The erosion of professional ethics in medicine has severe consequences for patients and society. Existing approaches often rely on retrospective analysis and lack the precision and timeliness needed to effectively identify and mitigate risks. Although patient online reviews offer a unique opportunity to proactively detect ethical issues by providing candid, unsolicited feedback on healthcare experiences, few studies have empirically established the link between patient reviews and ethical breaches in medicine. This research introduces a novel machine learning framework to derive text-based indicators of physicians’ professional ethics using online patient reviews. Our approach leverages large language models to extract ethics-related comments and employs few-shot contrastive learning to train multilabel classifiers. Empirical validation studies suggest that the ethical indicators can help predict a wide range of adverse outcomes including drug-related deaths, disciplinary actions, malpractice claims, and rent-seeking behaviors. Our framework offers promising avenues for proactively managing ethical risks in healthcare and other professional services.
AbbasiALiJAdjerohD, et al. (2019) Don’t mention it? Analyzing user-generated content signals for early adverse event warnings. Information Systems Research30(3): 1007–1028.
2.
AbrahamsASFanWWangGA, et al. (2015) An integrated text analytic framework for product defect discovery. Production and Operations Management24(6): 975–990.
3.
ArazOMChoiTOlsonDL, et al. (2020) Role of analytics for operational risk management in the era of big data. Decision Sciences51(6): 1320–1346.
4.
BandmanB (2003) The Moral Development of Health Care Professionals: Rational Decision Making in Health Care Ethics. Westport, CT: Praeger.
5.
BarreLGalloJMcDonaldJV (2019) Review of disciplinary actions regarding controlled substances, Rhode Island 2012–2017. Journal of Medical Regulation105(1): 22–27.
6.
BastaniHGohJBayatiM (2019) Evidence of upcoding in pay-for-performance programs. Management Science65(3): 1042–1060.
7.
BauderRKhoshgoftaarT (2018) Medicare fraud detection using random forest with class imbalanced big data. In: 2018 IEEE International Conference on Information Reuse and Integration (IRI), 06–09 July 2018, Salt Lake City, UT, pp. 80–87.
8.
BeauchampTL (2003) Methods and principles in biomedical ethics. Journal of Medical Ethics29(5): 269–274.
9.
BeckerGS (1968) Crime and punishment: An economic approach. Journal of Political Economy76(2): 169–217.
10.
BlenkinsoppJSnowdenNMannionR, et al. (2019) Whistleblowing over patient safety and care quality: A review of the literature. Journal of Health Organization and Management33(6): 737–756.
11.
BobroskeKFreemanMHuanL, et al. (2022) Curbing the opioid epidemic at its root: The effect of provider discordance after opioid initiation. Management Science68(3): 2003–2015.
12.
BrennanTARothmanDJBlankL, et al. (2006) Health industry practices that create conflicts of interest: A policy proposal for academic medical centers. Journal of the American Medical Directors Association295(4): 429–433.
13.
BrownTMannBRyderN, et al. (2020) Language models are few-shot learners. Advances in Neural Information Processing Systems33: 1877–1901.
14.
BuschRS (2012) Healthcare Fraud: Auditing and Detection Guide. Hoboken, NJ: John Wiley & Sons.
15.
ChadwickR (2016) Professional ethics. In: CraigE (ed) Routledge Encyclopedia of Philosophy (1st ed.). London:Routledge, pp. 314–315.
16.
ChalfinADanieliOHillisA, et al. (2016) Productivity and selection of human capital with machine learning. American Economic Review106(5): 124–127.
17.
ChandrasekaranASenotCBoyerKK (2012) Process management impact on clinical and experiential quality: Managing tensions between safe and patient-centered healthcare. Manufacturing & Service Operations Management14(4): 548–566.
18.
ChenTGuestrinC (2016) XGBoost: A scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco, CA: Association for Computing Machinery, pp.785–794.
19.
ChristianB (2020) The Alignment Problem: Machine Learning and Human Values. New York, NY: WW Norton & Company .
20.
CohenMAKunreutherH (2007) Operations Risk Management: Overview of Paul Kleindorfer’s contributions. Production and Operations Management16(5): 525–541.
21.
