Stop-Loss

Stop-Loss in Health Insurance and Provider Payment

Stop-loss moves the cost of health claims above a threshold away from the party that would otherwise pay them, usually onto an insurer or a public payer. In health economics the term has two common uses, cover that a self-funded employer or an at-risk provider group buys against very large claims, and the annual ceiling on a patient's own out-of-pocket payments within an insurance policy. Both versions trade protection against large losses for weaker cost incentives above the threshold, which is why stop-loss turns up in debates on capitation, risk adjustment and benefit design. The financing sense is meant here, not the stop-loss order used in trading.

Two kinds of ceiling on retained loss

A self-funded United States employer pays its employees' claims from its own resources and can buy stop-loss insurance to limit that exposure. The Department of Labor's Technical Release 2014-01 describes stop-loss contracts as covering claims costs above a set amount, the attachment point, for a single covered person or for aggregate claims over a period. The National Association of Insurance Commissioners (NAIC) white paper on stop-loss and self-funding calls the single-person layer specific cover, against claim severity, and the group layer aggregate cover, against claim frequency. The per-person layer and aggregate stop-loss are usually two components of one policy.

The second use concerns the insured person. Cutler and Zeckhauser, in their review of the anatomy of health insurance, list three cost-sharing features of service benefit policies, the deductible, the coinsurance rate and the stop loss amount, which they define as the maximum out-of-pocket payment by the person in a year. The two uses share one shape, each fixing the point beyond which further claims cost the protected party nothing, but the protected party and the regulation differ.

The payment rule for a per-person layer

Under specific stop-loss each covered person's eligible claims for the contract period are compared with the attachment point. A general form of the carrier's payment allows for partial reimbursement and a maximum payment.

$$L_i = \min\{\, r \times \max(C_i - K_i,\ 0),\ M \,\}$$

where $L_i$ is the stop-loss payment for person $i$, $C_i$ is that person's eligible claims in the contract period, $K_i$ is the attachment point that applies to the person, $r$ is the reimbursement share of claims above the attachment point (1 for full reimbursement) and $M$ is any maximum payment under the contract. The employer's retained claims for the person are $C_i - L_i$, which equals $\min(C_i, K_i)$ only when $r$ is 1 and no maximum is reached.

Attachment points need not be uniform. The NAIC white paper describes a laser specific attachment point, a higher attachment point for one or more individuals with known high-cost conditions or other identified risk factors. The same paper notes that, without an endorsement extending cover, a policy reimburses only claims incurred and paid in the same policy year.

For the patient version, the yearly out-of-pocket payment under a deductible, coinsurance and a stop-loss can be written as follows.

$$P(X) = \min\{\, \min(X, D) + c \times \max(X - D,\ 0),\ S \,\}$$

where $P(X)$ is the patient's out-of-pocket payment in the year, $X$ is the total covered bill, $D$ is the deductible, $c$ is the coinsurance rate and $S$ is the stop-loss amount. The insurer pays $X - P(X)$. The bill at which the ceiling binds is $X^{*} = D + (S - D)/c$, and above $X^{*}$ the patient's marginal price of covered care is zero.

Worked example with four members and a laser

All numbers in this example are illustrative. A self-funded employer's plan has four members whose eligible claims for the year are USD 8,000 (A), USD 30,000 (B), USD 90,000 (C) and USD 260,000 (D), a total of USD 388,000. Its specific stop-loss policy has an attachment point of USD 50,000, no maximum payment and no exclusions.

1. Full reimbursement and a uniform attachment point. With $r$ equal to 1, members A and B produce no recovery. Member C produces 90,000 − 50,000 = 40,000 and member D produces 260,000 − 50,000 = 210,000, so recoveries total 250,000. Retained claims are 8,000 + 30,000 + 50,000 + 50,000 = 138,000, which matches 388,000 − 250,000.

2. A laser on member D. With D's attachment point lasered to USD 150,000, D's recovery falls to 260,000 − 150,000 = 110,000. Recoveries become 40,000 + 110,000 = 150,000 and retained claims 388,000 − 150,000 = 238,000, so the laser hands USD 100,000 back to the employer.

