Skip to Content
logologo
AI Incident Database
Open TwitterOpen RSS FeedOpen FacebookOpen LinkedInOpen GitHub
Open Menu
Discover
Submit
  • Welcome to the AIID
  • Discover Incidents
  • Spatial View
  • Table View
  • List view
  • Entities
  • Taxonomies
  • Submit Incident Reports
  • Submission Leaderboard
  • Blog
  • AI News Digest
  • Risk Checklists
  • Random Incident
  • Sign Up
Collapse
Discover
Submit
  • Welcome to the AIID
  • Discover Incidents
  • Spatial View
  • Table View
  • List view
  • Entities
  • Taxonomies
  • Submit Incident Reports
  • Submission Leaderboard
  • Blog
  • AI News Digest
  • Risk Checklists
  • Random Incident
  • Sign Up
Collapse

Report 3363

Associated Incidents

Incident 60821 Report
UnitedHealth Accused of Deploying Allegedly Flawed AI to Deny Medical Coverage

Loading...
UnitedHealth sued for allegedly using AI to deny elderly patients care
news5cleveland.com · 2023

UnitedHealth Group is being sued for allegedly using an artificial intelligence algorithm to systematically deny elderly patients rehabilitative care.

The class action lawsuit, filed Tuesday in Minnesota district court, says the defendants — UnitedHealth Group, UnitedHealthCare and NaviHealth — illegally used an AI model called nH Predict in place of medical professionals despite knowing it had a 90% error rate. This helped the group continuously deny Medicare Advantage Plan patients the care their doctors said was necessary, such as nursing facility stays, by "predicting" what a patient "should" require instead of what they actually do, the lawsuit states.

It also says the defendants intentionally limited "their employees' discretion to deviate" from nH Predict, forcing them to meet dedicated targets of use or be disciplined and terminated.

More than 90% of patients who appealed the nH Predict's claim denial had the decision overturned, according to the lawsuit.

But the plaintiffs claim the health care companies knew only a tiny group of policyholders would appeal the denied claims and that most would either pay out-of-pocket for the care or forgo it altogether. This would create higher profit for the company while patients deplete their savings to continue care "all because an AI model 'disagrees' with their real live doctors' determinations," it states.

"Defendants bank on the patients' impaired conditions, lack of knowledge and lack of resources to appeal the erroneous AI-powered decisions," the lawsuit says.

UnitedHealth Group is the largest insurance company in the U.S. with more than 52.9 million customers. 

Two of those customers are the plaintiffs: Gene Lokken and Dale Henry Tetzloff, who are now both deceased.

Both plaintiffs were covered by a Medicare Advantage Plan provided by the defendants. This plan is supposed to cover post-acute care, which is medically necessary care for patients recovering from serious illnesses and injuries. This could include skilled care, therapy and other home health services, skilled nursing facilities, inpatient rehabilitation facilities and long-term care facilities.

But the plaintiffs say they were forced to pay out-of-pocket for the post-acute care their doctors deemed necessary for their health, prompting the lawsuit that seeks to stop the practice with a court order and award patients damages.

Their attorneys are pushing to represent tens of thousands of people who have similar stories, according to Reuters, which could amount to damage claims reaching billions of dollars.

UnitedHealth denied all claims that nH Predict is used for coverage determinations, according to HealthCare Drive.

Read the Source

Research

  • Defining an “AI Incident”
  • Defining an “AI Incident Response”
  • Database Roadmap
  • Related Work
  • Download Complete Database

Project and Community

  • About
  • Contact and Follow
  • Apps and Summaries
  • Editor’s Guide

Incidents

  • All Incidents in List Form
  • Flagged Incidents
  • Submission Queue
  • Classifications View
  • Taxonomies

2024 - AI Incident Database

  • Terms of use
  • Privacy Policy
  • Open twitterOpen githubOpen rssOpen facebookOpen linkedin
  • e1b50cd