Study guide
14.3 Advanced Analytics in Claims Operations & Fraud Detection
Chartered Property Casualty Underwriter CPCU
Track your progress as you read
Finished sections get a check mark.
1 Chapter 1: CPCU Program Structure, Matriculation & Exam Policies
1.1 CPCU Designation Pathway & Program Requirements 1.2 CPCU Matriculation, Experience & the Ethics Commitment 1.3 Virtual Exam Format, Policies & Test-Taking Strategy
2 Chapter 2: CPCU 500 — Risk Foundations, Principles & Analysis
2.1 Building Your Foundation: The Role of Risk Management and Insurance 2.2 Risk Concepts & Classifications 2.3 Enterprise Risk Management & ISO 31000 2.4 The Risk Management Process: Identification & Analysis
3 Chapter 3: CPCU 500 — Risk Treatment, Critical Thinking & Leadership
3.1 Risk Control Techniques & Loss Mitigation 3.2 Risk Financing: Retention vs. Transfer 3.3 Leading with Critical Thinking 3.4 Communicating and Collaborating as a Leader 3.5 Strategic Decision-Making & Leadership in Risk Management
4 Chapter 4: CPCU 520 — Insurer Value Chain & Marketing Systems
4.1 The Insurer Value Chain & Cross-Functional Operations 4.2 Property-Casualty Distribution Systems & Channels 4.3 Market Segmentation, Research & Strategic Positioning
5 Chapter 5: CPCU 520 — Commercial & Personal Lines Underwriting
5.1 Line Underwriting vs. Staff Underwriting Operations 5.2 The Underwriting Process & Decision Analysis 5.3 Adding Value Through Personal Lines Underwriting 5.4 Underwriting Portfolio Management & Capacity
6 Chapter 6: CPCU 520 — Claims Handling & Reinsurance Operations
6.1 Claims Operations & The Claim Handling Process 6.2 Reinsurance Functions & Strategic Value 6.3 Reinsurance Types, Treaties & Calculation Mechanics
7 Chapter 7: CPCU 520 — Ratemaking & Actuarial Foundations
7.1 Actuarial Ratemaking Goals & Regulatory Standards 7.2 Ratemaking Methods & Rate Indication Calculations 7.3 Loss Reserving & Loss Development Triangle Analysis
8 Chapter 8: CPCU 530 — Legal Foundations & Insurance Contract Law
8.1 Understanding Legal Fundamentals in Insurance 8.2 General Contract Elements in Insurance Contexts 8.3 Distinctive Legal Characteristics of Insurance Contracts 8.4 Judicial Interpretation & Policy Ambiguity Doctrines
9 Chapter 9: CPCU 530 — Property Ownership, Torts & Liability Exposures
9.1 Determining Property Ownership and Responsibility 9.2 Negligence Principles & Elements of a Tort 9.3 Legal Defenses to Negligence & Tort Reform 9.4 Strict Liability, Vicarious Liability & Intentional Torts
10 Chapter 10: CPCU 530 — Agency Law, Intermediaries & Resolving Disputes
10.1 Agency Creation & The Scope of Producer Authority 10.2 Fiduciary & Contractual Duties in the Agency Relationship 10.3 Intermediary Legal Liability & Errors and Omissions (E&O) 10.4 Resolving Disputes: Negotiation, ADR, and Litigation
11 Chapter 11: CPCU 540 — Insurer Financial Statements & SAP vs. GAAP
11.1 Statutory Accounting Principles (SAP) vs. GAAP 11.2 The NAIC Annual Statement Balance Sheet & Surplus 11.3 Insurer Income Statement & Statement of Cash Flows
12 Chapter 12: CPCU 540 — Insurer Financial Analysis & Ratio Evaluation
12.1 Underwriting Profitability Metrics & Combined Ratio 12.2 Operating Performance Ratios & Overall Profitability 12.3 Solvency, Capacity & NAIC IRIS Financial Ratios
13 Chapter 13: CPCU 540 — Insurer Investment Management & Capital Strategy
13.1 Insurer Investment Portfolios & Asset-Liability Management 13.2 Insurer Capital Structure, Surplus Notes & Alternative Capital 13.3 The NAIC Risk-Based Capital (RBC) Framework
14 Chapter 14: CPCU 550 — Insurance Data Analytics & Predictive Modeling
14.1 The Expanding Universe of Insurance Data 14.2 Predictive Analytics & Machine Learning in Underwriting 14.3 Advanced Analytics in Claims Operations & Fraud Detection
15 Chapter 15: CPCU 550 — Technology, Cyber Risk & Data Ethics
15.1 InsurTech Innovations & Digital Transformation 15.2 Cyber Risk Management & Insurer Operational Resilience 15.3 Ethical Data Governance, Algorithmic Bias & Regulatory Compliance
16 Chapter 16: CPCU 551 — Managing Commercial Property Risk
