Kentucky Ass'n of Health Plans, Inc. v. Nichols

KENNEDY, Circuit Judge,

dissenting.

I write separately with respect to Part III of the majority’s opinion to dissent from the majority’s conclusion that Ky. Rev.Stat. Ann. §§ 304.17A-2701 and 304.17A-171(2) (Banks-Baldwin 1995) fall within ERISA’s insurance savings clause. I believe that Kentucky’s any willing provider laws have little to do with insurance and are not saved from preemption by ERISA’s Insurance Savings Clause as they do not regulate insurance as a matter of common sense and fail all three of the McCarran-Ferguson factors.

I. Insurance Savings Clause

A.

The district court in the case at bar devoted only a sentence to the common *373sense test, merely concluding that, “the common-sense view is that the AWP statutes regulate the business of insurance.”2 Instead the court spent most of its time analyzing the McCarran-Ferguson factors, which significantly overlap the common sense test, being designed largely to guide a court’s common sense determination. See UNUM Life Ins. of Am. v. Ward, 526 U.S. 358, 119 S.Ct. 1380, 1389, 143 L.Ed.2d 462 (1999).

As the Supreme Court has generally started its savings clause analysis with the common sense test, I begin there as well. See id. at 366-68, 119 S.Ct. at 1386; See also Davies v. Centennial Life Ins. Co., 128 F.3d 934, 940-941 (6th Cir.1997). In doing so, I conclude that the Kentucky AWP laws do not meet the common sense test because they are directed at the contracts between benefit plans and third parties, rather than being specifically directed at the insurance industry. See Pilot Life Ins. Co. v. Dedeaux, 481 U.S. 41, 50, 107 S.Ct. 1549, 95 L.Ed.2d 39 (1987) (concluding that to fall within the savings clause a law must be specifically directed at the insurance industry, which in part requires that the law define the terms of the relationship between the insurer and the insured, rather than the insurer and a third party): The laws do not change the relationship between the insurer and insureds, as the same medical conditions are covered after the AWP laws as were insured before the passage of these provisions. The underwriting of risk, the traditional earmark of insurance, see Group Life & Health Ins. Co. v. Royal Drug Co., 440 U.S. 205, 211-12, 99 S.Ct. 1067, 1073-74, 59 L.Ed.2d 261 (1979), is in no way affected. Insureds are not free to attend the provider of then-choice, as the AWP laws merely require employee benefit plans to accept previously excluded doctors that are qualified, willing to join the plan, and agree to abide by its terms.

Contrary to the majority’s assertions, I believe it is also apparent that Kentucky’s AWP laws clearly target more than just members of the insurance industry.3 By their terms, Kentucky’s AWP laws apply to non-ERISA covered self-insured plans by defining health benefit plans to include “a self-insured plan ... to the extent permitted by ERISA.” §§ 304.17A-100(4)(a) and 304.17A-170(1). This definition includes self-insured plans not regulated by ERISA, such as government plans and church plans, which ERISA excludes from its coverage. See 29 U.S.C. § 1003(b). The result is that these self-insured plans, which, as a matter of common sense ought *374to be considered as operating outside the insurance industry, are subject to Kentucky’s AWP laws. C.f. CIGNA Healthplan of La., Inc. v. Louisiana, 82 F.3d 642, 649 (5th Cir.1996) (concluding that the statute was not saved as an insurance regulation because it was not limited to entities in the insurance industry as it applied to entities such as self-funded organizations, Taft-Hartley Trusts, or employers who establish and participate in self-funded trusts or programs as well as various intermediaries); Blue Cross & Blue Shield of Alabama v. Nielsen, 917 F.Supp. 1532, 1538-39 (N.D.Ala.1996) (holding that Alabama’s AWP statutes did not regulate insurance as a matter of common sense because they applied to all employee benefit plans, including self-funded plans, and were thus not specifically directed toward the insurance industry) (vacated in part on appeal by Blue Cross & Blue Shield of Alabama v. Nielsen, 142 F.3d 1375 (11th Cir.1998) due to mootness after the Alabama Supreme Court determined in a certified question that the law at issue did not apply to plaintiff); Stuart Circle, 995 F.2d at 503-04 (concluding that the AWP statute at issue, which only applied to insurance companies actively issuing insurance policies, met the common sense test as a law regulating insurance).

Significantly, the AWP provisions also apply to third parties that a self-insured ERISA plan hires to administer its plan benefits. As used in §§ 304.17A-110(3) and 304.17A-171, “health benefit plan” also includes Hospital, Health Service Corporation, and HMO contracts.4 Section 304.17A-270 uses the term “Health Insurer” instead of “Health Benefit Plan,” but similarly defines it to include any Insurance Company, HMO, or HSC.5 Thus, Kentucky’s AWP laws, by there very terms, prohibit all HMO’s, HSC’s and Insurance Companies from “discriminatfing] against any provider who is located within the geographic coverage area of the health benefit plan....” See §§ 304.17A-110(3) and 304.17A-171. The AWP laws would clearly apply to HMOs, HSCs, or Insurance Companies providing plan administration services, as there is nothing in the statute that would exclude them in such a scenario.

