Pet Health Data and Insurance Underwriting: Using EMR Access to Price Risk More Accurately
Pet Health Data and Insurance Underwriting: Using EMR Access to Price Risk More Accurately
Human health insurance was transformed by electronic health records. Pet insurance is a decade behind most underwriting still relies on breed, age, and a pet owner's self-reported health history. The MGAs that figure out how to access and use real pet health data will price risk better, reduce adverse selection, and win market share.
What Is the Current State of Pet Health Data in Underwriting?
Today, most pet insurance underwriting relies on self-reported information from pet owners breed, age, species, and pre-existing condition disclosures which is often inaccurate and susceptible to adverse selection. Veterinary medical records are rarely accessed, and there is no standardized system equivalent to human health insurance's electronic health records.
1. How Pet Insurance Underwrites Today
| Factor | Source | Accuracy |
|---|---|---|
| Breed | Pet owner self-report | Medium (misidentification common) |
| Age | Pet owner self-report | Medium (often estimated) |
| Species | Pet owner self-report | High |
| Location | Address/zip code | High |
| Pre-existing conditions | Pet owner disclosure | Low (under-reporting) |
| Prior claims | Prior insurer data (limited) | Medium |
| Vet history | Rarely accessed | N/A |
2. The Problem With Current Underwriting
- Adverse selection — Pet owners who know their pet is sick are more likely to buy insurance
- Under-disclosure — Owners may not know or report pre-existing conditions
- Breed-only pricing — Treats all Golden Retrievers the same regardless of individual health
- Claims surprise — High claims in early months indicate pre-existing conditions
3. The Data Opportunity
| Data Type | Current Access | Value for Underwriting |
|---|---|---|
| Breed-specific claims data | Available (industry data) | High — baseline pricing |
| Individual vet records | Very limited | Very high — individual risk |
| Genetic testing | Emerging | High — predictive risk |
| Wearable/activity data | Limited | Medium — health proxy |
| Prior claims history | Limited (CLUE for pets doesn't exist) | Very high — claims predictor |
| Vaccination history | Limited | Low-medium — care proxy |
What Does the Veterinary EMR Landscape Look Like?
The veterinary EMR landscape is highly fragmented, with 100+ systems across 30,000+ US vet practices and no interoperability standard equivalent to HL7/FHIR. Major players include IDEXX Neo/Cornerstone and Covetrus with limited API access, while newer startups like Shepherd and Digitail are API-first and more open to insurance data partnerships.
1. Major Veterinary EMR Systems
| System | Market Share | Data Access | Notes |
|---|---|---|---|
| IDEXX Neo/Cornerstone | Large | Limited API | Dominant in large practices |
| Covetrus (Pulse/AVImark) | Large | Limited | Being consolidated |
| eVetPractice | Medium | Cloud-based, API possible | Growing |
| Shepherd | Growing | Modern, API-first | Startup, insurance-friendly |
| Digitail | Growing | Cloud, modern | Startup, data-forward |
| PetDesk | Medium | Engagement platform | Pet owner-facing |
2. Data Access Challenges
| Challenge | Details |
|---|---|
| Fragmentation | 100+ EMR systems across 30,000+ US vet practices |
| No interoperability | No HL7/FHIR equivalent for veterinary data |
| Vet clinic ownership | Clinics own their data, no obligation to share |
| Privacy | Pet owner consent required for data sharing |
| Data quality | Inconsistent coding, free-text notes, varied formats |
| Cost | Data partnerships are expensive to establish |
What Data Sources and Partnerships Are Available?
Data sources range from low-cost pet owner self-reporting during enrollment to high-cost partnerships with veterinary chains like VCA and Banfield. A phased approach works best: start with pet owner data collection and your own claims database (available now), then pursue EMR vendor partnerships (6–18 months), and finally build direct EMR integrations with real-time data feeds (18–36 months).
1. Data Partnership Options
| Partner Type | Data Available | Access Model | Cost |
|---|---|---|---|
| Vet clinic chains (VCA, Banfield) | Patient records (with consent) | Partnership agreement | High |
| EMR vendors (IDEXX, Covetrus) | Aggregated health data | Data licensing | High |
| Pet health data startups | Aggregated, normalized data | API subscription | Medium |
| Pet DNA companies (Embark, Wisdom) | Genetic health markers | Partnership | Medium |
| Pet wearable companies | Activity/behavioral data | Partnership/API | Low-medium |
| Pet owner direct | Self-reported + uploaded records | Enrollment flow | Low |
2. Building Data Access
Phase 1: Pet Owner Data (Available Now)
- Collect health history during enrollment (questionnaire)
- Request vet records upload at enrollment
- Build claims database from your own policies
- Partner with vet clinics for referral + data sharing
Phase 2: Partnership Data (6–18 Months)
- Partner with veterinary chains for anonymized data
- License breed health data from research institutions
- Integrate with emerging pet health platforms
- Build relationships with EMR vendors
Phase 3: Integrated Data (18–36 Months)
- Direct EMR integration (with pet owner consent)
- Real-time health data from wearables
- Genetic risk data from DNA testing partners
- Predictive models using combined data sources
How Can Health Data Be Applied to Underwriting?
