hypothetical survival estimate from a black-box ML algorithm; clinicians were randomized to receive the ML prediction using absolute and\/or reference-dependent prognostic estimates. The primary outcome was prognostic accuracy relative to the LCPI.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Among 51 clinicians with complete responses, the median years in practice was 7 (interquartile range 3.5\u201319), 14 (27.5%) were female, 23 (45.1%) practiced in a community oncology setting, and baseline accuracy was 54.9% (95% confidence interval [CI] 47.0\u201362.8) across all vignettes. ML presentation improved accuracy (mean change relative to baseline 20.9%, 95% CI 13.9\u201327.9,\n                      <jats:italic toggle=\"yes\">P<\/jats:italic>\n                      \u2009&lt;\u20090.001). ML outputs using an absolute presentation strategy alone (mean change 27.4%, 95% 16.8\u201338.1,\n                      <jats:italic toggle=\"yes\">P<\/jats:italic>\n                      \u2009&lt;\u20090.001) or with reference dependence (mean change 33.4%, 95% 23.9\u201342.8,\n                      <jats:italic toggle=\"yes\">P<\/jats:italic>\n                      \u2009&lt;\u20090.001) improved accuracy, but reference dependence alone did not (mean change 2.0% [95% CI \u221211.1 to 15.0],\n                      <jats:italic toggle=\"yes\">P<\/jats:italic>\n                      \u2009=\u20090.77). ML presentation did not change the rates of recommending ACP nor PC referral (mean change 1.3% and 0.7%, respectively).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Limitations<\/jats:title>\n                    <jats:p>The singular use case of prognosis in mNSCLC, low initial response rate.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>ML-based assessments may improve prognostic accuracy but not result in changed decision making.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Implications<\/jats:title>\n                    <jats:p>ML prognostic algorithms prioritizing explainability and absolute prognoses may have greater impact on clinician decision making.<\/jats:p>\n                    <jats:p>Trial Registration: CT.gov: NCT06463977<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Highlights<\/jats:title>\n                    <jats:list list-type=\"bullet\">\n                      <jats:list-item>\n                        <jats:p>While machine learning (ML) algorithms may accurately predict mortality, the impact of prognostic ML on clinicians\u2019 prognostic accuracy and decision making and optimal presentation strategies for ML outputs are unclear.<\/jats:p>\n                      <\/jats:list-item>\n                      <jats:list-item>\n                        <jats:p>In this multicenter randomized survey study among vignettes of patients with advanced cancer, prognostic accuracy improved by 20.9% when clinicians reviewed vignettes with a hypothetical ML mortality risk prediction, with absolute risk presentation strategies resulting in greater accuracy gains than reference-dependent presentations alone.<\/jats:p>\n                      <\/jats:list-item>\n                      <jats:list-item>\n                        <jats:p>However, ML presentation did not change the rates of recommending advance care planning or palliative care referral (1.3% and 0.7%, respectively).<\/jats:p>\n                      <\/jats:list-item>\n                      <jats:list-item>\n                        <jats:p>ML-based prognostic assessments without explanations improve prognostic accuracy but do not change decisions around palliative care referral or advance care planning.