Although HMOs, HSCs, and Insurance Companies may accept risk in some situations, as third party administrators they would merely be contracting to handle paperwork and plan administration for a self-insured ERISA plan. While handling such administrative duties, however, these enti*375ties would be forced by Kentucky’s AWP laws to accept any willing provider into the plan, even though they were not underwriting any risk. The law in this instance is not directed at the business of insurance, as no insurance is involved. See Prudential Ins. Co., 154 F.3d at 829 (observing that the Arkansas AWP statute at issue defined health benefit plans so broadly as to include plan administrators and thus did not fit within a common sense view of a law directed specifically toward the insurance industry); Texas Pharmacy Ass’n v. Prudential Ins. Co. of America, 105 F.3d 1035, 1039 (5th Cir.1997) (noting that a self-insured employer would not be subject to Texas’ AWP statute, “but if the employer signed up with an HMO or PPO, those organizations would be subject to the statute, even if there is no insurance involved”); CIGNA Healthplan of La., Inc. 82 F.3d at 649 (concluding that Louisiana’s AWP law was obviously not limited to entities within the insurance industry, in part because it applied to “health care financiers, third party administrators, providers, or other intermediaries”). The only risk underwritten is that accepted by the ERISA self-insured plan, which under the “deemer clause” of ERISA § 514(b)(2)(B), 29 U.S.C. § 1144(b)(2)(B), cannot be treated as an insurance company for the purposes of state regulation. See FMC Corp. v. Holliday, 498 U.S. 52, 61, 111 S.Ct. 403, 409, 112 L.Ed.2d 356 (1990); Texas Pharmacy Ass’n, 105 F.3d at 1039. The common sense conclusion that can be drawn from the AWP statute’s coverage of entities clearly operating outside of the business of insurance is that the statute is concerned generally with regulating provider access to networks rather than specifically regulating the business of insurance.6 See Prudential Ins. Co., 154 F.3d at 829 (concluding that the AWP law at issue was “not a law directed at the insurance industry at all, but a law directed at regulation of broadly defined health benefit plans, only some of which fall within the insurance industry”).

The majority seeks tó refute this conclusion by arguing that Kentucky’s AWP law would not apply to entities performing only plan administration for self insurers, apparently on the grounds that to the extent that Kentucky’s AWP law seeks to do so, they would be preempted by ERISA. In support the majority cites Light v. Blue Cross and Blue Shield of Alabama, Inc., 790 F.2d 1247, 1248 (5th Cir.1986), which holds that state laws which apply to administrators of self insured plans may be preempted by ERISA and “if ERISA preempts state law, there is no applicable state law to which the administrator must conform.” Other courts have drawn a similarly unremarkable conclusion. See Insurance Bd. Under Social Ins. Plan of Bethlehem Steel Corp. v. Muir, 819 F.2d 408, 410-13 (3d Cir.1987) (concluding that Blue Cross and Blue Shield, which provided plan administration services for a self-insured employee benefit plan, were not engaged in the business of insurance and therefore that the state mandated benefit laws were preempted with respect to not only the underlying self-insured plan, but also the plan administrators).

Light, however, is inapposite to the proposition the majority ultimately seeks to advance, that is, if Kentucky’s AWP law does not apply to self-insurers, a fortiori, it also does not apply to entities administering the self-insured plan. The employee benefit plan at issue in Light contained a provision stating that “the contracts between South Central Bell and the plan administrator necessarily will conform to applicable state laws.” Id. at 1248. Plaintiffs argued that- since the plan adopted state law, the plan administrator was required to comply with that law. The Fifth Circuit began by noting that if ERISA preempted state law, there would be no state law for the administrator to conform *376to. The court then turned to Plaintiffs’ argument that the preemption provision of 29 U.S.C. § 1144(a) should not be read to apply to plan administrators. Plaintiffs asserted that because their action was against a plan administrator, it did not relate to an employee benefit plan, and therefore was not preempted. The court found no merit in plaintiffs’ argument because, “[ajbsent ERISA the state common law on which [plaintiffs] rely would provide causes of action for the improper handling of claims under benefit plans,” giving the law a “direct connection” with employee benefit plans. Id. at 1249. As a consequence, the court found § 1144 preempted plaintiffs’ state causes of action with respect to the administrator of their self-insured plan as well. Id.

In the case at bar, the majority states that based in part on reliance on Light, it does not believe that Kentucky’s AWP law could be enforced against plan administrators of self-insured plans. The majority appears to arrive at this conclusion due to its belief that, as in Light, Kentucky’s AWP statute would be preempted insofar as it applies to plan administrators, given that Kentucky’s AWP statute “relates to” employee benefit plans. However, having concluded that the entire AWP provision is preempted, the majority fails to apply savings clause analysis to the entire AWP provision. As I have discussed, Kentucky’s law, by its very terms, does attempt to regulate entities outside the insurance industry, through its regulation of HMO and HSC and Insurance Company contracts, which would include contracts purely for plan administration. The majority attempts to ignore this fact by summarily concluding that the law would not apply to plan administrators. However, in doing so the majority is apparently saying that the law is clearly preempted and not saved insofar as it applies to plan administrators, and then arguing that due to this preemption, the statute does not apply to entities outside the insurance industry. The majority cites no authority for its attempt to dissect Kentucky’s AWP provisions, dismissing some of the provision’s applications as preempted and then applying saving clause analysis only to those applications it wishes to retain. In applying the common sense test we should be looking at the whole provision, not determining whether portions of a provision are saved in some, but not all scenarios.

In sum, unlike Ward, where the notice-prejudice rule was by its very terms “directed specifically at the insurance industry and [was] applicable only to insurance contracts,” 526 U.S. at 366-68, 119 S.Ct. at 1386, Kentucky’s AWP laws clearly apply to entities and contracts outside the insurance industry.7 As a result, I must conclude that §§ 304.17A-110(3) and 304.17A-171(2) do not “satisfy the commonsense view as ... regulation[s] that hone in on the insurance industry” but rather the provisions “just have an impact on [that] industry.” See Ward, 526 U.S. at 368, 119 S.Ct. at 1387 (quoting Pilot Life, 481 U.S. at 50, 107 S.Ct. at 1554).

