NCSA Submits Public Comment on HCBS Quality Measure Set

PUBLIC COMMENT DEADLINE: MAY 28, 2026 · 11:59 PM ET — submit by 5 PM ET to be safe. Scroll to the action section ↓

VIA ELECTRONIC SUBMISSION at www.regulations.gov

Centers for Medicare & Medicaid Services
Department of Health and Human Services
Attention: CMS-2453-NC
P.O. Box 8016
Baltimore, MD 21244-8016

Re: Medicaid Program; 2028 Medicaid Home and Community-Based Services Quality Measure Set; CMS-2453-NC

Date: May 27, 2026

Dear Administrator Oz:

The National Council on Severe Autism (NCSA) appreciates the opportunity to comment on the proposed 2028 Medicaid HCBS Quality Measure Set (CMS-2453-NC). Quality measurement for Home and Community-Based Services should evaluate outcomes for the people each measure covers, not the categorical features of the setting or service models.


Background

On April 28, 2026, the Centers for Medicare & Medicaid Services (CMS) published the proposed 2028 Medicaid Home and Community-Based Services (HCBS) Quality Measure Set in the Federal Register, opening a thirty-day public comment period that closes May 28, 2026. The proposal contains 23 mandatory measures and 2 voluntary measures (CMS-2453-NC, Tables 4 and 5). It will be the first measure set states must report under 42 CFR § 441.311(c), beginning July 9, 2028, with stratified reporting phased in through 2032. CMS developed the proposed set after considering recommendations from the HCBS Quality Measure Set Review Workgroup (Mathematica, 2024-2026), a multi-stakeholder body convened in 2024 to review the 2024 set and recommend additions, removals, and modifications. The Workgroup issued its Final Report in 2025; CMS adopted some recommendations and declined others, with the rationale for each disposition documented in the notice. References to “the Workgroup” throughout these comments refer to that body.

NCSA represents individuals with severe forms of autism, including those with profound cognitive impairment, limited or no functional communication, complex medical needs, and intensive lifelong support requirements. The population we represent depends on Medicaid HCBS more than nearly any group the proposed set covers, and is among the most vulnerable to gaps in service delivery, workforce shortages, and measurement instruments that cannot capture their experience. Adults with severe autism cannot reliably self-report on experience-of-care surveys. Their families absorb the consequences when authorized services fail to arrive. Their stability depends on consistent, skilled direct support workers whose availability the proposed set does not measure. A quality measure set that cannot see this population cannot drive improvement for it.


Executive Summary

NCSA submits ten recommendations on the proposed set:

  1. Restore a mandatory measure of consumer-directed services.

  2. Add a mandatory measure of the gap between authorized and delivered service hours, disaggregated by acuity level using a validated multi-source support-intensity instrument.

  3. Stratify the five mandatory administrative and assessment measures by HCBS population group, with sub-stratification by support intensity tier for the IDD population.

  4. Add risk adjustment to LTSS-6, LTSS-7, and LTSS-8 that accounts for documented medical necessity in facility admissions and length of stay.

  5. Require a validated proxy or family-respondent protocol for participants who cannot self-report, with independent third-party investigation of safety incidents.

  6. Include all four Workgroup-recommended access measures as mandatory in the 2028 set.

  7. Restructure LTSS-1 and LTSS-2 so the numerator ties to delivered services traceable to the plan.

  8. Add a mandatory direct support workforce capacity measure covering vacancy rate, turnover rate, and wage adequacy benchmarked to acuity.

  9. Commission psychometric validation of POM and NCI-IDD for participants who cannot self-report, and add a validated Family-Reported Outcome Measure such as FROM-16.

  10. Require states to deploy modern data infrastructure capable of supporting these measurement asks, with federal technical assistance and matching support.

The proposed 2028 HCBS Quality Measure Set leaves the population of individuals affected by severe forms of autism and their families substantially invisible in the data CMS will use to drive state Medicaid HCBS improvement; the ten recommendations that follow describe what would close that gap.


