Can AI health-insurance fraud tools be trusted?
A review of 33 studies finds broad technical promise but little comparable evidence on external validation, cost, fairness or real-world deployment. A suspicious-claim score is not proof of fraud and still requires accountable human investigation.
The short version
3 key points
- 1.Seven databases produced 450 records; after 227 duplicates and screening, 33 peer-reviewed studies from 2016–2025 were included.
- 2.Recall appeared in 18 studies, accuracy in 17, F1 in 15, AUROC in 13 and precision-recall AUC in only three, making model comparisons unreliable for highly imbalanced fraud data.
- 3.Only three studies discussed generalisability, two limited interpretability and nine cost-effectiveness. The review could not perform a meta-analysis because datasets, fraud definitions and methods differed too much.
The answer: promising filters, not automated fraud verdicts
Machine-learning, deep-learning and graph-based systems often outperformed the comparators chosen in 33 reviewed studies. That makes AI a credible tool for ranking claims or networks for further investigation. It does not show that insurers can safely reject care, recover money or accuse a person or provider from a model score. Most evidence came from country-specific datasets with inconsistent fraud definitions and evaluation, while external validation, economic evidence and real-world governance were sparse.
Fraud is rare relative to legitimate claims. A model can report excellent overall accuracy by labelling almost everything legitimate, while still missing costly fraud or burdening honest patients and clinicians with false alarms. The operational question is therefore not whether one algorithm posts a high percentage. It is how many genuinely fraudulent cases are found, how many legitimate claims are disrupted, whether investigators can understand the alert, and whether the system produces net savings after review costs and appeals.[1]
How 450 records became 33 studies
The author followed PRISMA 2020 and searched Scopus, Web of Science, OVID Medline, Embase, PubMed, Dimensions and OpenAlex for peer-reviewed English-language work published from 2016 through March 2025. The search identified 450 records, removed 227 duplicates and screened 223 titles and abstracts. Forty reports were sought, three were unavailable and four failed the full-text criteria, leaving 33 studies for qualitative synthesis.
The review extracted dataset, country, fraud type, AI method, performance measures, funding and limitations. Twenty-four studies addressed fraudulent claims, while municipal and international insurance schemes appeared in six studies each. Machine learning was most common, followed by deep learning, graph methods and hybrids. The review did not pool a single effect because inputs, labels, comparators and metrics were too heterogeneous; that was the methodologically safer choice.[1]
Why the reported performance cannot form a league table
Recall was reported in 18 studies, accuracy in 17, precision in 16, F1 score in 15 and AUROC in 13. Only three reported precision-recall AUC, a particularly informative measure when positive cases are rare. Four reported specificity. Different train-test splits, resampling methods, class prevalences and definitions of confirmed fraud mean that a 95% accuracy in one dataset cannot be compared directly with an 85% result in another.
Thirty studies claimed improved performance over an existing method, but only three discussed generalisability and two highlighted limited interpretability. The pattern is consistent with a development literature rewarded for model improvement, not a deployment literature designed to expose failure. Fraud labels may themselves reflect past investigative priorities: if audits historically targeted some providers or regions more heavily, a model can learn enforcement patterns rather than underlying misconduct.[1]
The strongest practical finding is what remains unmeasured
Only nine studies addressed cost-effectiveness and eight runtime efficiency. The review found little consistent evidence on implementation cost, operational workflow, fairness, privacy, appeals or long-term financial sustainability. A model may raise recall while creating so many low-quality alerts that investigators spend more time, providers wait longer for payment and patients face delayed care. Technical performance is only one component of public value.
External validation was also weak. Many models were tied to the coding rules, reimbursement incentives and fraud patterns of China, the United States or another single system. A graph pattern that indicates collusion in one claims network may be routine referral structure in another. Before transfer, insurers need locked testing on local data from a later period and an independent organisation, with thresholds chosen around the real cost of misses and false accusations.[1]
What this means for patients, clinicians and investigators
Patients need due process when an algorithm affects authorisation, reimbursement or investigation. An alert should trigger proportionate review, not automatic denial. People must be told the reason for an adverse action in understandable terms and have a route to correct coding errors, identity misuse or unusual but legitimate treatment. Protected characteristics and proxies such as postcode, provider network and diagnosis mix require fairness testing because unequal scrutiny can become self-reinforcing.
Clinicians and providers need timely notification and a human reviewer who can understand context. Fraud teams should separate anomaly detection from evidence of intent, record which alerts led to confirmed findings and monitor whether investigators over-trust model rankings. Public payers and regulators should demand audit logs, access controls, data minimisation, drift monitoring and published error rates. Procurement language should never equate a suspicious claim with proven fraud.[1]
Review-level limits and what would change the assessment
The review included only English-language peer-reviewed studies through March 2025, so it may miss later, non-English and operational evidence. One author conducted the quality assessment twice rather than using independent reviewers, and titles and abstracts were screened with an AI-assisted platform before full eligibility review. The Drummond economic-evaluation checklist rated 31 studies good, yet the review simultaneously found that comprehensive cost analysis was usually absent; those broad ratings should not be read as proof of low bias or deployment readiness.