*377B.

Next I “consider [the] three factors employed to determine whether the regulation fits withiri the ‘business of insurance’ as that phrase is used in the McCarran-Ferguson Act: first whether the practice has the effect of transferring or spreading a policyholder’s risk; second, whether the practice is an integral part of the policy relationship between the insurer and the insured; and third, whether the practice is limited to entities within the insurance industry.” Ward, 526 U.S. at 367, 119 S.Ct. at 1386 (internal citations and quotations omitted). The McCarran-Ferguson factors only serve to reinforce my conclusion that Kentucky’s AWP laws are not saved from preemption by ERISA § 514(b). While none of the McCarran-Ferguson factors should be seen as dispositive, they provide helpful guideposts for savings clause analysis, lending depth and guidance to the common sense test, particularly in difficult cases such as the one we are presented with. See Ward, 526 U.S. at 372-74, 119 S.Ct. at 1389. With this understanding I begin by considering whether Kentucky’s AWP laws have the effect of transferring or spreading policyholder risk.

Appellees argue, and the majority agrees, that the district court was correct in finding that Kentucky’s AWP laws spread policyholder risk. The district court reached this conclusion because it believed the laws were similar to the mandated benefit laws that the Supreme Court found to spread risk in Metropolitan Life Ins. Co. v. Massachusetts, 471 U.S. 724, 105 S.Ct. 2380, 85 L.Ed.2d 728 (1985). According to the district court, risk was spread because Kentucky’s AWP laws made it more likely that an insurer would cover a visit to a provider that the policyholder may have paid to visit on his or her own, due to the policyholder’s preference for that particular provider.

I disagree with the district court’s attenuated risk spreading analysis and find its analogy to Metropolitan Life to be misplaced. Metropolitan Life involved a Massachusetts law which required insurers to provide certain minimum mental health benefits to Massachusetts residents. Id. at. 724, 105 S.Ct. 2380. While applying the McCarran-Ferguson factors to the Massachusetts law, the Supreme Court concluded that risk was being spread because the statute was intended “to effectuate the legislative judgment that the risk of mental-health care should be shared.” Id. at 743, 105 S.Ct. at 2391.

Mandated benefit laws like the 'One at issue in Metropolitan Life require an insurer to treat individuals with health care conditions that may not have been covered before the law. The effect of such laws is to shift a significant degree of risk from individuals who originally had no such coverage under the insurance policy, to the insurer. Although a closer question, the same can likely be said for mandated provider laws. Under such laws, an insurance company must provide coverage for treatment from any category of provider,8 which typically broadens insurance coverage to include forms of alternative medicine (e.g., chiropractic, massage therapy, acupuncture) not previously covered under the policy. The Ninth Circuit confronted such a law in Washington Physicians Serv. Ass’n v. Gregoire, 147 F.3d 1039 (1998). In analyzing whether the law fell under the savings clause as a law regulating insurance, the court began by recognizing that:

Because insurance is the business of spreading risk, Royal Drug, 440 U.S. at 205, 99 S.Ct. 1067, a state law that regulates the relationship between a carrier and a provider without affecting the risk borne by the insured is outside the definition of insurance regulation. Id. at 213-214, 99 S.Ct. 1067; Hahn v. Oregon *378Physicians Serv., 689 F.2d 840, 843-844 (9th Cir.1982).

Id. at 1045. The court concluded, however, that by expanding the kinds of treatment that an insurer must cover to include various types of alternative medicine, the risk that the insured might need alternative medical treatment was shifted to the insurance company. See Gregoire, 147 F.3d at 1046. In reaching this conclusion, the court distinguished Washington’s mandatory provider law from AWP laws, noting that the mandatory provider statute did not require any carrier to contract or deal with any particular provider; instead it merely forbade a carrier from excluding a particular class of provider. See Gregoire, 147 F.3d at 1046.9

AWP statutes such as Kentucky’s are different in purpose and effect than either the mandatory benefit or mandatory provider statutes respectively at issue in Metropolitan Life and Gregoire. Rather than shifting risk from policyholders to insurers, Kentucky’s AWP statutes merely prohibit benefit plans from excluding qualified providers who want to join the plan’s provider network and are able to meet the plan’s requirements. The risk assumed by the benefit plan under its policy, that the policyholder will require medical treatment, remains unaltered. The statute’s passage in no way alters the terms of the policyholder’s policies.10 The only contracts affected are those between the benefit plan and the providers already in the plan network. Kentucky’s AWP laws do not require that a single doctor be added to any benefit plan network; benefit plans may still maintain provider networks which require doctors to meet certain terms before joining and the plans may still require that policyholders see plan doctors for their medical costs to be covered. See Gary A. Francesconi, Erisa Preemption of “Any Willing Provider” Laws-An Essential Step Toward National Health Care Reform, 73 Wash. U.L.Q. 227, 248-49 (1995) (observing that with or without the any willing provider law, the insured’s access to certain providers is limited by the policy).

While doctors who meet the plans qualifications may independently decide to join the plan and the plan must accept them, many doctors may not meet the plan’s qualifications or may have no desire to join that particular plan. For example, in an area with multiple benefit plans, a relatively small percentage of additional qualified doctors may decide to join any particular plan network. In such a scenario, many doctors may already be members of several plan networks and may have no desire to enroll in additional networks. Even those qualified doctors who were previously excluded from all the plans in such an area would be unlikely to join every plan after the AWP law. The result is that although Kentucky’s AWP laws make it marginally more likely that a policyholder’s benefit plan network will contain their preferred doctor,11 they will still be restricted to the doctors in their benefit plan network regardless of the membership or nonmembership of their preferred doctor.

