NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) Committee on Disability in America; Field MJ, Jette AM, editors. The Future of Disability in America. Washington (DC): National Academies Press (US); 2007.

Cover of The Future of Disability in America

The Future of Disability in America.

Show details

9Coverage of Assistive Technologies and Personal Assistive Services

After retiring a few years ago, Mr. B led a vigorous independent life until he experienced a major stroke. Paralyzed on his right side, he now needs a wheelchair to move around his house, where he has lived alone since his wife died. The contractor who processes Medicare claims for this kind of medical equipment approved a manual wheelchair, but with his paralyzed right arm, Mr. B cannot self-propel it. He has asked for approval for a power wheelchair, but the contractor—citing Medicare rules—insists on additional documentation that the more expensive equipment is required. All this is time-consuming, complex, and frustrating, even though Mr. B’s rehabilitation team is helping. In the meantime, Mr. B is having difficulty getting around at home, and he and his daughter are worrying about what he will do if his request is not approved.

This story—recounted by an airline employee to a committee member who uses a scooter—illustrates the hurdles that individuals often face in trying to obtain coverage from a public or a private health plan for wheelchairs and other mobility assistive devices. Persistence in appealing denials and assistance from professionals in navigating complex policies and procedures are often necessary when someone wishes to have coverage for particular assistive devices and services approved—and even then, success is not assured.

This chapter focuses on the second issue in health care financing included in the committee’s charge: health insurance coverage of assistive technologies. Because personal assistance services sometimes serve similar purposes for people with disabilities, this chapter also examines coverage of these services. The discussion begins with a brief, general overview of the types of coverage policies and limitations that may affect access to services important to people with disabilities. The subsequent two sections describe insurance coverage of assistive technologies and services. A short section on access to services or equipment through programs not based on insurance follows. The chapter then considers the challenges of expanding coverage in an environment of high and increasing health care costs. The final section presents recommendations.

Overall, the committee found that many of the health plan coverage policies identified in the 1991 Institute of Medicine (IOM) report Disability in America continue to be targets of criticism more than 15 years later. The concept and interpretation of medical necessity remain controversial, as do statutory and regulatory limits on coverage for assistive technologies and personal care services. In many areas, health care financing policies have not evolved to support the independence and community participation of people with disabilities. One major exception is the expanded use by states of waivers of certain Medicaid regulations to cover a wider range of “home-and community-based” services as alternatives to institutional care. For the most part, as also described in Chapters 4, 6, and 8, this nation’s system of health care financing and delivery is not well designed to meet the needs of people with chronic health conditions or disabilities.

The question of whether a technology or service is covered by a health plan arises only for people who are insured. As described in the preceding chapter, although Medicare covers virtually all older adults, almost 14 percent of younger adults with a physical limitation and 8 percent of those with a limitation in activities of daily living (ADLs) are uninsured, as are almost 5 percent of children with special health care needs. For people who are uninsured or whose health plans do not pay for assistive technologies or personal assistance services, the options include going without; paying for the item personally (often called “out-of-pocket payment” or “self-payment”); or securing assistance from vocational rehabilitation, assistive technology, or other special public or private programs, some of which are described below.

Neither this chapter nor the previous one offers a comprehensive examination of health care financing issues relevant to people with disabilities, a task that would demand a report in itself. For example, this chapter does not examine coverage of occupational and physical therapy, including the concerns that have been raised that Medicare payment caps on these services could harm patients with mobility and other impairments.1 It also does not consider controversies over payment rules for specialized inpatient rehabilitation facilities that critics argue will deny many people with disabilities the appropriate intensity of care because the rules rely on an outdated or questionable set of diagnoses to define the patient mix that facilities must have to qualify for payment (see, e.g., Hackbarth [2003] and Thomas [2005]).2 The committee recognizes that inadequate provider payments levels and methods can affect the availability and quality of services and health care. Along with inadequate professional education, they can contribute to the problems that people with disabilities often report in finding health care professionals who understand their medical condition and its nonmedical dimensions, who appreciate their need for routine health promotion and primary care services, and who have accessible facilities and equipment.

The committee also recognizes that rising health care costs, particularly costs for Medicare and Medicaid, make some of the changes proposed in this chapter difficult. Decisions to pay for such changes by reducing spending elsewhere or by increasing taxes will present policy makers with hard choices—particularly in the context of pressure to control overall spending growth in federal health care programs. Successful steps to make health care services more efficient and to control fraudulent and abusive practices by providers and suppliers will help reduce spending but, in the committee’s view, are unlikely to suffice.


For an insured individual, coverage of any service or product depends on the answers to several questions. First, is the product or service explic itly covered (or excluded), or if it is not, is it covered (or excluded) under a broad category of services or products? Second, what general additional criteria or conditions apply to coverage? Third, does a given individual’s specific situation meet the requirements for coverage? The answers to these questions are not always easy for a patient, family member, or health care provider to determine. The answers may vary depending on the source of insurance and the specific type of assistive service or technology.

Medical Necessity and Coverage

By statute, Medicare is prohibited (unless it is explicitly authorized) from paying for services and items 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” (42 CFR 1395y(a)(1)(A)). The U.S. Congress has made certain statutory exceptions to the diagnosis and treatment restrictions, for example, by authorizing coverage of selected preventive services such as screening mammography. Otherwise, this “medical necessity” provision provides the basis both for decisions about coverage of whole categories of technologies or services (e.g., personal care services) and for decisions about coverage in individual cases. For example, a service or product may be covered for one individual and denied for another because the service or item was judged to be not reasonable and necessary given the latter individual’s diagnosis and condition (e.g., a power wheelchair for someone who can operate a manual wheelchair). (See also the discussion in Appendix D.)

Medical necessity criteria have proved particularly troublesome for people with disabilities seeking coverage of assistive technologies and personal care services. They may be invoked to deny payment for nonmedical services, such as assistance with bathing, or products, such as bathroom grab bars, that help people manage daily life or protect their health (e.g., by avoiding falls). Because the rules and procedures are complex, otherwise approvable claims may also be denied because beneficiaries or their providers did not correctly document the case for coverage or because the organizations administering claims for services applied policies inconsistently or incorrectly. Denials of claims for certain assistive technologies and services based on the failure to meet medical necessity criteria are disheartening and confusing and reduce people’s ability to function at home and in the community (Iezzoni and O’Day, 2006).

Federal law does not define medical necessity for the Medicaid program, and state definitions vary and may be broader than the Medicare definition (Rosenbaum et al., 2002; Perkins et al., 2004).3 Connecticut’s language is particularly broad, covering health care that is provided “to correct or diminish the adverse effects of a medical condition or mental illness; to assist an individual in attaining or maintaining an optimal level of health; [or] to diagnose a condition or prevent a medical condition from occurring” (Connecticut Department of Social Services, 2003, p. 2). A number of states refer to services for conditions that may cause or worsen a disability or that may prevent deterioration of a health condition (Rosenbaum et al., 2005).

Definitions of medically necessary care also vary for private health plans. Many use definitions and interpretations similar to those governing Medicare.

Medicare administrators have devoted considerable effort to developing guidance for providers and consumers about when a service will or will not be covered (CMS, 2006e). The decision memo on assistive mobility technologies discussed later in this chapter is in an example. Nonetheless, the great range of individual care needs and circumstances makes it difficult to eliminate subjective decision making and variability.

Other Coverage Limitations

Even if a service or a product is not excluded for lack of medical necessity, various provisions of public or private health plans may limit access. Examples include

  • restrictions that apply temporarily or indefinitely to someone with a preexisting condition (but the federal rules described in Chapter 8 limit the length of such restrictions in group insurance policies);
  • limits on the number of services (e.g., 20 home visits) that will be reimbursed;
  • caps on total reimbursement (e.g., $1,000 for home medical equipment);
  • restrictions on who can provide a service (e.g., excluding coverage for personal care provided by a family member); and
  • widespread requirements that individuals pay part of the cost of a service or product.

