In our initial discussions with the TEP and other stakeholders, we determined that "client functioning" is a potentially multifaceted concept in the context of Medicaid HCBS programs. Unlike other post-acute care services provided in community settings, the primary goal of HCBS programs is not just to restore functional ability compromised through injury or illness. Rather, HCBS programs provide the services and supports needed to allow participants to function in their homes and communities, regardless of impairments. A limited number of programs also include a rehabilitation component. Regardless, all programs strive to promote optimum client health and avoid preventable adverse health outcomes.
To accommodate these multiple dimensions, the domain of client functioning for the measure scan includes the following three subdomains: functioning in daily activities; social role functioning; and health promotion/prevention of adverse health outcomes. The first subdomain relates to support for and ability to engage in daily activities related to self-care and household activities, often referred to as Activities of Daily Living (ADLs) and Instrumental Activities of Daily Living (IADLs).2, 3 At the most basic level, a central endeavor of HCBS programs is enabling individuals with disabilities to undertake such daily activities as bathing, dressing, eating, getting around their homes and communities, using the telephone, buying groceries, and managing finances and housework.
Social role functioning extends the role of functioning to the social and community realm. It refers to individuals' supported ability to function as friends, family members, employees, students, activists, and general members of an integrated society. Finally, health functioning includes not only health promotionix and the receipt of recommended health care services, but also the avoidance of serious adverse or preventable health outcomes, such as injury, abuse, hospitalization, and death.
Functioning in Daily Activities
Two disparate, but related, constructs of functioning in daily activities emerged in the process of assembling the construct list (Table 1). The first encompasses changes in an individual's underlying ability to accomplish basic daily activities over time. This is often measured through external or self-reported assessment data regarding the ability to complete defined activities, such as taking a bath or shower, feeding oneself, or climbing stairs, without assistance. Many well-developed assessment instruments and scales have been designed to measure such underlying abilities; these tools can be used longitudinally.
The second construct is more focused on client outcome, rather than baseline ability, and relates to whether assistance for performing daily activities is available when needed. This notion is somewhat independent of functional ability, because it refers to successful completion of the activity with appropriate supports, regardless of the amount or type of support required. "Assistance" in this construct could include hands-on help, assistive technology, cuing, or supervision.
During both the initial exercise to assess importance and in subsequent written and oral comments, several TEP members argued that other key correlates of activity functioning were missing from these two constructs. These included pain that interfered with daily functioning, along with the concepts of fatigue, weakness, and depression. While not included in Table 1, this feedback suggests that AHRQ consider the full importance of these additional constructs during its planned measure development.x,4
Social Role Functioning
There are many aspects of the social role all individuals play in society, as members of families, social groups, workplaces, communities, and so forth. The support for individuals with disabilities to fulfill these roles was articulated by many stakeholders as an important measure of HCBS quality. To operationalize this notion of social participation, TEP members identified five complementary constructs. The first two, presence of friendships and maintenance of family relationships, capture support for interpersonal relationships. The second two, employment/productive activities for adults and school attendance for children, relate to meaningful and productive roles in society. In particular, the adverse relationship between disability and employment has been well established.5,6 The final, catch-all construct of community inclusion was intended to assess the degree to which program participants are included and integrated within the communities where they live.
To a certain degree, all the constructs listed above are outcomes measures. Failure to realize these outcomes reflects unmet need for the supports necessary to achieve them. One underlying construct we identified for all these dimensions of social role functioning was met or unmet need for transportation. This relationship was identified too late to add as an explicit construct; therefore, the universe of transportation measures was not systematically evaluated. However, AHRQ may want to consider the relationship between transportation and social role outcomes in measures to assess this subdomain.
Health Promotion/Preventing Adverse Health Outcomes
This subdomain comprises two complementary constructs: measures of interventions intended to promote positive health outcomes and counts of adverse, undesirable ones. Indeed, some stakeholders argued that these measure constructs belong more appropriately in the program performance domain. Others, however, saw a relationship between function in daily activities and in the social sphere and maximum health functioning. From a positive perspective, this means that a quality program facilitates receipt of recommended, age-appropriate preventive health care.
According to the input we received, it is equally important for HCBS programs to help prevent negative outcomes. These include serious, reportable events that reflect a consensus regarding client health and safety, such as suicide, assault, injurious falls, and harmful medication errors.xi It also includes more intensive health care services, such as hospitalization, which are potentially avoidable through appropriate community-based care.