CohenMC (2018) Big data and service operations. Production and Operations Management27(9): 1709–1723.
22.
ColemanCChouEKatz-SamuelsJ, et al. (2022) Similarity search for efficient active learning and search of rare concepts. AAAI36(6): 6402–6410.
23.
CuiRGallinoSMorenoA, et al. (2018) The operational value of social media information. Production and Operations Management27(10): 1749–1769.
24.
De BockKWCoussementKCaignyAD, et al. (2023) Explainable AI for operational research: A defining framework, methods, applications, and a research agenda. European Journal of Operational Research317(2): 249–272.
25.
DegenhardtLGrebelyJStoneJ, et al. (2019) Global patterns of opioid use and dependence: Harms to populations, interventions, and future action. Lancet394(10208): 1560–1579.
26.
DevlinJChangM-WLeeK, et al. (2019) BERT: Pre-training of deep bidirectional transformers for language understanding. In: Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers). Minneapolis, MN: Association for Computational Linguistics, pp. 4171–4186.
27.
DuBoisJMAndersonEEChibnallJT, et al. (2019) Serious ethical violations in medicine: A statistical and ethical analysis of 280 cases in the United States from 2008–2016. American Journal of Bioethics19(1): 16–34.
28.
DuBoisJMChibnallJTAndersonEE, et al. (2016) A mixed-method analysis of reports on 100 cases of improper prescribing of controlled substances. Journal of Drug Issues46(4): 457–472.
29.
EkinTFrigauLConversanoC (2021) Health care fraud classifiers in practice. Applied Stochastic Models in Business and Industry37(6): 1182–1199.
30.
FineschiVTurillazziECateniC (1997) The new Italian code of medical ethics. Journal of Medical Ethics23(4): 239–244.
31.
GalUHansenSLeeA (2022) Research perspectives: Toward theoretical rigor in ethical analysis: The case of algorithmic decision-making systems. Journal of the Association for Information Systems23(6): 1634–1661.
32.
Gallup (2023) Nurses retain top ethics rating in U.S., but below 2020 high. Gallup.com. Available at https://news.gallup.com/poll/467804/nurses-retain-top-ethics-rating-below-2020-high.aspx (accessed 22 November 2023).
33.
GaoGGreenwoodBNAgarwalR, et al. (2015) Vocal minority and silent majority: How do online ratings reflect population perceptions of quality. MIS Quarterly39(3): 565–590.
34.
GarbuttGDaviesP (2011) Should the practice of medicine be a deontological or utilitarian enterprise?Journal of Medical Ethics37(5): 267–270.
35.
GilesTHungermanDOostromT (2023) Opiates of the Masses? Deaths of Despair and the Decline of American Religion (No. w30840). Cambridge, MA: National Bureau of Economic Research.
36.
GourAAggarwalSKumarS (2022) Lending ears to unheard voices: An empirical analysis of user-generated content on social media. Production and Operations Management31(6): 2457–2476.
37.
HamCAlbertiKGMM (2002) The medical profession, the public, and the government. The British Medical Journal324(7341): 838–842.
38.
HaoH (2015) The development of online doctor reviews in China: An analysis of the largest online doctor review website in China. Journal of Medical Internet Research17(6): e134.
39.
HarcourtBE (2007) Against Prediction: Profiling, Policing, and Punishing in an Actuarial Age. Chicago, IL: University of Chicago Press.
40.
HarveyJHeinekeJLewisM (2016) Editorial for journal of operations management special issue on “professional service operations management (PSOM)”. Journal of Operations Management42–43(1): 4–8.
41.
HerlandMBauderRAKhoshgoftaarTM (2019) The effects of class rarity on the evaluation of supervised healthcare fraud detection models. Journal of Big Data6(1): 1263. DOI: https://doi.org/10.1186/s40537-019-0181-8.
42.
HerlandMKhoshgoftaarTMBauderRA (2018) Big data fraud detection using multiple medicare data sources. Journal of Big Data5(1): 1–21.
43.
HustedGLHustedJH (2005) Ethics: A patient-centered approach. In: DalyJ (ed) Professional Nursing: Concepts, Issues, and Challenges. New York, NY: Springer Publishing Company, 175–179.