16.1 Assessing the Need for Commercial Property Products 16.2 Commercial Property: Building & Personal Property Coverage Form 16.3 Business Income & Extra Expense Coverage Mechanics 16.4 Tailoring Commercial Property Coverage 16.5 Closing Coverage Gaps: Equipment Breakdown, Crime, and Builders' Risk 16.6 Protecting Property in Transit: Inland and Ocean Marine 16.7 Thinking Critically About Commercial Property Loss Exposures
17 Chapter 17: CPCU 552 — Managing Commercial Liability Risk
17.1 Commercial General Liability (CGL) Coverage & Policy Analysis 17.2 Selecting Business Auto Coverage 17.3 Covering Cyber Losses 17.4 Managing Workplace Risks: Workers Compensation, Employers Liability, and EPLI 17.5 Filling Liability Coverage Gaps: Umbrella, Professional, D&O, and Environmental 17.6 Thinking Critically About Commercial Liability Loss Exposures
18 Chapter 18: CPCU 555 — Advancing Personal Insurance Products
18.1 Homeowners Policy Forms & Property-Liability Coverages 18.2 The Personal Auto Policy (PAP): Coverages & Conditions 18.3 Providing Unique Personal Coverages 18.4 Personal Umbrella Policies & Personal Risk Management 18.5 Recommending Personal Investment Strategies
19 Chapter 19: CPCU 556 — Building a Competitive Edge in Personal Lines
19.1 Marketing Personal Lines Products 19.2 Analyzing Personal Auto Risks 19.3 Assessing Residential Property Risks 19.4 Evaluating Unique Personal Exposures 19.5 Developing Personal Lines Underwriting Strategies 19.6 Differentiating Personal Lines Products
20 Chapter 20: Ethics 311 & the CPCU Code of Professional Conduct
20.1 Ethical Decision Making: Ethics 311 and a Four-Step Framework 20.2 The CPCU Code of Professional Conduct & the Nine Canons 20.3 Enforcing the Code: Advisory Opinions, Discipline, and Conflicts in Practice
14.3 Advanced Analytics in Claims Operations & Fraud Detection
Key Takeaways Explain This Topic AI
Claims represent an insurer's largest single expenditure (consuming 70% to 80% of earned premiums in losses and LAE), making predictive triage, automated routing, and early severity identification essential to operational solvency.
Automated First Notice of Loss (FNOL) systems leverage predictive complexity scoring to immediately bifurcate claims, routing simple physical damage to express virtual photo-estimating while dispatching high-complexity casualty files to specialized senior adjusters.
Early severity prediction models identify 'creeper' claims—minor soft-tissue injuries that secretly threaten to breach policy limits—prompting early reserve adjustments and proactive settlement tenders to avoid bad-faith extra-contractual liability.
Special Investigation Units (SIUs) combine supervised fraud scoring with graph analytics (Social Network Analysis / SNA) to uncover organized crime rings, staged accident syndicates, and medical kickback networks that evade single-claim inspection.
Unstructured claims assets are operationalized through Computer Vision (automated vehicle and property damage appraisal via convolutional neural networks) and Natural Language Processing (NLP extraction of ICD-10 codes, legal threats, and sentiment from adjuster notes and medical records).
Last updated: September 2026
Advanced Analytics in Claims Operations & Fraud Detection
Quick Answer: Claims handling is the largest cost center in property-casualty insurance, accounting for 70% to 80% of every earned premium dollar in losses and Loss Adjustment Expenses (LAE). Modern claims analytics transforms operations from a reactive, manual posture into a predictive, proactive discipline. Core applications include Automated First Notice of Loss (FNOL) Complexity Scoring to bifurcate claims into express photo-estimating versus high-touch adjuster handling; Severity Prediction Models to detect "creeper" claims and prevent bad-faith policy limit breaches ; Social Network Analysis (SNA) and graph analytics in Special Investigation Units (SIU) to dismantle organized fraud rings; Litigation Propensity Scoring ; Computer Vision for automated damage appraisal; and Natural Language Processing (NLP) to extract diagnostic signals from unstructured medical and legal records.
