Preliminary Outcome Evaluation of the Balancing Incentive Program

Printer Friendly Version in PDF Format (33 PDF pages)

ABSTRACT

This is a follow-up to four earlier evaluation reports on the Balancing Incentive Program (BIP). The BIP, legislated in the 2010 Affordable Care Act, offered states temporary enhanced federal financial participation for Medicaid home and community-based services (HCBS). Participation was limited to states that, as of 2009, were allocating less than 50% of Medicaid spending on long-term services and supports (LTSS) toward HCBS. Twenty-one states applied and were approved to participate, although three states withdrew prior to the end date of the BIP. Participating states had to commit to achieving four program goals and file quarterly reports on progress toward these goals. First, states had to commit to reaching or exceeding a benchmark for spending on HCBS (50% of all Medicaid LTSS expenditures for all but one state, which was required to meet a benchmark of at least 25% spending on HCBS). The remaining three goals concerned development and implementation of administrative infrastructure: a no wrong door/single entry point (NWD/SEP) information and referral system, a standardized needs assessment, and conflict-free case management (CFCM). Participating states could also set voluntary state-specific goals and use BIP funding not needed to achieve the mandatory goals to meet their voluntary goals. This preliminary outcomes evaluation report indicates the extent to which participating states had completed or partially implemented the BIP's required infrastructure goals as well as states' own voluntary goals through September 30, 2015 (the end date of the BIP). Also reported are the percentages of LTSS spending allocated to HCBS compared to institutional care in BIP participating and non-participating states through September 30, 2014. When 2015 expenditure figures become available, a final outcomes report will be published.

DISCLAIMER: The opinions and views expressed in this report are those of the authors. They do not necessarily reflect the views of the Department of Health and Human Services, the contractor or any other funding organization.

TABLE OF CONTENTS

ACRONYMS

EXECUTIVE SUMMARY

1. INTRODUCTION

2. DATA AND METHODS

• Share of Long-Term Services and Supports Expenditures Spent on Home and Community-Based Services
• State Infrastructure Changes
• State Discretionary Goals

3. FINDINGS

4. DISCUSSION

REFERENCES

APPENDIX A: BALANCING INCENTIVE PROGRAM EXPERIENCE IN INDIANA AND LOUISIANA

NOTES

LIST OF EXHIBITS

• EXHIBIT 1: Research Questions and Data Sources
• EXHIBIT 2A: Medicaid LTSS Expenditures and the Percentage for HCBS by States Participating in the Balancing Incentive Program, FY2009 and FY2014
• EXHIBIT 2B: Change in Medicaid LTSS Expenditures and the Percentage for HCBS by States Participating in the Balancing Incentive Program, FY2000-FY2009 and FY2009-FY2014
• EXHIBIT 2C: HCBS as Proportion of Total LTSS Spending, Overall and by Population Group, States Participating in the Balancing Incentive Program, 2009 and 2014
• EXHIBIT 3: Achievement of Infrastructure Requirements by States Participating in the Balancing Incentive Program
• EXHIBIT 4: Achievement of Sustainability and Coordination Plans by States Participating in the Balancing Incentive Program
• EXHIBIT 5: Achievement of State Discretionary Goals by States Participating in the Balancing Incentive Program
• EXHIBIT A-1: Medicaid LTSS Expenditures and the Percentage for HCBS, FY2009 and FY2014
• EXHIBIT A-2: Achievement of Infrastructure Requirements: Indiana and Louisiana

ACRONYMS

The following acronyms are mentioned in this report and appendix.

EXECUTIVE SUMMARY

Long-term services and supports (LTSS) are used by people with disabilities or chronic health conditions who need help with activities of daily living (e.g., bathing, dressing, eating) or instrumental activities of daily living (e.g., preparing meals, managing money). Historically, the financing and delivery of Medicaid LTSS has favored institutional care over home and community-based services (HCBS), despite the fact that people with disabilities generally prefer to live in their communities. The Affordable Care Act included several initiatives designed to increase the use of Medicaid HCBS and improve the infrastructure for provision of those services. States that, in 2009, were spending less than 50% of total Medicaid LTSS expenditures on HCBS were eligible to participate in the Balancing Incentive Program. Participating states were expected to increase the share of LTSS dollars spent on HCBS and to improve the LTSS infrastructure to create a more consumer-friendly, consistent, and equitable system, in exchange for which they received an enhanced federal match rate for HCBS. The rate of the enhanced federal match and the targeted rate of HCBS expenditures were dependent on the baseline spending of the state. States spending less than 25% of LTSS dollars on HCBS at baseline received a 5% enhanced rate of Federal Medical Assistance Percentage (FMAP) and were required to increase HCBS spending to at least 25% of total LTSS dollars. States spending between 25% and 50% of LTSS on HCBS at baseline received a 2% enhanced FMAP and were required to spend at least 50% of LTSS dollars on HCBS. States were required to meet these expenditure targets by September 30, 2015.

In addition to increasing spending on HCBS, states participating in the Balancing Incentive Program were required to accomplish three infrastructure improvement goals: create a no wrong door/single entry point (NWD/SEP) system for people seeking LTSS; develop a core standardized assessment to be used with all populations; and ensure a conflict-free case management (CFCM) process. Although all states were required to address the same goals, they were afforded great flexibility in the means they used to accomplish those goals. This report describes the outcomes in the participating states in achieving these infrastructure goals by the end of the Balancing Incentive Program on September 30, 2015. It also presents preliminary outcome results for the achievement of expenditure targets; final outcome results are pending availability of expenditure data, anticipated in spring 2017.