Recommendations

1. The removal of HCBS-10 leaves the proposed set with no measure of self-direction.

The notice proposes to remove HCBS-10 (Self-Direction of Services and Supports among Medicaid Beneficiaries Receiving Long-Term Services and Supports (LTSS) through Managed Care Organizations) from the 2028 set entirely, citing feedback from interested parties that the measure is administratively burdensome and provides “unclear information on quality of care or outcomes” (CMS-2453-NC, Section II.B.).

For many families NCSA represents, consumer-direction is the only mechanism for securing the one-to-one or two-to-one workforce with complex transdisciplinary skillsets that an individual affected by severe forms of autism often requires. As a note, NCSA uses the term consumer-direction to avoid ambiguity, ensuring participants and their legal representatives, if applicable, are recognized decision-makers. Though CMS recognizes self-direction, consumer-direction, and participant-direction interchangeably, the community NCSA represents reports exclusion when self-direction is misinterpreted by agencies and managed care organizations as requiring a participant with the capacity for autonomous decisionmaking. Regardless of terminology, the substantive purpose of self-direction, as considered in HCBS-10, demands that NCSA strongly urge CMS to restore, retain, and improve measurement of this critical service option.

Agencies claim a portion of CMS reimbursement rates as administrative fees, and agency-based staffing routinely fails to deliver authorized hours for participants with significant behavioral and complex medical needs because agencies cannot recruit at the rates offered, cannot match consistent workers to individuals who need predictability, and decline complex cases at intake. Consumer-direction shifts control of administration (recruitment, training, termination, and related functions) to the participant or their representative, if applicable. When consumer-direction is elected, the consumer assumes the administrative burden unpaid, which allows direct workers access to higher wages without increasing CMS reimbursement rates. Removing the only measure of this category from a set covering 8.4 million HCBS beneficiaries signals that consumer-direction is not a quality concern. We disagree.

Recommendation 1: Restore a mandatory measure of consumer-directed services that captures both self-directed arrangements and arrangements directed by substitute decision-makers, and that assesses whether participants received authorized hours and whether budget authority enabled competitive wage rates. If CMS proceeds with the HCBS-10 removal, the final notice should commit to a successor measure for the 2030 set built on the consumer-direction framing.

2. The proposed set does not measure whether authorized services are actually delivered.

Beyond self-direction, the proposal leaves a second category of service delivery unmeasured: whether authorized services arrive at all. The 23 mandatory and 2 voluntary measures (CMS-2453-NC, Tables 4 and 5) contain no measure of the gap between services authorized in a person-centered plan and services actually delivered. LTSS-2 measures whether a plan exists; it does not measure whether the services described reach the participant. The most common HCBS failure NCSA hears from families is the gap between approval and delivery: authorized hours go unfilled because no direct support worker is available, the agency declines the case, or the rate is too low to recruit.

The Medicaid and CHIP Payment and Access Commission (MACPAC) documented this data gap over five years ago. State information technology (IT) and data analysis capacities “limit data-driven program policies and decision-making,” waiver expenditure data “do not provide the full picture needed to assess ID/DD [intellectual disability/developmental disability] services,” and siloes between state Medicaid and developmental disabilities agencies block sharing of program-critical data (Barth, Lewis, & Simmons, 2020). The gap remains unaddressed.

Recommendation 2: Add a mandatory measure of authorized-versus-delivered service hours. States should report this measure disaggregated by acuity level using a validated multi-source support-intensity instrument with item content shielded from public access to prevent system gaming. AAIDD’s Supports Intensity Scale™ (SIS) is the most widely adopted such instrument; state equivalents meeting the same multi-source and content-validity standards are acceptable. Most states currently administer such instruments through state-trained assessors or managed care plan staff; CMS should require phase-in of third-party administration to reduce conflict-of-interest risk and protect against gaming by entities with a financial stake in the assessment outcome. CMS already collects the underlying utilization data through Transformed Medicaid Statistical Information System (T-MSIS) Analytic Files. If methodology development cannot be completed for 2028, CMS should commit to inclusion in the 2030 set.

3. Required stratification by geography alone will render the severe autism population invisible.

Where the prior two sections describe measurement absences at the program level, the third concerns who appears in the data the proposal does produce. CMS proposes mandatory stratification of five measures by geography only (CMS-2453-NC, Section II.C.) and aggregate reporting across all HCBS programs, authorities, delivery systems, and managed care plans (Section II.D.). The combined effect is that participants with severe forms of autism—a cross-disability population—will be reported in a single aggregate alongside aging participants, participants with mild intellectual disability, physical disability, serious mental illness, and acquired brain injury, populations whose support profiles are not comparable.

