Analyzing use of home and community-based services (HCBS) at the national level by type of service can be challenging due to variation in Medicaid coverage, codes used on claims, and terminology across states. To support analysis of Medicaid-funded HCBS use, this brief describes a methodology for identifying and classifying HCBS claims reported to the Transformed Medicaid Statistical Information System by service category.

KEY POINTS

• Home and community-based services (HCBS), primarily funded through Medicaid, are a range of medical and non-medical services provided in home and community settings that support individuals with functional limitations to reside in these settings rather than in institutions.
• Assessing use of Medicaid-funded HCBS at the national level, particularly by type of service, can be challenging because of variation in state Medicaid coverage and the use of codes on claims for these services. While the Transformed Medicaid Statistical Information System (T-MSIS) has an HCBS taxonomy code set that allows states to report the type of service on claims, few states use the taxonomy codes, and the data element is considered unusable.
• Using procedure codes and other data elements found in the 2016-2020 T-MSIS Analytic Files Research Identifiable Files, we identified HCBS claims paid by Medicaid, classifying claims into 20 service categories.
• Claims for services in all 20 HCBS categories were identified across all years. Among states and territories that met the data quality criteria for inclusion in the analysis, 10 states, the District of Columbia, and the U.S. Virgin Islands had no claims for home health services, a mandatory service type for Medicaid programs, and the U.S. Virgin Islands had no personal care services claims. Few claims were identified for rent for live-in caregiver services.
• This methodology can be used to analyze Medicaid HCBS use by type of service across states despite variation in state Medicaid programs and reporting of services.

BACKGROUND

Home and community-based services (HCBS) are a range of medical and non-medical services, also referred to as long-term services and supports (LTSS), provided in home and community settings rather than in institutions.¹ Examples of HCBS include case management, personal care, home health, day services, and non-medical transportation. HCBS support a variety of people, including older adults and people with disabilities, who need help with routine activities of daily living.¹ Some HCBS are funded through private insurance and out-of-pocket spending, and unmet service and support needs are often met through unpaid assistance from family,^2,3 but Medicaid is the primary source of funding for HCBS and accounted for more than half of all HCBS spending in 2020.²

In recent decades, policy and programmatic efforts have incentivized use of HCBS over institutional LTSS, leading to increased use of HCBS consistent with many people’s preferences for receiving services in the community.⁴ Yet assessing HCBS use at the national level can be challenging because of variation in states’ Medicaid coverage policies. Under Medicaid, states are not required to cover most HCBS, but can choose to do so under a variety of optional authorities that differ by the types of services covered and eligibility criteria.⁵ These authorities include the 1915(c) HCBS waiver, 1915(i) HCBS state plan option, 1915(j) self-directed personal assistant services, 1915(k) Community First Choice, and Section 1115 demonstration waiver.^6,7 In addition, states can use the Money Follows the Person grant program and the Program of All-Inclusive Care for the Elderly (PACE) as additional pathways to cover HCBS.^8,9 This patchwork of pathways states use to cover HCBS leads to variation in eligibility, scope of benefits, and delivery systems across state Medicaid programs. For example, as of 2018, two-thirds of states offered the personal care state plan option, but far fewer elected additional optional state plan services.¹⁰ All states, however, offer at least one optional HCBS program or waiver.¹⁰

Variation in Medicaid HCBS programs across states creates challenges when analyzing HCBS use nationally, particularly by type of service, due to differences in reporting of services, in addition to differences in services offered across programs. Medicaid claims for HCBS often are not clearly labeled, and procedure codes are not consistent across states.¹¹ HCBS waiver services are often reported as “other” services, making it difficult to distinguish between different types of HCBS.¹¹ When services have more specific labels, they often vary by name across states. Some states use national codes, and others use state-specific codes that differ from one state to the next.¹¹ Inconsistencies in data labeling and procedure codes used stem from differing guidance given to providers and data processing procedures across states and, in some cases, across HCBS programs within a state.¹² In many states, multiple departments, with varying reporting requirements, oversee HCBS programs. To standardize reporting, the Centers for Medicare & Medicaid Services (CMS) developed a nationwide HCBS taxonomy code set and made it available for states to use when reporting Medicaid data to the Transformed Medicaid Statistical Information System (T-MSIS).¹¹ Currently, however, few states use the HCBS taxonomy codes, and the data element in T-MSIS is considered unusable.¹³

