How Risk Is Evaluated, Priced, and Managed

​

Healthcare pricing within PEO-sponsored and level-funded benefit programs is driven by structured underwriting, pooled risk mechanics, and regulatory requirements.

​

The information requested during quoting and renewal is not arbitrary—it is essential to accurately assess risk, align pricing with expected claims, and maintain long-term plan stability.

​

The following outlines the key components of underwriting, why specific data is required, and how it ultimately impacts pricing and outcomes.

​

​​

Why do healthcare carriers in some states — and all PEOs — require dependent-level information?

 

In certain small group markets (including NY, NJ, MA, and VT), state law requires carriers to collect dependent-level enrollment data. All PEO-sponsored plans now require it as well, regardless of where the company is headquartered.

​

There are three primary reasons:

 

1. Accurate Pricing

Premiums are structured by coverage tier (employee only, employee + spouse, employee + child(ren), family). Dependent-level data ensures premiums are calculated correctly and prevents under- or over-billing.

​

2. Eligibility & Compliance

Carriers must confirm that covered dependents meet eligibility requirements (legal spouse, domestic partner where applicable, age limits, etc.). This supports accurate claims processing, COBRA administration, and compliance with federal and state regulations.

 

3. Pooled Plan Administration (PEOs)

PEOs combine employees from multiple companies into a single benefits structure. Complete enrollment data allows carriers and PEOs to administer coverage accurately across the pooled population.

​

Importantly, this information is used solely for plan administration and compliance, handled securely, and not shared beyond what is required to administer benefits.

​

​​

Why do PEOs request a recent healthcare invoice and renewal information?

 

Underwriting is fundamentally about assessing risk.

 

While PEO underwriters review demographics, census data (average age, gender distribution, enrolled lives including dependents), location, industry, and plan design, a recent healthcare invoice or renewal provides a real-world view of how the market has already priced your group.

​

Your most recent renewal:

• Reflects actual claims activity and utilization trends

• Shows how your population is currently being assessed

• Helps align proposed PEO pricing with current market conditions

• Reduces the risk of post-implementation or renewal surprises

​​

If a renewal is approaching (within 60–90 days), it becomes especially relevant to pricing discussions.

​

All documentation is used strictly for underwriting and pricing alignment — not to identify or diagnose individual employees.

​

In short, recent renewal data allows for more reliable pricing and fewer surprises after implementation.

 

 

Why are minimum participation requirements required (typically ~50%)?

 

Minimum participation rules in the healthcare plan exist to prevent adverse selection.

​​

Adverse selection occurs when only employees who anticipate high healthcare usage enroll while healthier employees decline coverage.

 

Participation thresholds ensure:

​

• A balanced risk pool

• Predictable premium stability

• Protection against renewal volatility

• Fairness across employers in pooled arrangements

​​

Participation requirements ultimately protect employers from disproportionate cost increases and help maintain plan stability.

​​​​​​

​​​​

Why do PEOs and open market carriers for level-funded products require PHQs (Personal Health Questionnaires) for very small groups (< 5 employees)?

 

For very small groups (typically under five employees), Personal Health Questionnaires (PHQs) are often required because traditional demographic underwriting lacks actuarial credibility at that size.

​

With such limited enrollment, each individual represents a meaningful share of total expected claims. As a result, a single high-cost condition can materially distort projected loss ratios.

​

Fully insured vs. level-funded (key distinction)

• Fully insured (ACA-compliant small group):
  PHQs are not allowed. These plans are guaranteed issue and community rated, meaning no medical underwriting is permitted.

​​

• Level-funded / partially self-funded (PEO or open market):
  PHQs are typically required for very small groups because these plans involve risk-based underwriting.

​​

Why PHQs are used

​

Most PEO medical programs—and many level-funded arrangements in the open market—operate within large pooled structures (often level-funded or partially self-funded).

​

To protect the integrity of that broader risk pool, carriers apply additional underwriting diligence when a very small group enters the plan.

 

PHQs provide visibility into:

• Chronic condition prevalence

• Specialty pharmacy exposure

• Overall claims severity risk

 

These are key drivers that aren’t captured in a basic census.

​​

Important clarification

​

All individual medical information collected through PHQs is protected under HIPAA (Health Insurance Portability and Accountability Act) and is never disclosed to the employer.

​​

​​​

Are PEOs and carriers using AI to underwrite group health plans?

​

Yes. Many carriers and PEO-sponsored health programs now incorporate predictive analytics and advanced risk modeling into underwriting and renewal pricing decisions.

​

Platforms such as Gradient AI’s SAIL™ model (https://www.gradientai.com/) analyze de-identified historical medical and pharmacy claims data (when available), along with demographic inputs, to generate a normalized, forward-looking risk assessment. These tools allow carriers to evaluate projected severity, utilization trends, and overall claims exposure more precisely than traditional census-only underwriting.

​

Importantly, while specific carrier and PEO relationships are not typically publicly disclosed, platforms like Gradient AI are widely embedded across the underwriting ecosystem—including insurance carriers, TPAs, captives, and PEO-sponsored programs—meaning these models are often influencing pricing even when not explicitly visible to the employer.

​

What underwriters are actually evaluating

​

At a high level, these models generate normalized risk scores, typically benchmarked around 1.0 (average risk):

• > 1.0 → higher-than-average expected claims risk

• < 1.0 → lower-than-average expected claims risk

 

For example, a score of 1.18 indicates projected claims risk approximately 18% above baseline, while a score in the 0.80–0.90 range would generally be viewed as favorable.

 

These scores are driven by a combination of:

• Demographic factors (age, gender, geography)

• Morbidity indicators (chronic conditions, utilization patterns)

• Pharmacy exposure, particularly specialty drugs

 

Underwriters evaluate these inputs collectively to form a forward-looking view of expected claims over the next 12 months.

​

Why this matters for employers

​

This shift toward predictive underwriting has several implications:

• Pricing is increasingly data-driven and forward-looking, not just based on basic census data

• Two groups with similar demographics can be priced very differently based on underlying risk indicators

• Renewal outcomes are more closely tied to emerging claims trends, not just historical averages

 

For employers, this reinforces the importance of understanding how their population is being evaluated—not just what the final pricing is.

​

Important privacy clarification

​

All data used in these models is de-identified and handled in accordance with the Health Insurance Portability and Accountability Act.

​

Analytics vendors operate as HIPAA Business Associates, meaning they are legally required to protect health information and can only use it for approved underwriting and administrative purposes. Employers never see individual-level medical data.

​

Bottom line

​

Underwriting is increasingly driven by predictive, data-driven risk modeling—not just static demographics.

 

In many cases, employers are being evaluated using these models behind the scenes, even if it is not explicitly communicated—making it critical to understand not just pricing outcomes, but the underlying drivers behind them.

​​​