Health data enables a shift from blunt breed-and-age pricing to individual risk assessment. With access to veterinary records, you can verify pre-existing conditions (reducing adverse selection), score individual risk using claims history plus health data, implement dynamic pricing that rewards healthy behavior, and fast-track underwriting for pets with verified health status.
1. Individual Risk Assessment
With health data access, you could:
| Application | Data Needed | Impact |
|---|---|---|
| Pre-existing condition verification | Vet records | Reduce adverse selection |
| Individual risk scoring | Claims history + health data | More accurate pricing |
| Dynamic pricing | Ongoing health data | Reward healthy behavior |
| Fast-track underwriting | Verified health status | Better customer experience |
| Claims prediction | Historical health patterns | Loss ratio improvement |
2. Pricing Improvement
| Pricing Approach | Current | With Health Data |
|---|---|---|
| Breed-based | "All Goldens pay $50/month" | "Healthy Goldens pay $42, high-risk pay $58" |
| Age-based | "All 5-year-olds pay $X" | "Well-managed 5-year-olds pay less" |
| Pre-existing | "Self-reported exclusions" | "Verified condition history" |
| Risk selection | "Accept all breeds" | "Price individual risk accurately" |
3. Expected Improvement
| Metric | Without Health Data | With Health Data |
|---|---|---|
| Loss ratio | 60–70% | 55–62% |
| Adverse selection | Significant | Reduced |
| Pricing accuracy | ±15–20% | ±8–12% |
| Customer satisfaction | Healthy pets overpay | Fairer pricing |
| Competitive advantage | None (everyone same) | Significant |
What Are the Privacy, Compliance, and Ethical Requirements?
Pet health data is linked to pet owner personal information, making it subject to CCPA and evolving state privacy laws. You need explicit pet owner consent for any health data access, must practice data minimization, provide right-to-delete capabilities, and be transparent about what data influences underwriting decisions. Ethically, data should improve outcomes for pet owners not just MGA margins.
1. Data Privacy Requirements
| Requirement | Details |
|---|---|
| Pet owner consent | Explicit consent for health data access |
| CCPA compliance | Pet health data linked to pet owner (personal data) |
| State privacy laws | Varies by state, evolving rapidly |
| Data minimization | Collect only what's needed for underwriting |
| Data security | Encrypt, access control, audit logging |
| Data retention | Clear policies on how long data is kept |
| Right to delete | Pet owners can request data deletion |
2. Regulatory Considerations
- State DOIs may have positions on using health data in pricing
- Fair pricing requirements cannot use data to discriminate unfairly
- Transparency must disclose what data is used in underwriting
- Emerging regulations around AI/ML in insurance pricing
- Pet insurance-specific regulations (NAIC Model Act) may evolve
3. Ethical Guidelines
- Use data to price fairly, not to exclude
- Ensure data improves outcomes for pet owners (not just MGA margins)
- Be transparent about what data is collected and how it's used
- Allow pet owners to see and correct their data
- Don't penalize pet owners who decline to share data excessively
For veterinary data strategies and AI in claims, see our dedicated guides.
Frequently Asked Questions
1. Can you access veterinary records for underwriting?
Limited currently. Access comes through pet owner authorization, vet partnerships, data aggregators, and claims history. No standardized pet EMR system exists.
2. How does health data improve underwriting?
Enables individual risk assessment beyond breed/age. More accurate pricing, reduced adverse selection, better rates for healthy pets.
3. What are the challenges?
Fragmented EMR systems, no interoperability, privacy concerns, inconsistent data quality, and regulatory uncertainty.
4. What data is most valuable?
Prior claims history, diagnosed conditions, and breed-specific health markers. Wearable and genetic data are emerging sources.
5. What are the major veterinary EMR systems and how accessible is their data?
Major systems include IDEXX, Covetrus, and eVetPractice with limited API access. Newer startups like Shepherd and Digitail are API-first and more open to insurance data partnerships.
6. How can pet insurance MGAs build data access over time?
Follow a three-phase approach: collect pet owner data at enrollment now, pursue vet chain and EMR vendor partnerships at 6–18 months, and build direct EMR integrations with real-time data feeds at 18–36 months.
7. What improvement in loss ratio can health data provide?
Loss ratios can improve from 60–70% to 55–62%, pricing accuracy improves from plus or minus 15–20% to plus or minus 8–12%, and adverse selection is significantly reduced.
8. What are the privacy and ethical requirements for using pet health data?
You need explicit pet owner consent, CCPA compliance, data minimization, encryption, clear retention policies, right-to-delete support, and full transparency about how data influences underwriting decisions.
External Sources
Internal Links
- Explore Services → https://insurnest.com/services/
- Explore Solutions → https://insurnest.com/solutions/