The Supreme Court discussed the transfer of policyholder risk in detail in Group *379Life & Health Insurance Co. v. Royal Drug Co., 440 U.S. 205, 99 S.Ct. 1067, 59 L.Ed.2d 261 (1979). The Court was considering whether agreements between Blue Shield and participating pharmacies concerning the price that policyholders would pay for prescriptions, constituted the business of insurance. Under the agreements, the participating pharmacies would fill any Blue Shield policyholder’s prescription for a two-dollar charge. See id. at 209, 99 S.Ct. at 1072. The pharmacy was then entitled to reimbursement from Blue Shield for the cost of acquiring the drugs prescribed. See id. The Court found that, under the antitrust exemption provision of the McCarran-Ferguson Act, 15 U.S.C. § 1012(b), the arrangement did not constitute the “business of insurance” because Blue Shield’s agreement with the pharmacies merely limited the amount that it would have to pay to cover its policyholder risks. See id. at 214-15, 99 S.Ct. at 1075. While realizing that the policyholders might ultimately benefit in the form of lower rates charged by Blue Shield, the Court concluded that no additional policyholder risks were being underwritten. Instead, the Court saw the agreements as merely arrangements for the purchase of goods and services by Blue Shield and thus not the “business of insurance.” See id. at 214, 99 S.Ct. at 1075.

The majority attempts to distinguish Royal Drug on the grounds that “[ujnlike the health plans that are the subject of the Kentucky statute, the insurer in Royal Drug did not restrict, the number of-providers in question.” However, the insurer in Royal Drug merely offered to enter into agreements with each licensed pharmacy in Texas. This is significantly different from actually entering into such agreements, because, as the Supreme Court observed, “only pharmacies that can afford to distribute prescription drugs for less than this $2 markup can profitably participate in the plan.” Id. at 209, 99 S.Ct. at 1072. As a result, the insurer’s promise to enter into agreements with all pharmacies was essentially an empty one, as only larger, high volume pharmacies could afford to enter into such an agreement. Such a hollow offer provides no real basis to distinguish the case from the limited provider networks at issue in the case at bar.

Similarly, Kentucky’s AWP laws have almost no effect on the policyholder risk that insurers must underwrite. Like Blue Shield’s unchanged obligation to cover a policyholder’s prescriptions after entering into the pharmacy agreements, Kentucky insurers must cover the same medical procedures after the AWP law as they would have to if the AWP provision had not been enacted. It is true that the AWP laws would likely force insurers to pay higher rates, however, as in Royal Drug, this would only affect the price that insurers must pay for procedures covered in the policies they have issued, not the type of policyholder risks that they contractually must cover.12 Accord, Prudential Ins. Co., 154 F.3d at 828 (finding that the Arkansas AWP statute plainly did not spread risk, and noting that appellants did not even attempt to argue to the contrary).

In discussing the nature of insurance, the Royal Drug Court referred to SEC v. Variable Annuity Life Ins. Co., 359 U.S. 65, 79 S.Ct. 618, 3 L.Ed.2d 640 (1959), quoting it for the proposition that “the concept of insurance involves some investment risk-taking on the part of the company.” The Court noted thát while *380petitioners didn’t dispute this fact, they maintained that the pharmacy agreements did involve the underwriting of risk. To this end, they argued that Blue Shield accepted a premium and assumed the risk that policyholders would need prescription drugs by agreeing to enter into contracts with existing pharmacies to provide such drugs for promised reimbursement from Blue Shield. The Court observed that “the fallacy of petitioner’s position is that they confuse the obligations of Blue Shield under its insurance policies, which insure against the risk that the policyholders will be unable to pay for prescription drugs during the period of coverage, and the agreements between Blue Shield and the participating pharmacies, which serve only to minimize the costs Blue Shield incurs in fulfilling its underwriting obligations.” The court then noted that the benefit promised to the policyholders, that they would only have a two-dollar co-payment for any prescriptions, remained unchanged by the arrangement.

As a result the court concluded that Blue Shield’s arrangements left policyholders basically unconcerned (from a financial perspective) about the arrangements Blue Shield entered into with participating pharmacies. This is not to say that at least some participants would not be disappointed by the fact that their pharmacy of choice might not have been included, if for example it was not large enough to provide prescriptions for only a two-dollar markup. However, the financial risks that Blue Shield agreed to cover remained unchanged, leaving the policyholders without concern for Blue Shield’s cost savings agreements with the larger pharmacies.

As the majority notes, in the case at bar the policyholders were also unconcerned with the insurer’s compensation arrangements. However, the majority concludes that the policyholders were concerned with restrictions on their freedom of choice in seeking medical treatment. Ignoring for a moment the fact that Kentucky’s AWP laws do not allow policyholders the freedom to choose their own doctor, and thus only addresses this concern in at best a very tangential way, any concerns over freedom of choice are beside the point. The critical issue with respect to the risk spreading prong, as well as whether the law regulates insurance as a matter of common sense, is whether or not the law is related to the risks underwritten by the insurer. As the Court noted in Royal Drug, while discussing the meaning of “business of insurance” under the McCarran-Ferguson Act: “It is important, therefore, to observe at the outset that the statutory language in question here does not except the business of insurance companies from the scope of the antitrust laws. The exemption is for the ‘business of insurance,’ not the ‘business of insurers.’” Id. at 210-11, 99 S.Ct. at 1073. Because Kentucky’s AWP laws seek to merely regulate the “business of insurers” by dictating how they structure their provider networks, irrespective of the risks they underwrite, they should not qualify for savings clause protection.