Although 21 states were accepted into the Balancing Incentive Program, this report includes data for only 18 states. Three states ended their participation early and their data are excluded from the main report text. Nebraska began participation in October 2014, but left the program by March 2015 and did not submit any quarterly reports describing activities and outcomes achieved during the brief time it was involved; it therefore is excluded from this report. Two additional states, Indiana (began participation in September 2012) and Louisiana (began participation in August 2013), also ended participation in the Balancing Incentive Program early (by December 2014). Information from those two states is excluded from the results presented in this report, but is provided as Appendix A.

Data for this report are drawn from a variety of sources. Data are presented to compare the status of states' activities at that time with their status at baseline. These data allow assessment of whether states met the required goals of the program, as well as the progress that was made when the final goals were not met. Data from the states' Balancing Incentive Program Quarterly Reports were used to determine whether the state had completed all of the components required for each Balancing Incentive Program infrastructure reform by the end of the program, September 30, 2015. Expenditure data were compiled from Truven Health Analytics reports on Medicaid LTSS expenditures for FY2009, FY2012, and FY2014 (Eiken et al., 2010; Eiken et al., 2014; Eiken et al., 2016). These data capture the achievements in rebalancing expenditures as of September 30, 2014, 1 year prior to the end of the Balancing Incentive Program, and so are preliminary only.

This outcomes evaluation identified the following preliminary results that states achieved related to the goals of the Balancing Incentive Program.

• Total HCBS expenditures as a percentage of total LTSS expenditures for states participating in the Balancing Incentive Program rose from 40.7% of LTSS in FY2009 to 52.1% of LTSS in FY2014, 1 year before the end of the Balancing Incentive Program. Half of the participating states had exceeded the target threshold by this preliminary date, and all had increased the share of LTSS spending for HCBS.

• On average, among states taking part in the Balancing Incentive Program, the percentage point increase in LTSS expenditures spent on HCBS was greater during the 5 years following baseline than during the 5 years preceding the baseline, suggesting that the program enhanced states' efforts to shift LTSS spending toward the community.

• States taking part in the Balancing Incentive Program had a greater increase in HCBS spending as a share of total LTSS expenditures than did states that were eligible but not participating in the program.

• Among states taking part in the Balancing Incentive Program, the share of LTSS spending on HCBS was much greater for people with intellectual or developmental disabilities (I/DD) than it was among older people and people with physical disabilities in all states except Mississippi, where spending on HCBS was low in both groups. This pattern was true both at baseline and 5 years later, and was observed for states not taking part in the Balancing Incentive Program, whether or not they were eligible to do so, as well as those taking part in the Balancing Incentive Program.

• The share of LTSS dollars expended for HCBS versus institutional care is higher for people with serious mental illness or severe emotional disturbance (SMI/SED) than for older adults and people with physical disabilities, but lower than for people with I/DD. Again, this same pattern was observed in states whether or not they were eligible for or participating in the Balancing Incentive Program.

• By the end of the Balancing Incentive Program, 14 of the 18 participating states had achieved all of the required infrastructure changes. States were most successful in developing protocols to ensure CFCM (achieved by 17 states). Fourteen of the 18 states were able to complete the work of creating an NWD/SEP system by the end of the program.

• A challenge score was calculated to indicate the amount of time (months) from enrollment through the end of the Balancing Incentive Program relative to the amount of work states needed to do to achieve the required goals (infrastructure and expenditure). The challenge score was only weakly correlated with the states' abilities to meet the required goals.

• Sixteen states were able to achieve requirements for developing plans for sustainability and coordination of the NWD/SEP systems with the states' Health Information Exchange Information Technology systems.

• Several states identified discretionary goals at the time of application to the program. Among the six states that had indicated a goal to expand state plan HCBS options, five states indicated progress in achieving this goal (Connecticut, Maryland, Mississippi, New York, and Texas). Among the five states that had indicated a plan to expand mental health services, four states showed progress in achieving this goal (Arkansas, Georgia, New York, and Ohio).

Overall, preliminary findings indicated that participating states made significant progress toward meeting the goals of the Balancing Incentive Program, but were not all able to achieve the required infrastructure improvement goals in the time available. Several states have requested additional time from the Centers for Medicare and Medicaid Services (CMS) to compete the required work, and CMS continues to work with states to ensure that these goals are eventually achieved. Although expenditure data are only preliminary, when compared with other eligible states, participation in the Balancing Incentive Program appears to have had a positive impact on progress toward spending goals overall and several states have achieved the required expenditure goals a year ahead of the deadline. States were not asked to address expenditure patterns by population and, indeed, states' participation in the Balancing Incentive Program did not affect the historical pattern of greater HCBS spending for people with I/DD and less spending on HCBS for older adults and people with physical disabilities. Data for people with SMI/SED were not reported for the earlier years, and so it is not possible to observe trends in expenditures patterns for that population.