NCSA families report states have reduced hours during brief periods of improvement, despite severe/profound autism being a lifelong condition in need of intensive supports across the lifespan. Aggregate data does not accurately capture the unique needs of this population. Symptom improvement under service is not evidence the service is no longer required. Participants phasing out of services appear as a measured success in HCBS aggregate data. For our population, a reduction in hours due to brief periods of stability triggers regression and costly emergency intervention. This kind of common experience among NCSA families is hidden in aggregate data.

A state can post strong aggregate scores while underserving the population NCSA represents because better-served populations within the same aggregate mask the gap.

Recommendation 3: Stratify the five mandatory administrative and assessment measures by HCBS population group, using the categories CMS itself references (older adults; adults with IDD; adults with physical disabilities; adults with serious mental illness; adults with acquired brain injury; cross-disability populations). For the IDD stratum, sub-stratify by support intensity tier using the same instrument criteria specified in Recommendation 2: an evidence-based and validated multi-source support-intensity instrument with content shielded from public access, with phase-in of third-party administration. Apply cell-size suppression protections consistent with the privacy and small-cell-size considerations CMS raises in Section II.C. CMS should specify that acuity-stratified data identifies under-service of high-support-need populations, not to justify service-hour reductions for participants whose stability depends on the service. Stratify survey measures by population group where the survey instrument supports it.

4. Facility admission and length-of-stay measures (LTSS-6, LTSS-7, LTSS-8) require risk adjustment for medical necessity.

Where stratification renders the severe autism population invisible at the aggregate level, three facility-related measures actively penalize states for serving them. LTSS-6, LTSS-7, and LTSS-8 score facility admissions and length of stay as inverse quality indicators. Their technical specifications (CMS-2453-NC Table 4) contain no risk adjustment for medical necessity. For some participants with severe autism, including those with treatment-resistant aggression, severe self-injury, life-threatening pica, severe elopement risk, or chronic catatonia, admission to a higher level of care may be the appropriate clinical disposition. The Supreme Court’s decision in Olmstead v. L.C., 527 U.S. 581 (1999), interpreting the Americans with Disabilities Act’s integration mandate, requires placement in the most integrated setting appropriate to the individual’s assessed needs. The Court emphasized that “nothing in the ADA or its implementing regulations condones termination of institutional settings for persons unable to handle or benefit from community settings” (527 U.S. at 601-02). For some participants whose needs cannot be safely met in community placement, a facility may be the clinically appropriate setting. Unrisk-adjusted, the proposed measures create a direct incentive for states to deny medically necessary placements to protect reported quality scores.

Recommendation 4: Add risk adjustment to LTSS-6, LTSS-7, and LTSS-8 that accounts for documented medical necessity. At minimum, exclude from the numerator admissions and stays preceded by a determination that community placement could not safely meet the participant’s assessed needs. Alternatively, report a separate “appropriate facility admission” sub-rate alongside the total rate.

5. The proposed reliance on adult participant-reported surveys leaves NCSA’s population systematically underrepresented.

The facility measures fail one subset of severe autism participants by penalizing necessary admissions. The participant-reported survey measures fail a different subset by relying on instruments those participants cannot use. Eighteen of the 23 mandatory measures draw from participant-reported experience-of-care surveys: the Consumer Assessment of Healthcare Providers and Systems Home and Community-Based Services survey (HCBS CAHPS), the National Core Indicators-Aging and Disabilities survey (NCI-AD), the National Core Indicators-Intellectual and Developmental Disabilities In-Person Survey (NCI-IDD), and Personal Outcome Measures (POM). The NCI-IDD In-Person Survey and POM are administered to adults 18 and older with IDD; the four surveys, as currently administered, are adult-only instruments. The proposed set contains no measures for children, youth, or young adults receiving HCBS.

For adults affected by severe forms of autism, direct self-report is either impossible or methodologically unreliable. The proposed set does not specify proxy methodology, does not require oversampling of populations less likely to self-report, and does not provide a mechanism by which their experiences reach the reported measure values.