1. Transforming the Claims Value Chain
In property-casualty insurance, the claims department is the ultimate "moment of truth" for policyholders. It is also the carrier's primary financial drain. Operational excellence in claims is defined by two competing objectives:
Indemnity Accuracy: Paying exactly what is contractually owed—neither underpaying (which invites bad-faith litigation) nor overpaying (which inflates the loss ratio).
Loss Adjustment Expense (LAE) Containment: Minimizing the administrative, legal, and operational costs required to investigate, evaluate, and settle claims. THE CLAIMS ANALYTICS REVOLUTION
- TRADITIONAL REACTIVE CLAIMS PREDICTIVE ANALYTIC CLAIMS
- ┌───────────────────────────────┐ ┌───────────────────────────────┐
- │ • Manual mail/phone FNOL │ │ • Digital & telematics FNOL │
- │ • FIFO claim file assignment │ │ • Instant algorithmic triage │
- │ • Late-developing reserves │ ───► │ • Early severity forecasting │
- │ • Intuition-based fraud checks│ │ • Graph-based fraud detection │
- │ • Reactive legal defense │ │ • Litigation propensity score │
- │ • Physical field inspections │ │ • Computer vision photo estim.│
- └───────────────────────────────┘ └───────────────────────────────┘
- 2. First Notice of Loss (FNOL) Routing & Complexity Scoring
The claims lifecycle begins at First Notice of Loss (FNOL) . Traditionally, claims were assigned on a round-robin or first-in, first-out (FIFO) basis to the next available general adjuster, regardless of underlying complexity. This created severe operational bottlenecks: experienced adjusters were bogged down processing routine fender-benders, while junior adjusters mishandled multi-party casualty claims with catastrophic exposure.
Modern claims systems implement Algorithmic FNOL Complexity Scoring at the exact moment of initial intake: FNOL COMPLEXITY SCORING & TRIAGE
- [Digital FNOL Ingestion]
- (Mobile App, Crash Telematics eCall, Web Portal, Call Center)
- │
- ▼
- [Predictive Complexity Engine]
- │
- ┌───────────────────────────┼───────────────────────────┐
- ▼ ▼ ▼
- [EXPRESS PATH] [STANDARD PATH] [SPECIALIZED PATH]
- (Complexity: 0 - 25) (Complexity: 26 - 65) (Complexity: 66 - 100)
- Drivable vehicle • Moderate collision • Multi-vehicle pileup
- Single vehicle / clear liab• Disputed intersection • Bodily injury / ambulance
- Property only (no injury) • Moderate physical damage • Commercial truck involved
- Straight to Computer • Assigned to Staff Field • Dispatched to Senior Major
- Vision Photo-Estimating Adjuster Case / Litigation Unit
- Scoring Indicators at FNOL
The predictive complexity engine evaluates dozens of risk features within seconds of report:
Impact Telematics: Delta-V (velocity change at impact), airbag deployment, rollover sensor activation, vehicle towed from scene.
Accident Topology: Number of vehicles involved, presence of commercial tractor-trailers, pedestrian involvement, intersection vs. highway crash.
Injury Indicators: Ambulance dispatch, emergency room transport, complaints of neck/back numbness, pedestrian or cyclist strikes.
Jurisdictional Venue: Whether the claim occurred in a judicial venue known for high litigation rates and runaway jury verdicts.
3. Severity Prediction Models & Early Reserving Adequacy
One of the most dangerous phenomena in casualty claims is the "Creeper Claim" (or "Jumper Claim"). These are claims initially reported as minor soft-tissue sprains or low-speed impacts where the initial case reserve is set at $5,000. Over the next 18 to 36 months, the claimant undergoes continuous chiropractic treatment, followed by epidural steroid injections, an unexpected lumbar MRI revealing disc herniation, and ultimately a two-level spinal fusion surgery.
