American Journal of Law & Medicine

Paying for Prevention: Challenges to Health Insurance Coverage for Biomedical HIV Prevention in the United States


Because this Article deals in part with emerging technologies, coverage determinations by private insurers will turn on not only contract interpretation, but also health technology assessment (HTA). This section discusses procedures for determining coverage under employer-based health insurance, because approximately 90% of privately-insured policyholders are insured under this type of plan. (234) These plans consist of employer-purchased commercial group insurance or employer-funded medical care, often administered by a commercial insurer, (235) Employer-based plans are regulated by an overlapping system of internal policies, state laws, and federal laws, principally the Employee Retirement Income Security Act (ERISA) and the ACA. (236) Coverage determinations are also shaped by state and federal insurance mandates, which are enforceable under ERISA. (237) Under the ACA, individual and small-group insurance plans sold in state exchanges must provide coverage for services determined to be "essential health benefits." (238) Where no mandate exists, however, insurance coverage will be a question of contract interpretation. This section discusses principles of interpretation for substantive coverage decisions and the use of HTA.

1. Interpreting Insurance Contracts

Private insurers' coverage decisions for biomedical HIV prevention interventions are a question of contract interpretation. Benefits vary widely under self-funded employer plans and policies purchased by individuals. Group insurance, however, accounts for the majority of health insurance, (239) and most policies are multi-peril plans that provide coverage for a variety of conditions that qualify as "sickness" or "disease," often naming certain conditions to be excluded or included. (240) These multi-peril plans require that a policyholder meet two requirements to obtain coverage for care: (1) he or she must have an "illness, disease, sickness, injury or ... pathological state" that occurs while the policy is in force (or within a window after the policy comes into force), and (2) the expense must be related to that condition and almost always must be "medically necessary" to his or her health. (241) Litigation since the 1960s has blurred these criteria in overall determinations of "medical necessity," (242) a requirement originally imposed by private insurers and managed care organizations to control costs. (243) When determining whether a request for coverage meets the first requirement, courts have construed "sickness" and "illness" to have a plain-language, popular meaning (244) namely, a condition affecting one's usual activities (245) and sickness is deemed to last until one is cured. (246) "Disease," on the other hand, refers to "an alteration in the state of the body or of its organs or tissues, interrupting or disturbing the performance of its vital functions, and causing or threatening pain or weakness." (247) "Medical necessity" is generally understood to denote care that is "medically appropriate or medically beneficial." (248)

In coverage litigation, health insurance contracts tend to be construed in favor of coverage. (249) The construction of this contractual language generally rests on "the meaning conveyed to the average person applying for such insurance and reading such provisions." (250) Although clear language is not liberally interpreted, most courts follow the principle of contra proferentum to interpret ambiguous provisions in favor of the policyholder, (251) or seek to fulfill the policyholder's "reasonable expectations."252 While some courts have suggested that ERISA preempts state contract interpretation rules for employer-based health insurance contracts, (253) others have held that contra proferentum remains applicable when "normal principles of contract construction are unavailing." (254) Any exclusions of coverage must be "conspicuous, plain, and clear" and drafted "such that a reasonable purchaser of insurance would have both noticed and understood it." (255)

A policyholder denied coverage has recourse to several mechanisms of review. Under the ACA (which reinforced preexisting state laws, (256) ERISA provisions, (257) and Department of Labor regulations (258)), insurers must provide an internal claims review and an external review process; external review procedures must meet the consumer protection standards set by the National Association of Insurance Commissioners and applicable state standards. (259) Beneficiaries may also access judicial review under either ERISA (260) or the ACA. (261) Review is usually de novo; (262) if the plan, however, grants the administrator the authority to determine eligibility, decisions may only be reviewable based on abuse of discretion. (263) Research suggests that approximately half of judicial (264) and external reviews (265) end to decide in favor of coverage.