The limitation is acute for the safety measures. HCBS CAHPS: Personal Safety and Respect Composite (CMS Measure Inventory Tool [CMIT] ID 00095-03-C-LTSS) and POM: People Are Free from Abuse and Neglect (CMIT 01822-01-C-LTSS) both rely on participant self-report. The population most vulnerable to abuse and neglect in HCBS is the population least able to self-report it. A safety measure that excludes nonverbal cross-disability participants measures the safety of the participants safest to ask, not the participants most at risk.

Recommendation 5: Require a validated proxy or family-respondent protocol for participants who cannot self-report, and require states to report the proxy-response rate alongside the measure value so a state cannot post strong scores by excluding the hardest-to-capture participants. For the safety measures, require independent third-party investigation of reported incidents, with protocols designed to detect abuse and neglect of nonverbal cross-disability participants. NCSA supports CMS’s commitment to develop Children, Youth, and Young Adult (CYYA) measures and asks that the eventual set include validated family-respondent methodology from the outset.

6. CMS rejected the Workgroup’s recommendation to add four substantive access measures.

The participant-reported survey gap above describes whose voices the proposed set misses. The Workgroup also identified four measure topics whose absence cuts across populations. The Workgroup recommended adding four NCI-AD and NCI-IDD measures (CMS-2453-NC, Table 3): access to mental health services, access to needed assistive equipment, knowing whom to contact about a complaint, and knowing whom to talk to about changing services. CMS declined to include them, citing the absence of conceptually comparable measures across the four surveys, which CMS argues would introduce inconsistencies in cross-state reporting (CMS-2453-NC, Section II.A). NCSA agrees cross-survey comparability matters. The four measures the Workgroup recommended address access dimensions for which no comparable measures currently exist precisely because the surveys have historically under-measured these domains. The proper response is to develop comparable measures, not to omit the only ones the Workgroup identified.

Access to mental health services is acutely consequential for severe autism. Psychiatric comorbidity is common, behavioral crises drive emergency department (ED) use and facility admissions, and community psychiatric providers willing to treat patients with profound cognitive impairment are scarce. A 2021 nationwide survey of 714 US physicians found that only 56.5 percent strongly agreed they welcomed patients with disability into their practices, only 40.7 percent were very confident in providing the same quality of care, and 82.4 percent believed people with significant disability have worse quality of life than nondisabled people (Iezzoni et al., 2021). The scarcity of community psychiatric capacity is a documented driver of emergency department use and inpatient admission for psychiatric crisis intervention among adults with intellectual and developmental disabilities, who experience ED utilization at substantially higher rates than the general population (Durbin et al., 2018).

Assistive and enabling technology and equipment, including communication devices, carries comparable weight: it determines whether participants can communicate, manage daily activities, and engage with behavioral support. Sixteen states have formally adopted Technology First initiatives that prioritize technology access for people with intellectual and developmental disabilities as the first option for support (Friedman, 2023). State HCBS policy moves toward expanding accountability for assistive technology access; the federal measure set should follow that trajectory by including the assistive equipment measure the Workgroup recommended, not by leaving access unmeasured. Knowing how to register a complaint or change providers is a basic accountability function the proposed set leaves unmeasured.

Recommendation 6: Include all four Workgroup-recommended measures as mandatory in the 2028 set. If CMS declines any specific measure, the final notice should provide measure-specific rationale demonstrating that the cross-survey alignment problem cannot be solved through measure development.

7. LTSS-1 and LTSS-2 should be restructured to measure delivered services, not solely plan existence.

The access measures in Section 6 are missing from the proposal. LTSS-1 and LTSS-2 are present but measure the wrong thing. The Workgroup recommended removing LTSS-1, LTSS-2, and LTSS-3 (CMS-2453-NC, Table 3). CMS removed LTSS-3 but proposes keeping LTSS-1 and LTSS-2 as mandatory. Person-centered service plans matter substantively: they document what services a participant requires, what supports those services should provide, and how the participant’s preferences shape delivery. The problem with LTSS-1 and LTSS-2 as currently structured is that they measure whether a plan exists with documented elements, not whether the services described in the plan reach the participant. The Workgroup identified the limited correlation between plan documentation and delivered services; the solution is to fix the measures, not to abandon person-centered planning as a quality dimension.