<\/jats:p>\n                      <\/jats:list-item>\n                    <\/jats:list>\n                  <\/jats:sec>","DOI":"10.1177\/0272989x251349489","type":"journal-article","created":{"date-parts":[[2025,7,4]],"date-time":"2025-07-04T06:27:43Z","timestamp":1751610463000},"page":"690-702","update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":1,"title":["The Impact of Machine Learning Mortality Risk Prediction on Clinician Prognostic Accuracy and Decision Support: A Randomized Vignette Study"],"prefix":"10.1177","volume":"45","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2692-6306","authenticated-orcid":false,"given":"Ravi B.","family":"Parikh","sequence":"first","affiliation":[{"name":"Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA"},{"name":"Department of Medical Ethics and Health Policy, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA"},{"name":"Penn Center for Cancer Care Innovation, Abramson Cancer Center, University of Pennsylvania, Philadelphia, PA, USA"},{"name":"Corporal Michael J. Crescenz VA Medical Center, Philadelphia, PA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9966-1973","authenticated-orcid":false,"given":"William J.","family":"Ferrell","sequence":"additional","affiliation":[{"name":"Department of Medical Ethics and Health Policy, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA"},{"name":"Penn Center for Cancer Care Innovation, Abramson Cancer Center, University of Pennsylvania, Philadelphia, PA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anthony","family":"Girard","sequence":"additional","affiliation":[{"name":"Department of Medical Ethics and Health Policy, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jenna","family":"White","sequence":"additional","affiliation":[{"name":"Department of Medical Ethics and Health Policy, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA"},{"name":"Penn Center for Cancer Care Innovation, Abramson Cancer Center, University of Pennsylvania, Philadelphia, PA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sophia","family":"Fang","sequence":"additional","affiliation":[{"name":"Department of Operations Research and Financial Engineering, Princeton University, Princeton, NJ, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Justin E.","family":"Bekelman","sequence":"additional","affiliation":[{"name":"Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA"},{"name":"Department of Medical Ethics and Health Policy, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA"},{"name":"Penn Center for Cancer Care Innovation, Abramson Cancer Center, University of Pennsylvania, Philadelphia, PA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3775-9462","authenticated-orcid":false,"given":"Marilyn M.","family":"Schapira","sequence":"additional","affiliation":[{"name":"Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA"},{"name":"Penn Center for Cancer Care Innovation, Abramson Cancer Center, University of Pennsylvania, Philadelphia, PA, USA"},{"name":"Corporal Michael J. Crescenz VA Medical Center, Philadelphia, PA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"179","published-online":{"date-parts":[[2025,7,4]]},"reference":[{"key":"e_1_3_3_2_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-018-0300-7"},{"key":"e_1_3_3_3_2","doi-asserted-by":"publisher","DOI":"10.1097\/CCM.0b013e3181fb7b1c"},{"key":"e_1_3_3_4_2","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.320.7233.469"},{"key":"e_1_3_3_5_2","doi-asserted-by":"publisher","DOI":"10.1177\/0269216314526272"},{"key":"e_1_3_3_6_2","doi-asserted-by":"publisher","DOI":"10.1001\/jamainternmed.2014.5271"},{"key":"e_1_3_3_7_2","doi-asserted-by":"publisher","DOI":"10.1001\/jama.300.14.1665"},{"key":"e_1_3_3_8_2","doi-asserted-by":"publisher","DOI":"10.1056\/NEJMsb1305469"},{"key":"e_1_3_3_9_2","doi-asserted-by":"publisher","DOI":"10.1001\/jamaoncol.2015.1976"},{"key":"e_1_3_3_10_2","unstructured":"Committee on Approaching Death: Addressing Key End of Life Issues Institute of Medicine. Dying in America: Improving Quality and Honoring Individual Preferences Near the End of Life. National Academies Press (US); 2015. http:\/\/www.ncbi.nlm.nih.gov\/books\/NBK285681\/ [Accessed September 29 2022]"},{"key":"e_1_3_3_11_2","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0226597"},{"key":"e_1_3_3_12_2","doi-asserted-by":"publisher","DOI":"10.1177\/107327481502200415"},{"key":"e_1_3_3_13_2","doi-asserted-by":"publisher","DOI":"10.1200\/JCO.2015.61.9239"},{"issue":"20","key":"e_1_3_3_14_2","article-title":"The