2. Coverage of New Interventions and Health Technology Assessment (266)

When new medical interventions arise, such as a new vaccine, insurers increasingly use HTA in determining coverage. HTA is a process that "examin[es] ... properties of a medical technology used in [healthcare], such as safety, efficacy, feasibility, and indications for use, cost, and cost-effectiveness, as well as social, economic, and ethical consequences." (267) The use of HTA has grown along with trends toward priority-setting and cost-containment among healthcare payers, as well as evidence-based medicine by practitioners. (268) Compared to public insurers, private insurers have made greater advancements in HTA, (269) including setting procedures for case-by-case analyses, establishing internal HTA units, and subscribing to private assessment services. (270) In 2001, a national survey found that HTA reports were the most frequently used sources of information for reviewing new interventions; (271) a more recent assessment of six of the seven largest national U.S. insurers and five smaller regional insurers found that every plan relied on both internal and external HTA when making certain coverage decisions. (272)

Despite widespread use, HTA procedures vary widely, (273) and efforts to create national HTA institutions have been unsustainable. (274) Approximately sixty-four U.S. entities are organized to perform HTA. (275) The largest may be the Technology Evaluation Center run by Blue Cross/Blue Shield in partnership with Kaiser Permanente, (276) which assesses approximately thirty new technologies per year. HTA reports by this unit contribute to local coverage determinations that affect approximately 25% of the U.S. population. (277) An analysis of HTAs comparing this center's recommendations with Medicare determinations found that both processes examined not only effectiveness, but also demand among beneficiaries and potential for overall health benefit. (278)

Many have commented on the use of cost-effectiveness criteria to inform coverage policy. Although HTA in the United States "consistently fails to incorporate a standard method of economic analysis," (279) cost concerns inform coverage decisions in 90% of plans. (280) Mechanisms for addressing costs include formal cost-effectiveness analysis, requiring selective preauthorization, explicitly excluding coverage, or requiring that the policyholder use less costly interventions before accessing more expensive care. (281) Some commentators have noted that private insurers face a conflict of interest in deciding whether to cover expensive new technologies, given incentives for containing costs (282) and the confidential nature of most decisions, (283) For public insurers, political pressure and publicity may mitigate this conflict. (284) Regulations promulgated under the ACA have attempted to minimize conflicts of interest in coverage decisions by private insurers. (285)


Government-sponsored health insurance programs in the United States include Medicare, Medicaid, the State Children's Health Insurance Program, Tricare, the Veteran's Administration, the Indian Health Service, and other public clinics and state programs. Public insurers' coverage decisions largely mirror decisions by private insurers--these public plans echo the requirement of medical necessity, provide procedures for coverage determinations and appeals, use HTA to weigh new technologies, and issue determinations that are subject to judicial review. This section discusses Medicaid and Medicare as examples of public insurers, describing the principles for making coverage decisions and the use of HTA.

1. Interpretation of Statutorily Authorized Benefits

Coverage under public plans depends on whether the treatment falls under the applicable statute and accompanying regulations. Like private insurers, Medicare and Medicaid rely on determinations of medical necessity. (286) The federal Medicaid statute does not impose this restriction directly, (287) but it allows states to "place appropriate limits on a service based on such criteria as medical necessity or on utilization control procedures." (288) Restrictive coverage decisions are one method of controlling Medicaid costs. (289) Medicare authorizes coverage for many categories of services, but excludes coverage for treatments that are "not reasonable and necessary for the diagnosis or treatment of illness or injury or to improve the functioning of a malformed body member." (290) Coverage decisions occur at several levels, including case-by-case decisions about individual claims, local coverage determinations by Medicare contractors, and national coverage determinations by the Centers for Medicare and Medicaid Services (CMS). (291)

States vary in their rules for reviewing adverse claims determinations under Medicaid, but all beneficiaries are entitled to a hearing before denial is final. (292) The standard of review is whether the decision was "supported by substantial evidence in the record," unless the beneficiary requests de novo review. (293) Although the federal Medicaid statute does not specifically offer judicial review, beneficiaries may make civil claims for coverage denials under other laws, such as 42 U.S.C. [section] 1983. (294) Medicare beneficiaries can appeal individual claims denials to a series of reviewers, (295) including de novo review by an administrative law judge, (296) culminating in de novo review by the Medicare Appeals Council. (297) Review by the Medicare Appeals Council may be followed by judicial review depending on the amount in controversy; upon judicial review, agency findings of fact are conclusive "if supported by substantial evidence." (298) Both Medicaid and Medicare beneficiaries can appeal local and national coverage determinations to an administrative law judge, to the Appeals Board, and ultimately to the courts for judicial review. (299)