Recommendation 7: Restructure LTSS-1 and LTSS-2 so the numerator ties to delivered services traceable to the plan, or pair them with the new authorized-versus-delivered measure proposed in Recommendation 2.

8. The proposed set contains no measure of the direct support workforce.

The plan-existence problem in Sections 2 and 7 reflects a deeper gap: the proposed set never measures the workforce that turns plans into delivered services. Across the 23 mandatory and 2 voluntary measures, no measure addresses the capacity, stability, or wage adequacy of the direct support workforce that delivers HCBS. State quality scores cannot meaningfully reflect HCBS quality without measuring the workforce that delivers it.

Workforce capacity is the primary determinant of whether authorized services arrive for the severe autism population. NCSA’s 2025 Family Caregiver Survey of 1,289 families documented the gap: 69 percent of respondents need staff trained in non-verbal communication, 67 percent in daily living support, and 60 percent in managing severe aggression. Fifteen percent cannot find appropriately trained staff at all, and only 21 percent report reliable in-home staffing of any kind (NCSA, 2025).

Recommendation 8: Add a mandatory direct support workforce capacity measure. At minimum, the measure should include the direct support professional (DSP) vacancy rate, the DSP turnover rate, and the average DSP wage rate benchmarked to the skill, training, and acuity-level of the participants the DSP serves rather than to a flat living-wage floor. States should report all three sub-measures statewide and stratify by HCBS program, where feasible. CMS should also commit to development, for the 2030 set, of an HCBS provider acceptance rate measure stratified by support intensity tier.

9. The proposed measure set excludes both participants who cannot self-report and the families of affected participants.

Workforce measurement (Section 8) describes a missing dimension at the system level. The next gap sits at the instrument level: even the data the proposed set does collect cannot reach the people NCSA represents or their families. CMS’s four domains for participant-reported measures (community inclusion, person-centered care, safety, and transportation) align with what NCSA’s population needs measured (CMS-2453-NC, page 21). The gaps in the proposed set sit at the instrument level. The proposed Patient-Reported Outcome Measures (PROMs), POM and NCI-IDD, are not validated for participants who cannot self-report. The proposed set also contains no Family-Reported Outcome Measure (FROM) capturing family member experience as an outcome in itself. Together these omissions leave the severe autism population and its families systematically invisible. Regulation at 42 CFR § 441.312(c)(3)(iv) requires measures to reflect “an evidenced-based process including testing, validation, and consensus among interested parties.” Neither the PROM gap nor the FROM gap is closed by what CMS currently proposes.

NCI-IDD’s published methodology acknowledges the PROM gap directly. The NCI-IDD In-Person Survey permits proxy respondents only for the observable Section II questions (Community Inclusion, Choices, Respect/Rights, Access to Needed Services), explicitly excluding the subjective experience questions in Section I where proxy reliability cannot be assumed (NCI-IDD, 2023). Peer-reviewed scoping reviews of proxy versus self-report in intellectual disability find mixed agreement across the literature, with subjective well-being consistently demonstrating the weakest correlation between respondents (Hithersay et al., 2022). Proxy measurement of internal states in people with intellectual disability remains an unresolved methodological problem in the field (Webb et al., 2024). Research on related quality-of-life instruments has found that scores stratify significantly by disability severity and that participants with severe or profound intellectual disability are systematically underrepresented in psychometric validation samples (Dunne et al., 2022). Research on cognitively complex populations more broadly has documented the same accessibility barriers and demonstrated that co-designed multimodal PROMs can address them when the population’s needs are integrated from the start (Kernaghan et al., 2026).

The proposed set also contains no Family-Reported Outcome Measure capturing the impact of HCBS quality on family members and informal caregivers. For the severe autism population, family caregivers absorb the consequences of HCBS service gaps directly: when authorized hours go unfilled, family members provide the substitute care. Family member quality of life is therefore an HCBS outcome in its own right, beyond serving as a proxy for participant experience. A 2026 study of 129 family members and partners of people with autism, using the validated Family Reported Outcome Measure (FROM-16), found a mean total score of 18.5, indicating “a very large effect” on family quality of life (Shah et al., 2026). The authors concluded that the impact “underscores the need for comprehensive family-centred care.” FROM-16 has been mapped to standard EQ-5D health utility values in a separate study, allowing family impact to be integrated into health economic evaluation (Shah et al., 2024). The instrument exists, is validated, and could be incorporated into the HCBS Quality Measure Set.