Table A.V.1b in Appendix V shows candidate measures for each construct within the three client functioning subdomains that meet the threshold criteria illustrated in Figure 1. It is significant that no construct is without at least one candidate measure. However, there are some notable themes, as well as discernible gaps in the universe of potential functioning metrics, that merit discussion.
As noted earlier, the goal of HCBS programs is typically to support participants' daily activity functioning. Therefore, measures based on client reports of unmet need for assistance to perform daily activities can be used to assess if a State's Medicaid HCBS program is providing appropriate, quality supports to enable functioning in daily activities. As shown in Table A.V.1b, these types of measures are available; some have been tested across multiple populations and several are currently used by State Medicaid programs. Furthermore, later feedback from stakeholders suggests these measures may be more relevant in HCBS programs than measures of change in underlying functional ability. In addition, research suggests that quality measures based on change in daily activity functioning can be problematic.7
With regard to social role functioning, we found multiple measures for each of the relevant constructs included on the short list. Their presence in several consumer surveys and measure sets validates our construct of this subdomain. These constructs, however, are still quite broad, and the candidate measures within each capture different dimensions of the overall construct. More work may be required to determine which metrics are the most important and relevant assessments of HCBS quality. For example, "employment" can be measured as bivariate (employed vs. not employed), type of employment (supported, competitive, part-time, etc.), in terms of hours worked or wages earned, or as satisfaction and choice with regard to current position.
There was consensus among stakeholders that health promotion for HCBS participants is important, and we found that several guidelines for what constitutes appropriate health promotion have been developed. These include generic guidelines, such as the Healthcare Effectiveness Data and Information Set (HEDIS) measure set for managed care settingsxii and the U.S. Preventive Services Task Force's (USPSTF) recommendations of evidence-based preventive services for primary care (the latter designed for the asymptomatic population).xiii Others are specific to populations and settings, such as the Home-Based Primary Care Quality Indicator Set for community-dwelling elders,8 HRSA's primary care recommendations for adults with HIV/AIDS,xiv and Wilkinson, et al.'s recent publication adapting the USPSTF recommendations for adults with intellectual disabilities.9
There is no consensus set of recommended preventive health care services applicable across Medicaid HCBS programs. Perhaps even more than with other constructs, it is an open question as to whether a single set can reflect the diverse needs of Medicaid HCBS recipients. Furthermore, there was no consensus regarding the actionability of such recommendations. Nor was there consensus on whether Medicaid programs can reasonably be held accountable for their implementation, given that HCBS program participants' health care may be delivered by a number of different providers and reimbursed by multiple payers.xv,10
Assuming that AHRQ does not make new preventive criteria a part of the USPSTF effort, a suggested alternative approach may be to capture the broader issue of access to basic health care services as part of a higher quality HCBS system. Measuring the availability of health care services would address the critical nature of services to overall client functioning, while still recognizing the role of care providers in determining the specific services that are appropriate for each individual. Measures based on client reports would assess if a State's Medicaid HCBS program is providing the appropriate supports to facilitate availability of health care services to clients.
Consistency and consensus are lacking with regard to HCBS-appropriate definitions of serious reportable events. This is particularly true for the non-intellectually disabled HCBS populations.11 In addition, most of the measures endorsed by the National Quality Forum and implemented by the States, for example, apply to institutional or residential settings only.12 More work is needed to develop definitions and measures that could work across community-based populations and State systems and be supported by a broad range of stakeholders.
Finally, with regard to avoidable hospitalizations, AHRQ has developed a set of Prevention Quality Indicators designed to measure hospitalizations that could have been prevented with appropriate community-based care.xvi These measures enjoy wide support (including endorsement by the National Quality Forum) and have been adopted for use in several States.13 However, their appropriateness to an HCBS population, with its range of disabling conditions and higher health risks, has not been assessed.
The Prevention Quality Indicators would require additional testing and/or modification to determine their appropriateness as quality measures for Medicaid HCBS programs. There is also an issue of whether the pertinent State administrative data sources (e.g., hospital discharge/encounter data) could be accessed in a timely manner to calculate these measures. Indeed, several TEP members raised concerns about the appropriateness of such a metric in evaluating Medicaid HCBS programs, as well as the ability of such programs to capture data on hospitalization, let alone determine which hospitalizations may have been avoidable.