2. Coverage of New Interventions and Health Technology Assessment

Public insurers' coverage decisions on new technologies are frequently publicized, and Medicare builds in procedures for public comment during centralized coverage determinations. (300) Public comment almost invariably adds pressure in favor of coverage, (301) drawing the attention of program beneficiaries, (302) drug and device manufacturers, (303) advocacy groups, other administrative agencies, and lawmakers. (304) Medicare beneficiaries are a "large and politically powerful constituency," while Medicaid beneficiaries may be less politically influential. (305) Although formal cost-effectiveness analysis is rare, (306) Medicare appears to scrutinize high-cost technologies more closely for effectiveness. (307)

Medicaid coverage determinations vary by state, and the system "has taken little initiative in implementing HTA with cost-effectiveness analysis into some of its coverage programs." (308) Nevertheless, many state contractors use some form of HTA, (309) often with assistance from clinical consultants and external organizations. (310)

Most Medicare coverage determinations are made locally, either through informal assignment of a billing code, or through a formal local coverage determination proceeding. (311) Procedures for local coverage determinations vary, but usually include literature reviews, contacting other Medicare contractors, and consulting advisory committees. (312) Decisions that "restrict or substantially change" existing policies usually require a public commentary period. (313) Coverage decisions may be requested by manufacturers, healthcare providers, and beneficiaries. (314) National coverage determinations are issued by CMS in approximately 25% of new billing code assignments, setting coverage policy for all contractors. (315) The Medicare Coverage Advisory Committee advises these decisions, (316) and the Agency for Healthcare Research and Quality can provide additional guidance. (317) For coverage, a technology must fall under a statutory benefit category and must not be excluded by another statutory provision. (318) The relevant inquiry is whether the technology "is safe and effective per the Food and Drug Administration regulatory process," and whether "adequate evidence leads [CMS] to conclude that the health technology improves net health outcomes." (319)


This Part identifies legal and economic grounds on which public and private insurers may limit or refuse coverage for HIV prevention technologies. These grounds may support coverage denials for individuals or for entire groups of policyholders. There is some overlap among these justifications, but courts have upheld each as a distinct basis for coverage refusal. Insurers' most persuasive arguments for denying coverage may be contractual or statutory interpretations of "medical necessity" that exclude preventive care. Although insurers may--and potentially should--provide coverage due to market factors, cost-effectiveness, or public policies favoring public health, there may be slim legal basis to require favorable coverage determinations under existing interpretations.


The primary obstacle to insurance coverage for biomedical HIV prevention is establishing that these technologies are medically necessary. (320) Coverage will depend on whether the contractual or statutory definition of medical necessity includes the prevention of disease. The majority of private insurance contracts and public insurance provisions may exclude preventive care.

1. Interpretations of Medical Necessity

Medical necessity determinations are central to healthcare cost control efforts in the United States. (321) The conception of medical necessity arose in private insurance plans after World War II, and the Medicare statute incorporated this restriction into public insurance in 1965. (322) Insurers, particularly managed care organizations, (323) have increasingly used medical necessity exclusions to reduce costs, (324) and litigation over medical necessity denials is common. (325) Public concern over coverage denials spurred state statutes providing for external review of insurer decisions, (326) and the ACA took up this concern by mandating procedures for external review. (327) Essentially every public and private health insurer reviews claims for medical necessity; (328) to be enforceable, however, a medical necessity provision must be included in the contract. (329) Medical necessity definitions are often complemented by specified treatment exclusions, including care that is "experimental," "investigational," "cosmetic," or "for the convenience of the patient." (330)

Commentators have widely critiqued insurers' interpretations of medical necessity, (331) and Sara Rosenbaum and colleagues have collected an extensive list of definitions in use. (332) Contractual definitions are often nonspecific, (333) and contracts may incorporate guidelines outside the policy that are consistent with general medical standards. (334) "Multidimensional definitions" are common practice, and these generally permit insurers "to select among the treatments that ostensibly are all appropriate in favor of one that is the safest, the least costly, and not only for the convenience of the member or provider." (335) As Rosenbaum and colleagues have found, most medical necessity definitions extend only to the treatment and diagnosis of illness or disease; (336) despite notable exceptions and guidance drafted by the American Medical Association (337) and National Health Law Program, (338) few definitions explicitly extend to cover care necessary "to prevent" or "for the purpose of preventing" illness or disease.

