A Commentary on the Medicare Low Vision Rehabilitation Demonstration Project
Print edition page number(s) 69-75
This document reflects the opinions of the authors and does not reflect the opinion or policy of the U.S. Department of Veterans Affairs or any other organization.
Despite the best efforts of the Centers for Medicare and Medicaid Services (CMS) with regard to the field of visual impairment and blindness, and despite the concerted efforts of the members of the field to make the most of the Medicare Low Vision Rehabilitation Demonstration Project, there are intrinsic flaws in scope and design of the project that preclude its successful implementation and may even undermine the services that have been extended by Medicare to beneficiaries with visual impairments in the last decade. To set the stage and provide a context for this argument, we begin with some background information and a review of events leading up to the April 1, 2006 initiation of the five-year demonstration project.
Organization of Medicare
The Medicare system is currently administered by 16 regional carrier directors, who are responsible for regions that encompass from one to 12, not necessarily contiguous, U.S. states. Regional carriers have considerable autonomy in coverage decisions with respect to services for which CMS has no policy. Medicare billing is submitted through codes representing a disease, impairment, or treatment, which are specifically approved or not approved for reimbursement.
Reimbursement for vision rehabilitation outside the demonstration project
Currently, 13 of the 16 regional carriers are known to be providing Medicare reimbursement for vision rehabilitation services performed by physicians or occupational therapists. Five are reimbursing in all of the states in their regions and eight in some of the states in their regions, for a total of at least 26 states. Some carriers have created local coverage policies while others simply follow the Program Memorandum of 2002, which provided guidelines for reimbursing vision rehabilitation services, as described further in the CMS Program Memorandum of May 2002 section of this essay.
Regional carriers develop policies in response to requests for reimbursement for a particular service. If no provider in the region is billing for that service, the carrier is unlikely to have a stated policy. Furthermore, if a carrier reimburses for a particular service in any state in its region, it is likely that it will also reimburse for that service in its other states if requested. Therefore, when there is no record of vision rehabilitation services in a particular state being reimbursed, but its regional carrier reimburses for vision rehabilitation services in other parts of the region, the likelihood is that no providers are submitting billings for such services. To our knowledge, in every state in which a physician or occupational therapist has requested reimbursement for vision rehabilitation with reference to the May 2002 program memorandum, reimbursement has been forthcoming.
Background and sequence of events
Mid-1990s: Existing services could not meet the need
In the mid-1990s, existing state and agency services were unable to meet the rehabilitation needs of the emergent and ever-increasing population experiencing vision loss from age-related macular degeneration (AMD) for three main reasons. First, many state and private service agencies required recipients to be legally blind to qualify for Medicare-funded rehabilitation, whereas those with gradually progressing central vision loss from AMD require vision rehabilitation training long before their vision deteriorates enough to qualify as legal blindness. Second, most state and many private programs were designed to meet the educational and employment needs of younger individuals, whereas seniors with AMD have very different physical, ergonomic, and psychosocial comorbidities and rehabilitation needs and skills. Third, most programs were budgeted for a small number of clients, whereas the population of visually impaired seniors numbers in the millions.
Gaining regional Medicare coverage for vision rehabilitation
By demonstrating the fact that existing services could not meet vision rehabilitation needs, ophthalmologists successfully negotiated with the Medicare regional carrier directors in the mid-1990s to add codes for visual impairment to the list of impairment codes approved for rehabilitation services within the Medicare system. This Medicare reimbursement for vision rehabilitation services--first gained in Kansas and Florida, then in Michigan and Illinois--was groundbreaking in two respects: it was the first time Medicare recognized visual impairments as functional deficits qualifying for rehabilitation and it was the first time rehabilitation services were extended through any agency to seniors with visual acuities of 20/70 or less or central scotomas.
As required for Medicare-reimbursed rehabilitation, vision rehabilitation services are provided by physician-therapist teams. The three groups of therapists recognized by Medicare as providers of medical services and granted Medicare provider numbers are occupational therapists, physical therapists, and speech therapists. Their recognition by Medicare is based not on state licensing but rather on the integration of their training within the medical system, with clinical rotations performed in hospitals. Some states, Michigan included, do not license occupational therapists but rather simply register them; nevertheless, occupational therapists in all states are reimbursed by Medicare for services provided in medical sites and in patients' homes and communities.