Most such restrictions are intended to control programs costs. The intent may be to discourage spending on services of minimal benefit or to limit fraudulent or abusive practices by health care providers and vendors.4 Some coverage restrictions have quality-of-care objectives. For example, coverage of complex, high-risk procedures may be limited to specifically approved facilities (e.g., a Medicare-approved center for heart or lung transplantation). Likewise, as described in Chapter 7, coverage and marketing of complex medical equipment may require U.S. Food and Drug Administration approval on the basis of evidence of safety and perhaps effectiveness. The rejection of a service or product for coverage on the basis of evidence of harm or ineffectiveness also serves quality-of-care objectives.5 Yet another example of a coverage-related policy that serves quality objectives is the recently issued quality standards for suppliers of durable medical equipment, prosthetics, orthotics, and medical supplies (CMS, 2006c).


Chapter 7 defined assistive technologies, consistent with the Assistive Technology Act of 1998, as any item, piece of equipment, or product system, whether it is acquired commercially, modified, or customized, that is used to increase, maintain, or improve the functional capabilities of individuals with disabilities. It also noted that the broad legislative language intentionally permitted programs created by the legislation to cover general use products if, for a given individual, such a product worked as well as or better than for a specially designed product. That breadth does not extend to federal or private health insurance plans.

As described by Iezzoni and O’Day (2006), assistive technology “re sides at that fractious border between covered and uncovered benefits” (p. 150). Wolff and colleagues (2005) observed that, although most health plans cover some assistive technology, “its predominant use in daily functioning rather than for therapeutic purposes has contributed to ambiguity in health insurance coverage. Coverage policies are typically stringent, and coverage disputes in this area are among the most common and problematic” (Wolff et al., 2005, p. 1140).

A 2001 University of Michigan survey of people with disabilities found that approximately 40 percent of people who obtained assistive technologies paid for them out of pocket (Carlson and Berland, 2002).6 Private health insurance and Medicare were the primary third-party sources of payment, and each was mentioned as a source of payment by between 15 and 20 percent of respondents. Free provision, Medicaid, and the U.S. Department of Veterans Affairs were each mentioned as a source of payment by between 5 and 10 percent of respondents. As reported in Chapter 7, about 40 percent of respondents reported that the use of assistive technologies had reduced their need for help from another person. Nevertheless, the users of assistive technologies cite a lack of funding and a lack of information about appropriate technologies as barriers to access (NCD, 2000b).

Freiman and colleagues (2006) recently reviewed public funding for an array of assistive technologies. Table 9-1 summarizes their conclusions about the extent of public insurance coverage of several kinds of technologies. It shows a very mixed pattern of coverage with the U.S. Department of Veterans Affairs most consistently offering coverage (see also PAI [2002]).

TABLE 9-1. Summary of Public Health Plan Coverage of Assistive Technology.


Summary of Public Health Plan Coverage of Assistive Technology.


Medicare covers certain medically necessary assistive technologies under its statutory benefits for durable medical equipment, prosthetic devices, and orthotic devices (42 U.S.C. 1395(k), (m), and (x)). It specifically excludes coverage of certain assistive items, notably, hearing aids and eyeglasses (except that one pair of glasses is covered following cataract surgery) (42 U.S.C. 1395y). The statutory definition of durable medical equipment includes some items, such as oxygen tents, that are not usually considered assistive technologies.

The federal officials and private contractors who administer Medi care have developed interpretations of statutes and regulations to guide decisions about coverage in individual cases. Examples of assistive items excluded under such guidance as not meeting the criteria for coverage are bath seats, grab bars, humidifiers, home elevators, and wheelchair ramps (Medicare Carriers Manual, Section 2100.1). They are considered convenience items, not medical items, even though some of this equipment may help prevent falls and otherwise protect an individual’s health.

One critical statutory requirement is that durable medical equipment must involve use in the home. Based on the statute, regulations define durable medical equipment as “equipment furnished by a supplier or a home health agency that: (1) can withstand repeated use; (2) is primarily and customarily used to serve a medical purpose; (3) generally is not useful to an individual in the absence of an illness or injury; and (4) is appropriate for use in the home” (42 CFR 414.202) (emphasis added).7 As discussed below, the home use provision has been restrictively interpreted to limit coverage of mobility assistive equipment that helps beneficiaries get around outside the home. Coverage may require a prescription, and repairs to equipment are usually covered only if the equipment is rented (DATI, 2005).

A recent example of administrative guidance on coverage determinations involved a 2005 decision memorandum that was intended to clarify the interpretation of the statute and regulations related to mobility assistive devices, including manual and power wheelchairs (CMS, 2005b). It replaced a more restrictive proposal that had been prompted by concerns about rapidly increasing spending for power wheelchairs and specific instances of fraud and abuse (see, e.g., GAO [2004]).8 Although the memorandum liberalized the policy in some respects, it left critical restrictions in place, as described below. The document, which was developed with the involvement of other governmental agencies and outside experts, cited evidence identified through a literature review that the equipment was reasonable and necessary for beneficiaries with mobility limitations. The memorandum stressed that determinations of coverage should include assessment of “the technology that most appropriately addresses the beneficiary’s needs as determined by the coverage criteria” (CMS, 2005b, p. 9). Clearly, such an assessment involves subjective judgments that can lead to variations in individual coverage decisions.

Unfortunately, based on the statutory reference to durable medical equipment used in the home, the memorandum focused on individual needs related to ADLs, which are primarily performed in the home, rather than instrumental activities of daily living, which include some activities (e.g., shopping and traveling) that occur outside the home. Thus a power wheelchair would be covered for an individual who needs the device both for personal care activities in the home and for travel outside the home but not for an individual who requires the device only outside the home. A person might also be denied coverage if the device could not actually be used in the home because of the home’s small size or floor plan. Furthermore, on the basis of a particularly restrictive interpretation of the home use criterion, coverage may be limited to a type of wheelchair or other equipment that will not allow people to travel easily or safely outside the home. Thus the statutory and regulatory criteria for coverage of assistive technologies are quite different from those for coverage of surgical and medical interventions. For example, hip replacement surgery would not be challenged for individuals simply because pain limited their activities outside but not inside their home.9 (The necessity of surgery might still be challenged for other reasons.)

The “use-in-the-home” language of the statute has become a central issue in efforts to revise Medicare policies on durable medical equipment. Critics of current interpretations argued that the language was intended to distinguish between equipment suitable for home use and equipment that required an institutional setting. As one advocacy group has argued: “[t]he intent of the ‘in the patient’s home’ clause was not to make elderly and disabled persons prisoners in their own homes” (UCP, 2006, unpaged).

Clearly, the “use-in-the-home” restriction, especially as it is now interpreted, is inconsistent with the values of independence and community participation that have become widely recognized elsewhere in federal health and social policies since Medicare was enacted. Nonetheless, efforts to change the statute have, to date, not succeeded. Also, although the Centers for Medicare and Medicaid Services (CMS) was to consider the appropriateness of the “in-home” criterion as part of President George W. Bush’s New Freedom Initiative and in light of efforts to promote work and community living, CMS action seems to be limited to congressionally directed demonstration projects (DHHS, 2002).10

The evidence base for many assistive technologies and services important to people with disabilities is limited, which points to the need for more research in this area (Clancy and Andresen, 2002). In a paper that outlines the many financial obstacles to accessible assistive technologies, the American Academy of Physical Medicine and Rehabilitation observed that “to be fair, Medicare is often asked to approve coverage and reimbursement of assistive technology with little clinical data [or] outcomes studies” (AAPM&R/FPM&R, 2003, p. 5). As noted above, CMS judged that the evidence base supported the coverage of much mobility assistive equipment for people with limitations in ADLs.

Criticism is often directed at government officials and Medicare contractors for their restrictive coverage decisions, but certain coverage restrictions appear to be so firmly based in the Medicare statute that statutory change is needed to permit easier access to assistive technologies that support independence and productivity. The home use criterion for durable medical equipment is one example, at least as it relates to durable medical equipment that cannot actually be used inside the home but that can help people get around in the community. The statutory language related to medically necessary care is another example. Given the statute, more expansive administrative interpretations can only go so far.