Historically, CMS used other data sources, including CMS-64 data and CMS 372 data, to report national and state HCBS expenditures in their Medicaid LTSS Annual Expenditures Reports,¹⁴ as well as counts of 1915(c) waiver program participants nationally and by state.¹⁵ However, the data sources used for the analyses presented in these reports are at the aggregate level, rather than the enrollee level, and do not allow for detailed analysis of the use of specific types of services (e.g., case management, personal care services, or non-medical transportation) within certain HCBS programs or authorities (e.g., 1915(c) waiver programs or 1915(i) HCBS state plan option). In 2022, CMS published a brief with estimated counts of HCBS users in 2019, calculated using data from the T-MSIS Analytic Files (TAF).¹⁶ However, the brief presents HCBS user estimates by service categories used in the LTSS expenditures reports and categories defined in Section 9817 of the American Rescue Plan Act of 2021, and does not disaggregate users within these categories by type of service used.

To support research analyzing Medicaid-funded HCBS use by type of service across states, we assessed whether a modified version of an HCBS taxonomy developed for use with Medicaid Analytic eXtract (MAX) enrollment and claims data submitted to the Medicaid Statistical Information System (MSIS) (the data collection system CMS used prior to T-MSIS) could be used to identify and classify HCBS claims submitted to T-MSIS.¹¹ This issue brief describes methods to identify and classify HCBS claims by type of service using a taxonomy of 20 service categories.

METHODS

HCBS Categories

To create an HCBS taxonomy for use with TAF RIF data (TAF RIF HCBS taxonomy), we started with 17 service categories, shown in Table 1, previously used in the HCBS taxonomy developed by Truven Health and Mathematica for use with MAX data (MAX HCBS taxonomy).^11,18,19 When reviewing these 17 categories from the MAX HCBS taxonomy,^iv we found that home health services and personal care services in a home-based or school-based setting were grouped into two categories: round-the-clock services and home-based services. To enable analysis of these three service types separately from other HCBS, we created three new HCBS categories: home health, personal care services, and school-based personal care services. These three new categories are not mutually exclusive with the MAX HCBS taxonomy categories, meaning claims for certain services could be classified as round-the-clock services or home-based services and as home health, personal care, or school-based personal care services. The Appendix has a table listing each of the categories in the TAF RIF HCBS taxonomy and the services or groups of services included in each category.

Table 1

HCBS Categories from the MAX HCBS Taxonomy.

Classification of Claims

To identify and classify TAF RIF claims for HCBS, we used national Healthcare Common Procedure Coding System (HCPCS) procedure codes, Current Procedural Terminology (CPT) procedure codes, and state-specific procedure codes when we could identify them. In addition, we considered procedure code modifiers, place of service codes, and type of service codes on claims. Procedure code modifiers provide more detail about the service that might not be obvious from the procedure code. If a procedure code could fall into multiple service categories, we used modifiers and place of service and type of service codes to assign the service to only one category.

For the 17 HCBS categories from the MAX HCBS taxonomy, we assigned claims to each category using methods originally developed by Mathematica and Truven Health to create the MAX HCBS taxonomy.¹¹ Use of this methodology was possible because procedure codes, procedure code modifiers, place of service codes, and type of service codes present in MAX are also present in the TAF RIF. Figure 1 shows the four ways that we determined a claim was for a home and community-based service and to which category it belonged.

First, we identified state-specific procedure codes, giving them precedence over national procedure codes, and used them to classify claims (Figure 1, Pathway 1). In the MAX HCBS taxonomy, 15 states had state-specific codes for HCBS.²⁰ Because states can add or retire state-specific codes, we reviewed the most frequently occurring state-specific codes in the OT file paid by a program covering HCBS.^v This investigation identified additional state-specific codes for HCBS in 42 states. In total, 46 states had state-specific codes for HCBS. The five states for which we did not find any state-specific codes were Alabama, Arizona, Illinois, Kentucky, and Nebraska. Puerto Rico and the U.S. Virgin Islands did not have any territory-specific codes for HCBS.