1. INTRODUCTION

Long-term services and supports (LTSS) are used by people with disabilities or chronic health conditions who need help with activities of daily living (e.g., bathing, dressing, eating) or instrumental activities of daily living (e.g., preparing meals, managing money). Historically, the financing and delivery of Medicaid LTSS has favored institutional care over home and community-based services (HCBS), despite the fact that people with disabilities generally prefer to live in their communities. The Affordable Care Act (ACA) included several initiatives designed to increase the use of Medicaid HCBS and improve the infrastructure for provision of those services. States that, in 2009, were spending less than 50% of total Medicaid LTSS expenditures on HCBS were eligible to participate in the Balancing Incentive Program. Participating states were expected to increase the share of LTSS dollars spent on HCBS and to improve the LTSS infrastructure to create a more consumer-friendly, consistent, and equitable system, in exchange for which they received an enhanced federal match rate for HCBS. The rate of the enhanced federal match and the targeted rate of HCBS expenditures were dependent on the baseline spending of the state. States spending less than 25% of LTSS dollars on HCBS at baseline received a 5% enhanced rate of Federal Medical Assistance Percentage (FMAP) and were required to increase HCBS spending to at least 25% of total LTSS dollars. States spending between 25% and 50% of LTSS on HCBS at baseline received a 2% enhanced FMAP and were required to spend at least 50% of LTSS dollars on HCBS. States were required to meet these expenditure targets by September 30, 2015.

In addition to increasing spending on HCBS, states participating in the Balancing Incentive Program were required to accomplish three infrastructure improvement goals: create a no wrong door/single entry point (NWD/SEP) system for people seeking LTSS; develop a core standardized assessment (CSA) to be used with all populations; and ensure a conflict-free case management (CFCM) process. Although all states were required to address the same goals, they were afforded great flexibility in the means they used to accomplish those goals. This report describes the outcomes in the participating states in achieving these infrastructure goals by the end of the Balancing Incentive Program on September 30, 2015. It also presents preliminary outcome results for the achievement of expenditure targets; final outcome results are pending availability of expenditure data, anticipated in spring 2017.

Specifically, this report addresses the overarching question: Can a relatively small increase in the FMAP support meaningful change in the way that LTSS are provided, such that there are meaningful structural changes in the system and shifts in expenditures to better support HCBS? This analysis assesses the success that states participating in the Balancing Incentive Program had achieved in meeting the statutorily-established goals, as well as their success in achieving their state-specific goals. With regard to the required goals, we compare how well the Balancing Incentive Program states have done relative to each other and overall. This report also presents information to compare spending on HCBS among states participating in the Balancing Incentive Program with two other groups of states: (1) those that were eligible for the Balancing Incentive Program but did not participate; and (2) those that were ineligible for the Balancing Incentive Program on the basis of the share of Medicaid LTSS expenditures spent on HCBS. Expenditure data are presented as of September 2014; a final report will be prepared after data through the end of September 2015 become available (anticipated June 2017).

2. DATA AND METHODS

Data for this report are drawn from a variety of sources to represent the situation in states as of the end of the Balancing Incentive Program period (September 30, 2015) or the most current data available (for expenditures, as of September 30, 2014). Data are presented to compare the status of states' activities at that time with their status at baseline. These data allow assessment of whether states met the required goals of the Program, as well as the progress that was made when the final goals were not met. Exhibit 1 shows the specific research questions we addressed and the data sources used for each.

Although 21 states were accepted into the Balancing Incentive Program, this report includes data for only 18 states. Three states ended their participation early and their data are excluded from the main report text. Nebraska began participation in October 2014, but ended participation by March 2015 and did not submit any quarterly reports describing its activities and outcomes achieved during the brief time the state was involved. Two additional states, Indiana (started September 2012) and Louisiana (started August 2013), also ended their participation early. This report includes information for these two states as an Appendix A, but excludes them from the body of the report.

Because states began participating in the Balancing Incentive Program at various points, baseline is defined in various ways. For expenditures, baseline for all states is defined as 2009. This reflects the legislative requirement that eligibility for the Balancing Incentive Program was determined based on having spent less than 50% of state Medicaid LTSS expenditures for HCBS in 2009. Therefore, although states began participating in the Balancing Incentive Program at different times, 2009 is treated as the baseline year for assessing progress toward rebalancing of expenditures. The amount of enhanced FMAP available to states also was determined on the basis of HCBS expenditures in 2009. For purposes other than assessing expenditures, the baseline point is defined as the date of application to participate in the Balancing Incentive Program.

Share of Long-Term Services and Supports Expenditures Spent on Home and Community-Based Services

The assessment of outcomes related to patterns of spending is only preliminary at this time because expenditure data for FY2015, the final year of the Balancing Incentive Program, are not yet available; only data through FY2014 are available. This report will be updated when data for FY2015 become available. Expenditure data were compiled from Truven Health Analytics' reports on Medicaid LTSS expenditures for FY2009, FY2012, and FY2014 (Eiken et al., 2010; Eiken et al., 2014; Eiken et al., 2016). These sources primarily use data from Form CMS-64 Quarterly Expense Reports, which are audited state claims data detailing aggregate spending. CMS-64 reports are submitted to the Centers for Medicare and Medicaid Services (CMS) to determine federal matching reimbursement for each state. CMS-64 expenditure data are from the CMS Medicaid and Children's Health Insurance Program Budget Expenditure Systems.

