UnitedHealth's Optum Developing New AI-Powered Risk Scoring System for Medicare Advantage

UnitedHealth's Optum Developing New AI-Powered Risk Scoring System for Medicare Advantage

UnitedHealth's Optum is developing a new artificial intelligence-powered risk scoring system for Medicare Advantage, which aims to calculate the health status and potential costs of Medicare patients. This system would use AI to assess patient data and determine risk scores, potentially impacting Medicare Advantage payments.

The development of this system comes amid controversy, as UnitedHealth faces a lawsuit alleging that its AI tool, NaviHealth, was used to deny coverage for patients in skilled nursing facilities, contradicting doctors' orders. The lawsuit claims the AI model doesn't account for individual patient needs and relies on data from similar patients.

UnitedHealth has previously faced allegations of inflating risk scores to increase Medicare Advantage payments. A whistleblower lawsuit claimed the company encouraged upcoding and ignored concerns about accuracy. Despite these concerns, UnitedHealth Group emphasizes the importance of responsible AI use, with guiding principles focusing on reliability, fairness, accountability, transparency, and privacy.

Optum provides analytics consulting services to improve business performance in healthcare, including risk adjustment and quality programs. Its risk adjustment platform helps insurers understand risk score trends and opportunities. The development of the new AI-powered risk scoring system highlights the growing role of AI in healthcare and the need for careful consideration of its potential impact on patient care and healthcare financing.

About the author

TOOLHUNT

Effortlessly find the right tools for the job.

TOOLHUNT

Great! You’ve successfully signed up.

Welcome back! You've successfully signed in.

You've successfully subscribed to TOOLHUNT.

Success! Check your email for magic link to sign-in.

Success! Your billing info has been updated.

Your billing was not updated.