When contesting a medical necessity denial, the policyholder has the burden of proving medical necessity, (339) and a court may consider all relevant evidence bearing on the question. (340) Courts have interpreted "necessary" variously to mean "appropriate" (favoring policyholders) (341) or, more rarely, "indispensable, essential, unavoidable, compulsory, or required" (favoring insurers). (342) Although the judgment of the treating physician receives substantial weight in coverage determinations, (343) outweighing the opinions of clinicians who have not personally examined the patient, (344) insurers may permissibly disagree with the physician (345) as long as the denial is not arbitrary. (346)

2. Application to Biomedical HIV Prevention

When an insurance contract or statutory provision does not specifically include preventive care, policyholders seeking coverage for biomedical HIV prevention will face challenges meeting insurers' criteria for medical necessity. As detailed above, insurers will conduct a two-part inquiry to determine that (1) the policyholder has an "injury," "illness," "sickness," or "disease," and (2) the care is "medically appropriate or medically beneficial" to treat the condition. (347) Insurers may come to different determinations for policyholders who are HIV-negative and seeking to avoid initial infection, compared to policyholders who are HIV-infected and seeking to avoid infecting others. Determinations will likely also differ based on whether the intervention serves an exclusively preventive purpose, compared to interventions that are dual-purpose for HIV prevention and another diagnostic or treatment purpose.

a. Exclusively Preventive Technologies

When an HIV prevention technology is exclusively preventive, insurers may seek to deny coverage based on the first prong of medical necessity. Policyholders who are not infected with HIV are seeking preventive technologies precisely because they are healthy; they lack an injury, illness, sickness, or disease as needed to trigger coverage. Even when a policy does not specify the need for sickness or illness, policy language that limits coverage to treatment and diagnosis is another means of codifying the first prong of medical necessity. Individuals who receive treatment and diagnostic care are presumed to have an illness or disease, and courts have upheld medical necessity denials when policyholders do not present proof of a "condition" requiring treatment. (348) Coverage denials of contraception have relied on medical necessity definitions, (349) and the same arguments may be raised for HIV prevention technologies. Although insurers may cover less costly preventive care due to market demand and cost-effectiveness, expensive prophylactic interventions are more likely to fail medical necessity criteria. (350)

If insurers deny coverage based on lack of an illness or disease, policyholders may argue that the risk of HIV infection is itself a pathology. Although biological vulnerability to HIV is a universal human condition (and unlikely to be deemed a pathology), elevated risk of HIV infection--a risk above that of the general population--may present a closer case. Individuals who engage in risky behaviors may attribute some of these actions to underlying morbidities, such as a substance use disorder or a psychological condition like depression. Reasoning this way, elevated HIV risk could be reclassified as a symptom of another illness, and policyholders could argue that HIV prevention technologies are in fact treatment for that symptom.

Although this argument would expand coverage, it has disturbing implications, including the potential for increased discrimination against vulnerable communities. For example, a majority of new HIV infections in the United States are attributable to sex between men. (351) Classifying same-sex attraction and behavior as pathological has contributed to discrimination against individuals who identify as gay, lesbian, and transgender. (352) Arguments that define elevated HIV risk as an illness or symptom thereof would exacerbate this problem. Policyholders may also experience increased HIV risk due to biology (e.g., women are at higher risk from heterosexual sex (353)) or local epidemiology (e.g., in communities where prevalence is high). But the argument that it is pathological to be female or part of a high-prevalence community can also be used to stigmatize members of those groups. …

Log in to your account to read this article – and millions more.