The majority points to Stuart Circle Hosp. Corp. v. Aetna Health Management, 995 F.2d 500 (4th Cir.1993), as support for its contention that Kentucky’s AWP laws do transfer risk. Stuart Circle involved a Virginia statute which required insurers to accept any willing provider if they established preferred provider networks. The court concluded that while the statute “related to” ERISA plans, it was not preempted, because it fell within ERISA’s savings clause as a law regulating insurance. Id. at 504. After finding the common sense test satisfied, in part because the law was located within Virginia’s insurance code,13 the court moved on to the *381McCarran-Ferguson factors. Discussing whether the law transferred or spread policyholder risk, the court concluded that the Virginia statute “affects the type and 'cost of treatment available to an insured.” Id. The court reached this conclusion because it believed that without the AWP law, policyholders might on occasion attend a doctor outside of their plan network, due to a personal preference for that doctor, and in doing so, would be forced to shoulder all or part of the cost. See id.

I am unpersuaded by the majority’s argument that Stuart Circle’s attenuated risk spreading rationale should be applied to Kentucky’s AWP laws. As I have discussed, Kentucky’s AWP laws do not require health benefit plans to include a single additional provider unless the qualified provider, not the policyholder, decides to join the plan. Policyholders are not necessarily any better off than they were before the law was passed, only benefitting from the law if their provider is both willing to join their particular provider network and able to meet its requirements. The insurance policy and the contingencies it underwrites, i.e., the risk that insured will need medical treatment for a condition covered under the policy, remain the same, regardless of Kentucky’s AWP law.14 There is no shifting of the risk between the policyholder and insurer that a specific medical contingency will occur.

The majority attempts to escape from Royal Drug’s teaching by relying on dicta from American Drug Stores, Inc. v. Harvard Pilgrim Health Care, Inc., 973 F.Supp. 60 (D.Mass.1997), in which the court questioned “whether the holding of Royal Drug may be translated into the ERISA preemption context at all.” Id. at 72. The American Drug Stores court was referring to the fact that Royal Drug interpreted the scope of the antitrust exemption under § 2(b) of the McCarran-Ferguson Act, but did not address §§ 1 and 2(a) of the act, which preserved a State’s ability to regulate the “business of insurance” from Commerce Clause attack. See Royal Drug, 440 U.S. at 219 n. 18, 99 S.Ct. at 1077.15 However, the Supreme Court has not indicated that it is troubled by this distinction. As recently as last term, the Court employed the McCarran-Ferguson factors it created in Royal Drug and im*382ported into ERISA preemption analysis in Metropolitan Life, without any indication that its past precedent interpreting whether a law “regulates insurance” under ERISA’s savings clause was called into doubt. See UNUM Life Ins. Co. of America v. Ward, 526 U.S. 358, 373, 119 S.Ct. 1380, 1388, 143 L.Ed.2d 462 (1999).16 Until the Court indicates that its Royal Drug business of insurance test is no longer appropriate in the ERISA savings clause context, something that it has yet to do, I believe we must continue to apply the test as dictated by the Court’s precedent.

Moving to the second McCarran-Fergu-son factor, I consider whether Kentucky’s AWP laws affect an integral part of the policy relationship between the insurer and insured. The majority asserts that the district court was correct in finding that Kentucky’s AWP laws dictate a substantive term of the contract between the insurer and insured and are thus an integral part of this relationship. As support for this proposition, the majority again cites Stuart Circle, which concluded that because Virginia’s AWP law affected treatment and cost (through the same attenuated manner in which the court concluded risk was spread) it was integral to the insurer-insured relationship. 995 F.2d 500, 503.

Again I find Stuart Circle unconvincing. The effect of Kentucky’s AWP laws center on the insurer-provider relationship. The terms of the insurer-insured relationship are only affected in a very indirect manner, making it difficult to see the AWP laws as integral to that relationship. See Prudential Ins. Co., 154 F.3d at 830 (finding the Arkansas AWP law not to be integral to the insurer-insured relationship, as it defined only the terms between the insurer and providers, and distinguishing Gregoire because Washington’s any category of provider law did affect the insurer-insured relationship by expanding the kinds of treatment the policy must cover); Cf., Pilot Life Co. v. Dedeaux, 481 U.S. 41, 50-51, 107 S.Ct. 1549, 1554-55, 95 L.Ed.2d 39 (1987) (noting that Mississippi’s law of bad faith, in contrast to the mandated benefits law in Metropolitan Life, did not define the terms between the insurer and insured, and thus only affected the insurer-insured relationship in an attenuated way). The Court’s discussion of California’s notice-prejudice rule in UNUM Life Ins. Co. of America v. Ward, 526 U.S. at 372-76, 119 S.Ct. at 1389-90 (1999), provides an illuminating example of what qualifies as an integral policy relationship under the McCarran-Ferguson factors. The Supreme Court observed that:

[California’s notice-prejudice rule] serves as an integral part of the policy relationship between the insurer and insured. Metropolitan Life, 471 U.S. at 743, 105 S.Ct. 2380. California’s rule changes the bargain between insurer and insured; it effectively creates a mandatory contract term that requires the insurer to prove prejudice before enforcing a timeliness-of-elaim provision.