Under those policies, the services of other vision rehabilitation professionals, including that of certified vision rehabilitation therapists, certified low vision therapists, and orientation and mobility (O&M) specialists, could also be billed to Medicare using the supervising physicians' Medicare numbers. This arrangement is referred to within the medical system as "incident-to" billing, which always requires that the services are provided in the same location as the physician.
Other regions follow suit
In the late 1990s, when ophthalmologists met with CMS officials to advocate for a national coverage policy for vision rehabilitation, CMS officials suggested that the most efficient and effective way of gaining national coverage would be to negotiate regional coverage policies, one by one. Following this approach, as noted earlier, by May 2002, Medicare carriers in at least 26 states were providing Medicare reimbursement for rehabilitation services for beneficiaries with visual impairments on the same basis as for rehabilitation for any other impairment.
While Medicare coverage for vision rehabilitation was being successfully extended, region by region, across the United States, legislation was initiated to mandate CMS reimbursement for vision rehabilitation services nationally and CMS recognition of certified low vision therapists, certified vision rehabilitation therapists, and O&M specialists as providers of medical vision rehabilitation services on par with occupational therapists. In recognition of the long history of service of certified vision rehabilitation therapists and O&M specialists and of the fact that the certified low vision therapist specialty was designed by the Low Vision Division of the Association for Education and Rehabilitation of the Blind and Visually Impaired to work with physicians, the American Academy of Ophthalmology and the American Optometric Association supported the legislation.
CMS program memorandum of May 2002
On May 29, 2002, the central administration of CMS published a program memorandum that set forth guidelines for reimbursable vision rehabilitation services (Centers for Medicare and Medicaid Services, 2002). These guidelines included medical need and potential to benefit; services performed by a medical doctor, an occupational therapist, or by others billed "incident to" the supervising physician; an outside time frame of three months; and precise documentation and billing. Since the program memorandum confirmed that vision rehabilitation is a service covered by CMS, it prevented regional carrier directors from denying reimbursement for vision rehabilitation. As with all coverage, however, carriers can question individual billings and deny for cause. In practical terms, the program memorandum confirmed that CMS would provide reimbursement for vision rehabilitation to Medicare beneficiaries across the United States, with some variation possible in the details of coverage among regional carriers (some, for example, required a brief mental evaluation and others did not).
The American Academy of Ophthalmology and the American Optometric Association continued their original support of the proposed legislation, even though national coverage was available, as of May 2002, to physicians and occupational therapists in any site and to certified low vision therapists, certified vision rehabilitation therapists, and O&M specialists on an "incident-to" basis. The only remaining restriction was that their services, like all "incident-to" services, could not be provided in patients' homes and communities.
Legislative advocacy: Facts, fiction, and friction
In our opinion, however, those initiating the legislation, did not acknowledge that the program memorandum represented a significant development in CMS policy, and often continued to advocate for the legislation on the basis that there was no CMS reimbursement for vision rehabilitation at all, although the real remaining issue, in our opinion, was the site restrictions for service providers who were not occupational therapists. This resulted in much confusion, and prompted animated discussions about the definition of vision rehabilitation as differentiated from blind rehabilitation, and whether occupational therapists specializing in vision rehabilitation, certified vision rehabilitation therapists, O&M specialists, and certified low vision therapists are all qualified to provide vision rehabilitation services for seniors, and if not, who is.
Vision rehabilitation for older adults
Occupational therapists had been gaining certification as low vision therapists in small but steady numbers and bringing to vision rehabilitation their knowledge of the range of physical, ergonomic, psychosocial, and cognitive deficits, medication effects, and other comorbidities that are common to seniors. Technological advances were also offering practitioners an increased understanding of the impact on function of different patterns of central vision loss, prompting new strategies for rehabilitation that are particularly appropriate for seniors with central vision loss.