Expenditures are reported by service category, allowing the data to be identified as HCBS or institutional LTSS expenditures. Expenditures for institutional LTSS include spending for nursing facilities, intermediate care facilities for individuals with intellectual disabilities, mental health facilities, mental health disproportionate share hospital payments, or unspecified, institutional managed LTSS. HCBS expenditures include spending for personal care, home health, Program of All-Inclusive Care for the Elderly, rehabilitative services, private duty nursing, health homes, case management, Community First Choice (1915(k) state plan option), 1915(c) waivers, 1915(i) HCBS, 1915(i) personal care and HCBS, Money Follows the Person demonstration, case management, and HCBS managed care authorities. Although data from CMS-64 reports are considered reliable, there are some limitations. Prior to FY2010, rehabilitative services, private duty nursing, managed LTSS, and HCBS under Section 1915(i) could not be identified from CMS-64 data. These services, therefore, are omitted in data from years prior to 2010.¹ RTI used these data to calculate trends in expenditures for the 5-year period prior to the baseline (2004-2009) and for the 5-year period from baseline to through 2014, the most recent year for which expenditure data are available. Data are presented overall, and by population subgroups.

State Infrastructure Changes

The baseline period for issues other than expenditures is defined on a state-specific basis. Although expenditure data share a common baseline year of 2009, the baseline for the infrastructure components refers to the situation existing in the state at the time of application. Depending on the state, this baseline period ranged from 2012 to 2014. Information for these aspects of the state infrastructure is drawn from a variety of sources, including the Balancing Incentive Program State Quarterly Reports, Balancing Incentive Program Technical Assistance profiles of each state program, and State Balancing Incentive Program Applications and Work Plans.

Data from the states' Balancing Incentive Program Quarterly reports were used to determine whether the state had completed all of the components required for each Balancing Incentive Program infrastructure reform by September 30, 2015. The quarterly reports provided information on a series of subtasks used to demonstrate state progress and implementation of the infrastructure reforms. We extracted data for each subtask to identify the percentage of each subtask completed by September 30, 2015, its completion date, and whether the states incurred any delays compared against the timetables for completion originally proposed by the states when they submitted their applications. We also compared the extent to which states completed all of their infrastructure requirements to the actual work and time they had to achieve these required infrastructure goals, as reported by the state's individual challenge score (Wiener et al., 2015).

No Wrong Door/Single Entry Point. The establishment of an NWD/SEP system is designed to make it easier for beneficiaries to access the LTSS system. To fulfill this requirement, the NWD/SEP staff coordinate completion of the functional assessment, completion of the financial eligibility assessment, final eligibility determinations, enrollment in services, and setup of supports for individuals with LTSS needs (Mission Analytics, 2013). Data from the states' Balancing Incentive Program Quarterly Reports indicated the status of 17 specific subtasks required to demonstrate achievement of this infrastructure goal. These subtasks included the following:

1. Develop standardized informational materials that NWD/SEP systems provide to individuals.

2. Train all participating staff on eligibility determination and enrollment processes.

3. Develop a detailed system design for the process to guide a person through assessment and eligibility determination (i.e., single eligibility coordinator, case management system, or otherwise coordinated process).

4. Select a vendor to develop the automated system for the process to guide a person through assessment and eligibility determination (i.e., single eligibility coordinator, case management system, or otherwise coordinated process) (the state may choose to develop the system internally).

5. Pilot implementation and testing of the process to guide a person through assessment and eligibility determination (i.e., single eligibility coordinator, case management system, or otherwise coordinated process).

6. Create process to guide a person through assessment and eligibility determination (i.e., single eligibility coordinator, case management system, or otherwise coordinated process) is implemented statewide.

7. Provide system updates for the process to guide a person through assessment and eligibility determination (i.e., single eligibility coordinator, case management system, or otherwise coordinated process).

8. Develop and implement a memorandum of understanding across the Medicaid Agency, operating agencies, and the NWD/SEPs.

9. Identify service shed coverage of all NWD/SEPs (i.e., percent of state's population covered by all NWD/SEPs).

10. Ensure that NWD/SEPs are accessible to older adults and individuals with disabilities.

11. Register a domain name for a community LTSS informational website, which provides the right to link content to a Uniform Resource Locator.

12. Develop and incorporate content for the informational website.

13. Incorporate the Level I screen into the informational website (recommended, not required).

14. Contract for a 1-800 number service.

15. Train staff on answering phones, providing information, and conducting the Level I screen.

16. Develop an advertising plan.

17. Implement an advertising plan to inform individuals of the NWD/SEP.

We reviewed each state's quarterly reports to determine whether all 17 of the required NWD/SEP subtasks were completed by September 30, 2015.

Core Standardized Assessment. The second required component of the Balancing Incentive Program was the development of a CSA to ensure that a consistent set of information is collected for all populations receiving LTSS. Although states were required to collect a core set of domains and items for all populations, the assessment instruments and process could vary across populations. However, the assessment for any given population was required to be consistent across the state--it could not vary by region or program. Successful development of a CSA included the following subtasks:

1. Develop questions for the Level I screen, a preliminary determination of likely functional and financial eligibility for the program.

2. Incorporate additional domains and topics into assessments, if necessary.

3. Train staff members at NWD/SEP systems to coordinate the CSA, to ensure that the CSA is used in a uniform manner throughout the state.

4. Identify qualified personnel to conduct the CSA.

After reviewing the status of each CSA subtask in the state quarterly reports, we reported whether the state had completed all of the required CSA components by September 30, 2015.

Conflict-Free Case Management. The third key required infrastructure improvement in the Balancing Incentive Program was the establishment of protocols to ensure CFCM, by removing or mitigating potential conflicts of interest regarding conducting assessments and developing care plans and the provision of services. Requirements for CFCM are not unique to the Balancing Incentive Program. Similar, but not identical, requirements are included as part of the Community First Choice (Section 1915(k)) provisions of the ACA and in the Medicaid Program Final Rule on State Plan Home and Community-Based Services (CMS, 2014). States were required to report on one subtask to indicate whether they had achieved the CFCM infrastructure requirement: establishment of a protocol for removing or mitigating conflict of interest. As with the other infrastructure goals, we identified whether each state had completed this subtask by September 30, 2015.