Medicaid Coverage

Medicare is a national program, despite some variability in interpretation of Medicare coverage policies or guidelines by the private contractors who administer the program.11 Medicaid, in contrast, operates under federal standards that give states considerable authority to determine whether individuals should have access to assistive technologies. Under federal law, home health services are a required element of state Medicaid programs, and certain medical equipment and supplies are a required part of home health care coverage (42 CFR 440.70). Even when a Medicaid program covers assistive technologies, it may still impose various restrictions, such as caps on the amount that will be paid for equipment.

Some items, such as hearing aids and prosthetic devices, are optional for most adults but are required for children under the Medicaid’s Early Periodic Screening, Detection, and Treatment (EPSDT) benefit. (The State Children’s Health Insurance Program [SCHIP] option, which covers many children with special health care needs, does not include EPSDT benefits as such.) As summarized in Box 9-1, the EPSDT benefit covers an extensive array of services for children, including health care that “must be made available for treatment or other measures to correct or ameliorate defects and physical and mental illnesses or conditions discovered by the screening services” (CMS, 2005e, unpaged; see also the discussion of EPSDT in Chapters 4 and 8). Many of these services are particularly relevant for children with disabilities (see, e.g., Smith et al. [2000]). As described later in this chapter, the passage of the Deficit Reduction Act of 2005 has raised some concerns about children’s continued access to EPSDT services.

Box Icon

BOX 9-1

Medicaid’s Early Periodic Screening, Detection, and Treatment Benefits. (as required under Section 1905(r) of the Social Security Act) Screening Services (all of the following):

As discussed further in the section on personal assistance services, the U.S. Congress has authorized waivers of certain Medicaid requirements in a variety of situations. Nearly all states have received waivers for home- and community-based service programs that are intended to facilitate care in the community for people who would otherwise be at high risk of needing institutional care. Most waiver programs for adults age 65 and over include some coverage of home modifications (e.g., installation of grab bars and widening of doors to accommodate wheelchairs) and some products (e.g., personal emergency response systems) that are excluded under conventional Medicaid coverage guidelines (Freiman et al., 2006).

Although most discussions of assistive technologies focus on people living in the community, these technologies may also benefit people living in residential care facilities, including nursing homes. Examples range from wheelchairs and hearing aids to products that assist with cognition and “wander management” systems for people with dementia. A recent report identified several barriers to the use of such technologies in institutions, including a perceived lack of financial resources (Freedman et al., 2005a). It noted that Medicaid and Medicare do not specifically reimburse organizations for the devices that they supply to the residents of nursing homes, although the programs do pay for covered devices that individuals obtain directly from equipment vendors.

Private Health Insurance

The committee found little specific information on private health plan coverage of different kinds of assistive technologies.12 In general, people covered by private health plans will often encounter problems with coverage restrictions and medical necessity determinations similar to those that that Medicare beneficiaries face (Iezzoni, 2003; Iezzoni and O’Day, 2006).

A few states have mandates that require private insurance plans to cover certain assistive technologies, primarily hearing aids and prostheses, subject to various restrictions. For example, seven states (Connecticut, Kentucky, Louisiana, Maryland, Minnesota, Missouri, and Oklahoma) require coverage of hearing aids for children and one (Rhode Island) requires coverage for children and adults (ASHA, 2007). The reach of these mandates is, however, limited by federal law that exempts self-insured employer health plans from such mandates and from many other state regulations (Mendelsohn, 2006).

In addition to outright exclusions, private health plans often place fairly low limits (e.g., $1,500) on the maximum amount that they will pay for assistive equipment during a year. They may also limit how often they will pay for an item (e.g., once every 5 years) (Iezzoni, 2003).


Even with advances in the value, availability, and use of assistive technologies that can reduce the need for some kinds of caregiving, human caregiving remains a crucial resource that allows many people, particularly those with serious disabilities, to live independently and to participate in community life. In a 2004 Harris poll, of the respondents who reported the need for personal assistance, 77 percent reported that they relied on family or friends and 29 percent reported that they relied on paid assistance (Harris Interactive, 2004b). In 2005, about 7 percent of all kinds of paid home care (including skilled care) was paid for out of pocket and another 7 percent was covered by private insurance; Medicare paid for about 27 percent, and Medicaid paid for 55 percent (Komisar and Thompson, 2007).

LeBlanc and colleagues have observed that “personal care is a complex construct, known by a variety of names, overlapping with existing service systems, blurring the lines between skilled and unskilled, and between formal and informal home care … [and] … evolving in different ways across the States, many of which continually make changes in their programs” (LeBlanc et al., 2001 p. 2). The term “personal assistance” seems to be preferred by consumer groups, but Medicare and Medicaid policies generally refer to personal care services or personal care attendants. As used here, personal assistance services or personal care services support individuals who are limited in their ability to perform basic daily activities such as bathing, eating, and dressing. Such services may also include assistance with shopping, light housekeeping, meal preparation, bill paying, and other activities.

Medicare and Private Insurance

Medicare does not cover personal assistance or personal care services as such. Nonmedical personal care services may, however, be provided by home health aides in the course of providing covered home health services (Box 9-2). Such care can be provided only if beneficiaries are also receiving skilled services, which, in turn, are available only if beneficiaries are determined to be homebound. Medicare rules define homebound to mean that a beneficiary has “a normal inability to leave home,” such that trips outside the home require “a considerable and taxing effort by the individual” (such as relying on personal assistance or assistive technology) and are “infrequent or of relatively short duration” (see 42 U.S.C. 1395n(a)(2)(F)). Legislation passed in 2000 provided that absences for health care treatment (including services at day care centers) will not disqualify a beneficiary from being considered homebound under Medicare, Medicaid, and SCHIP.13

Box Icon

BOX 9-2

Medicare Coverage of Personal Care Services as a Home Health Service. Home health services are covered under Medicare Parts A and B. To qualify for home health care, a person must need skilled care, be homebound, and have a plan of care ordered by a physician. (more...)

During the early 1990s, the use of the Medicare home health benefit increased significantly, particularly the use of home health aide services. In the words of the Medicare Payment Assessment Commission, the benefit “increasingly began to resemble long-term care” rather than care covered under other Medicare benefits for post-acute care (MedPAC, 2004, p. 142).14 Reflecting concern about the sharply increased expenditures and allegations of fraud and abuse by home health care providers, the U.S. Congress changed provider payment methods and eligibility criteria. Spending dropped substantially after the implementation of these changes. The Medicare home health benefit now involves “less of the maintenance of chronically ill or disabled people over time” at a low intensity of care (MedPAC, 2004, pp. 145–146).

Private coverage of personal care services is also limited. Some (but not all) private Medicare supplemental insurance policies will pay for a limited number of “at-home recovery” visits to help beneficiaries with ADLs after an illness, injury, or surgery (CMS, 2006b).

The committee found no systematic information on employer-sponsored or private health insurance coverage of personal assistance services. Private long-term care insurance may cover personal care services, but, again, the committee found no systematic information on the extent of such coverage. Again, much personal care is provided by family members on an unpaid basis or is paid for out of pocket by individuals or family members.


Home health care is a required benefit under Medicaid (Box 9-3), but personal care services for adults are an optional benefit that many states offer.15 As of 2004, 26 states and the District of Columbia covered personal care services as an optional benefit under their Medicaid state plans (Summer and Ihara, 2005). Any optional benefit included in a state plan must be available statewide to “categorically needy” beneficiaries.16 States may also cover personal care services under special waiver arrangements that encourage home- and community-based services as an alternative to institutional care. Use of such waivers has expanded substantially since publication of the 1991 IOM report, in part as a result of the U.S. Supreme Court’s 1999 decision in Olmstead v. L.C. That decision held that states may have to provide community-based services for people with disabilities rather than care in less integrated institutional settings (see Appendix D). For children, personal care services are covered under the EPSDT benefit (Smith et al., 2000). In 1988, the U.S. Congress required that Medicaid pay for covered services provided to eligible children in public schools (Herz, 2006).17 The variety of ways that states can cover personal assistance services and the frequency of state coverage changes make it difficult to summarize state policies on these services.

Box Icon

BOX 9-3

Personal Assistance Services and Medicaid Home Health Benefits. Federal policy requires that state Medicaid plans include home health services for eligible people age 21 and over who qualify for nursing home care. The Medicaid statute, unlike the Medicare (more...)