Next, we reviewed national procedure codes (HCPCS and CPT procedure codes), using them to identify and classify claims for HCBS that were not identified using state-specific procedure codes. For claims with national procedure codes, there are three ways to identify an HCBS claim: using the procedure code only, using the procedure code and other codes, or using the procedure code and codes indicating a service was paid by an HCBS program.

Figure 1

Pathways for Classifying Claims for HCBS. HCBS = home and community-based services; OT = Other Services; CPT = Current Procedural Terminology; HCPCS = Healthcare Common Procedure Coding System. Note:

For claims with national procedure codes in the MAX HCBS taxonomy that met the criteria for HCBS and contained enough information to classify a claim (e.g., G0154--services of skilled nurse in home health setting), we used only the procedure code to identify and classify the claim (Figure 1, Pathway 2). But if a claim could be mapped to multiple service categories based on a procedure code alone, we used procedure code modifiers and place of service and type of service codes to determine to which service category to assign the claim (Figure 1, Pathway 3).^vi,vii For example, procedure code T2025 is a generic code for waiver services, but, when accompanied by a procedure code modifier such as TD for services from a registered nurse, HE for behavioral health services, or GP for physical therapy, the procedure code and accompanying modifier can be used to assign the claim to the appropriate service category.

For certain categories of HCBS, we required that the service must be paid for by a program covering HCBS (e.g., Sections 1915(c), 1915(k), 1915(i), and 1915(j), Money Follows the Person, and PACE) (Figure 1, Pathway 4).^viii This additional criterion was necessary due to a key difference between the TAF RIF HCBS taxonomy and the MAX HCBS taxonomy. For the MAX HCBS taxonomy, Truven Health and Mathematica first identified only Section 1915(c) fee-for-service wavier service claims prior to classifying claims by service type.¹¹ When identifying HCBS claims, we started with all claims in the TAF RIF, including claims for services that were provided outside of home and community-based settings. Therefore, we needed the additional criterion to more precisely identify whether a claim was for a service provided in a home or community-based setting.

Claims information used to determine whether the service was provided by a program covering HCBS included program type, benefit type, HCBS service, and waiver type codes (Table 2). The claim had to have at least one of these codes to be considered a claim for HCBS. For example, the procedure code H0006 (alcohol and/or drug services, case management) might or might not refer to a home and community-based service. To confirm the claim was for HCBS, and classify it as case management, the claim had to have at least one of the relevant codes in Table 2.

Table 2

TAF RIF Data Elements and Relevant Codes for Identifying Claims Paid by an HCBS Program.

For the three new HCBS categories not in the MAX HCBS taxonomy (home health services, personal care services, and school-based personal care services), we used similar data elements to identify and classify services under these categories (Table 3). With this classification, home health services, personal care services, and school-based personal care services are not required to be paid by an HCBS program (i.e., they were identified through only pathways 1, 2, or 3 in Figure 1).^ix Services in these three categories are not mutually exclusive with the original MAX HCBS taxonomy categories; these services are also counted in and are a subset of services in the round-the-clock or home-based service categories.

Table 3

New HCBS Categories and the Specifications Used to Identify Claims in These Categories.

FINDINGS

Using the methods described in this issue brief, we identified Medicaid HCBS claims in all states and territories that are reporting to T-MSIS and were not excluded from the analysis due to data quality issues. The TAF RIF taxonomy identified HCBS claims in all service categories in all years.^x However, 10 states (Arkansas, California, Florida, Maine, Minnesota, Mississippi, Oklahoma, Oregon, South Dakota, and Vermont), the District of Columbia, and the U.S. Virgin Islands did not have any claims for home health services, a mandatory service type for Medicaid programs, and the U.S. Virgin Islands did not have any personal care service or school-based personal care service claims. There were few claims for rent for live-in caregiver services (for which the taxonomy only contains state-specific procedure codes from Minnesota and New York), and it is recommended that TAF RIF users do not include this category in their analyses.