By the time the adjuster realizes the magnitude of the claim, the exposure has exploded from $5,000 to $750,000, blowing past policy limits and causing severe actuarial reserve deficits. THE "CREEPER CLAIM" TRAJECTORY
- $800K ──┐ [Actual Claim Settlement]
- │ $750,000
- $600K ──┤ ▲
- │ ╱
- $400K ──┤ ╱
- │ [Spinal Surgery] ╱
- $200K ──┤ $250,000 ╱
- │ [MRI Herniation] ▲ ╱
- $0K ──┴───[Initial] ─────── $45,000 ──────────────┼──────╱──────────────────►
- Reserve: $5K Month 6 Month 18 Month 30
- ───► PREDICTIVE SEVERITY ENGINE detects pattern at Day 14 ──► Sets Reserve to $500K
- Early Severity Forecasting
Machine learning severity models analyze early claim attributes—such as claimant age, pre-existing comorbidities, physical impact dynamics, and specific medical billing codes—to predict the ultimate claim payout within the first 14 to 30 days of reporting.
Preventing Bad-Faith Policy Limit Breaches
In third-party liability insurance, the insurer owes a fiduciary duty to protect its insured from an excess judgment. If a claimant sustains severe injuries and offers to settle within the policy limit (e.g., a $100,000 policy limit demand), an insurer that unreasonably refuses or delays the settlement can be held liable for bad faith . If the case proceeds to trial and the jury awards $3,000,000, the insurer may be forced to pay the entire $3,000,000 out of its own surplus, completely stripping away the policy limit defense.
Predictive severity models alert claims managers to high policy-limit breach probability early, prompting timely settlement tenders that shield policyholders and insurers from catastrophic bad-faith exposure.
4. Special Investigation Unit (SIU) Fraud Analytics
According to industry estimates, insurance fraud costs property-casualty carriers tens of billions of dollars annually, adding an estimated $400 to $700 per year in extra premiums to the average American family's insurance bills. Fraud is divided into two broad categories:
Hard Fraud: Deliberate, premeditated fabrications designed to collect insurance payouts. Examples: Staged multi-vehicle collisions, arson for profit, phantom passengers who were never in the car, and fabricated burglaries.
Soft Fraud (Opportunistic Fraud): The exaggeration or padding of an otherwise legitimate claim. Examples: Claiming pre-existing hail dents occurred during a recent storm, inflating lost wage claims, or medical providers "upcoding" routine physical therapy sessions. THE FRAUD SPECTRUM
- HARD FRAUD SOFT FRAUD
- ┌───────────────────────────────┐ ┌───────────────────────────────┐
- │ • Premeditated, staged events │ │ • Opportunistic padding │
- │ • Arson for profit │ vs. │ • Pre-existing damage claims │
- │ • Phantom claimants & cars │ │ • Exaggerated wage losses │
- │ • Organized crime syndicates │ │ • Inflated medical treatments │
- │ • 100% fabricated losses │ │ • Legitimate event exaggerated│
- └───────────────────────────────┘ └───────────────────────────────┘
- Social Network Analysis (SNA) and Graph Analytics
Traditional fraud detection evaluated claims in isolation, looking for individual red flags (e.g., policy bound 3 days before total loss). However, modern sophisticated fraud is orchestrated by organized crime syndicates —collaborative networks of corrupt medical clinics, unethical personal injury attorneys, dishonest auto body repair shops, and "cappers" (individuals paid to stage accidents).
When evaluated one-by-one, each individual claim appears completely innocuous: different policyholders, different accident dates, and different vehicles. However, Social Network Analysis (SNA) maps these relationships using graph databases (nodes representing entities, edges representing connections): GRAPH ANALYTICS FRAUD RING DETECTION
- [Accident 1] ──► (Claimant A) ──────┐
- │ │
- ▼ ▼
- [Tow Yard X] [Medical Clinic M] ◄─── (Doctor D)
- ▲ ▲
- │ │
- [Accident 2] ──► (Claimant B) ──────┘
- │
- ▼
- [Attorney L] ◄───── (Capper C) ──────► [Body Shop B]
- ▲ ▲
- │ │
- [Accident 3] ──► (Claimant C) ──────────────────────────────┘
By executing graph algorithms (e.g., degree centrality, community detection, shared entity linkage), the SNA engine identifies that:
- Three separate multi-vehicle accidents over six months all utilized the exact same towing operator,
- All claimants were treated at the exact same physical therapy clinic by the same attending physician,
- All claimants retained the exact same personal injury law firm, and
Two of the claimants share a bank routing account number or residential address.