A 2004 survey of the states that offer personal care services as an optional Medicaid benefit found that 4 of 24 responding states limited coverage to the home (Summer and Ihara, 2005). Policies in other states varied, with some states covering services in day care settings, relatives’ homes, and workplaces. All states offering the benefit included assistance with dressing, bathing, laundry, and toileting. Most states covered assistance with eating, meal preparation, grooming, mobility, shopping, and housekeeping. Fifteen of the responding states limited the number of hours of covered services.

In addition, seventeen of the states had some provision for a consumer to direct the services that he or she receives, for example, by preparing service plans or supervising workers (Summer and Ihara, 2005). Three states provided consumers with an individual budget for services and allowed them to purchase services within that budget. Most states (21) had training programs for personal care workers, and a majority (16) reported shortages of such workers. Three-quarters of states allowed payments for services provided by family members (conditioned on the receipt of training). The estimated cost per beneficiary ranged from less than $1,500 in two states to more than $10,000 in eight states.

When states cover personal assistance services under a Medicaid waiver, they may depart from certain Medicaid rules that apply to required or optional benefits, for example, rules that require coverage for categorically needy individuals in the entire state and all categorically needy beneficiaries. The Home- and Community-Based Service (HCBS) waiver program, which was authorized in 1981 under Section 1915c of the Social Security Act, encourages states to cover alternatives to institutional long-term care for beneficiaries who meet the admission criteria for such care (or who would do so in the absence of home- and community-based services). The alternative services are supposed to be cost neutral, which is defined to mean that the average cost per participant is not to exceed the average cost of nursing home care for a person with similar needs. According to CMS, 48 states and the District of Columbia have at least one HCBS waiver, and an additional state has a similar program operating under another waiver authority (CMS, 2007). Many states have several waivers, and the total number of programs exceeds 285. In 1998–1999, 45 states included some coverage of personal care services under at least one HCBS waiver program (Le Blanc et al., 2001). Waiver-based programs may also cover other nonmedical services, for example, home modifications and respite care.

Waiver-based programs can set caps on spending and on the number of people to be served (slots). In 2002, slightly more than 400,000 slots were available across all states, but this amount is less than the number of people wishing to participate (Reester et al., 2004). Of 171 programs, 69 programs had waiting lists totaling more than 155,000 people. The waiting times for people with traumatic brain injuries and children with special needs averaged over 20 months.

Collectively, HCBS waivers accounted for 30 percent of Medicaid long-term care spending in 2001 (up from 15 percent in 1992) and 66 percent of spending on community-based services (up from 37 percent in 1992) (Reester et al., 2004). Much of the growth in the HCBS waiver program is due to the extension of home- and community-based services to people with mental retardation or developmental disabilities, who accounted for 38 percent of the program participants and almost 75 percent of the spending in 2001 (compared with approximately two-thirds of spending in 1991) (Litvak and Kennedy, 1991; Reester et al., 2004). As noted above, the U.S. Supreme Court’s decision in Olmstead v. L.C. has helped spur this growth, although the growth began before that 1999 decision was handed down. People with other disabilities and beneficiaries age 65 and over together accounted for 54 percent of participants and 21 percent of HCBS waiver program spending.

In addition to the programs described above, the Tax Equity and Financial Responsibility Act of 1982 gave states the option of providing community-based services to children with physical or mental disabilities if the children would be eligible for Medicaid institutional services but could be cared for at home at a cost that does not exceed the estimated cost of institutional care (Bazelon Center for Mental Health Law, 2002). A child who qualifies receives a Medicaid card and is eligible for regular Medicaid child services. States cannot set caps on the number of children who can be served under the waiver or limit coverage to children with certain diagnoses or children in certain geographic areas. In 2002, 20 states had programs under this option (often called the Katie Beckett option after the child whose situation prompted the provision) (Peters, 2005). Ten states did not include children with diagnoses of mental disabilities, despite federal requirements prohibiting such limitations (Peters, 2005). (Some children may also be eligible for direct services under the Maternal and Child Health Bureau’s Title V state grants program [Markus et al., 2004].)

A recent review of evaluations of the benefits and costs of home- and community-based services reported that the programs tended to increase both costs and benefits (Grabowski, 2006). (The statutes creating the programs emphasize cost neutrality, hence the emphasis on costs as well as benefits in the evaluations.) Benefits included increased social activity and interaction, greater caregiver and recipient satisfaction, and reduced levels of unmet needs. The costs increased because the savings achieved from the substitution of less expensive services for some nursing home residents (or potential nursing home residents) were offset by the increased spending for people who likely would not have received nursing home care in the absence of the program. The review noted that achieving program savings has become harder as noninstitutional care has become more common, thereby shrinking the pool of people for whom institutional care costs can be reduced. On the basis of recent evaluations, the review concluded that, overall, the evidence about program cost-effectiveness was relatively weak (because of the use of research designs that did not adequately control for nonrandom selection into intervention and comparison groups) and accumulating slowly. The author of the review argued against cutbacks in home- and community-based services, given the strong consumer and public preference for noninstitutional care, but he also proposed the use of research strategies that would better identify efficient program and policy strategies.

An evaluation of North Carolina’s waiver program attempted to control for selection effects (e.g., the choice of participants on the basis of their expected costs) through the use of a complex statistical analysis (Van Houtven and Domino, 2005). The analysts concluded that the program decreased nursing home and hospital costs (which are not always considered in cost neutrality assessments) for participants compared with those for nonparticipants. It had no effect on total costs, once differences in health status were taken into account.

Despite some concerns about the quality of care and program oversight,18 the U.S. Congress included provisions to expand coverage for certain Medicaid home- and community-based services in the Deficit Reduc tion Act of 2005. The law now permits states to provide all services allowed under the HCBS waiver program without getting a formal waiver when the services are for people age 65 and over or for people with disabilities who have incomes up to 150 percent of the poverty level (Crowley, 2006). Beneficiaries do not have to meet the criteria for institutional care. The services do not have to be provided statewide, and enrollment can be capped.

In addition, the act allows states to undertake “cash and counseling” programs without obtaining a waiver (Crowley, 2006). The details may vary among the programs in different states, but these programs usually give a consumer (or a representative) a cash allowance or budget based on a needs assessment. They allow the consumer to direct his or her own personal care and certain other services. The programs also provide consumer counseling and other assistance (e.g., provision of an independent service that actually manages the payments to workers and monitors use). Researchers have evaluated the first three states to implement cash and counseling demonstration projects (before the waiver was implemented).19 One evaluation reported “unambiguous evidence that Cash and Counseling improved the amount and quality of paid personal assistance from the perspective of consumers, with no discernible adverse effects on safety or health” (Carlson et al., 2005). A companion evaluation reported that the costs per beneficiary were 15 to 20 percent higher than the costs for each individual in the control group, in part because individuals in the intervention groups were more likely to receive services for which they were eligible (Dale and Brown, 2005). In two of three states, the costs for the intervention group were not significantly different from those projected if the individuals had received the agency services authorized in their baseline care plan.

Although the Deficit Reduction Act included features that may benefit people with disabilities, it also included several provisions that may shift costs to Medicaid beneficiaries, potentially limiting their access to needed services (Kaiser Commission, 2006a). Almost all of the expected savings (deficit reductions) in the legislation would come from cuts in benefits or from higher beneficiary costs in the form of premiums and copayments. Before passage of the law, states could not require premium payments for most Medicaid beneficiaries nor impose more than nominal copayments. Up to 9 million beneficiaries (half of them children) now could face such new costs. Although beneficiaries with disabilities and those who are dually eligible for Medicare and Medicaid are exempt, the legislation also allows states to offer alternatives to the traditional Medicaid coverage that may have less extensive rehabilitation benefits. Concerns about this legislation are discussed further below.


To various degrees, depending on an individual’s age, place of residence, type of disability, and other characteristics, people who need assistive technologies or services may find noninsurance programs helpful. Some of these programs are privately sponsored. Others are funded by federal or state governments.