Other Outcomes. In addition to the required expenditure and infrastructure goals, states were required to develop plans for sustainability and for coordination of their NWD/SEP systems with the states' Health Information Exchange (HIE) Information Technology (IT) systems. States included information about their progress toward these goals as part of their quarterly progress reports.

State Discretionary Goals

In addition to the required goals, states had the opportunity to set goals of their own choosing at the time of their application. These typically were goals the states planned to achieve with the use of the enhanced FMAP funds. Discretionary state goals were identified from state Balancing Incentive Program applications and from a report on states' planned use of enhanced FMAP (Mission Analytics Group/Balancing Incentive Program Technical Assistance Center, 2014). These goals included eliminating waiting lists for HCBS waivers; expanding state plan HCBS programs to serve new populations and/or more individuals; expanding mental health services; increasing payment rates for HCBS; supporting transitions from institutions to the community; and improving quality measurement (Wiener et al., 2015).

States varied in the ways that they reported progress toward their discretionary goals. We identified a number of potential indicators and data sources that might be used to determine the extent to which states achieved these goals, and barriers to use of these data (Karon et al., 2015). At this time, data for most of these potential measures are not available for the time period of interest. Measures and data sources were available to address two of the discretionary outcomes: increasing access to HCBS and expanding mental health services.

Data from the National Association of States United for Aging and Disabilities (NASUAD) Medicaid Integration Tracker, Medicaid.gov, and Truven Health Analytics' LTSS expenditure reports were used to assess whether states had increased access to HCBS options by adopting certain Medicaid HCBS state plan options. Data from Truven Health Analytics Medicaid LTSS expenditure reports also were used to determine whether states had expanded access to mental health services by increasing LTSS expenditures for individuals with mental health disabilities. For each measure, we compared the information available in 2012--when most states had started to implement the Balancing Incentive Program--to the most recent data available for the particular measure. Depending on the data source, the comparison data year was either 2014 or 2016.

3. FINDINGS

Research Question 1: How successful were states in meeting the required target for the share of LTSS expenditures attributed to HCBS?

As described previously, states were eligible for the Balancing Incentive Program if, in FY2009, they spent less than 50% of LTSS dollars on HCBS. Within that requirement, there was significant variation in the amount of spending increases needed by states to reach the desired benchmarks. Exhibit 2A provides information on the baseline (FY2009) expenditures and on expenditures in FY2014, the most recent year for which data are available. Data include the total amounts spent on LTSS in each year, and the percentage of those expenditures that were for HCBS in each of those years.

Total HCBS expenditures for states participating in the Balancing Incentive Program were 40.7% of LTSS expenditures (a weighted average) in the baseline year (FY2009). This represented a state average of 39.4% of LTSS expenditures going toward HCBS, ranging from 26.0% of LTSS spending for HCBS in New Jersey to 49.1% in Maine. By the end of FY2014 (the most current time for which data are available, 1 year before the end of the Balancing Incentive Program), total HCBS expenditures accounted for 52.1% of LTSS expenditures in participating states and the average spending rate across these states was 50.2%. However, only half of the participating states have met or exceeded their required threshold as of this preliminary final report. Mississippi is excluded from these percentages because they had a lower threshold to meet.

States not participating in the Balancing Incentive Program can be split into two distinct groups: (1) those that were eligible (i.e., spent less than 50% of LTSS on HCBS in FY2009) but did not participate in the program; and (2) those that were ineligible (i.e., spent more than 50% of LTSS dollars on HCBS in FY2009). At baseline, the rate of spending on HCBS was comparable for states taking part in the program (40.7% of total expenditures, or 39.4% state average) and those that were eligible but not participating (38.6% of total expenditures, 41.9% state average). Although both sets of states increased their rate of spending on HCBS, there was a greater increase among states participating in the Balancing Incentive Program (11.4 percentage point increase to 52.1% of total expenditures, or 10.8 percentage point increase in state average spending) than in those states that were eligible but did not participate in the program (6.1 percentage point increase to 44.7% of total expenditures, 4.3 percentage point increase to 46.2% state average spending rate).

The rate of spending on HCBS also increased among those states that were ineligible because they already had met the expenditure threshold at the baseline period. Among those states, the total share of LTSS spending that went toward HCBS increased by 6.9 percentage points to 66.4%, and the average state rate of HCBS spending increased to 65.2%.

There are many factors that may contribute to the growth in HCBS as a share of LTSS expenditures. Over the 5-year period prior to the baseline year (FY2004-FY2009), states were already shifting LTSS expenditures toward HCBS. The Balancing Incentive Program was designed to encourage that movement. Exhibit 2B compares the growth in HCBS as a share of LTSS expenditures during that 5-year pre-baseline period with the 5-year period from baseline through the most recent period for which data are available.