Id. In sharp contrast, Kentucky’s AWP provisions leave the contract terms between the insurer and insured, unaltered. The relationships directly affected by the law are those existing between insurers and third parties (i.e., medical providers). As discussed above, the medical risks covered by the policy remain the same. Thus, even if an insured’s preferred provider decides to join the insured’s network, and complies with its terms in doing so, the medical coverage that the insurer has contracted to underwrite remains unchanged. Unlike the mandated benefit laws at issue *383in Metropolitan Life, the mandated provider law in Gregoire, or the mandatory notice-prejudice rule in Ward, Kentucky’s AWP laws do not force the insurer to offer a benefit to insureds that was not available before the law. Rathér; Kentucky’s AWP laws merely force insurers to potentially make additional contractual arrangements with providers they might otherwise exclude. The medical conditions covered remain unaffected and the insureds are still limited to the plan’s network of providers. Therefore, I must conclude that Kentucky’s AWP law is not integral to the insured-insurer relationship.

Finally, I consider whether the Kentucky AWP laws are limited to entities within the insurance industry. As discussed under the common sense Test, I do not believe this to be the case. The law not only regulates entities that fall outside the traditional definition of insurer, it also extends to include entities in no way involved in underwriting risks. In fact, a review of the statute shows that while it may affect the way that some insurance companies run their business, it has nothing to do with the underwriting of risk, the traditional earmark of insurance. See Royal Drug, 440 U.S. at 211-12, 99 S.Ct. at 1073-74. Accordingly, I believe that Kentucky’s AWP laws fail the third prong of the McCarran-Ferguson test as well.

In sum, I am forced to conclude that §§ 304.17A-110(3) and 304.17A-171(2), Kentucky’s AWP laws, are not saved from preemption as laws that regulate the business of insurance, because under ERISA § 514(b), they fail to meet not only the common sense test, but also all of the McCarran-Ferguson factors.17 While federalism concerns prohibit federal courts from lightly preempting acts of a state legislature, I agree with the Eighth Circuit’s observation in Prudential Ins. Co. that, “it is for Congress, not the courts, to reassess ERISA in light of modern insurance practices and the national debate over health care.” Prudential Ins. Co., 154 F.3d at 829-30.

II. Severability

Because I believe that Kentucky’s AWP provisions, Ky.Rev.Stat. Ann. § 304.17A-110(3) and 304.17A-171(2) (Banks-Baldwin 1995), are preempted by 29 U.S.C. § 1144(a) and not saved by the savings clause of 29 U.S.C. § 1144(b)(2)(A), I write separately to address whether the laws should be invalidated in their entirety, or whether the offending provisions of the statutes may be severed in whole or part. While never explicitly stating so, as dis-cusséd above, the majority appears to agree that Kentucky’s AWP laws are preempted at least in their application to HMOs and HSCs who are performing administrative or related duties for self-insured employee benefit plans. However, despite implicitly drawing this conclusion, the majority still concludes that the AWP law is saved by the insurance savings clausé. In doing so, the majority fails to mention, let alone discuss, how it severs Kentucky’s AWP provision in such a manner that it no longer applies to third party plan administrators' performing administrative functions for self-insured plans, thereby allowing it to conclude that the provision is saved in all other applications.

The Supreme Court has often stated that the inquiry into whether a statute is severable is essentially an inquiry into legislative intent. See, e.g., Minnesota v. Mille Lacs Band of Chippewa Indians, 526 U.S. 172, 119 S.Ct. 1187, 143 L.Ed.2d 270 (1999); Zobel v. Williams, 457 U.S. 55, 64, 102 S.Ct. 2309, 2315, 72 L.Ed.2d 672 (1982); Regan v. Time, Inc., 468 U.S. 641, 653, 104 S.Ct. 3262, 3269, 82 L.Ed.2d 487 (1984) (plurality opinion). In discerning the legislature’s intent, the Court has directed that:

Unless it is evident that the legislature would not have enacted those provisions which are within its power, independent*384ly of that which is not, the invalid part may be dropped if what is left is fully operative as a law. Champlin Refining Co. v. Corp Comm’n of Okla., 286 U.S. 210, 234, 52 S.Ct. 559, 76 L.Ed. 1062 (1932). See also Alaska Airlines, Inc. v. Brock, 480 U.S. 678, 684, 107 S.Ct. 1476, 94 L.Ed.2d 661 (1987); Regan v. Time, Inc., supra, at 653, 104 S.Ct. 3262.

Mille Lacs Band of Chippewa Indians, 526 U.S. at 191, 119 S.Ct. at 1198. Therefore, as evidence of the legislature’s intent I look to Kentucky’s severability statute, which states:

It shall be considered that it is the intent of the general assembly in enacting any statute, that if any part of the statute be held unconstitutional the remaining parts shall remain in force, unless the statute provides otherwise, or unless the remaining parts are so essentially and inseparably connected with and dependent upon the unconstitutional part that it is apparent that the general assembly would not have enacted the remaining parts without the unconstitutional part, or unless the remaining parts, standing alone, are incomplete and incapable of being executed in accordance with the intent of the general assembly.

Ky.Rev.Stat. Ann. § 446.090. The statute indicates that if any part of a Kentucky statute is found unconstitutional, it is the legislature’s intent that the remaining provisions be saved. However, in drafting the statute the legislature also recognized that situations will exist where severing the offending parts cannot be considered as it would make it impossible to carry out the original legislative intent.