CONCLUSION

Building off an HCBS taxonomy developed for use with MAX data, we developed an approach to identifying and classifying HCBS claims in the TAF RIF data. This methodology captures claims for HCBS provided under various HCBS authorities, not just 1915(c) waiver programs, and may be used to analyze Medicaid-funded HCBS use across states. Until the quality of the HCBS taxonomy data element in T-MSIS improves, researchers may use the TAF RIF HCBS taxonomy to assess HCBS use by type of service. However, it should be noted that this methodology has not been validated. Future work could include validating the TAF RIF HCBS taxonomy and using the taxonomy to analyze HCBS use by type of service and relationships between HCBS use, other health care use, and health and person-centered outcomes.

APPENDIX. TAF RIF HCBS TAXONOMY CATEGORIES AND RELATED SERVICES

REFERENCES

1.

“Home and Community Based Services.” Medicaid​.gov. Accessed: July 8, 2022. Available: https://www​.medicaid​.gov/medicaid/home-community-based-services/index.html.

2.

M. O’Malley Watts, M. Musumeci, & M. Ammula. (2022). Medicaid Home and Community-Based Services: People Served and Spending During COVID-19. San Francisco, CA: Kaiser Family Foundation. Available: https://www​.kff.org/report-section​/medicaid-home-community-based-services-people-served-and-spending-during-covid-19-issue-brief/.

3.

J. Caldwell. Reducing Costs for Families and States by Increasing Access to Home- and Community-Based Services. Waltham, MA: Community Living Policy Center, Brandeis University. Available: https://heller​.brandeis​.edu/community-living-policy​/images/reducing-cost-hcbs​.pdf.

4.

N. Thach, & J.M. Wiener. (2018). An Overview of Long-Term Services and Supports and Medicaid: Final Report. Washington, DC: U.S. Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation. Available: https://aspe​.hhs.gov​/reports/overview-long-term-services-supports-medicaid-final-report-0.

5.

“HCBS Technical Assistance Website: Authority Comparison Chart.” hcbs-ta​.org. Accessed: September 21, 2022. Available: https://www​.hcbs-ta.org​/authority-comparison-chart.

6.

“Home and Community Based Services Authorities.” Medicaid​.gov. Accessed: September 21, 2022. Available: https://www​.medicaid​.gov/medicaid/home-community-based-services​/home-community-based-services-authorities/index.html.

7.

M. Guth, & M. Musumeci. (2021). State Options to Expand Medicaid HCBS: Examples and Evaluations of Section 1115 Waivers. San Francisco, CA: Kaiser Family Foundation. Available: https://www​.kff.org/medicaid​/issue-brief​/state-options-to-expand-medicaid-hcbs-examples-evaluations-of-section-1115-waivers/.

8.

“Money Follows the Person.” Medicaid​.gov. Accessed: August 8, 2022. Available: https://www​.medicaid​.gov/medicaid/long-term-services-supports​/money-follows-person/index.html.

9.

“Program of All-Inclusive Care for the Elderly.” Medicaid​.gov. Accessed: August 26, 2022. Available: https://www​.medicaid​.gov/medicaid/long-term-services-supports​/program-all-inclusive-care-elderly​/index.html.

10.

M. Musumeci, M. O’Malley Watts, & P. Chidambaram. (2020). Key State Policy Choices about Medicaid Home and Community-Based Services. San Francisco, CA: Kaiser Family Foundation. Available: https://www​.kff.org/medicaid​/issue-brief​/key-state-policy-choices-about-medicaid-home-and-community-based-services/.

11.

V. Peebles, & A. Bohl. (2013). The HCBS Taxonomy: A New Language for Classifying Home- and Community-Based Services. Baltimore, MD: Centers for Medicare & Medicaid Services. Available: https://www​.cms.gov/Research-Statistics-Data-and-Systems​/Computer-Data-and-Systems​/MedicaidDataSourcesGenInfo​/Downloads/MAX_IB19_Taxonomy​.pdf. [PMC free article: PMC4204915] [PubMed: 25343057]

12.

A. Christ, & T. Huyenh-Cho. (2021). Using Data for Good: Toward More Equitable Home and Community-Based Services in Medi-Cal. Oakland, CA: California Healthcare Foundation. Available: https://www​.chcf.org​/publication/toward-more-equitable-home-community-based-services-medi-cal/.