On average, the percentage point increase was greater, among states taking part in the Balancing Incentive Program, during the 5 years following baseline than during the 5 years preceding the baseline, suggesting that the Balancing Incentive Program further spurred states' efforts to shift LTSS spending toward the community. A majority of states participating in the program (11 of 18 states) had a larger percentage point change in HCBS expenditures as proportion of LTSS spending from the period from 2009 to 2014 than they did in the pre-baseline period of 2004-2009, but this pattern was not found consistently among the states. The greatest percentage point increases from 2009 to 2014 were in Arkansas (19.8), Ohio (19.3), and Maryland (18.4). Maine had a -1.0 percentage point change in HCBS expenditures as proportion of LTSS spending from 2009 to 2014, whereas it had an 8.1 percentage point increase from 2004 to 2009.

The average total percentage point increase was greater for states participating in the Balancing Incentive Program than for other states in both periods of time. There was a 10.1 percentage point increase in the years between 2009 and 2014 for states participating in the Balancing Incentive Program compared with only 4.9 percentage points for states that were eligible but not participating, and only 2.8 percentage points increase among states that were ineligible.

Although the percentage of Medicaid LTSS spent on HCBS has been increasing overall, various population groups have had different experiences. A key outcome of interest is how the Balancing Incentive Program affected different population groups. Exhibit 2C presents the share of all LTSS spent on HCBS in total, and for three key population groups: (1) a combined group of older adults and people with physical disabilities; (2) the group of people with intellectual or developmental disabilities (I/DD); and (3) people with serious mental illness (SMI) or severe emotional disturbance (SED). Data for this latter group are available only for 2014.

Among states taking part in the Balancing Incentive Program, the share of LTSS spending on HCBS was much greater for people with I/DD than for older people and people with physical disabilities in all states except Mississippi, where spending on HCBS was low in both groups. This pattern was true both at baseline and 5 years later for states that took part in the Balancing Incentive Program and those that did not. Only among states that were ineligible for the program did the share of LTSS spending on HCBS for older adults and people with physical disabilities exceed 50%. By contrast, the share of LTSS spending on HCBS for people with I/DD exceeded 60% for all groups of states, in both time periods. Spending was also typically higher among states taking part in the Balancing Incentive Program for people with I/DD than for people with SMI or SED, but to a lesser degree. In three states (Georgia, Iowa, and Maryland), the share of spending on the I/DD and SMI/SED populations was nearly equal, and in four states (Arkansas, Illinois, Mississippi, and Ohio), it was greater for the SMI/SED population.

There was a sizeable range among states in the share of LTSS spent on HCBS for the different populations by FY2014. Among older people and adults with physical disabilities, the share of spending on HCBS ranged from 12.5% (Kentucky) to 55.1% (Texas, the only state taking part in the Balancing Incentive Program that spent greater than 50% of LTSS on HCBS for this population). For people with I/DD, spending on HCBS ranged from 20.1% (Mississippi) to 99.7% (New Hampshire). This pattern of greater spending on HCBS for people with I/DD than for older adults and people with physical disabilities is consistent with previous practices in Medicaid, and was largely unchanged by participation in the Balancing Incentive Program. For people with SMI/SED, spending on HCBS ranged from 2.9% (Connecticut) to 98.7% (Maryland).

Research Question 2: How successful were states in achieving the required infrastructure changes?

By the end of the Balancing Incentive Program, 14 of the 18 participating states had achieved all of the required infrastructure changes (Exhibit 3). Nevada had not finalized any of the structural changes for the three required goals.

Most states were successful in developing protocols to ensure CFCM. Only Nevada requested more time to achieve this goal. Unlike many states that developed mitigation strategies, Nevada sought to achieve total separation between service provision and case management. Nevada reported that it had made 80% progress toward this ambitious goal, and had requested an extension of time from CMS through November 31, 2015, to enable the state to hire and train additional care managers.

Three states (Illinois, Nevada, and Pennsylvania) completed some but not all of the required work of developing and implementing a CSA. Illinois and Nevada had completed the work of CSA development, but had not completed training staff members at the NWD/SEPs to coordinate the CSA. Illinois anticipated completing this activity by mid-2016. Nevada reported that it had completed 40% of this activity as of the end of the Balancing Incentive Program, and had requested an extension of time from CMS, through the end of 2015. Pennsylvania also sought an extension from CMS that would allow the state until February 29, 2016, to complete the task of incorporating additional domains and topics into the assessments.

Fourteen of the 18 states were able to complete development of an NWD/SEP system by the end of the Balancing Incentive Program. Illinois needed more time to complete many of the tasks associated with this goal; it anticipated completion by July 2016. Nevada requested extensions from CMS to complete many of the required tasks. Depending on the specific task (e.g., training all staff), extensions were requested for a few months or for as much as a full year.

New Jersey developed all of the required materials and trained staff, but had not completed tasks related to selection of a vendor to develop the automated system to guide an individual through the assessment and eligibility determination process, nor had it tested that process. The state sought CMS approval for an extension through the end of 2015; however, CMS considered this goal finalized because the state was able to meet the structural change requirements without the IT enhancements (personal communication with CMS, August 1, 2016). Ohio experienced delays in testing the process and having it fully implemented, with anticipated completion by September 1, 2016, to test and October 1, 2016, to implement; Ohio also experienced delays in the development and incorporation of content into a website. Pennsylvania requested extensions through the end of March 2016 to complete many of the tasks associated with this goal.

Full completion of the goals of the Balancing Incentive Program included both achieving the required infrastructure goals and the required level of spending for HCBS. By September 2014, 1 year prior to the end of the Balancing Incentive Program, nine states completed all of those goals.