For children, the Individuals with Disabilities Education Act (IDEA), as amended, includes important provisions related to assistive technology (Hager and Smith, 2003). In essence, IDEA requires that if states and school districts accept federal education funds, they must make available to children with disabilities a free, appropriate public education, including special education and related services that allow a child to benefit from such education (see Chapter 4). Each child is to have an individual education program or plan that identifies the services that need to be provided without charge, as determined on a case-by-case basis. Covered devices (e.g., specially adapted laptop computers) may sometimes be used at home as well as at school if that is educationally necessary. The school district often retains ownership of assistive devices that have not been personally prescribed, which means that a young person may lose access to the technology once he or she graduates from high school. The “related services” provision of IDEA also extends to personal care services. In 2004, more than 6.8 million children ages 3 to 21 received services under IDEA (U.S. Department of Education, 2005).

The federal Assistive Technology Act does not directly fund the provision of equipment but, rather, supports technology development, training for rehabilitation engineers and technicians, technical assistance and advocacy, and research on needs for assistive technology. In addition, it requires states that receive grants under the act (as all states do) support awareness programs, provide outreach, provide loans, and engage in certain other activities to promote assistive technologies.

Other programs, including Vocational Rehabilitation programs, may also provide access to assistive technologies, as summarized in Table 9-2. These programs generally do not have assistive technology as their primary focus (Freiman et al., 2006). In addition to the programs listed, state work ers’ compensation programs require coverage of some assistive technologies, although the details vary (Allen, 1998).

TABLE 9-2. Summary of Financial Assistance for Assistive Technology in Selected Federal Programs.


Summary of Financial Assistance for Assistive Technology in Selected Federal Programs.

The committee did not locate a similar overview of programs that support financial access to personal assistance services. It did identify a number of federal and state programs that include some direct or indirect support for such services, usually as part of a broader program of home-and community-based services. Box 9-4 lists some examples. In addition, the National Institute on Disability and Rehabilitation Research has funded a number of relevant research and data collection programs, including the Center for Personal Assistance Services at the University of California at San Francisco (UCSF, undated).

Box Icon

BOX 9-4

Examples of Federal Programs (Other Than Medicare and Medicaid) That Provide Some Support for Personal Assistance Services. The Aid and Attendance program and the Homebound program of the U.S. Department of Veterans Affairs allow veterans to receive a (more...)

In addition, nearly all states have programs that use “state-only” funds (i.e., funds not used to obtain federal matching funds) to support home-and community-based services that supplement and fill gaps in Medicaid services (Kitchener et al., 2006). Most programs are relatively small but provide more flexibility than has been allowed under Medicaid rules and waivers.

A patchwork of charitable, advocacy, and other private organizations (many of which receive support from the federal government or state governments) also help people obtain assistive technologies and, less often, personal care services. Some, such as most Centers for Independent Living, encompass a range of conditions. Others focus on particular chronic conditions or disabilities, such as spinal cord injuries, multiple sclerosis, and cerebral palsy. One example is The Wheelchair Recycler, whose founder received Christopher Reeve’s first wheelchair from the Christopher Reeve Foundation and recycled the parts to at least three different people whose chairs needed replacement parts and repairs (New Mobility Magazine, 2006).


As discussed in Chapter 8, the future will bring increasing stresses on this country’s system of financing health care services for individuals with and without disabilities. The specter of rising costs is a major constraint on proposals to expand coverage of assistive services and technologies and, indeed, poses the threat of cutbacks in public and private health care programs that may particularly affect people with disabilities.

For all the difficulties and problems that have thus far confronted this country’s system of financing health care services, including services for individuals with disabilities, the situation will surely become more challenging in the future. Although some challenges stem from those with philosophical opposition to government- or employer-based health insurance programs, the most widely understood threat relates to health care costs that continue to rise at rates higher than general inflation. Other problems stem from the link between employment and health insurance and from competitive market dynamics that discourage insurers from offering coverage to people with higher-than-average health care costs.

Rising Costs

For decades, health care costs per capita have grown faster than the gross domestic product (GDP), and these costs have thus consumed an ever larger share of GDP—with an increase from 7.2 percent of GDP in 1979 to 16 percent in 2004 (Smith et al., 2006). By 2015 the share of the GDP accounted for by health care spending is projected to reach 20 percent (Borger et al., 2006). In addition, Medicare and Medicaid account for an increasing share of federal spending, and this shift will accelerate as the baby boom generation becomes eligible for Medicare (CBO, 2005a). Policy makers will eventually have to confront growing deficits that are, among other consequences, projected to exhaust the Medicare Part A Trust Fund by 2018 (CMS, 2006a).

Among the states, Medicaid is consuming an increasing share of state budgets, 22 percent in fiscal year 2004, which is an amount that slightly exceeds state spending for elementary and secondary education (NASBO, 2006). Most of the projected growth in Medicaid expenditures is linked to spending for older people and people with disabilities (Ku, 2003).

Health care expenditures will continue to rise for several reasons. As described in Chapter 1, the aging of the population will lead to a substantial increase in the number of people who are in the age group that is at the highest risk of disability. Even if rates of disability continue to decline modestly in the future (and this is hardly assured, as discussed in Chapter 3), total spending for Medicare beneficiaries is likely to continue to increase (Chernew et al, 2005b; see also Cutler and Meara [1999] and CMS [2006g]). The pace of expensive innovation in medical care is more likely to continue or increase than abate. Although the aging of the population, price inflation, and other factors contribute to increased health care expenditures, advances in medicine are generally viewed as the strongest force behind the growth in health care spending—one that has proved difficult to constrain in the long term through cost control strategies (see, e.g., Newhouse [1992], Cutler [1995], Chernew et al. [2004], and Bodenheimer [2005]).

Responses to Rising Costs and Implications for People with Disabilities

Since the enactment of Medicare, rising costs have limited policy makers’ willingness to expand Medicare coverage to new groups (e.g., those ages 55 to 64) or a broader range of services (e.g., personal assistance services). The recent extension of Medicare coverage to include prescription drugs is one of the few expansions since the program’s start. (Other expansions include benefits for hospice care and benefits for specified preventive services, such as certain kinds of cancer screening.)

Currently, actions or proposals to increase or establish cost sharing for Medicaid beneficiaries are generating controversy, as are a range of other cutbacks in Medicaid programs that may have a disproportionate impact on people with disabilities. Given the number of states, the number of options, the complexity of the policies and proposals (at both the federal and the state levels), and the early stage of implementation, it is very difficult to gauge or even catalog what is happening across the states. One suggestion at the end of this chapter is for more systematic tracking of changes in federal and state Medicaid policies and programs that may affect adults and children with disabilities.

Examples of a few worrisome developments in the Medicaid program include provisions in the Deficit Reduction Act of 2005 that removed the general prohibition against the imposition of premiums and cost sharing on Medicaid beneficiaries (Kaiser Commission, 2006a). For families with incomes over 150 percent of the federal poverty level, the total can reach 5 percent of the family’s income in a month or a fiscal quarter. (For a family of four, the poverty level in 2006 was $20,000 [DHHS, 2006b].) Although children (under age 18) who are in mandatory Medicaid categories are exempt, adult beneficiaries with disabilities may be affected.

The Deficit Reduction Act also allows the mandatory or voluntary enrollment of beneficiaries in “benchmark” health plans that are not required to offer the benefits traditionally required for Medicaid beneficiaries. In violation of statutory requirements, it appears that some states have been enrolling beneficiaries who are exempt from mandatory Medicaid enroll ment in benchmark plans without offering them a choice or a clear explanation of the differences between the regular Medicaid benefits and those offered by the benchmark plan (Solomon, 2006). A benchmark plan can be “modeled on (or equivalent to) benefit options offered to state and federal employees or the benefits provided by the state’s largest HMO [health maintenance organization]” or simply be approved as “appropriate” by the secretary of the U.S. Department of Health and Human Services (Solomon, 2006, p. 2). The state employee plan could, in principle, be a bare bones option created by the state to provide a minimal benefit benchmark plan for Medicaid beneficiaries (Mann and Guyer, 2005). Given that a significant proportion of Medicaid beneficiaries with disabilities may not be in the categories theoretically exempt from enrollment in benchmark plans (Rosenbaum, 2006) and given concerns about the adequacy of enforcement of protections by CMS, beneficiaries with disabilities may be more vulnerable to benefit reductions than an initial review of the legislation would suggest.