Recommendation 9: CMS should: (a) commission psychometric validation of POM and NCI-IDD for participants who cannot self-report, including those with severe forms of autism, profound intellectual disability, and comparable lifelong intensive support needs, before requiring those instruments as the primary measures of HCBS quality for this population; (b) add a validated Family-Reported Outcome Measure such as FROM-16, or commission development of a comparable FROM instrument, to capture family member quality of life impact as a distinct HCBS outcome; (c) where validation cannot be completed for the 2028 set, require states to report the proxy-response rate alongside each measure value (consistent with Recommendation 5) and the participant exclusion rate, defined as the percentage of eligible participants whose responses could not be obtained at all; and (d) commit to inclusion in the 2030 set of validated instruments, co-designed multimodal PROMs where appropriate, and a family-reported outcome measure.

The recommendations above require data infrastructure capable of supporting them. The final recommendation turns to that infrastructure.

10. The measure set design assumes administrative burden that 2026 data infrastructure can defeat.

Several interested parties in the docket have argued that more granular measurement, broader stratification, and faster reporting timelines would impose excessive administrative burden on states. The argument reflects the data infrastructure many states currently operate on. It does not reflect what current technology makes possible. Peer-reviewed research demonstrates that artificial intelligence and big data analytics applied to health insurance systems can reduce administrative delays, improve claims processing efficiency, and optimize resource allocation (Ara et al., 2025). Automated multimodal classification of healthcare data has achieved 98.57 percent accuracy in recent benchmarks (Jaya & Mahalakshmi, 2025), supporting the feasibility of large-scale extraction from case management records. Edge computing architectures further enable real-time data optimization across distributed healthcare systems (Liu & Zhang, 2023).

CMS itself has reached the same conclusion. The Health Technology Ecosystem initiative states that “outdated infrastructure and disconnected data have made it harder, not easier, to get quality care and achieve positive long-term outcomes” (CMS Health Technology Ecosystem, 2025). The accompanying CMS Interoperability Framework defines criteria for CMS-Aligned Networks that allow payers, including CMS itself, to query networks for clinical data tied to claims and for specific quality data elements necessary for payment or health care operations (CMS Interoperability Framework, 2025). State Medicaid agencies in Alabama, Louisiana and Pennsylvania have pledged to connect their Medicaid systems to CMS Aligned Networks (CMS Health Technology Ecosystem: States, 2026).

CMS has also published the CMS AI Playbook to guide agency adoption of artificial intelligence. Patient-facing applications from major technology and healthcare companies, including Apple, Google, Microsoft, OpenAI, Anthropic, CVS Health, and UnitedHealth Group, have pledged to connect to CMS Aligned Networks under the Ecosystem’s Kill the Clipboard and Conversational AI Assistants categories.

The MACPAC findings cited in Recommendation 2 (Barth, Lewis, & Simmons, 2020) recommended data infrastructure improvements five years ago. The peer-reviewed literature and CMS’s own initiatives together establish that the technical potential to support the measurement asks in the recommendations above exists today. CMS cannot consistently commit to modernizing the digital health ecosystem while accepting that infrastructure-based quality reporting is too burdensome to require. The burden objection should drive infrastructure investment, not constrain what HCBS quality data states are required to produce.

Recommendation 10: Require states to deploy modern data infrastructure capable of supporting the reporting asks in the recommendations above, with federal technical assistance and federal match support for the underlying systems. The final notice should establish that administrative burden is a near-term implementation question to be solved through infrastructure investment, not a permanent constraint on what HCBS quality data states are required to produce.


Closing

The proposed 2028 HCBS Quality Measure Set leaves the population of individuals affected by severe forms of autism and their families substantially invisible in the data CMS will use to drive state Medicaid HCBS improvement. The set measures community integration without measuring whether services arrive. It measures the existence of plans without measuring delivery of those plans, and scores facility admissions as failure without distinguishing medically necessary placements. It excludes participants who cannot self-report from the experience measures, rendering the severe autism population invisible in aggregate state results. It removes the only self-direction measure, declines four Workgroup-recommended access measures, and contains no measure of the direct support workforce that delivers HCBS. 

