Research – Ziad Obermeyer

The Health Costs of Cost-Sharing, forthcoming in QJE (with Amitabh Chandra and Evan Flack). Media: Vox, Modern Healthcare.

For a more complete list, please see my Google Scholar or PubMed pages.

Health economics + Machine learning

Diagnosing Physician Error: A Machine Learning Approach to Low-Value Health Care, QJE 2022 (with Sendhil Mullainathan). Git repo here. Slide deck here. Summary and comment (with Kate Baicker) in JAMA Health Forum here. Awards: International Health Economics Association Arrow Award 2023.
Predictive modeling of US healthcare spending in late life, Science 2018 (with Liran Einav, Amy Finkelstein, and Sendhil Mullainathan)
Does Machine Learning Automate Moral Hazard and Error? AER: P&P 2017 (with Sendhil Mullainathan)
Early death after discharge from emergency departments, BMJ 2017(with Brent Cohn, Michael Wilson, Bapu Jena, and David Cutler)
Prediction Policy Problems, AER: P&P 2015 (with Jon Kleinberg, Jens Ludwig, and Sendhil Mullainathan)

Medicine + Machine learning

Pooled testing efficiency increases with test frequency, PNAS 2022 (with Ned Augenblick, Jon Kolstad, and Ao Wang). Media: MIT Tech Review
Variation in common laboratory test results caused by ambient temperature, Med (Cell Press) 2021 (with Devin Pope)
An algorithmic approach to unexplained pain disparities in underserved populations, Nature Medicine 2021 (with Emma Pierson, David Cutler, Jure Leskovec, and Sendhil Mullainathan). Git repo here. Slide deck here. Media: Wired, MIT Tech Review, STAT, NPR’s Science Friday featured segment, FT Tech Tonic
A Comparison of Patient History- and EKG-based Cardiac Risk Scores, AMIA 2019 (with Andy Miller and Sendhil Mullainathan)
Direct Uncertainty Prediction for Medical Second Opinions, ICML 2019 (with Maithra Raghu, Katy Blumer, Rory Sayres, Robert Kleinberg, Sendhil Mullainathan, and Jon Kleinberg)
Regularizing Latent Variable Models with Discriminative Models, ICML 2019 (with Andy Miller, Jon Cunningham, and Sendhil Mullainathan)
Individuals’ body temperatures vary meaningfully and predict mortality, BMJ 2017 (with Jasmeet Samra and Sendhil Mullainathan)

Algorithmic bias

Race adjustments in clinical algorithms can help correct for racial disparities in data quality. PNAS 2024 (with Anna Zink and Emma Pierson). Invited PNAS commentary by John Mullahy: Clinical decisions, patient race, and flawed data.
A ‘Playbook‘ for identifying and fixing algorithmic bias. Media: STAT News. Regulatory: FTC PrivacyCon 2021, ONC Annual Meeting, NCQA Digital Quality Summit, CA Dept. of Fair Employment & Housing hearing
On the Inequity of Predicting A While Hoping for B. AER: P&P 2021 (with Sendhil Mullainathan)
Racial bias in allocation of COVID-19 relief funding, JAMA 2020 (with Pragya Kakani, Amitabh Chandra, and Sendhil Mullainathan). Git repo here. Media: Wall Street Journal, STAT
Dissecting racial bias in an algorithm used to manage the health of populations, Science 2019 (with Brian Powers, Christine Vogeli, and Sendhil Mullainathan). Git repo here. Media: Wall Street Journal, Washington Post, LA Times, AP, Guardian, Telegraph, Bloomberg, Wired, Scientific American, STAT, Nature, Breitbart (!), and many others. Awards: Altmetric 2019 Top 100, ‘Editors’ Pick’: STAT Madness 2020, Willard G. Manning Memorial Award for the Best Research in Health Econometrics 2021, Financial Times Responsible Business Education Award. Regulatory action: The study prompted the NY state insurance regulator and Sens. Booker and Wyden to initiate inquiries into algorithmic bias (in WSJ, Wired).

Machine learning — perspectives