The system instantly flags the entire cluster as an organized staged accident ring , transferring the files to the Special Investigation Unit (SIU) and state law enforcement authorities for joint prosecution.
5. Claim Litigation Propensity Models
Attorney representation dramatically inflates claims costs. Once a claimant retains legal counsel, average indemnity payouts double or triple, cycle times stretch from months into years, and defense legal fees (Defense and Cost Containment LAE) compound rapidly.
Why Do Claimants Hire Attorneys?
Actuarial studies reveal that claimants rarely hire attorneys immediately after an accident. Instead, attorney retention is typically triggered by operational friction and communication failures :
Adjuster failing to return phone calls for 72 hours.
Delays in authorizing a rental replacement vehicle.
Dispute over a $500 property damage deductible.
Perceived cold, adversarial, or dismissive adjuster tone.
Predictive Intervention Protocols
Litigation Propensity Models score incoming claims on a 0 to 100 scale, estimating the probability that the claimant will seek legal representation. When a claim receives a high litigation propensity score, the claims engine triggers proactive intervention:
High-Touch Assignment: Routes the file to an adjuster renowned for empathetic, clear customer communication.
Expedited Property Settlement: Authorizes immediate payout for uncontested vehicle damage and rental coverage, eliminating the customer's initial frustration.
Advance Medical Payments: Issues immediate payment for out-of-pocket medical co-pays, resolving the claimant's financial anxiety before they seek billboard litigation attorneys.
6. Computer Vision & Automated Damage Appraisal
Computer Vision has revolutionized auto collision and property damage assessment. Using deep Convolutional Neural Networks (CNNs) trained on tens of millions of historical damaged vehicle photographs, automated estimating engines can appraise physical damage from smartphone photos uploaded by the policyholder. COMPUTER VISION APPRAISAL WORKFLOW
- [Policyholder Smartphone Photos] (Uploaded via Mobile App)
- │
- ▼
- [Convolutional Neural Network (CNN) Parsing]
- │
- ┌───────────────────────────┼───────────────────────────┐
- ▼ ▼ ▼
- [Part Identification] [Damage Classification] [Decision Engine]
- Front bumper cover • Minor scratch • Repair vs. Replace logic
- Left headlight assembly • Deep tear / rupture • CCC / Mitchell parts db
- Quarter panel / hood • Structural frame impact • Localized labor rates
- │
- ▼
- [Automated Repair Estimate Generated]
- (If total < $3,500 and no frame damage: Instant Payout to App)
- The Operational Trade-Off: Virtual Estimating vs. Physical Teardown
While computer vision achieves remarkable speed for minor external damage (e.g., scraped bumpers, cracked tail lights), it has inherent physical limitations:
Hidden Sub-Structural Damage: Surface photos cannot reveal bent suspension components, compromised cooling radiators, or micro-cracks in high-strength steel chassis behind exterior plastic bumper covers.
Supplement Inflation: If a vehicle is authorized for repair based purely on a superficial photo estimate, the body shop frequently discovers extensive hidden damage upon physical teardown, resulting in multiple costly "supplements" that erase initial administrative savings.
Optimal Hybrid Deployment: Carriers restrict straight-through automated photo payouts to low-speed, single-point impact claims under a defined monetary threshold (e.g., $3,000 to $5,000), requiring physical adjuster inspection whenever airbags deploy or structural frame rails are impacted.
7. Natural Language Processing (NLP) in Claims Operations
Up to 80% of actionable claims data resides in unstructured text documents: adjuster activity logs, police accident narratives, independent medical examinations (IMEs), physical therapy billings, and formal legal demand letters.
Core NLP Applications in Claims
Named Entity Recognition (NER): Automatically scans unstructured medical records to identify and extract International Classification of Diseases (ICD-10) diagnosis codes, Current Procedural Terminology (CPT) treatment codes, and surgical recommendations.
Demand Letter Information Extraction: When plaintiff attorneys issue formal policy limit demand packages (often hundreds of pages of scanned PDFs), NLP algorithms ingest the documents, immediately extracting policy limit numbers, statutory deadline dates, itemized medical specials, and allegations of bad faith, preventing missed legal response windows.
