Glossary

Model drift

When a clinical AI model quietly gets worse because the world it was trained on has moved on — what model drift is, why healthcare feels it acutely, and how it is monitored. As of July 2026.

By Jonas WeirReviewed by Jonas Weir · editorial reviewUpdated

The short version

  • Model drift is the gradual decline in a deployed clinical AI model's accuracy when the real-world data it sees no longer matches the data it learned from.
  • It is a recognised deployment failure: in April 2020 a hospital sepsis-alerting model was switched off after COVID-19 shifted the patient mix enough to trigger spurious alerts.
  • Discrimination can look stable while calibration silently drifts — one study saw kidney-injury models keep their ranking ability yet steadily over-predict risk across nine years.
  • Regulators now expect deployed models to be monitored across their lifecycle, which is why post-market surveillance of AI has become a discipline of its own.

Model drift is the gradual decline in a deployed clinical AI model's accuracy when the real-world data it sees no longer matches the data it was trained on. As patients, clinical practice, coding conventions, and instruments change, the model's predictions grow steadily less reliable — often without any visible warning. As of July 2026.

Why it matters in healthcare

Most machine-learning systems are validated once, at a single moment, on data already collected. Healthcare then keeps moving. New treatments change who ends up in hospital; a coding update relabels a diagnosis; a scanner is replaced; referral patterns shift with the seasons. Each of these can pull live data away from the training distribution. Researchers describe this formally as dataset shift — the situation where "a machine-learning system underperforms because of a mismatch between the data set with which it was developed and the data on which it is deployed" 1. Because clinical models inform decisions about real people, silent degradation carries direct patient-safety weight, which is why drift is treated as a governance problem rather than a modelling footnote.

How it works

Drift usually arrives through one of three doors. In covariate or data drift, the mix of inputs changes while their meaning holds. In label drift, the base rate of the outcome moves. In concept drift, the relationship between inputs and outcome itself changes. A study of emergency-department admission models captured the last two together, noting that "the underlying data distributions characterising episodes can change with time (data drift), and so can the relationship between episode characteristics and associated clinical outcomes" 3. Any of the three can degrade performance, and they frequently travel together.

Where it appears today

The clearest cautionary case is a sepsis-alerting model at a US academic health system: in April 2020 it "had to be deactivated because of spurious alerting" after COVID-19 changed patient characteristics beyond what the model had learned 1. Slower drift is just as real. Across nine years of Veterans Affairs data, seven acute-kidney-injury models kept their ability to rank patients but increasingly over-predicted risk — "discrimination was maintained for all models" while "calibration declined" as event rates rose 4. The failure was invisible to a discrimination-only check and obvious once calibration was tracked over time.

Common misunderstandings

Drift is the same as a one-time validation gap. It is a distinct problem. A widely deployed proprietary sepsis model scored an area under the curve of only 0.63 and missed 67% of sepsis cases when tested outside its origin site 2 — a spatial gap between development and a new setting, present from day one, rather than a temporal decline. Both matter, but they are diagnosed and fixed differently, and confusing them leads teams to re-validate when they should be monitoring.

Good discrimination means the model is fine. The kidney-injury work shows a model can keep its ranking ability while its probabilities drift out of alignment 4. Watching only an AUROC will miss it; the answer is to monitor calibration alongside discrimination.

Drift is a surprise. It is expected. FDA, Health Canada, and the MHRA jointly advise that "deployed models have the capability to be monitored" with controls to manage "degradation of the model (for example, dataset drift)" 5. The systematic version of that monitoring — continuous post-market surveillance of AI performance and equity — is now called algorithmovigilance 6.

Related terms

Drift is best understood alongside model calibration, which is often the first metric to slip; external validation, the spatial cousin of temporal drift; and algorithmovigilance, the discipline of catching both after deployment. For the wider picture of how far clinical AI has spread and where it strains, see our AI in healthcare statistics.

Questions & answers

  • What is model drift?

    Model drift is the gradual decline in a deployed AI model's accuracy that happens when the real-world data it encounters stops matching the data it was trained on. In healthcare, patient populations, clinical practice, coding conventions, and instruments all change over time, so a model that was accurate at launch can grow steadily less reliable without anyone changing a line of its code.

  • What is the difference between model drift and data drift?

    Data drift describes the input side: the distribution of the data feeding the model has moved. Model drift is the consequence — the model's outputs become less accurate as a result. Data drift is one of the main causes of model drift, alongside changes in the relationship between inputs and outcomes, sometimes called concept drift.

  • How is model drift detected in clinical AI?

    Through ongoing monitoring after deployment: tracking input distributions, comparing predicted risks against observed event rates, and watching calibration as closely as discrimination. Regulators now expect this kind of lifecycle surveillance for AI-enabled devices, and the practice has its own name — algorithmovigilance.

Sources

  1. Finlayson SG, Subbaswamy A, Singh K, et al. The Clinician and Dataset Shift in Artificial Intelligence. N Engl J Med. 2021;385(3):283-286. doi.org/10.1056/NEJMc2104626
  2. Wong A, Otles E, Donnelly JP, et al. External Validation of a Widely Implemented Proprietary Sepsis Prediction Model in Hospitalized Patients. JAMA Intern Med. 2021;181(8):1065-1070. doi.org/10.1001/jamainternmed.2021.2626
  3. Duckworth C, Chmiel FP, Burns DK, et al. Using explainable machine learning to characterise data drift and detect emergent health risks for emergency department admissions during COVID-19. Sci Rep. 2021;11(1):23017. doi.org/10.1038/s41598-021-02481-y
  4. Davis SE, Lasko TA, Chen G, Siew ED, Matheny ME. Calibration drift in regression and machine learning models for acute kidney injury. J Am Med Inform Assoc. 2017;24(6):1052-1061. doi.org/10.1093/jamia/ocx030
  5. US FDA, Health Canada, MHRA. Good Machine Learning Practice for Medical Device Development: Guiding Principles. October 2021. www.fda.gov/medical-devices/software-medical-device-samd/good-machine-learning-practice-medical-device-development-guiding-principles
  6. Embi PJ. Algorithmovigilance—Advancing Methods to Analyze and Monitor Artificial Intelligence–Driven Health Care for Effectiveness and Equity. JAMA Netw Open. 2021;4(4):e214622. doi.org/10.1001/jamanetworkopen.2021.4622