study to understand prognoses and preferences for outcomes and risks of treatments (SUPPORT). The SUPPORT Principal Investigators","volume":"274","author":"A controlled trial to improve care for seriously ill hospitalized patients","year":"1995","unstructured":"A controlled trial to improve care for seriously ill hospitalized patients. The study to understand prognoses and preferences for outcomes and risks of treatments (SUPPORT). The SUPPORT Principal Investigators. JAMA. 1995;274(20):1591\u20138.","journal-title":"JAMA"},{"key":"e_1_3_3_15_2","doi-asserted-by":"publisher","DOI":"10.1001\/jamaoncol.2022.6303"},{"key":"e_1_3_3_16_2","doi-asserted-by":"publisher","DOI":"10.1200\/OP.22.00128"},{"key":"e_1_3_3_17_2","doi-asserted-by":"publisher","DOI":"10.1111\/hex.12196"},{"key":"e_1_3_3_18_2","doi-asserted-by":"publisher","DOI":"10.1089\/jpm.2022.0095"},{"key":"e_1_3_3_19_2","doi-asserted-by":"publisher","DOI":"10.1007\/s00520-021-06774-w"},{"key":"e_1_3_3_20_2","doi-asserted-by":"publisher","DOI":"10.1001\/jama.2023.22295"},{"key":"e_1_3_3_21_2","doi-asserted-by":"publisher","DOI":"10.1016\/bs.hesbe.2018.07.003"},{"key":"e_1_3_3_22_2","doi-asserted-by":"publisher","DOI":"10.2307\/1914185"},{"key":"e_1_3_3_23_2","doi-asserted-by":"publisher","DOI":"10.1038\/bjc.2017.232"},{"key":"e_1_3_3_24_2","doi-asserted-by":"publisher","DOI":"10.1200\/JCO.2016.70.1474"},{"key":"e_1_3_3_25_2","doi-asserted-by":"publisher","DOI":"10.1056\/NEJMoa1000678"},{"key":"e_1_3_3_26_2","doi-asserted-by":"publisher","DOI":"10.1200\/EDBK_175474"},{"key":"e_1_3_3_27_2","doi-asserted-by":"publisher","DOI":"10.1200\/EDBK_201211"},{"key":"e_1_3_3_28_2","doi-asserted-by":"publisher","DOI":"10.1136\/bmjopen-2023-081661"},{"key":"e_1_3_3_29_2","doi-asserted-by":"publisher","DOI":"10.1089\/jpm.2019.0580"},{"key":"e_1_3_3_30_2","doi-asserted-by":"publisher","DOI":"10.1377\/hlthaff.2014.1055"},{"key":"e_1_3_3_31_2","doi-asserted-by":"publisher","DOI":"10.3102\/10769986211017480"},{"key":"e_1_3_3_32_2","doi-asserted-by":"publisher","DOI":"10.1080\/00031305.1992.10475842"},{"key":"e_1_3_3_33_2","doi-asserted-by":"publisher","DOI":"10.1201\/b14835-21"},{"key":"e_1_3_3_34_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-024-02850-w"},{"key":"e_1_3_3_35_2","doi-asserted-by":"crossref","unstructured":"Agarwal N Moehring A Rajpurkar P Salz T. Combining human expertise with artificial intelligence: experimental evidence from radiology. National Bureau of Economic Research Working Paper. 2023. NBER Working Paper No. w31422. https:\/\/ssrn.com\/abstract=4505053","DOI":"10.3386\/w31422"},{"key":"e_1_3_3_36_2","doi-asserted-by":"publisher","DOI":"10.1038\/s43856-021-00069-1"},{"key":"e_1_3_3_37_2","doi-asserted-by":"publisher","DOI":"10.1200\/JCO.2022.40.28_suppl.202"},{"key":"e_1_3_3_38_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-023-31404-2"},{"key":"e_1_3_3_39_2","doi-asserted-by":"publisher","DOI":"10.1007\/s11606-006-0075-2"},{"key":"e_1_3_3_40_2","doi-asserted-by":"publisher","DOI":"10.1145\/3313831.3376590"},{"key":"e_1_3_3_41_2","doi-asserted-by":"publisher","DOI":"10.1200\/JOP.18.00516"}],"container-title":["Medical Decision Making"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/0272989X251349489","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/0272989X251349489","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/0272989X251349489","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T18:03:32Z","timestamp":1777658612000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/0272989X251349489"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,4]]},"references-count":40,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2025,8]]}},"alternative-id":["10.1177\/0272989X251349489"],"URL":"https:\/\/doi.org\/10.1177\/0272989x251349489","relation":{},"ISSN":["0272-989X","1552-681X"],"issn-type":[{"value":"0272-989X","type":"print"},{"value":"1552-681X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,4]]}}}