3. A 90% reimbursement share without a laser. With $r$ equal to 0.9, recoveries are 0.9 × 40,000 = 36,000 for C and 0.9 × 210,000 = 189,000 for D, a total of 225,000. Retained claims are 388,000 − 225,000 = 163,000, which is the 138,000 of step 1 plus 10% of the 250,000 excess.

4. Adding the premium. At a premium of USD 45,000 for the step 1 policy, the employer's total cost is 138,000 + 45,000 = 183,000. Total claims are still USD 388,000, and the contract only changes who pays them.

5. The patient version. A policy has a deductible of USD 500, coinsurance of 20% and a stop-loss of USD 2,000. For a USD 3,000 bill the patient pays 500 + 0.2 × 2,500 = 1,000, below the ceiling. For a USD 30,000 bill the uncapped amount would be 500 + 0.2 × 29,500 = 6,400, so the stop-loss binds, the patient pays 2,000 and the insurer pays 28,000. The ceiling binds at a bill of 500 + 1,500 / 0.2 = 8,000.

6. Interpretation. The attachment point alone does not fix what the employer bears, since retained claims here range between USD 138,000 and USD 238,000 under the same USD 50,000 headline attachment point. For the patient, covered care beyond a bill of USD 8,000 is free at the point of use for the rest of the year.

Why a threshold contract appeals to a risk-averse buyer

Arrow's 1963 analysis of medical care gives the theory. He showed that if an insurer will sell any policy at a premium that depends only on its actuarial value, a risk-averse buyer chooses 100 per cent coverage above a deductible minimum, which is the stop-loss form. If the insurer is also risk-averse, the Pareto optimal policy involves some coinsurance above the minimum, which corresponds to a reimbursement share below 100%. Arrow named administrative costs and capital tied up by irregular payments as reasons premiums exceed actuarial value.

Under expected utility theory the value of the cover rises with the buyer's risk aversion and the spread of its claims. The carrier's expected payment for one person under full reimbursement is $E[(C - K)_+]$, the expected value of claims above the attachment point, which depends on the shape of the upper tail of the claims distribution as well as on its mean. Groups with the same average cost can therefore face very different stop-loss prices, and a small group, with less risk pooling, values the cover more.

Stop-loss for providers paid by capitation

When a payer shifts financial risk to physicians, stop-loss limits how much of that risk they carry. The Medicare Advantage rules on physician incentive plans (42 CFR 422.208) set a risk threshold of 25% of potential payments for risk based on referral services, and physicians above it must have aggregate or per-patient stop-loss protection. Aggregate protection must cover 90% of the costs of referral services that exceed 25% of potential payments, and per-patient protection at least 90% of referral costs above a deductible that depends on the size of the patient panel.

To set those deductibles, CMS applies the central limit theorem to a sample of fee-for-service Medicare claims and finds, with 98% confidence, the point at which a multi-specialty group of a given panel size would lose no more than 25% of potential payments through referral services. The method reflects the law of large numbers, since average cost per patient varies less in a larger panel.

Hospital payment has a similar layer. Under Medicare inpatient prospective payment, 42 CFR 412.84 bases the outlier payment for an extraordinarily high-cost case on 80% of the difference between the hospital's adjusted cost for the discharge and the outlier threshold, or 90% for burn cases. A fixed price per diagnosis-related group with such a layer behaves like case-based payment with stop-loss at an 80% or 90% reimbursement share.

Risk sharing alongside risk adjustment

Where competing insurers receive risk-adjusted capitation payments, stop-loss style payments are one form of risk sharing between insurers and the sponsor or regulator. Cutler and Zeckhauser describe paying plans on actual experience, such as USD 50,000 extra for each organ transplant patient, as after-the-fact reinsurance. Van Barneveld and colleagues state the trade-off directly, since risk sharing reduces an insurer's incentives for selection as well as its incentives for efficiency.