I begin my severability inquiry by addressing whether portions of Ky.Rev.Stat. Ann. §§ 304.17A-110(3) and 304.17A-171(2) may be severed in such a way that the provisions would not be preempted in their entirety. Because I believe that §§ 304.17A-110(3) and 304.17A-171(2) contain an impermissible connection with ERISA covered plans, there is no way to save the provisions themselves from preemption by severing particular parts while leaving the rest intact. Even if it were possible to do so, merely removing any prohibited “references to” ERISA plans would not suffice. The Kentucky provisions impermissible “connection with” ERISA plans would continue, as ERISA plans would still be effectively prohibited from offering limited provider panels, as none would be available to them. The only way that §§ 304.17A-110(3) and 304.17A-171(2) could be saved, would be to add language effectively allowing health care entities to offer limited provider networks to ERISA covered plans. Because both Kentucky and federal case law prohibit us from adding language to a state statute to remove its conflict with federal law, see, e.g., Musselman v. Commonwealth, 705 S.W.2d 476, 477 (Ky.1986), Eubanks v. Wilkinson, 937 F.2d 1118, 1120 (6th Cir.1991), I conclude that §§ 304.17A-110(3) and 304.17A-171(2) are preempted in their entirety. See Prudential Ins. Co., 154 F.3d at 832 (concluding that the AWP statute at issue was preempted in its entirety); Texas Pharmacy Ass’n, 105 F.3d at 1039 (drawing the same conclusion).

Finally, I address whether the preempted provisions, provision (3) of § 304.17A-110 and provision (2) of § 304.17A-171, may be severed from their respective statutory sections or whether the sections must also be preempted. Kentucky’s sev-erability statute indicates that unconstitutional provisions may presumptively be severed, unless severing the offending provisions makes it impossible to execute the remaining provisions of a statute in the manner the legislature intended. It is clear that §§ 304.17A-110 and 304.17A-171 can be executed in accordance with the legislature’s original intention without provision (3) and (2) respectively. Accordingly, while I conclude that §§ 304.17A-110(3) and 304.17A-171(2) should be preempted in their entirety, the statutory sections they reside in should remain unaffected.

. As noted in the majority's opinion, § 304.17A-110(3) was the provision that was actually before the district court. However, both the parties have agreed that this appeal will govern § 304.17A-270 as well, as it is essentially a recodification of § 304.17A-110(3). Accordingly, unless otherwise noted I would apply my analysis of § 304.17A-110(3) to § 304.17A-270 as well.

. The district court in the companion case of Community Health Partners, Inc. v. Commonwealth of Kentucky, 14 F.Supp.2d 991 (1998) devoted more room to its discussion of the common sense test. The court concluded that the statute satisfied the common sense test because it affected specific terms of the insurance policy, was located in the insurance code and regulated insurers or insurer related entities. Id. at 1001-02. While the district court conducted a careful analysis, I.reach a contrary conclusion on all points for the reasons explained below.

. In addition to traditional insurance companies, Kentucky’s AWP laws apply to entities such as HMO’s and Health Service Corporations. These entities might be argued to fall within the insurance industry if they have agreed to accept the risk that the covered individual will need medical services in exchange for a fee. See Group Life & Health Ins. v. Royal Drug, 440 U.S. 205, 211, 99 S.Ct. 1067, 1073, 59 L.Ed.2d 261 (1979) (stating that, ”[t]he primary elements of an insurance contract are the spreading and underwriting of policyholder risk"). As such, they may be seen as underwriting policyholder risk and then covering that risk with the provision of medical services, rather than by paying for those services as a traditional insurer would. Given the somewhat subtle distinction in this difference, it could be argued that some HMO's and HSC’s are engaged in the business of insurance. See. Washington Physicians Serv. Ass’n v. Gregoire, 147 F.3d at 1045 (adopting this view); but see Texas Pharmacy Ass’n, 105 F.3d at 1038-39 (concluding that the Texas statute was not limited to entities within the insurance industry, as it applied to entities such as HMO’s and PPO’s). However, as Kentucky’s AWP laws also apply to entities that are clearly outside the business of insurance, I believe that it is unnecessary for us to decide this issue.

. Section 304.17A-100(4)(a) defined "Health Benefit Plan” as including:

Any hospital or medical expense policy or certificate; nonprofit hospital, medical-surgical, and health service corporation contract or certificate; a health benefit plan offered by a provider-sponsored integrated health delivery network; a self-insured plan or a plan provided by a multiple employer welfare arrangement, to the extent permitted by ERISA; health maintenance organization contract; and standard and supplemental health benefit plan as established by KRS 304.17A-160.

Ky.Rev.Stat. Ann. § 304.17A-100(4)(a). Section 304.17A-170 supplies a nearly identical definition of "Health Benefit Plan” (by reference to § 304.17A-005) for § 304.17A-171.

. As observed in part I of the majority opinion, § 304.17A-270 recodified § 304.17A-110(3), merely substituting "Health Insurer” for "Health Benefit Plan,” and leaving the AWP provision otherwise unchanged. Section 304.17A-005 defines "Insurer,” as used in § 304.17A-270, as:

[A]ny insurance company; health maintenance organization; self-insurer or multiple employer welfare arrangement not exempt from state regulation by ERISA; provider sponsored integrated health delivery network; self-insured employer organized association, or nonprofit hospital medical-surgical, dental, or health service corporation authorized to transact health insurance business in Kentucky.

Ky.Rev.Stat. Ann. § 304.17A-005 (Banks-Baldwin 1999). This definition largely tracks repealed § 304.17A-100(4)(a)'s definition of "Health Benefit Plan.” Despite the majority's claim that the substitution of the term "Health Insurer” for "Health Benefit Plan” in the current statute excludes administrators under contract with benefits plans from the scope of the statutes, the definition of "insurer” continues to include plan administrators.