States varied in how much work they needed to do to achieve the required infrastructure goals. Some states had more of the required infrastructure components in place than did others when they began participating in the Balancing Incentive Program. States also enrolled at different points in time, so that they varied in how much time they had to complete the required work. Exhibit 3 includes a challenge score, calculated to indicate the amount of time (months) from enrollment through the end of the Balancing Incentive Program relative to the amount of work states needed to do to achieve the required goals (infrastructure and expenditure) (Wiener et al., 2015). A lower challenge score indicates a greater challenge (i.e., less time available to complete more work). The challenge score was only weakly correlated with the states' abilities to meet the required goals. Six states (Georgia, Iowa, Kentucky, Massachusetts, Mississippi, and New Jersey) with challenge scores below the median value (i.e., with less time to accomplish more) had met all of the required goals. All but one of the states (Nevada) with the highest challenge scores (i.e., having the most time to accomplish what they needed) had met all of the required goals. Nevada was a particularly notable exception--it had one of the highest challenge scores (i.e., showing much time to complete few tasks) and had made progress, but needed to request an extension to complete the required goals.

In addition to the specific structural requirements, states were required to develop plans for sustainability and coordination of their NWD/SEP systems with the states' HIE IT systems. Sixteen states were able to achieve both planning requirements (Exhibit 4). Nevada and New Jersey were not able to complete plans for their NWD/SEP systems to coordinate with their HIE IT systems. Nevada requested CMS approval for an extension through the end of 2015 to complete its coordination plans. New Jersey completed all of the required NWD/SEP system changes, although they were unable to coordinate it with their HIE IT system as it was undergoing redesign; CMS considered New Jersey's changes adequate without the HIE IT coordination.

Research Question 3: How successful were states in achieving other goals that they set for themselves?

In addition to the required goals of the Balancing Incentive Program, several states identified discretionary goals, including: (1) expansion of state plan HCBS options to serve more people or new populations; and (2) expansion of mental health services.

Among the six states that had indicated at baseline that they had a goal to expand state plan HCBS options, five states indicated progress in achieving this goal (Exhibit 5). Connecticut, Maryland, Mississippi, and Texas expanded state plan HCBS options by adopting 1915(i) state plan programs. Two states (Maryland and New York) adopted health homes and three states (Maryland, New York, and Texas) also adopted 1915(k) state plan programs to expand access to state plan HCBS options. Arkansas was the only state indicating plans to expand state plan HCBS options and showed some progress toward that goal. The state was engaged with planning its 1915(i) state plan and health home programs but has since ended its progress in implementing the two programs (CMS correspondence).

Among the five states that had a plan at baseline to expand mental health services, four states showed progress in achieving this goal. Comparing the LTSS spending for individuals with mental health disabilities in 2012 (baseline year) and 2014, Illinois was the only state that had decreased LTSS spending for this population. However, LTSS spending for individuals with mental health disabilities is a limited indicator that a state has expanded mental health services, as it gives no insight into the number of individuals served or the scope of services received.

4. DISCUSSION

The Balancing Incentive Program established by the ACA was designed to help states provide a greater share of LTSS through HCBS while improving the LTSS infrastructure to create a more consumer-friendly, consistent, and equitable system. This report describes the outcomes of the work done by the participating states to achieve these goals. Findings from this outcomes evaluation are summarized below.

States increased the share of LTSS spending on HCBS. Total HCBS expenditures as a percentage of total LTSS expenditures for states participating in the Balancing Incentive Program rose from 40.7% of LTSS in FY2009 to 52.1% of LTSS in FY2014, 1 year before the end of the Balancing Incentive Program. Half of the participating states had exceeded the target threshold by this preliminary date, and all had increased the share of LTSS spending for HCBS.

States that participated in the Balancing Incentive Program had a greater increase in HCBS spending than did other states. Compared with states that were eligible for the Balancing Incentive Program but did not participate, participating states had a higher percentage point increase in HCBS spending. In total, these states also showed a greater increase in HCBS spending over time than did other states when comparing the 5 years before the baseline year to the 5 years after. A majority of states participating in the Balancing Incentive Program (11 of 18 states) had a larger percentage point change in HCBS expenditures as proportion of LTSS spending from the period of 2009-2014 than they did in the baseline period of 2004-2009. These achievements were the intention of the Balancing Incentive Program, which targeted states with HCBS spending under 50% of LTSS.

The share of LTSS expenditures spent on HCBS varied by population. In nearly all states, HCBS spending was a greater share of LTSS expenditures for people with I/DDthan for older adults or people with physical disabilities. This is true across time and regardless of participation in the Balancing Incentive Program. Spending was also typically higher among states taking part in the Balancing Incentive Program for people with I/DD than for people with SMI or SED, but to a lesser degree and with several exceptions. For example, both Georgia and Iowa had a higher share of spending on the SMI/SED population than on the I/DD population.

A majority of states achieved the required infrastructure changes. Fourteen of the 18 participating states achieved all of the required infrastructure changes, with the most success in the area of CFCM. The most common area of difficulty was completing the NWD/SEP requirements.

States' abilities to achieve the required goals seemed to have little to do with their status at baseline. A challenge score was calculated to indicate how much work states had to do to achieve the required goals, taking into account the amount of time from enrollment through the end of the Balancing Incentive Program and baseline status on each of the required goals. The challenge score was only weakly correlated with the states' abilities to meet the required goals.

Most states were able to develop plans for sustainability and coordination of their NWD/SEP systems. States were required to develop plans for sustainability and coordination of their NWD/SEP systems with the states' HIE IT systems. Sixteen states achieved both requirements and 18 totalcompleted their sustainability plans.

