In early 2024, a reinsurer issued a quiet underwriting memo that reclassified how diabetes codes were treated under long-term disability (LTD) reinsurance treaties. The memo did not change any policy language, but it changed the risk model underneath it. Within months, one employer-sponsored LTD claim filed by a mid-career professional with well-controlled type 2 diabetes was initially approved, then retroactively reviewed, and ultimately denied. The policyholder had not committed fraud. The employer had not switched carriers. The only thing that changed was the actuarial lens through which the claim was viewed.

This case is not isolated; a similar pattern has emerged in other blocks of business, where coding changes have led to unexpected denials. For the insured, the impact is often invisible until a claim is filed. The story of this one client illustrates how a technical rule change can ripple through the entire disability insurance ecosystem, and why both employers and individuals need to pay closer attention to the fine print of reinsurance arrangements.

The Reinsurer's Rule Change That Shifted One Client's Claims

The memo, dated January 2024, revised the reinsurer's underwriting guidelines for diabetes codes under its disability treaties. Specifically, it tightened the definition of "well-controlled" diabetes by lowering the acceptable HbA1c threshold from 7.0% to 6.8% and adding new requirements for continuous medication adherence. The change was framed as a response to new clinical data suggesting that even well-controlled type 2 diabetes carries higher long-term disability risk than previously assumed.

For the client in question, a 47-year-old operations manager at a regional logistics firm with 250 employees, the impact was immediate. His policy had been in force for six years, and he had filed a claim in March 2024 after a back injury prevented him from performing his job duties. The claim was initially approved, and he received two monthly benefit payments of $4,200 each. Then, in June, the reinsurer flagged his file for review under the new diabetes guidelines.

The review centered on his medical records, which showed an HbA1c level of 7.1%—within the old guidelines but above the new threshold. The reinsurer determined that his diabetes was no longer considered "well-controlled" under the revised definition, and because the claim was still within the elimination period, it was retroactively denied. The employer's HR department was caught in the middle, unable to explain to the employee why his coverage had evaporated.

The policy language itself had not changed. The definitions of disability, elimination period, and benefit amount were identical to what had been in place for years. The change was purely in the underlying risk model that the reinsurer used to price the block of business. This is a critical point: reinsurers do not need to rewrite policies to alter coverage; they simply adjust the assumptions that determine which claims are paid.

How Underwriting Rules Actually Move Premiums

Underwriting rules are the inputs that actuaries use to set premiums. When a reinsurer changes its diabetes coding definitions, it is effectively repricing the risk associated with that condition. This repricing can happen at the treaty level, affecting an entire block of policies, or at the individual level, when a specific policy is up for renewal. In either case, the loss ratio—the ratio of claims paid to premiums collected—is the key metric that drives the change.

If a reinsurer determines that diabetes claims are costing more than expected, it will either increase premiums or tighten underwriting criteria. The 2024 memo was a tightening of criteria, but it also had a premium effect. Some estimates suggest that for blocks with a high concentration of diabetic policyholders, premium shifts of 10–20% were possible over the following renewal cycle. These shifts are rarely communicated to individual insureds, who see only their own premium statement, if they see any change at all.

For employer-sponsored LTD plans, the impact is even more diffuse. The employer negotiates the contract with the primary insurer, who in turn buys reinsurance. The employer may not even know that a reinsurer changed its rules, unless the insurer passes along a rate increase. In the case described here, the employer did not receive a rate increase; instead, the reinsurer's rule change simply meant that one claim was no longer covered. This is the asymmetry of reinsurance: the rules are invisible until a claim is denied.

Actuaries, not claims adjusters, are the ones who set these rules. Adjusters apply the policy language, but the policy language is shaped by actuarial assumptions about risk. When those assumptions change, the practical meaning of the policy changes, even if the words stay the same. This is why the case is so instructive: it shows that the real power in insurance lies not in the claims department, but in the actuarial models that price the risk.

Diabetes Codes: The Underwriting Grey Zone

Diabetes is a chronic condition with a wide spectrum of severity. Type 2 diabetes, which accounts for roughly 90–95% of all cases, is often managed with lifestyle changes and oral medications. Type 1 diabetes, by contrast, requires insulin therapy and carries a higher risk of complications. Underwriters have long treated these two types differently, but the grey zone lies in how they define "well-controlled" for type 2.