13.

K.J. Caswell, T.A. Waidmann, & K. Wei. (2021). Measuring Medicaid Service Utilization among Dual Medicare-Medicaid Enrollees using Fee-for-Service and Encounter Claims. Washington, DC: Urban Institute. Available: https://www​.urban.org​/sites/default/files​/publication/104788​/measuring-medicaid-service-utilization-among-dual-medicare-medicaid-enrollees-using-fee-for-service-and-encounter-claims-t-msis-analytic-files-data-quality_0​.pdf.

14.

C. Murray, A. Tourtellotte, D. Lipson, & A. Wysocki. (2021). Medicaid Long Term Services and Supports Annual Expenditures Report: Federal Fiscal Year 2019. Chicago, IL: Mathematica. Accessed: August 7, 2023. Available: https://www​.medicaid​.gov/medicaid/long-term-services-supports​/downloads/ltssexpenditures2019.pdf.

15.

J. Ross, K. Liao, & A. Wysocki. (2021). Medicaid Section 1915(c) Waiver Programs Annual Expenditures and Beneficiaries Report: Analysis of CMS 372 Annual Reports, 2015-2017. Chicago, IL: Mathematica. Accessed: August 7, 2023. Available: https://www​.medicaid​.gov/sites/default/files​/2021-01/cms-372-report-2015-2017​.pdf.

16.

M.Y. Kim, E. Weizenegger, & A. Wysocki. (2022). Medicaid Beneficiaries Who Use Long-Term Services and Supports: 2019. Chicago, IL: Mathematica. Accessed: August 7, 2023. Available: https://www​.medicaid​.gov/medicaid/long-term-services-supports​/downloads/ltss-user-brief-2019.pdf.

17.

“Transformed Medicaid Statistical Information System (T-MSIS) Analytic Files (TAF).” Medicaid​.gov. Accessed: September 21, 2022. Available: https://www​.medicaid​.gov/medicaid/data-systems​/macbis/medicaid-chip-research-files​/transformed-medicaid-statistical-information-system-t-msis-analytic-files-taf/index.html.

18.

S. Eiken. (2012). HCBS Taxonomy Development. Presented at: NASUAD HCBS Conference. Available: http://www​.nasuad.org​/documentation/hcbs_2012​/HCBS%202012%20Presentations1​/HCBS%202012​%20Presentations1​/Thursday/1130/Potomac%202​/2012HCBSTaxonomyConfSlides.pdf.

19.

S. Tribe. (2012). Online Training and HCBS Taxonomy. Presented at: NASUAD HCBS Conference. Available: http://www​.nasuad.org​/documentation/nasuad_materials​/airs%20conference/PDFs/6​.%20NASUAD​%20I&R​%20Support%20Center​%20Services,%20Online​%20Training%20and%20HCBS%20Taxonomy.pdf.

20.

“MAX Data for 2012-2015.” (2016). Baltimore, MD: Centers for Medicare & Medicaid Services. Available: https://www​.cms.gov/Research-Statistics-Data-and-Systems​/Computer-Data-and-Systems​/MedicaidDataSourcesGenInfo​/MAXGeneralInformation. [PubMed: 26110197]

Joy Rooney, M.P.P., Carol Irvin, Ph.D., Jeral Self, Ph.D., Anna Pickrell, M.P.H., and Elizabeth Greener, Ph.D., work at Mathematica.

William Haltermann III, M.P.P., and Judith Dey, Ph.D., work in the Office of Behavioral Health, Disability, and Aging Policy in the Office of the Assistant Secretary for Planning and Evaluation.

Suggested citation:

Rooney, J., Irvin, C., Self, J., Pickrell, A., Greener, E., Haltermann, W. III & Dey, J. Identifying and Classifying Medicaid Home and Community-Based Services Claims in the Transformed Medicaid Statistical Information System, 2016-2020 (Issue Brief). Washington, DC: Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Health and Human Services. September 18, 2023.All material appearing in this report is in the public domain and may be reproduced or copied without permission; citation as to source, however, is appreciated.