NCSA has coordinated a congressional sign-on letter to the CMS Administrator that echoes several of these concerns and requests acuity-disaggregated reporting of HCBS waiver utilization. NCSA will continue this work through its July 2026 policy convening in Washington, D.C., where featured speakers will address Family-Reported Outcomes, modern IT infrastructure improving access to health resources, importance of risk-adjusted rates, and more. 

We urge CMS to act on the ten recommendations above before finalizing the 2028 set..

Respectfully,

/s/
Jackie Kancir
Executive Director
National Council on Severe Autism
www.NCSAutism.org

Your Voice Counts: Submit Your Own Public Comment

Free Downloads
NCSA’s Public Comment PDF
Editable Sample Letter to practice drafting your own public comment

Tips for Submitting Your Own Public Comment

  • Make it personal. Your family's story, in your own words, is what makes the difference.

  • Share your experience. What does HCBS services look like for your family?

  • Share your recommendations. What does CMS need to measure so your family's experience shows up in the data?

  • Be specific. One concrete example beats five general claims. Name your state. Say what service was authorized, what did not arrive, what happened next.

  • Be respectful. Focus on policy, not personalities. Civil comments carry weight.

  • Comments become public record. Do not include children's names, medical details, addresses, or anything else you would not want public. Do NOT put your address or phone in the comment body for this docket.

When you are happy with your comment, click the next button to be redirected to the page to submit your public comment.

submit your public comment

References

Regulatory and CMS sources

Centers for Medicare & Medicaid Services. (2025). CMS AI Playbook. https://ai.cms.gov/CMS-AI-Playbook.pdf

Centers for Medicare & Medicaid Services. (2025). Health Technology Ecosystem. https://www.cms.gov/priorities/health-technology-ecosystem/overview

Centers for Medicare & Medicaid Services. (2025). Interoperability Framework. https://www.cms.gov/health-technology-ecosystem/interoperability-framework

Centers for Medicare & Medicaid Services. (2026). Health Technology Ecosystem: Early Adopters — States. https://www.cms.gov/health-tech-ecosystem/early-adopters/states

Centers for Medicare & Medicaid Services. (2026). Medicaid Program; 2028 Medicaid Home and Community-Based Services Quality Measure Set [Notice with comment period; CMS-2453-NC]. Federal Register, 91(81), 22823. https://www.federalregister.gov/documents/2026/04/28/2026-08190/medicaid-program-2028-medicaid-home-and-community-based-services-quality-measure-set

HCBS Quality Measure Set, 42 C.F.R. § 441.312 (2024).

Case law

Olmstead v. L.C., 527 U.S. 581 (1999). https://supreme.justia.com/cases/federal/us/527/581/

Peer-reviewed literature

Ara, J., Anjum, A. T., Chambugong, L., & Chowdhury, R. A. (2025). Harnessing artificial intelligence and big data analytics to enhance premium optimization and utilization efficiency in health insurance systems. Journal of Business and Management Studies, 7(7). https://doi.org/10.32996/jbms.2025.7.7.6

Barth, S., Lewis, S., & Simmons, T. (2020). Medicaid services for people with intellectual or developmental disabilities: Evolution of addressing service needs and preferences. Health Management Associates report to the Medicaid and CHIP Payment and Access Commission. https://www.macpac.gov/wp-content/uploads/2021/01/Medicaid-Services-for-People-with-Intellectual-or-Developmental-Disabilities-%E2%80%93-Evolution-of-Addressing-Service-Needs-and-Preferences.pdf

Dunne, L., Hogan, A., O’Toole, A., & Carney, P. (2022). Quality of life outcomes in a community cohort of adults with an intellectual disability using the Personal Outcome Scale. Frontiers in Rehabilitation Sciences, 3, 941959. https://pmc.ncbi.nlm.nih.gov/articles/PMC9397816/