Another worrisome action is the implementation by the state of Tennessee of a new, highly restrictive definition of medical necessity that appears to allow managed care plans great leeway in disapproving services on the basis of their costs (see, e.g., Mann [2005] and Cha [2006]). To cite one more example, CMS has approved a West Virginia plan that could reduce or eliminate Medicaid benefits for beneficiaries who fail to keep medical appointments, do not follow treatment regimens, and otherwise fail to take “personal responsibility” for their health (Bishop and Brodkey, 2006; Eckholm, 2006). How transportation difficulties, mental health problems, disagreements between the patient and the physician, and other common complications will be considered is not clear. It also not clear how the program might affect beneficiaries with disabilities, including those who are supposed to be exempt from the requirements.

Beyond these public-sector responses, health care cost inflation has also contributed to a decline in the percentage of employers offering health insurance, although employer decisions are also sensitive to the health of the overall U.S. or state economy, labor markets, and other conditions (Kaiser Commission and Health Research and Educational Trust, 2005). The decline has been most significant for smaller employers, which, in any case, have been less likely to offer health insurance than large companies (Gencarelli, 2005; Kaiser Commission and Health Research and Educational Trust, 2005). (In 2005, 98 percent of firms with 200 or more workers offered employee health insurance, whereas 59 percent of smaller firms did so.) A recent analysis suggested that “more than half of the decline in [health insurance] coverage rates experienced over the 1990s is attributable to the increase in [private] health insurance premiums” (Chernew et al., 2005a, p. 1021). As discussed above, employees with chronic health conditions or disabilities who must seek individually purchased insurance are very likely to find that coverage is either unaffordable or unavailable.

As health insurance coverage costs have grown, employers have also shifted more of the costs to employees.20 Between 2001 and 2005, average premium payments by employees increased from $30 to $51 per month for individual coverage and from $149 to $226 per month for family coverage (Kaiser Commission and Health Research and Educational Trust, 2005). For most health plan options, average annual deductibles more than doubled during this 5-year period, reaching average levels of nearly $325 for preferred provider plans and more than $600 for conventional health plans. The percentage of employers who offered what are termed high-deductible health plans (a deductible of $1,000 or more for an individual plan) grew from 5 percent in 2003 to 20 percent in 2005 The average deductible in these plans was $1,870 for single coverage and nearly $3,700 for family coverage.

High-deductible plans—or even an end to employer-based health insurance—have been promoted by some as a way of increasing personal responsibility for one’s health and health care use. For people with moderate to significant health care needs and low or modest incomes, the burden of increased cost sharing can be substantial. For example, one analysis estimated that with a $500 deductible, almost one-quarter of hospitalized patients would have out-of-pocket costs that exceeded 10 percent of their income; a $2,500 deductible would lead to that result for approximately two-thirds of patients (Trude, 2003, as analyzed by Davis [2004]).

Beyond simply transferring some costs from the insurer to the insured, increased cost sharing is expected to reduce the amount of health care that people use. Unfortunately, research suggests that cost sharing decreases the use of beneficial as well as nonbeneficial or unnecessary services (see, e.g., Newhouse [1993] and Tamblyn et al. [2001]). For people with chronic conditions who need medications, physician management, and other services, cost sharing may reduce patient adherence to medication regimens and compromise other management of these conditions (see, e.g., Dor and Encinosa [2004] and Federman et al. [2001]). Such a reduction in adherence could lead to worse outcomes for individuals with disabilities and could increase the incidence or severity of disabilities in individuals with chronic diseases.

Various strategies have been proposed and, to a limited degree, implemented or tested as alternatives to strategies that simply increase the financial burdens on people with serious chronic conditions or disabilities. The promotion of Medicare, Medicaid, and private managed care plans, particularly in the 1980s and 1990s, is an example. These plans often have broader coverage (e.g., including prescription drugs) than other health plans but usually place restrictions on the patient’s choice of health care provider. Analysts and advocates have expressed concerns that health plans restrict access to services important to people with disabilities to control costs and discourage enrollment by higher-risk individuals (see, e.g., DeJong and Sutton [1998] and Beatty et al. [2005]). Some cite a backlash against managed care controls on access to providers and services as one reason for the renewed interest in cost sharing as a cost containment strategy (see, e.g., Robinson [2002]).

More recently, the use of disease management programs have been promoted as a way to improve outcomes and reduce costs for people with specific chronic health conditions that collectively account for a significant share of health care costs.21 The programs, which are quite variable, typically emphasize education and other means to increase patient adherence to diet, medication, exercise, or other regimens supported by clinical evidence. To date, evidence of the effectiveness of such disease management programs in improving outcomes or reducing costs is limited and uneven, and observers have expressed concern that the emphasis on limiting costs may undercut the programs’ potential to improve health outcomes (Short et al., 2003; Brown and Chen, 2004; CBO, 2004; Gold et al., 2005). The federal government has a number of disease management demonstration projects under way, and these may provide findings to help guide the design of successful programs (CMS, 2005f).22 Even if the adoption of such programs improves health status and reduces the progression of a chronic condition to a disability, the effects on health status and costs are not likely to offset the demographic trends cited elsewhere in this report.

Another proposed approach, under the rubric of “value-based insurance design,” has sought to relate the level of patient cost sharing to the benefit of a drug or other service for a particular category of patients (see, e.g., Fendrick and Chernew [2006]). For example, the cost sharing required for a particular service would be lower for people with conditions for which research on the condition has documented that the service provides value but would be higher for other people with other conditions. The goal would be to constrain the inappropriate use of services by some people and increase the appropriate of services by others. Discussions of this strategy have focused on people with chronic conditions, but the approach could be extended to people with high-cost disabilities.

Rising costs for health care clearly do not preclude expansions in covered services or in covered populations, especially in good economic times. A case in point is SCHIP, as is the extension of Medicare coverage to prescription drug benefits noted earlier in this chapter. Notwithstanding these policy changes, rising health care costs will make it increasingly difficult to expand access to coverage for the services needed by individuals with disabilities and may in fact threaten the existing coverage provided through SCHIP and other programs (see, e.g., Broaddus and Park [2007] and Kenney and Yee [2007]).

As health care costs continue to increase and the population grows older, the United States will face several important and difficult trade-offs. First, decisions about the share of societal resources that should be allocated to publicly funded health insurance programs such as Medicare and Medicaid are critical and are intimately related to overall tax and fiscal policies. In addition to the direct taxes that help support Medicare and Medicaid, tax subsidies for private health insurance and health care in general must be considered as part of the total resource allocation picture for governments. Although an increasing share of the country’s resources has been going to health care, this cannot continue without limit.

Second, within the health care sector, decisions will have to be made about how to allocate resources among alternative uses or needs. The offsetting of increases in spending for one group (e.g., children) with decreases in spending for other groups (e.g., working-age adults) is hardly new, but the future is likely to bring more difficult trade-offs, often in the face of limited evidence about the potential consequences. Generational stresses are likely to become even more apparent. A recent assessment of the value of the increased medical spending in the United States between 1960 and 2000 concluded that “the money spent has provided good value” but noted that the cost per additional year of life achieved has increased sharply for older age groups compared with that for newborns and children (Cutler et al., 2006, p. 926).

Third, in the private sector, continued or accelerated cost escalation could see an unraveling of employer-sponsored private health insurance. As noted in Chapter 8, fewer employers are offering health insurance benefits, and the percentage of working-age people without health insurance has been increasing.

This is the environment in which people with disabilities and their advocates seek to protect and expand access to important services simultaneously, particularly within Medicare and Medicaid. Certainly, disagreements exist about the role of governments in spreading the cost burdens of chronic illness and disability, but the major constraint on the expansion—and, indeed, the maintenance—of covered services for people with disabilities is anxiety about costs. This anxiety—combined with disappointment at the checkered history of various cost-containment strategies—stands as a barrier to expansive policy changes for public and private insurance programs.