Confidence would rise with shared benchmark definitions, later-period and cross-system external validation, precision-recall reporting, documented human review and prospective evaluation of savings, delays, appeals and unequal error. The review received no financial support and declared no commercial conflict. Its defensible conclusion is narrower than a sales claim: AI can help prioritise investigation, but health-insurance organisations still lack comparable evidence that these systems are fair, economical and reliable at scale.[1]
What this means for people
- Patients can face delayed care or reimbursement when false alerts are treated as verdicts, so human review and appeal routes are essential.
- Clinicians and providers need transparent reasons, timely correction of coding errors and protection from self-reinforcing audit bias.
- Fraud investigators need alert quality, confirmed outcomes and workload measures—not accuracy alone—to judge whether a system helps.
Global context
The 33 studies came from varied reimbursement systems, with research concentrated in China, the United States and India. Coding rules, fraud opportunities, audit practice and due-process requirements differ, so performance is not portable without local external validation.
What the evidence does not yet show
- English-only peer-reviewed literature through March 2025
- Large heterogeneity prevented quantitative meta-analysis
- One author performed study selection synthesis and repeated the quality assessment
- External validation, interpretability, fairness and governance were rarely evaluated
- Cost-effectiveness and real-world operational outcomes were sparse
What to watch next
- Independent later-period validation across different insurance systems
- Precision-recall performance and confirmed-fraud yield at operational thresholds
- Effects on payment delay, care access, appeals and investigation workload
- Fairness audits by patient and provider characteristics
- Transparent human review, explanation and contestability requirements
Living evidence record
Impact record IAI-0601PZM
Evidence stage
Studied
Confidence
Supported
Reporting basis
Source analysis
Independent or research support
Present
Record status
Monitoring
Last checked
15 October 2026
Source trail
1 direct source across 1 source type.
People impact
Documented in this record.
Uncertainty
Limits and next checks are explicit.
Stages describe the evidence available—not whether a technology is good or bad. See the public method.
Single-source reporting disclosure
This record analyses one direct source. It can establish what Frontiers in Public Health published or reported, but it is not independent corroboration of every performance claim or predicted outcome. The confidence label will change only when broader evidence is added.
Evidence trail
Sources used for this report
Links checked 15 October 2026
Research question, topics and regions
What peer-reviewed evidence shows about AI for detecting health-insurance fraud and what is missing for responsible deployment
Topics: Health insurance · Fraud detection · Machine learning · Claims · Fairness · Governance
Regions: Saudi Arabia · Global
This report is labelled source analysis. We summarise and analyse source material in our own words; company statements remain attributed claims until independently supported. Translated summaries preserve the meaning of the original source and link back to it. Read our editorial standards.
Watch the background, then check the evidence
UNESCO
AI-generated illustration · not a video frame
YouTube loads only when you choose. Press play in the player to start. If playback is unavailable, use “Watch on YouTube” below.
AI, jobs & human choice · Background explainer
Published 2024-10-14 · 11 min 27 sec
UNESCO Talks: Interview with 2024 Nobel laureate Professor Daron Acemoglu
Economist Daron Acemoglu discusses how choices about AI affect workers, inequality and society. Pair the interview with our comparison of task exposure, hiring and skills evidence.
Evidence check: An expert interview from 2024, offering a perspective on policy and work rather than a measured forecast of job losses.
Source checked 14 October 2026. Check available captions on the publisher’s video page.
The Impact Brief · Free
Keep following the stories that matter to you.
Get our weekday briefing on what changed and why it matters. Choose your topics below.
Continue the story
There’s more to this story
Finance & Business
Did five AI prototypes create $1 trillion for MENA?
No. A new World Economic Forum and Bain report built five early prototypes and models more than $1 trillion of possible value by 2030, but it does not report region-wide deployment or realised gains.
10 min · 3 sources
Finance & Business
Can machine learning show income inequality before surveys arrive?
An OECD working paper combines machine learning with tax-benefit simulation to estimate EU income inequality sooner. It improves on ordinary regression, but simple no-change and non-ML benchmarks remain competitive; estimates are preliminary and cannot determine an individual household's benefit entitlement.
6 min · 2 sources
Finance & Business
If 74% of companies see AI returns, why have only 13% scaled as planned?
BearingPoint surveyed 1,050 senior leaders across Europe, the United States and China. The results suggest that AI can create value, but they measure executives' reports—not audited profit caused by AI.
6 min · 2 sources
Reader commentary
What does this mean where you are?
Tell us what you have seen at work, in your community or in everyday life. A question is welcome too. No account is needed; we review notes before publishing them.
Explore commentary across the portal →Published reader notes
0No published reader notes yet. You can start the evidence-led discussion above.
Prefer a private correction or response? Contact the newsroom.