Learn more

Detect, recover and prevent claim overpayments

Healthcare claim payments are complex, leaving ample room for inaccurate payments to slip through. Health plans can implement Optum® Data Mining, both pre- and post-pay, to help improve payment accuracy, reduce overpayments and save on medical and administrative costs without requesting provider documentation.

Our robust Data Mining solution helps:

Facilitate compliance with provider contracts
Identify and resolve outlier billing errors
Conduct reviews on adjudication and coding errors
Prevent future errors and overpayments

Maximize savings with pre- and post-pay Data Mining solutions

Blend technology and human intelligence

Our 4-tier analytic process and expert analysts review and deny errant claims before payment and identify overpayments already reimbursed.

Find errors across multiple categories

We identify and prevent claim errors including duplicates, billing issues, coordination of benefits and contract administration.

Shift from recovery to prevention

Our advanced analytics pinpoint the root cause of claim overpayments and apply learnings to prevent future claim errors.

Increase transparency through collaboration

We share concepts and analytics with health plans to help optimize their programs and work with providers to explain and prevent errors.

Drive more value from your data mining solution

We bring unmatched data mining expertise and experience to help health plans achieve payment accuracy and reduce healthcare costs. Our results:

More than 1 million claims reviewed daily*
More than $8 billion in annual pre- and post-pay savings*
9 million industry rules, regulations and policies*
More than 1,200 algorithms to generate insights and drive greater accuracy*

Learn more

Frequently asked questions about data mining solutions

With our Data Mining solution, health plans can:

Pinpoint billing errors
Uncover outliers
Identify trends
Highlight systemic issues that lead to improper payments

Data mining uses advanced analytics and AI to review large volumes of claims and detect issues that may not be visible through standard reviews. Through deeper analysis, it identifies the root causes — such as coding inconsistencies, policy misinterpretations or operational gaps — allowing health plans to implement targeted corrective actions.

Learn more about how AI is helping health plans identify behavioral anomalies and outliers.

By analyzing patterns across claim submissions, historical data and provider behavior, data mining can help health plans predict risk and identify claims likely to contain errors. This proactive capability aims to:

Improve the accuracy of claim determinations
Reduce downstream rework
Enhance overall payment integrity

The insights generated also support staff training, policy updates and process improvements that prevent recurrent inaccuracies.

Claims editing applies rules to identify known issues before payment and post‑pay audits focus on recovering improper payments after the fact. Data mining, however, goes further by identifying complex or emerging patterns indicative of hidden fraud, waste or abuse. It uses machine learning and advanced statistical approaches to detect outliers and anomalies that traditional methods may miss. Because it operates across the entire claims lifecycle, data mining strengthens both pre- and post‑payment accuracy.

Data mining may help reduce unnecessary documentation by using advanced models to target only claims with a high likelihood of error or risk. This helps to:

Lower provider abrasion
Preserve clinical staff time
Keep operational workflows efficient

In addition, feedback loops — where providers are informed of recurring errors — promote education and reduce the frequency of future record requests.

Data mining often combines deterministic rules, predictive machine learning models, anomaly-detection algorithms and AI‑assisted review tools:

Machine learning identifies patterns and predicts questionable claims.
Anomaly detection identifies unusual behavior and outliers.
Generative AI can help refine targeting criteria, summarize complex patterns and help automate portions of the review cycle for greater scalability and accuracy.