Assessing Expansions of Coverage in the Context of Rising Costs

Policy makers, advocates for people with disabilities, economists, health care services researchers, and others explicitly or implicitly apply various criteria in proposing what services and products should be covered by Medicare, Medicaid, or other private health care plans. A thorough discussion of principles for assessing (and, more controversially, ranking or setting priorities for) coverage of health care and related services is beyond the scope of this report (see, e.g., Patrick and Erickson [1993] and IOM [2000a]). Box 9-5 lists several commonly cited criteria for evaluating or determining coverage of specific services or products.

Box Icon

BOX 9-5

Criteria Used to Determine Coverage of Specific Services or Products. Evidence of effectiveness in improving health and well-being. Evidence of cost-effectiveness (generally or practically interpreted to mean that the cost associated with an additional (more...)

The list of coverage criteria does not include political acceptability. As a practical matter, however, public officials are sensitive to voter and interest group views. Government officials have proposed making cost-effectiveness a consideration in Medicare decision making but have, so far, put the proposal aside after negative public reaction. Acceptability is, to various degrees, an issue for employers who see health insurance as a recruitment and employee relations asset (as well as a cost).

Turning the criteria listed above into a workable decision-making process about coverage is not easy. Particularly for a well-accepted service for which coverage has not traditionally been questioned, a lack of positive evidence of effectiveness is generally not regarded as a barrier to coverage. Indeed, for such services, policy makers may be unresponsive or may even reject evidence of harm or ineffectiveness in response to objections from consumers or providers.

In any case, many widely used and accepted services have not been assessed in any rigorous way or have not been assessed for all the conditions and patient populations for which they are used. Testing existing as well as new technologies and services would require extremely large increments in research budgets, assuming that valid, feasible, and ethical evaluations could be constructed for all such services.

Finally, although the committee did not examine the principles that should be used to determine the expansion of health insurance coverage more generally, it notes that another recent IOM report set forth such principles (IOM, 2004d). Box 9-6 lists the principles. These principles address a number of concerns identified in this and the preceding chapter—for example, discontinuous coverage and affordability—that are particularly serious for people with serious chronic health conditions or disabilities. The committee recognizes that these principles represent aspirations that may be difficult to fulfill, given the cost pressures discussed above and competing political interests and priorities.

Box Icon

BOX 9-6

Key Principles to Guide the Extension of Health Insurance Coverage. Health care coverage should be universal. Health care coverage should be continuous.


The 1991 IOM report on disability singled out limited coverage of assistive technologies and personal care services as major shortcomings in the ability of the American health care financing system to meet the needs of people with disabilities. It particularly cited Medicare’s “outmoded” concepts of medical necessity as it applied to these services (IOM, 1991, p. 257). Consistent with the 1991 report’s focus on the primary prevention of disability, its recommendations about health care financing focused on expanded services for mothers and children and on the development of new models of health promotion appropriate for people with disabilities.

For the most part, the problems identified in 1991 continue. Interpretations of medical necessity still do not recognize the special health care needs of many people with disabilities. Likewise, the statutory requirements that durable medical equipment be suitable for “use in the home” and its narrow interpretation by CMS are inconsistent with the emphasis on independence and community integration exemplified in the Americans with Disabilities Act and other policies.

At the same time, as described in this chapter and Chapter 8, recent years have brought some progress in expanding health insurance coverage of assistive services and technologies. For example, the expansion of Medicaid home- and community-based services has increased resources for services that support independence and participation and that help people avoid institutional care. SCHIP, a federal-state program, has increased the number of children with public insurance coverage. The committee commends state and federal policy makers and those who have worked to achieve the goals of these policy changes.

Many of the restrictions on assistive services and technologies appear to stem in part from concerns about instances of fraud by health care providers and vendors and, to a lesser extent, the abandonment of equipment by consumers. Concerns about fraud in certain areas are legitimate, as are the positive objectives of providing care that is cost-effective and no more costly than necessary. Excessive attention to program integrity can, how ever, lead to barriers to beneficial services and to avoidable suffering, loss of independence, and restricted lives. Efforts to link expanded access to net cost savings (as in some Medicaid waiver programs) have been difficult to construct in ways that actually produce such savings. The broad challenge is to find better ways to remove harsh restrictions on assistive services and technologies without offering the proverbial “blank check.”

Although this country’s current health care financing system is not one that policy makers would likely design if they were starting anew, the system as it currently exists is a political and practical reality. Many people with disabilities and their advocates may seek broad reform or radical change (e.g., universal health coverage, the elimination of the employment-insurance link, and the federalization of Medicaid), but they also advocate for more modest or incremental changes, for example, steps to improve access to assistive services and technologies through health insurance and other programs and evaluations to assess the cost-effectiveness of different ways of improving access and outcomes. The innovative wheelchair leasing arrangement described in Chapter 7 is an example of the kinds of innovative approaches that need more systematic attention and evaluation.

The committee encourages the U.S. Congress to take a new look at how the statutory provisions on medically necessary care and their interpretation might be updated in light of scientific and technological advances and new understandings of health and disability. It also encourages continuing research to support clinical practice and coverage decisions based on evidence of effectiveness. At the same time that outmoded restrictions are discarded, continued efforts to prevent and detect fraudulent wasteful practices by providers and vendors remain important, but such efforts must be made with care to avoid actions that limit access. This committee’s recommendations are directed primarily at the Medicare and Medicaid programs, but the committee encourages private health plans to make similar adjustments in their policies relating to durable medical equipment.

Recommendation 9.1:The U.S. Congress and the U.S. Department of Health and Human Services should begin a process of revising Medicare and Medicaid laws and regulations and other relevant policies to make needed assistive services and technologies more available to people with disabilities and to put more emphasis on beneficiaries’ functional capacities, quality of life, and ability to participate in work, school, and other areas of community life. Priorities include

  • eliminating or modifying Medicare’s “in-home-use” requirement for durable medical equipment and revising coverage criteria to consider the contribution of equipment to an individual’s independence and participation in community life;
  • evaluating new approaches for supplying assistive technologies (such as time-limited rentals and recycling of used equipment) and providing timely and appropriate equipment repairs; and
  • continuing research to assess and improve the appropriateness, quality, and cost-effectiveness of the assistive services and technologies provided to people with disabilities.

In addition to concerns about Medicare, the committee was concerned about the flood of changes in state Medicaid programs noted earlier in this chapter. Tracking Medicaid policies and practices is difficult, especially given the complexity of the changes directed or permitted under the Balanced Budget Act of 2003 and the Deficit Reduction Act of 2005. Analyzing the potential effects of these and other changes on people with disabilities (including changes that come with intended safeguards for this population) adds further complexity. Nonetheless, systematic tracking and analysis are important, given the particular vulnerability of Medicaid beneficiaries with disabilities—both children and adults—to reductions in coverage and increases in beneficiary (or parent) responsibilities for understanding and following complex rules.

Much research on strategies to balance access, quality, and costs has focused on care for people with high-cost chronic illnesses. These strategies include the disease management and value-based insurance strategies described earlier in this chapter and the chronic care programs discussed in Chapter 4. This research and these programs could be refined to consider more explicitly and fully the relevance and consequences of such strategies for people with various degrees of physical, mental, and cognitive impairment. The special needs plans described in Chapter 8 constitute one narrowly targeted approach, but less restrictive approaches also need to be evaluated.

This chapter and preceding chapters have suggested a number of specific topics for research. The next chapter considers more generally the organization and financing of disability and rehabilitation research. It reiterates the messages of the 1991 IOM report on disability that research in this area is underfunded and inadequately coordinated.