Adjuster Diary Sentiment and Activity Monitoring: Analyzes adjuster file notes for compliance red flags, such as unexplained 30-day contact lapses, unresolved claimant inquiries, or hostile adjuster language that could compromise the carrier in bad-faith discovery.
- 8. Comprehensive Worked Scenario: Metro Mutual's End-to-End Analytics Deployment
- The Incident
On a rainy Tuesday morning at 8:15 AM, Sarah , an insured of Metro Mutual , is involved in a three-vehicle chain-reaction collision on an interstate highway. Her connected vehicle's telematics system triggers an automated eCall crash notification transmitting vehicle telemetry to Metro Mutual's claims engine:
Delta-V impact: 16.5 mph (Moderate front collision).
Airbag deployment: False.
Seatbelts latched: True (Driver and front passenger).
Vehicle drivable: False (Radiator breached). METRO MUTUAL REAL-TIME CLAIMS ENGINE
- Telematics Crash Alert (eCall) ──► FNOL Complexity Engine (Score: 42) ──► Dispatches Tow Truck
- ──► Authorizes Rental Car
- │
- ▼
- Photographs Uploaded ───────────► Computer Vision CNN ─────────────────► $4,850 Physical Damage
- ──► Structural Alert: Frame Clear
- │
- ▼
- Claimant Diary & Medical Text ──► NLP & Graph Engine (SNA) ────────────► Litigation Score: 18 (Low)
- ──► SIU Fraud Index: 4 (Clean)
- │
- ▼
- Outcome: Direct Virtual Repair Authorization + Fast-Track Settlement in 4 Days (LAE reduced by 60%)
- The Algorithmic Claims Execution
Instant FNOL Complexity Scoring & Triage:
The claims engine scores the collision at a Complexity Index of 42 . Because the vehicle is non-drivable, the system immediately dispatches an automated roadside towing service to transport the vehicle to a certified network repair facility, while sending a digital voucher for a rental vehicle directly to Sarah's smartphone.
Computer Vision Damage Assessment:
At the repair facility, high-resolution photos are uploaded to the platform. The convolutional neural network identifies damage to the front bumper cover, radiator support, and left headlight assembly, generating an itemized repair estimate of $4,850 matched against local parts inventories.
Severity & Litigation Propensity Monitoring:
Two days post-accident, the passenger reports neck soreness. The NLP engine parses the clinic intake notes: "Cervical strain, no radiculopathy, full range of motion preserved." The severity model estimates ultimate bodily injury exposure at $3,200, while the litigation propensity score remains very low (18/100) due to immediate customer satisfaction with the auto repair and rental vehicle.
SIU Fraud & Social Network Verification:
The graph analytics engine evaluates all parties, body shops, and medical clinics involved in the crash against Metro Mutual's enterprise fraud database. Zero shared nodes or suspicious historical linkages are detected (Fraud Index: 4/100).
Enterprise Outcome:
The physical damage claim is repaired and paid within 4 days. The minor bodily injury claim settles amicably for $3,500 with zero attorney involvement. Metro Mutual resolves the entire file with an indemnity accuracy of 100% while reducing total Loss Adjustment Expenses by 60% compared to legacy manual workflows.
Common Exam Traps in Claims and Fraud Analytics
Caution
Trap 1: The Fraud Score Denial Trap An insurer cannot deny an insurance claim or cancel a policy solely because a predictive fraud model generated a high fraud score. A high fraud score merely serves as an internal triage mechanism to refer the file to the Special Investigation Unit (SIU). Any claim denial must be grounded in concrete, independent, verifiable factual evidence developed through human investigation, or the carrier commits bad faith.
Warning
Trap 2: Computer Vision Cannot Overrule Physical Frame Inspection Examination questions may ask whether computer vision photo-estimating completely eliminates the need for physical body shop teardowns. It does not. Photo estimation is highly reliable for superficial sheet metal and plastic bumper covers, but it cannot detect internal chassis, suspension, or structural mechanical damage obscured beneath panels.
Note
Trap 3: Litigation Propensity Models and Fair Settlement Statutes An insurer cannot use a litigation propensity model to stonewall or delay claims filed by claimants who have hired attorneys. State Unfair Claims Settlement Practices Acts require prompt, fair, and equitable settlement negotiations regardless of whether a claimant is represented by legal counsel.