Fiction becomes fact: CMS clarifies who may provide services
Between May 2002 and June 2005, anyone the physician designated, including certified low vision therapists, certified vision rehabilitation therapists, and O&M specialists, could be reimbursed for services provided "incident to" the physician. In June 2005, physical therapists, concerned about athletic trainers encroaching on their field, prompted a CMS clarification stating that under any circumstances, incident to or not, only occupational therapists, physical therapists, and speech therapists could be reimbursed by CMS for billing rehabilitation codes for any impairment. With this ruling, myth became fact: certified low vision therapists, certified vision rehabilitation therapists, and O&M specialists could not be reimbursed by CMS at all.
Congress directs CMS to launch a demonstration project
In December 2003, in response to the previously proposed legislation, Congress signed into law the Medicare Prescription Drug, Improvement, and Modernization Act of 2003 (PL 108-173), which directed CMS to carry out a Low Vision Rehabilitation Demonstration Project to assess the impact of adding certified low vision therapists, certified vision rehabilitation therapists, and O&M specialists as Medicare-reimbursed providers of vision rehabilitation services. Although not mentioned in the Congressional mandate, CMS has described its objective in the demonstration project as establishing whether or not vision rehabilitation can be provided as a "budget-neutral" service: whether CMS will save enough money as a result of vision rehabilitation--for example, in fewer hip fractures caused by falls resulting from impaired vision--to pay for vision rehabilitation services.
Current state of the demonstration project
Currently, the Low Vision Rehabilitation Demonstration Project includes services in New Hampshire, North Carolina, Kansas, and Washington states, and in the metropolitan areas and surrounding suburbs of Atlanta, Georgia and New York City. In those areas certified low vision therapists, certified vision rehabilitation therapists, and O&M specialists, as well as occupational therapists, may provide up to 12 hours of service per patient per year, as needed. Services are billed by the supervising physicians, and occupational therapists may also independently bill CMS for services provided. CMS reimbursement for certified low vision therapists, certified vision rehabilitation therapists, and O&M specialists is provided at a rate that is roughly half the reimbursement provided to occupational therapists inside and outside the demonstration project: certified vision rehabilitation therapists are provided $12.81 per 15-minute unit of individual therapy time with a patient, certified low vision therapists and O&M specialists are provided $14.97 per unit, and occupational therapists are provided $28.04 per unit. The demonstration areas are to be matched with comparison areas under the same CMS regional carrier but not within the demonstration boundaries. Data are to be collected by CMS from CMS billing records for both the demonstration areas and the comparison areas and provided to the evaluation team at Brandeis University in Waltham, Massachusetts. We reviewed the final document describing the demonstration project and discovered significant, indeed fatal, flaws in its basic design.
Core flaws in the project design
There are a number of core flaws in the project design that undermine or preclude its successful implementation.
1. The project design requires data regarding the eligibility for vision rehabilitation services of all beneficiaries in the demonstration and comparison areas but these data are not available and cannot be gathered by the methodology of the project. The project design requires that for appropriate comparisons it must identify all seniors with Medicare who are eligible, by visual acuity standards, to receive vision rehabilitation services. The project design specifies it will identify these beneficiaries by finding visual impairment codes in their Medicare billing records. The problem is that visual impairment codes are only used for vision rehabilitation, so such codes are found only in the records of individuals who have received vision rehabilitation services. For all services but vision rehabilitation, ophthalmologists and optometrists use disease codes, for example "AMD" or "glaucoma." Disease codes do not indicate visual acuity or field; for example, a person with glaucoma who has perfect acuity and field and an individual with glaucoma who has complete loss of vision can only be distinguished using impairment codes rather than disease codes. Therefore, it is not possible to identify from Medicare billing records the population of seniors with Medicare who are eligible for vision rehabilitation services; rather, the only beneficiaries who can be identified are those who have already received vision rehabilitation services whose impairment has been coded. The November 2006 design report by Bishop and colleagues states: "It would be biasing and undermine our analyses to limit comparison subjects to those who also received low vision services" (Bishop et al., 2006, p. 8).
2. The project design calls for the timing of vision loss of beneficiaries to be identified. Bishop and colleagues described how the beneficiary survey sample will be selected from beneficiaries identified through claims based on qualifying low vision diagnosis:
We will include only beneficiaries who have recently experienced vision loss in the second eye ... We will develop and document a method to assure that the low vision diagnosis is new within the last year (Bishop et al., 2006, p. 12).