44.
JanssenAZhangX (2023) Retail pharmacies and drug diversion during the opioid epidemic. American Economic Review113(1): 1–33.
45.
JiangSGarnettRMoseleyB (2019) Cost effective active search. Advances in Neural Information Processing Systems 32: 3880–3889.
46.
JimenezMDFosterD (1998) The importance of compliance programs for the health care industry. University of Miami Business Law Review7: 503.
47.
JoglekarNRDaviesJAndersonEG (2016) The role of industry studies and public policies in production and operations management. Production and Operations Management25(12): 1977–2001.
KapteinM (2008) Developing a measure of unethical behavior in the workplace: A stakeholder perspective. Journal of Management34(5): 978–1008.
50.
KcDKimTTLiuJ (2022) Electronic prescription monitoring and the opioid epidemic. Production and Operations Management31(11): 4057–4074.
51.
KcDSScholtesSTerwieschC (2020) Empirical research in healthcare operations: Past research, present understanding, and future opportunities. Manufacturing & Service Operations Management22(1): 73–83.
52.
KoDMaiFShanZ, et al. (2019) Operational efficiency and patient-centered health care: A view from online physician reviews. Journal of Operations Management65(4): 353–379.
KraussCDoXAHuckN (2017) Deep neural networks, gradient-boosted trees, random forests: Statistical arbitrage on the S&P 500. European Journal of Operational Research259(2): 689–702.
55.
KumarARaghunathanAJonesRM, et al. (2021) Fine-tuning can distort pretrained features and underperform out-of-distribution. International Conference on Learning Representations.
56.
KumaraswamyNMarkeyMKEkinT, et al. (2022) Healthcare fraud data mining methods: A look back and look ahead. Perspectives in Health Information Management19(1): 1i.
57.
LantzyS.AndersonD.. 2020. Can consumers use online reviews to avoid unsuitable doctors? Evidence from RateMDs.com and the Federation of State Medical Boards. Decision Science51(4): 962–984.
58.
LiRTobeyMMayorgaME, et al. (2023) Detecting human trafficking: Automated classification of online customer reviews of massage businesses. Manufacturing & Service Operations Management25(3): 1051–1065.
59.
LiXGuoPLianZ (2016) Quality-speed competition in customer-intensive services with boundedly rational customers. Production Operation Management25(11): 1885–1901.
60.
LiuRHuangJZhangZ (2023) Tracking disclosure change trajectories for financial fraud detection. Production and Operation Management32(2): 584–602.
61.
LuSFRuiH (2018) Can we trust online physician ratings? Evidence from cardiac surgeons in Florida. Management Science64(6): 2557–2573.
62.
LuceyCSoubaW (2010) Perspective: The problem with the problem of professionalism. Academy Medical85(6): 1018–1024.
63.
LundbergSMLeeS-I (2017) A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems30: 4768–4777.
64.
LuoF.2021. State-level economic costs of opioid use disorder and fatal opioid overdose — United States, 2017. Morbidity and Mortality Weekly Report70.
65.
MacyB (2018) Dopesick: Dealers, Doctors and the Drug Company that Addicted America. New York, NY: Bloomsbury Publishing.
66.
ManaryMPBouldingWStaelinR, et al. (2013) The patient experience and health outcomes. New England Journal of Medicine368(3): 201–203.
67.
MandalJPonnambathDKParijaSC (2016) Utilitarian and deontological ethics in medicine. Tropical Parasitology6(1): 5–7.
68.
MarkouPCorstenD (2021) Financial and operational risk management: Inventory effects in the gold mining industry. Production and Operation Management30(12): 4635–4655.
MejiaJMankadSGopalA (2021) Service quality using text mining: Measurement and consequences. Manufacturing & Service Operations Management23(6): 1354–1372.
71.
MikolovTSutskeverIChenK, et al. (2013) Distributed representations of words and phrases and their compositionality. Advances in Neural Information Processing Systems26: 3111–3119.
72.
MohriMRostamizadehATalwalkarA (2018) Foundations of Machine Learning, 2nd edn.Cambridge, MA: The MIT Press.
73.