Responding to rising costs and evidence of improper billing, the Balanced Budget Act of 1997 provided for two dollar-amount caps per beneficiary for outpatient therapy services that were not provided in a hospital outpatient setting: one cap for occupational therapy and another cap for physical therapy services and speech language pathology services combined (GAO, 2005). Although the Centers for Medicare and Medicaid Services is supposed to develop an outpatient therapy payment system that recognizes patient needs and justifies individual waivers of the caps, the agency has, according to the Government Accountability Office (GAO, 2005), lacked the data necessary to do so. The U.S. Congress has provided temporary moratoria on the caps for most of the period since they were implemented in 1999. The Deficit Reduction Act of 2005 (which was enacted in February 2006) provided that exceptions to the cap could be granted for medically necessary services under certain circumstances for calendar year 2006 (CMS, 2006i; Vettleson, 2006). For example, certain conditions or diagnoses, such as hip replacement or aphasia, qualify for an automatic exception to the therapy caps when they are supported by a documented need for services exceeding the cap. The act provided for a manual exception process for beneficiaries who have conditions or diagnoses that are not covered by the automatic exception and who also have a documented need for services exceeding the cap amount.


Medicare rules require that 75 percent of the admissions of inpatient rehabilitation facilities be concentrated in 1 of 10 diagnoses (the “75 percent rule”). Full implementation of the rules has been delayed, but the rules are set to go into effect fully in 2007. Providers have argued that the 10 diagnoses (which were defined in 1983 as a way of excluding these facilities from application of the rules of the hospital prospective payment system) are no longer sufficient to describe the core patient population needing their services. Advocates for people with disabilities have argued that the rule will force many people into settings where they receive less intensive health care, such as nursing homes (Thomas, 2005). In 2003, the Medicare Payment Advisory Commission (Hackbarth, 2003) reported that more than 85 percent of facilities were estimated to be out of compliance with the requirement.


One exception to the discretion of states to define medical necessity is the legislation that created the Early Periodic Screening, Diagnostic, and Treatment benefit for children. It specifies that children are entitled to “necessary … diagnostic services, treatment and other measures … to correct or ameliorate defects and physical and mental illnesses and conditions discovered by the screening services” (Section 1905(r)(5) of the Social Security Act, cited by Peters [2006]).


A recent Government Accountability Office report cited an estimate by the Centers for Medicare and Medicaid services that Medicare made about $700 million in improper payments for durable medical equipment, prostheses, orthotics, and supplies between April 1, 2005, and March 31, 2006 (GAO, 2007). The report defined improper payments as resulting from clerical errors, the misinterpretation of program rules, fraud (“an intentional act or representation to deceive with knowledge that the action or representation could result in gain”), and abuse (which “typically involves actions that are inconsistent with acceptable business and medical practices and result in unnecessary cost”) (p. 1).


Consistent with the Medicare statute, those administering Medicare and other health insurance programs have increasingly sought evidence of effectiveness before covering new procedures (e.g., lung volume reduction surgery) or products (e.g., implantable automatic defibrillators) (Strongin, 2001; IOM, 2000a; CMS, 2005c CMS, 2006e). Only limited evidence supporting the effectiveness of existing covered services may be available, but a lack of evidence is typically a more prominent issue for new technologies. In some cases, coverage may be restricted to those participating in a clinical trial that is intended to provide evidence that Medicare can use to make a coverage decision. Coverage approval criteria for Medicare do not include evidence of cost-effectiveness, although the implementation of such a criterion has been proposed (CMS/HCFA, 2000; Garber, 2004; Neumann et al., 2005).


The Medical Expenditure Panel Survey, which is used for many purposes, divides spending on assistive technologies in ways that make it difficult to determine the source of payment in any comprehensive way (Freiman et al., 2006). For example, Medicaid spending through home- and community-based service waivers (including payments for assistive technologies) is categorized under “other personal health care,” and other spending may be included in either the durable medical equipment category or the other medical equipment category, both of which include items other than assistive technologies.


In 2001, approximately 6.5 percent of Medicare beneficiaries living in the community used the durable medical equipment benefit to obtain a mobility-related assistive device: 2.6 percent for a manual wheelchair, 2.4 percent for a walker, 1.0 percent for a cane, and 0.5 percent for a power wheelchair (Wolff et al., 2005). Beneficiaries obtaining wheelchairs were younger and had more limitations in ADLs than beneficiaries obtaining canes or walkers but were less likely to have been hospitalized.


The new policy replaced the past policy that limited coverage to individuals who were “bed or chair confined.”


Also, as pointed out by the American Academy of Physical Medicine and Rehabilitation (AAPM&R/FPM&R, 2003), the home use criterion is not found in the statutory discussion of prosthetics.


CMS was also to consider allowing trial rental periods for expensive equipment that is usually purchased and allowing equipment to be furnished to people in skilled nursing facilities for up to 1 month prior to discharge to provide time for equipment adjustments, training, and practice in using the equipment (DHHS, 2002).


Medicare policies do, however, differ in a number of respects for beneficiaries enrolled in the traditional fee-for-service program and those enrolled in Medicare-approved private health plans, in that the latter may choose to cover some services not usually covered.


A survey by McManus (2001) of 98 private health plans (half health maintenance organizations and half preferred provider organizations) of coverage of hearing services for children found that only 11 percent explicitly covered hearing aids (and most had cost caps or other limits) and more than 80 percent specifically excluded them. Two plans covered assistive communication devices under a state-mandated benefit for early intervention services, but most other plans included no specific information about coverage.


In 2002, CMS issued guidance that provided additional examples of acceptable trips outside the home and emphasized that the person’s situation should be considered “over a period of time rather than for short periods within the home health stay” (Medicare Home Health Agency Manual, §§204.1-204.2) (CMS, 2002).


In contrast to the limits on the number of days of care in skilled nursing facilities that Medicare covers, the Medicare statute sets no numerical limit on the number of home health visits.


Personal care services are defined in the State Medicaid Manual (LeBlanc et al., 2001).


Under the statute, state programs must cover “categorically needy” groups, which include people who met requirements for the Aid to Families with Dependent Children program as of 1996, children under age 6 whose family income is at or below 133 percent of the federal poverty level and children ages 7 to 18 in families with incomes at or below the poverty level, Supplemental Security Incomerecipients in most states, the recipients of adoption or foster care assistance under Title IV of the Social Security Act, and certain low-income Medicare beneficiaries. States have the option of providing Medicaid coverage for an extensive list of other groups, including certain working-age individuals with disabilities (CMS, undated).


This provision has generated considerable controversy, some of which is related to conflicting regulations from the U.S. Department of Education and the Centers for Medicare and Medicaid Services on whether Medicaid is the payer of first or last resort. In fiscal year 2005, Medicaid paid about $2.1 billion for school-based services and another $0.8 billion for school administrative expenses (Herz, 2006).


Given the rapid growth of the HCBS waiver programs, it is perhaps not surprising that a 2003 government report identified quality-of-care problems, including a failure to provide needed services and inadequate case management (GAO, 2003b; see also GAO [2002]). The report also criticized the oversight provided by CMS.


These demonstrations were conducted under the Medicaid research and demonstration waiver program, which is authorized under Section 1115 of the Social Security Act (Kaiser Commission, 2001). The cash and counseling demonstration projects and evaluations were jointly sponsored by the Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Health and Human Services and the private Robert Wood Johnson Foundation (RWJF, 2004). People were randomly assigned to receive a cash allowance for personal care services or to have services provided and reimbursed by an agency.


Blumenthal (2006, p. 196) has observed that “the ultimate form of cost shifting to employees is to drop health insurance altogether.” As described above, the percentage of employers (especially small employers) offering health insurance coverage has dropped in recent years.


“Disease management is a system of coordinated health care interventions and communications for populations with conditions in which patient self-care efforts are significant. Disease management supports the physician or practitioner/patient relationship and plan of care, emphasizes prevention of exacerbations and complications utilizing evidence-based practice guidelines and patient empowerment strategies, and evaluates clinical, humanistic, and economic outcomes on an going basis with the goal of improving overall health” (CBO, 2004, p. 19).


Depending on their structure and incentives, disease management programs may be an example of a so-called pay-for-performance program. This label covers a number of initiatives that provide financial incentives for health care providers to meet patient outcome, efficiency, or other goals (e.g., adoption of electronic patient information systems) (see, e.g., CMS [2005f], MedPAC [2005b], and Rosenthal [2005]). Notwithstanding the considerable enthusiasm for pay-for-performance programs, evidence about their effects on people with disabilities or potentially disabling conditions is still accumulating.

Copyright © 2007, National Academy of Sciences.
Bookshelf ID: NBK11441


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (2.9M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...