This data, however, cannot be gathered by the project's methods and is not available in Medicare records. The "low vision diagnosis" (that is, the visual impairment code indicating the diagnosis of low vision) CMS plans to use as evidence of the project's success or failure does not reflect or correlate with the time of onset of low vision in either eye. This is because the visual impairment code appears only when the vision rehabilitation services are received and there is no correlation between the time in which vision rehabilitation services are provided and the time of onset of visual impairment. The only time-of-onset data that are available from CMS records is the first time the disease diagnosis appears (for example AMD or glaucoma), but such data do not provide an index of visual impairment.
3. The project design proposes to identify all beneficiaries who have received "standard low vision services," but it has failed to recognize that not all services use visual impairment codes. Again, this means that the data on which CMS is depending cannot be found using the methods outlined in the project, since vision rehabilitation services performed by occupational therapists are billed using visual impairment codes, but optometrists who provide only low vision evaluations and devices do not usually use these codes (and at this time optometrists are the larger group serving people with low vision). Therefore, the largest group of beneficiaries receiving some level of low vision services will not be identifiable.
4. The project design requires that demonstration areas have existing medical vision rehabilitation services. Although the project design depends on pre-demonstration claims data for vision rehabilitation services, this does not appear to have been considered when the six demonstration sites were selected.
Since national reimbursement for vision rehabilitation has been available only since the program memorandum was issued in May 2002, a number of regions have yet to develop vision rehabilitation services under Medicare. According to the evaluation of the project, it acknowledges that the lack of participating regions is a problem in the following statement:
One potential concern is a low number of incident rehabilitation service users. Preliminary analysis ... for 2004 shows small numbers of users in several demonstration counties. Low numbers have an impact on power, so this will be monitored closely during [the] recruiting phase (Bishop et al., 2006, p. 15).
CMS has been surprised at the low numbers of individuals participating in the project. Sites most likely to provide substantial participation would be those with established vision rehabilitation services. Indeed, the guidelines for participation in the focus groups planned by project evaluators require that the therapist be in practice for one year prior to the demonstration project:
We will complete focus groups with Low Vision Rehabilitation providers for each of the six demonstration and six comparison study sites ... Only providers who have been in practice in the designated site for at least one year prior to the study will be invited to participate (Bishop et al., 2006, p. 23).
In spite of this requirement, sites were chosen that had few to no preexisting services.
5. The project design is not consistent with the medical model of vision rehabilitation, which teams an ophthalmologist or optometrist with a therapist. The project design calls for all ophthalmologists and optometrists to order and bill for the services of therapists, whether in their employ or not, with no requirements to provide low vision evaluations or to supervise or collaborate in any way with the therapists. We understand that CMS feels it cannot require or even suggest a low vision evaluation because some CMS regional carriers do not reimburse for low vision evaluations. However, the design presents a conflict with the responsible participation of ophthalmologists, optometrists, occupational therapists and certified low vision therapists, and should be seen as a problem for other vision rehabilitation professionals, as well.
The conflict occurs because best practices for such professionals require the supervising ophthalmologists or optometrists to perform low vision evaluations; order therapy sessions; provide recommendations for devices, therapy strategies, and resources; and supervise, collaborate, and communicate with therapists on an ongoing basis. This is the relationship stipulated between physicians and occupational therapists and between physicians and certified low vision therapists. If an individual is totally blind, a low vision evaluation is not necessary or useful, but it is stunningly the reverse for sighted individuals with partial vision loss. Understanding the disease causing the vision loss and the specific parameters of the remaining vision are critical to optimal rehabilitation. The acuity, field parameters, contrast sensitivity, scotoma size and position, and preferred retinal locus position with respect to the scotoma are all necessary information gleaned from a low vision evaluation and provided to the therapist.
Two of the important keys to making such short-term vision rehabilitation effective for this population of seniors with low vision, however, that are not reflected in the demonstration project are a low vision evaluation by a physician with the results given to the therapist to guide the training, and ongoing dialog and collaboration between the physician and the therapist.