NairANicolaeMNarasimhanR (2013) Examining the impact of clinical quality and clinical flexibility on cardiology unit performance—Does experiential quality act as a specialized complementary asset?Journal of Operation Management31(7–8): 505–522.
74.
NEJM Catalyst (2018) What Is Risk Management in Healthcare?Catalyst Carryover4(2).
75.
OliverJ (2023, October 13) ‘I felt trapped’: Caroline Ellison steps out of Sam Bankman-Fried’s shadow in star witness turn. Financial Times.
76.
PapadakisMATeheraniABanachMA, et al. (2005) Disciplinary action by medical boards and prior behavior in medical school. New England Journal of Medicine353(25): 2673–2682.
77.
ParsonsCASulaemanJTitmanS (2018) The geography of financial misconduct: The geography of financial misconduct. Journal of Finance73(5): 2087–2137.
78.
PengDXYeYFengB, et al. (2020) Impacts of hospital complexity on experiential quality: Mitigating roles of information technology. Decision Science51(3): 500–541.
79.
RehurekRSojkaP (2011) Gensim—Python framework for vector space modelling. NLP Centre, Faculty of Informatics, Masaryk University, Brno, Czech Republic3(2): 2.
80.
ReimersNGurevychI (2019) Sentence-BERT: Sentence embeddings using Siamese BERT-networks. In: Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). Hong Kong: Association for Computational Linguistics, pp. 3982–3992.
81.
RobertsonSMStanleyMACullyJA, et al. (2012) Positive emotional health and diabetes care: Concepts, measurement, and clinical implications. Psychosomatics53(1): 1–12.
82.
RolandMRaoSRSibbaldB, et al. (2011) Professional values and reported behaviours of doctors in the USA and UK: Quantitative survey. BMJ Quality & Safety20(6): 515–521.
83.
RosensteinAHO’DanielM (2008) A survey of the impact of disruptive behaviors and communication defects on patient safety. Joint Commission Journal on Quality and Patient Safety34(8): 464–471.
84.
RutkowLChangH-YDaubresseM, et al. (2015) Effect of Florida’s prescription drug monitoring program and pill mill laws on opioid prescribing and use. JAMA Internal Medicine175(10): 1642–1649.
85.
Saar-TsechanskyMProvostF (2007) Decision-centric active learning of binary-outcome models. Information Systems Research18(1): 4–22.
86.
SaifeeDH(Eric) ZhengZBardhanIR, et al. (2020) Are online reviews of physicians reliable indicators of clinical outcomes? A focus on chronic disease management. Information Systems Research31(4): 1282–1300.
87.
SongKTanXQinT, et al. (2020, November 2) MPNet: Masked and permuted pre-training for language understanding. arXiv preprint arXiv:2004.09297.
88.
TamineLGoeuriotL (2021) Semantic information retrieval on medical texts: Research challenges, survey, and open issues. ACM Computing Surveys54(7): 146:1–146:38.
89.
TunstallLReimersNJoUES, et al.2022. Efficient few-shot learning without prompts. In arXiv preprint arXiv:2209.11055.
90.
WangXWuQLaiG, et al. (2019) Offering discretionary healthcare services with medical consumption. Production and Operations Management28(9): 2291–2304.
91.
WhitakerBFLaura BeilH (2023, September 12). How Columbia ignored women, undermined prosecutors and protected a predator for more than 20 years. ProPublica.
XieJZhangZLiuX, et al. (2021) Unveiling the hidden truth of drug addiction: A social media approach using similarity network-based deep learning. Journal of Management Information System38(1): 166–195.
94.
YanLPedraza-MartinezAJ (2019) Social media for disaster management: Operational value of the social conversation. Production and Operations Management28(10): 2514–2532.
95.
YangCCYangHJiangL (2014) Postmarketing drug safety surveillance using publicly available health-consumer-contributed content in social media. ACM Transaction Management Information System5(1): 1–21.
96.
ZhangXDuQZhangZ (2022) A theory-driven machine learning system for financial disinformation detection. Production and Operations Management31(8): 3160–3179.
97.
ZhaoWLiuQBGuoX, et al. (2022) Quid pro quo in online medical consultation? Investigating the effects of small monetary gifts from patients. Production and Operations Management31(4): 1698–1718.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