. The fact that §§ 304.17A-110(3) and 304.17A-171(2) happen to be codified amongst Kentucky's insurance provisions does not alter this conclusion, as it in no way assures that the law is specifically directed toward the insurance industry.

. In an attempt to dismiss the problem of partial preemption (insofar as applied only to plan administrators), the majority appears to draw a distinction between Kentucky's AWP statute and the AWP statute at issue in Prudential Ins. Co., on the grounds that Kentucky’s statute does not specifically state that it applies to "plan administrators.” However, §§ 304.17A-110(3) and 304.17A-171(2) apply to "Health Benefit Plans,” defined to include HMO and HSC contracts. Similarly, § 304.17A-270 applies to "Health Insurers,” defined to include all Insurance Companies, HMOs and HSCs. These definitions in no way limit the applicability of Kentucky’s AWP provisions to entities underwriting risk. Consequently, if a self-insurer entered into a contract with an HMO, HSC, or Insurance Company solely for the provision of plan administration, there is no reason that the AWP laws would not apply. A plain reading of the AWP provision compels this conclusion regardless of the law's applicability to the self-insured entity the HMO or HSC contracted with. Consequently, one can only conclude that the AWP provisions are preempted, as by their terms they apply directly to HMO’s and HSC’s that have merely agreed to operate as administrators of self-insured employee benefit plans.

. The insurance company’s ability to use limited provided networks remains unaffected by such a law.

. As to the issue of whether AWP laws spread risk, the Gregoire court stated that it expressed no opinion. Gregoire, 147 F.3d at 1047.

. The majority mysteriously concludes that the provision changes the policy between the insured and insurer, “not in terms of covered medical conditions, but in terms of covered treatment by health care providers.’’ However, both the medical risks and corresponding treatment covered by the insurer remain unaltered by Kentucky’s AWP law. Consequently, I assume that by this statement the majority really means that by potentially expanding the number of plan doctors that one could go to receive treatment for a covered condition, the AWP law somehow changes the contract between the insurer and insured.

.If their doctor met the plans qualifications, had wanted to join the plan, but was denied membership, that doctor could no longer be kept out of the network if he or she still wished to join.

. Under Kentucky’s AWP law, benefit plans can no longer guarantee a limited panel of providers to which other doctors will not be added to handle the plan’s patients. To the extent that contracts between benefit plans and providers contained such exclusivity provisions, under the majority’s holding they will no longer be enforceable. While the likely tesult is an increase in the price that insurers must pay, this is not because of an increase in risk, but rather it is due to a decrease in the volume discount an insurer can command if policyholders- are spread across a potentially larger network of plan doctors. See Group Life & Health Ins. Co. v. Royal Drug Co., 440 U.S. 205, 213-15, 99 S.Ct. 1067, 1074-76, 59 L.Ed.2d 261 (1979).

. As indicated above, this fact should have little to do with determining whether a law is specifically directed at the insurance industry, as any law may be inserted within the insur-anee provisions of a state code. Further, the Virginia statute at issue, § 38.2-3407, only applied to "insurers,” defined within the Chapter to mean "an insurance company" *381which is itself defined as "any company engaged in the business of making contracts of insurance." Va.Code Ann. § 38.2-100. Thus, Virginia's AWP statute was more narrow in scope than Kentucky's, quite specifically limiting its application to the term "insurance companies,” defined by the statute to include only companies actively issuing' insurance contracts. Consequently, Virginia’s statute would clearly not apply to entities engaged in only administrative functions for an employee benefit plan.

. The conclusion that Kentucky's AWP laws are not the business of insurance is supported by the following rhetorical question which the Supreme Court posed in Royal Drug. The court addressed a scenario which essentially involved limited provider networks, stating:

Suppose, for example, that an insurance company entered into a contract with a large retail drug chain whereby its policyholders could obtain drugs under their policies only from stores operated by this chain. The justification for such an agreement would be administrative and bulk-purchase savings resulting from obtaining all of the ' company’s drug needs form a single dealer. Even though these cost savings might ultimately be reflected in lower premiums to policyholders, would such a contract be the "business of insurance?” Or suppose that the insurance company should decide to acquire the chain of drug stores in order to lower still further its costs of meeting its obligations to its policyholders. Such an acquisition would surely not be the "business of insurance.”

440 U.S. at 215, 99 S.Ct. at 1075. If such agreements are not the "business of insurance,” then it follows that a state's attempts to regulate these agreements would not qualify as such either.

. The majority appears to be persuaded by this reasoning. I do not draw the, same conclusion. However, even if the meaning of "business of insurance” was broadened beyond the contractual relationship between the policyholder and insurer, I do not believe that Kentucky's AWP law would qualify. As I have explained elsewhere, Kentucky's AWP laws have at most a speculative and tangential effect on this relationship.

. The district court in Community Health Partners v. Commonwealth of Kentucky, 14 F.Supp.2d 991 (W.D.Ky.1998) observed that United States v. Fabe, 508 U.S. 491, 113 S.Ct. 2202, 124 L.Ed.2d 449 (1993) might be argued to change and broaden savings clause analysis in the ERISA context. However, other federal courts have not reached this conclusion and the Supreme Courts most recent opinion in Ward gives no indication that the Court’s savings clause analysis has changed. See also, Prudential Ins. Co., 154 F.3d at 827-828 (concluding that Fabe had no effect on savings clause analysis).

. As we indicated above, the parties have agreed that our holding with regard to § 304.17A-110(3) will serve to determine whether § 304.17A-270 is preempted as well.