HbA1c is the standard biomarker for blood sugar control. The American Diabetes Association recommends a target of below 7% for most adults, but that is a clinical guideline, not an underwriting rule. Reinsurers set their own thresholds, and those thresholds have been creeping lower over the past decade. The 2024 memo lowered the threshold from 7.0% to 6.8%, a seemingly small change that has outsized effects on who qualifies as "well-controlled."

Compounding the issue is the introduction of new ICD-10 codes in 2023 that allow for more granular classification of diabetes subtypes. These codes differentiate between diabetes with and without complications, and they also capture whether the patient is on insulin. Reinsurers have been quick to adopt these codes, using them to segment risk more precisely. But precision cuts both ways: a policyholder who was previously lumped into a broad "type 2" category might now be classified into a higher-risk subgroup, triggering a claim review.

In the case of the operations manager, his medical records showed well-controlled diabetes under the old definition, but the new codes allowed the reinsurer to see that he had a history of intermittent medication lapses. That history was not material to his back injury, but it was material to his overall disability risk profile. The reinsurer argued that the lapses indicated a higher likelihood of future diabetes-related complications, which could extend his disability duration.

The Claims Denial That Exposed the Gap

The denial letter that the operations manager received in July 2024 was a model of bureaucratic opacity. It cited "a change in the underwriting guidelines applicable to your condition" and stated that his claim no longer met the definition of "total disability" under the policy. The letter did not explain what had changed, nor did it offer any avenue for appeal beyond a generic process.

The employer's HR department was unprepared. They had not been notified of the reinsurer's rule change, and they had no documentation to share with the employee. When the employee's attorney requested the underwriting guidelines, the insurer initially refused, citing proprietary business information. It was only after the attorney threatened to file a complaint with the state insurance department that the guidelines were produced, redacted heavily.

The appeals process hinged on the code history. The employee's medical records showed that his HbA1c had been below 7.0% for the two years prior to the claim, but the new threshold was 6.8%. His most recent reading, taken three months before the injury, was 7.1%. The insurer argued that this was a temporary lapse, but the reinsurer's rules did not allow for temporary lapses; they required sustained control over the past 12 months.

No fraud was alleged. The employee had been honest about his diabetes on his application, and his medical records were accurate. The denial was purely a function of the rule change. The case ultimately settled, with the insurer paying a portion of the claim as a goodwill gesture, but the settlement was confidential. The employee lost roughly six months of benefits, and the employer lost a valued team member who resigned shortly after.

Tech Did Not Cause This—But It Amplified It

It would be easy to blame the denial on automated systems, but the core issue was not technological. The reinsurer's rule change was a manual decision made by actuaries, not an algorithm. However, technology did amplify the speed and scope of the change. Automated claims systems can now flag mismatches between a policyholder's medical codes and the latest underwriting guidelines in real time, which is how the operations manager's file was flagged for review within days of the memo's release.

AI underwriting tools, which are increasingly used by both primary insurers and reinsurers, can process vast amounts of medical data and apply new rules retroactively. For example, a 2023 study by the actuarial firm Milliman found that AI-driven underwriting tools can reduce the time to apply new guidelines from months to weeks, and they are being adopted by major reinsurers like Swiss Re and Munich Re. This is a departure from the past, when manual review might have taken months and been applied only to new claims. Now, a rule change can be applied to the entire in-force block within weeks. The speed is a feature, not a bug, from the reinsurer's perspective, but it creates a significant risk for policyholders.

Yet the technology is not the root cause. The root cause is the actuarial model itself, which treats policyholders as statistical risks rather than individuals. An algorithm did not decide that the operations manager's diabetes made him a worse risk; it simply applied the new rule that the actuaries had set. The disruption is in the speed of change, not the logic behind it.

This distinction matters because it points to where the fix lies. If the problem were purely technological, then better algorithms might solve it. But the problem is fundamental to how reinsurance works: risk is transferred, pooled, and repriced on a portfolio basis. Individual policyholders are collateral damage when the portfolio is rebalanced.