Durbin, A., Balogh, R., Lin, E., Wilton, A. S., & Lunsky, Y. (2018). Emergency department use: Common presenting issues and continuity of care for individuals with and without intellectual and developmental disabilities. Journal of Autism and Developmental Disorders, 48(10), 3542-3550. https://link.springer.com/article/10.1007/s10803-018-3633-7

Friedman, C. (2023). Technology first: The impact on technology availability for people with intellectual and developmental disabilities. Journal of Applied Research in Intellectual Disabilities, 36(6), 1251-1263. https://onlinelibrary.wiley.com/doi/10.1111/jar.13144

Hithersay, R., Hamburg, S., Knight, B., & Strydom, A. (2022). Proxy-report in individuals with intellectual disability: A scoping review. Journal of Applied Research in Intellectual Disabilities. https://pmc.ncbi.nlm.nih.gov/articles/PMC9378492/

Iezzoni, L. I., Rao, S. R., Ressalam, J., Bolcic-Jankovic, D., Agaronnik, N. D., Donelan, K., Lagu, T., & Campbell, E. G. (2021). Physicians’ perceptions of people with disability and their health care. Health Affairs, 40(2), 297-306. https://www.healthaffairs.org/doi/10.1377/hlthaff.2020.01452

Jaya, J. A., & Mahalakshmi, K. (2025). Efficient healthcare data classification through gazelle-inspired optimization and multimodal deep learning. Knowledge and Information Systems, 67, 7263-7293. https://doi.org/10.1007/s10115-025-02424-3

Kernaghan, D., Egan, K., Lennon, M., & Maguire, R. (2026). Co-designing and evaluating a multimodal digital application to enable people with dementia to self-report quality of life patient-reported outcome measures. Journal of Participatory Medicine, 18, e87565. https://doi.org/10.2196/87565

Liu, Z., & Zhang, X. (2023). Multimedia and multimodal sensing with edge computing for personalized healthcare supply chain system data optimization. Personal and Ubiquitous Computing, 27, 955-972. https://doi.org/10.1007/s00779-022-01679-9

Shah, R., Salek, M. S., Finlay, A. Y., Kay, R., Nixon, S. J., Otwombe, K., Ali, F. M., & Ingram, J. R. (2024). Mapping of Family Reported Outcome Measure (FROM-16) scores to EQ-5D: Algorithm to calculate utility values. Quality of Life Research, 33, 1107-1119. https://doi.org/10.1007/s11136-023-03590-z

Shah, R., Finlay, A. Y., Ali, F. M., Otwombe, K., Nixon, S. J., Nixon, M.-E., Ingram, J. R., & Salek, S. (2026). Family members/partners of people with autism experience a major impact on quality of life as measured by the Family-Reported Outcome Measure (FROM-16). Journal of Patient-Reported Outcomes, 10(Suppl 1), abstract 207.5. 32nd Annual Conference of the International Society for Quality of Life Research. https://link.springer.com/article/10.1186/s41687-026-01004-5#Sec74859625

Webb, J., Stanton, M., Bourne, T., & Donaldson, L. (2024). The use of proxy measurement of internal states in people with intellectual disability: A scoping review. Journal of Applied Research in Intellectual Disabilities. https://onlinelibrary.wiley.com/doi/10.1111/jar.13241

Instrument-steward sources

American Association on Intellectual and Developmental Disabilities. (2023). Supports Intensity Scale (SIS). https://www.aaidd.org/sis

Mathematica Policy Research. (2024-2026). HCBS Quality Measure Set Review Workgroup. https://www.mathematica.org/features/hcbsqmsreview

Mathematica Policy Research. (2025). 2028 HCBS Quality Measure Set Review Workgroup final report. https://www.mathematica.org/-/media/internet/features/2026/hcbs-quality-measure-set/2028hcbsqmsreview-final-report.pdf

National Association of State Directors of Developmental Disabilities Services & Human Services Research Institute. (2023). National Core Indicators IDD In-Person Survey 21-22 overview and methodology. https://idd.nationalcoreindicators.org/wp-content/uploads/2023/05/IPS-21-22-NCI-IDD-Overview-and-Methodology_FINAL5_12.pdf

NCSA sources

National Council on Severe Autism. (2025). National Family Caregiver Survey [n = 1,289]. https://www.ncsautism.org/survey-results