HHS · no agency ID · record R0741

Enhanced Direct Enrollment Outlier Detection

Department of Health and Human Services · HHS/CMS/CCIIO

Overview

Development stage: In development or acquisitionSource: a) Pre-deployment – The use case is in a development or acquisition status.
High-impact designation: Presumed high-impact, determined notSource: b) Presumed high-impact, but determined not high impact
Impact justification: The Enhanced Direct Enrollment Outlier Detection use-case is not considered high-impact as it does not serve as the basis for decisions or actions that affect civil rights, liberties, privacy, critical resources, safety, or strategic assets. The AI models within this use-case use machine learning techniques to detect anomalous or inconsistent patterns relative to standard partner actions and other application channels outside of the EDE pathway. The findings are shared with the agency divisions in the form of different reports and tables. The data alone is not enough to determine fraud, however, can be used by CMS in tandem with other data to determine if CMS should take any corrective actions. CMS makes all determinations on actions.
Start date: Not reported
Withheld from public reporting: NoSource: a) No

Mission

Topic area: Information Technology
Operational functions: Fraud, waste and improper payments Derived by keyword rules; see methodology

Problem the AI is intended to solve

(Marketplace) Enhanced Direct Enrollment (EDE) allows consumers to apply for and enroll in an exchange plan directly through an approved partner's UI, without being redirected through the Healthcare.gov application. These partner systems directly interface with the APIs developed by the FFE. As EDE partners gain more control over their application process, the FFE must ensure program integrity.

Expected benefits

Implement ML to identify anomalies/quality issues with partner-submitted person, application, and policy data?.

System outputs

Ensure FFE EDE program integrity

Technology

AI classification: Classical / predictive MLSource: Classical/Predictive Machine Learning: Models trained on data to make predictions or classifications based on identified patterns or relatio…
System name(s): Not reported
Custom-developed code: Not reported
Public source code: Not reported

Sourcing

How it was built: Not reported
Vendor (as reported): Not reported
Vendors (standardized): None on the standard list

A vendor is the supplier named by the agency. It does not identify the underlying model or AI technology.

Data and privacy

Involves PII: Not reported
Privacy Impact Assessment: Not reported
Authorization to Operate: Not reported
Demographic features: Not reported
Training and evaluation data: Not reported
Federal Data Catalog: Not reported

Governance

0 of 8 minimum-practice questions answered. A blank answer means the agency reported nothing; it does not mean the practice is absent.

Pre-deployment testingNo answer reported

AI impact assessmentNo answer reported

Independent reviewNo answer reported

Ongoing monitoringNo answer reported

Operator trainingNo answer reported

Fail-safeNo answer reported

Appeal processNo answer reported

User and public consultationNo answer reported

Potential impacts and how they were identified

Not reported