What Policyholders and Employers Can Actually Do

For individual policyholders, the first line of defense is to read the fine print on how conditions are defined. Many disability policies include a clause that allows the insurer to update underwriting guidelines, but few policyholders understand what that means. Specifically, look for clauses like "right to amend" or "underwriting guidelines subject to change" in the policy document. It is worth asking the insurer, before a claim is filed, whether any guideline changes have been made that could affect the policy. For example, you could ask: "Have any changes been made to the underwriting guidelines for my condition in the past year? If so, can you provide a summary of those changes?"

Employers, who are the actual purchasers of group LTD coverage, have more leverage. At renewal, they can ask the broker to inquire about any reinsurer rule changes that might affect the block. Some employers have negotiated contracts that require the insurer to notify them of material changes in reinsurance terms. This is not standard, but it is possible. For instance, an employer could request a "reinsurance change notification" provision in the contract, which would require the insurer to inform them of any changes to the reinsurance treaty that could affect claims.

Another option is to consider level-funded or self-insured arrangements, where the employer assumes more risk directly. In these structures, the employer can negotiate directly with a reinsurer, or even bypass reinsurance altogether for certain conditions. This gives the employer more control over how underwriting rules are applied, but it also exposes them to more volatility.

Brokers can also negotiate for more granularity in how codes are applied. For example, an employer might ask that the insurer not use the newest ICD-10 codes for underwriting purposes, or that they use a more lenient HbA1c threshold. These are not standard concessions, but in a competitive market, some insurers are willing to offer them.

The Trade-Off: Transparency vs. Gaming

While the call for more transparency in underwriting rules seems straightforward, there is a significant trade-off: revealing too much could allow policyholders or employers to game the system. If underwriting guidelines are fully disclosed, individuals might adjust their behavior or medical records to meet the criteria, even if that means concealing relevant health information. For example, if the HbA1c threshold is known, a policyholder might delay a test until after a claim is filed, or a doctor might be pressured to record a lower reading. This could lead to adverse selection, where only the healthiest individuals are insured, driving up costs for everyone.

Moreover, proprietary underwriting models are a competitive advantage for reinsurers. Sharing them could erode their market position and reduce innovation. Actuaries spend years developing sophisticated risk models, and they are not eager to give away their intellectual property. This is why guidelines are often treated as trade secrets, even when they have a direct impact on policyholders.

Balancing transparency with the need to prevent gaming is a delicate act. Some reinsurers have begun to publish high-level summaries of their guidelines, but they stop short of revealing the precise thresholds or algorithms. This middle ground allows policyholders to understand the general direction of underwriting changes without enabling manipulation. The challenge is to find the right level of detail that informs without inviting abuse.

The Takeaway: Risk Models Are Living Documents

The case of the operations manager is a reminder that insurance policies are not static contracts; they are repriced continuously based on the latest risk models. Reinsurers manage portfolios, not individuals, and their decisions are driven by loss ratios and capital requirements. When a rule changes, the policyholder is often the last to know.

Transparency is the weakest link in this system. Reinsurers are not required to disclose their underwriting guidelines, and primary insurers often treat them as trade secrets. This asymmetry of information leaves policyholders at a disadvantage, especially when a claim is denied based on a rule they never saw.

Actuaries have a responsibility to communicate model changes more clearly, both to their clients and to the insureds who are affected. This does not mean revealing proprietary formulas, but it does mean explaining the rationale for changes and providing a reasonable transition period. Some reinsurers have begun to do this voluntarily, but it is not yet the norm.

The future of disability insurance is likely to be more granular, with more data and more precise risk classification. That granularity can lead to fairer pricing, but it also creates more opportunities for unexpected denials. As the industry moves in this direction, the lesson from this case is simple: read the fine print, ask questions at renewal, and assume that the rules can change at any time.

One concrete step forward is for industry associations, such as the American Academy of Actuaries, to develop a code of conduct for underwriting changes, including requirements for advance notice and grandfathering of existing claims. This would not eliminate the problem, but it would give policyholders a fighting chance. Until then, the onus is on individuals and employers to stay informed and advocate for themselves.

This article is for informational purposes only and does not constitute professional advice. Consult a qualified insurance professional for guidance on your specific situation.