Their 1998 comparison of mandatory pooling set excess-of-loss pooling, a stop-loss layer paid from a mandatory common pool, against high-risk pooling and proportional pooling. Holding fixed the percentage of total costs for which an insurer remains at risk, high-risk pooling gave the greatest reduction in incentives for cream skimming. Their 2001 study notes that some countries use outlier or proportional risk sharing to supplement demographic capitation, and found that other forms gave better trade-offs, which matters for social health insurance funds paying competing insurers.

How employer stop-loss is regulated in the United States

The NAIC Stop Loss Insurance Model Act, where a state adopts it, bars policies with an annual per-individual attachment point below USD 20,000, sets separate aggregate floors, and bars stop-loss from providing direct coverage of an individual's health care expenses. Technical Release 2014-01 adds that stop-loss does not usually guarantee the payment of benefits to plan participants.

Stop-loss in economic evaluation and budget impact

The perspective decides how stop-loss enters a model. From the societal perspective a carrier's recovery paid to the employer is a transfer, and the resource costs are the claims, the loading for administration and capital, and any change in care caused by altered incentives. From the employer's perspective, premiums and retained claims count and recoveries reduce cost.

In a budget impact analysis, an expensive drug takes the patient past any out-of-pocket ceiling quickly, leaving the payer with almost all of the marginal cost.

Limits and common misreadings

The most frequent error is to treat the attachment point as a firm cap on cost, when reimbursement shares, lasers, maximum payments, exclusions and the incurred-and-paid basis all leave some large claims with the employer, and the premium is paid regardless. Above the attachment point the employer or provider bears little or none of the marginal cost of claims in the contract year, apart from renewal pricing, which weakens the incentive to manage high-cost care. This mirrors the patient-side moral hazard that arises once an out-of-pocket ceiling binds. Set against that, the protection lets smaller organisations accept capitated contracts and reduces the gain from avoiding expensive patients. The two meanings should also be kept apart, since a patient's stop-loss limits hardship from out-of-pocket costs, whereas organisational stop-loss limits the risk retained by whoever pays the claims.

Sources

• Arrow KJ. Uncertainty and the welfare economics of medical care. American Economic Review. 1963;53(5):941-973. Section IV and Appendix.
• Cutler DM, Zeckhauser RJ. The anatomy of health insurance. In: Culyer AJ, Newhouse JP, eds. Handbook of Health Economics. Vol 1A. Elsevier; 2000:563-643. doi:10.1016/S1574-0064(00)80170-5. Working paper version, NBER Working Paper 7176, 1999. https://www.nber.org/papers/w7176
• National Association of Insurance Commissioners. Stop Loss Insurance Model Act (Model 92). Adopted 1995, amended 1999. https://content.naic.org/sites/default/files/model-law-092.pdf
• National Association of Insurance Commissioners. Stop Loss Insurance, Self-Funding and the ACA. White paper. 2015. https://content.naic.org/sites/default/files/inline-files/SLI_SF.pdf
• Code of Federal Regulations. 42 CFR 412.84, Payment for extraordinarily high-cost cases (cost outliers). eCFR, accessed 28 October 2026. https://www.ecfr.gov/current/title-42/section-412.84
• Code of Federal Regulations. 42 CFR 422.208, Physician incentive plans, requirements and limitations. eCFR, accessed 28 October 2026. https://www.ecfr.gov/current/title-42/section-422.208
• US Department of Labor, Employee Benefits Security Administration. Guidance on State Regulation of Stop-Loss Insurance. Technical Release 2014-01. 6 November 2014. https://www.dol.gov/agencies/ebsa/employers-and-advisers/guidance/technical-releases/14-01
• van Barneveld EM, Lamers LM, van Vliet RCJA, van de Ven WPMM. Mandatory pooling as a supplement to risk-adjusted capitation payments in a competitive health insurance market. Social Science and Medicine. 1998;47(2):223-232. doi:10.1016/S0277-9536(98)00056-2
• van Barneveld EM, Lamers LM, van Vliet RCJA, van de Ven WPMM. Risk sharing as a supplement to imperfect capitation: a tradeoff between selection and efficiency. Journal of Health Economics. 2001;20(2):147-168. doi:10.1016/S0167-6296(00)00077-1