6. The design protocol lacks the specificity, with respect to variables, data analysis, and quantitative analysis, required to pass the peer review process that is required for other federally funded research protocols. The design does not specify the variables it will use, its data analysis includes no power analysis and no estimates of sample size, and there is no indication of what qualitative analysis will be employed with respect to the focus groups. With respect to variables, the following is more reflective of a letter of intent than a final research design:
To match these comparison beneficiaries, we need to determine an efficient set of independent variables, probably no more than 4 or 5 in number, which have high association or predictive relationship with outcomes of interest ... If our plan to identify and employ a small set of strongly predictive independent variables proves impractical because no group of highly predictive variables exists, alternative matching approaches will be tried (Bishop et al., 2006, pp. 6-7).
The demonstration project should be completely restructured or terminated
Given the design of the demonstration project and the erroneous assumptions on which it appears to have been based, it is not a surprise that it has fallen short of its participation goals. The recent changes in November 2007 to expand the geographic areas and increase the maximum hours of service may increase participation but do not address the core design flaws. The possibilities for remediation appear to be:
- Redesign the project to address the concerns discussed above and redesign the evaluation to restrict it to comparison data actually available to CMS--for example, the number of beneficiaries receiving rehabilitation services before versus during the project. or
- Terminate the project and revise the legislation to reflect the current state of national reimbursement for vision rehabilitation and to limit its scope to gaining privileges for certified low vision therapists, certified O&M specialists, and certified vision rehabilitation therapists, which may allow passage without an expensive and cumbersome demonstration project.
We believe that either the project will implode of its own internal inconsistencies or, if it proceeds as designed, a negative outcome is virtually assured. In either case, the field of vision rehabilitation is not well served by the project, nor are the vision rehabilitation professionals who would otherwise stand to benefit by expanded privileges with respect to Medicare reimbursement.
Centers for Medicare and Medicaid Services. (2002, May 29). Provider education article: Medicare coverage of rehabilitation services for beneficiaries with vision loss. Program memorandum: Intermediaries/carriers (Change request 2083, Transmittal AB-02-078). Washington, DC: U.S. Department of Health and Human Services.
Bishop, C., Tennstedt, S., Ritter, G., Perloff, J., McGraw, S., Caswell, C., Stoddard, A., & Horwitz, J. (2006). Evaluation of the low vision rehabilitation demonstration: Design report (CMS contract no. 500-00-0031). Waltham, MA: Schneider Institute for Health Policy, Heller School for Social Policy and Management, Brandeis University.
Make Your Opinions Known
Readers are encouraged to respond to this Speaker's Corner column by visiting the JVIB message board, <www.afb.org/jvib_message_board.asp>, to engage in an online discussion.
Lylas Mogk, M.D., M.Ed., director, Visual Rehabilitation and Research Center, Henry Ford Health System, 29200 Schoolcraft, Livonia, MI 48150; e-mail: <firstname.lastname@example.org>. Gale R. Watson, M.Ed., CLVT, acting director, Blind Rehabilitation Service, Rehabilitation Services, Office of Patient Care Services, Veterans Health Administration, Washington, DC; and vision rehabilitation program manager, Atlanta VA Medical Center, 1670 Clairmont Road, Decatur, GA 30033; e-mail: <email@example.com>. Michael Williams, Ph.D., rehabilitation planning specialist and data manager, Blind Rehabilitation Service, Rehabilitation Services, Office of Patient Care Services, Veterans Health Administration, Washington, DC; and research scientist, Rehabilitation Research & Development Center of Excellence for Aging Veterans with Vision Loss, Atlanta VA Medical Center; e-mail: <firstname.lastname@example.org>.
Download braille-ready file
Download ASCII text file (ASCII files are for download only)
Previous Article | Next Article | Table of Contents
There are 0 comments on this article.
The Journal of Visual Impairment & Blindness (JVIB)--the international, interdisciplinary journal of record on blindness and visual impairment that publishes research and practice
and serves as a forum for the exchange of ideas, airing of controversies, and discussion of issues--is copyright © 2014 American Foundation for the Blind. All rights reserved.
If you would like to give us feedback, please contact us at email@example.com.