Most states made progress in completing discretionary goals. Five of the six states that had set discretionary goals to expand state plan HCBS options made progress in achieving this goal (Connecticut, Maryland, Mississippi, New York, and Texas). Four of the five states that planned to expand mental health services made progress in doing so (Arkansas, Georgia, New York, and Ohio).

States made significant efforts to achieve the goals of the Balancing Incentive Program, but were not always able to achieve these goals by the end of the demonstration period. CMS granted several states extensions of time to achieve the required goals and/or to continue spending enhanced FMAP funds received.

The findings reported here were primarily obtained from review of Truven Health Analytics reports on Medicaid LTSS expenditures for FY2009, FY2012, and FY2014 and states' quarterly reports. Data from state quarterly reports were somewhat limited, and it is possible that states completed more of the infrastructure requirements than those described here. Nonetheless, these preliminary outcome results indicate that participating states were generally successful in achieving the goal to increase the share of LTSS expenditures for HCBS, and made progress in developing infrastructure reforms to support the increased community-based spending. The extensions granted by CMS reflect support for states as they continue to achieve their infrastructure goals. The additional support needed may indicate the challenges states face as they shift their priorities toward HCBS for individuals with LTSS needs.

REFERENCES

Eiken, S., Sredl, K., Burwell, B., & Gold, L. (2010). Medicaid long-term care expenditures in FY 2009. Thomson Reuters. Retrieved from http://nasuad.org/hcbs/article/medicaid-long-term-care-expenditures-fy-2009.

Eiken, S., Sredl, K., Gold, L., Kasten, J., Burwell, B., & Saucier, P. (2014). Medicaid expenditures for long-term services and supports in FFY 2012. Report submitted to Centers for Medicare and Medicaid Services. Truven Health Analytics.Retrieved from https://www.medicaid.gov/medicaid/ltss/downloads/ltss-expenditures-2012.pdf.

Eiken, S., Sredl, K., Burwell, B., & Saucier, P. (2015).Medicaid Expenditures for Long-Term Services and Supports (LTSS) in FY 2013: Home and Community-Based Services were a Majority of LTSS Spending. Truven Health Analytics.

Eiken, S., Sredl, K., Burwell, B., & Saucier, P. (2016).Medicaid expenditures for long-term services and supports (LTSS) in FY 2014: Managed LTSS reached 15 percent of LTSS spending. Truven Health Analytics. Retrieved from https://www.medicaid.gov/medicaid/ltss/downloads/ltss-expenditures-2014.pdf.

George, E., Personal communication [Centers for Medicare and Medicaid Services representative]. August 1, 2016.

Karon, S., McGinn-Shapiro, M., Lyda-McDonald, B., Thach, T., & Wiener, J. (2015). Task 3 of the Evaluation of the Balancing Incentive Program: Assess Feasibility and Identify/Propose Potential Indicators and Data Resources. RTI International, report submitted to the U.S. Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation.

Wiener, J., Karon, S., McGinn-Shapiro, M., Lyda-McDonald, B., Thach, T., Justice, D., et al. (2015). Descriptive overview and summary of Balancing Incentive Program participating states at baseline. RTI International. Retrieved from https://aspe.hhs.gov/execsum/descriptive-overview-and-summary-balancing-incentive-program-participating-states-baseline.

APPENDIX A. BALANCING INCENTIVE PROGRAM EXPERIENCE IN INDIANA AND LOUISIANA

Indiana and Louisiana ended their participation in the Balancing Incentive Program early (by December 2014); as result, we have excluded data for those states from this outcomes report, and instead present the information here. Indiana began participation in September 2012 and Louisiana began August 2013. From baseline (FY2009) to FY2014, both states had modest increases in HCBS as a share of total LTSS expenditures (see Exhibit A-1). Indiana increased HCBS expenditures as share of LTSS expenditures from 30.6% in 2009 to 31.1% in 2014, and Louisiana increased from 36.4% to 39.3%. In Indiana, small increases also were seen in the share of LTSS spent on HCBS for two key populations: older people and those with physical disabilities,and people with I/DD. In Louisiana, the HCBS share of LTSS expenditures for older people and those with physical disabilities decreased slightly from 32.4% in 2009 to 30.7% in 2014 whereas the HCBS share of LTSS spending on people withI/DD increased from 46.7% to 55%. As was seen in other states, the share of LTSS expenditures used for HCBS in Indiana and Louisiana was greater for people with I/DD. In both states, that figure exceeded 50% by 2014.

By the end of their participation in the Balancing Incentive Program, neither Indiana nor Louisiana had completed work toward the required structural change goals (see Exhibit A-2). While Indiana did not complete any of the requirements, Louisiana successfully implemented a CSA tool and process, as well as CFCM. Because they ended participation early, neither state completed a sustainability plan, but they both completed a coordination plan for the NWD/SEP and HIE IT systems.

In addition to the required goals of the Balancing Incentive Program, some states identified discretionary goals. Indiana and Louisiana both included expansion of mental health services as a discretionary goal. Indiana made progress toward this goal during their time in the Balancing Incentive Program. Louisiana decreased LTSS spending for individuals with mental health disabilities from 2012 to 2014; however, spending on mental health disabilities is a limited indicator that a state has expanded mental health services, as it gives no insight into the number of individuals served or the scope of services received.

NOTES

1. The CMS-64 data which are used in this report do not necessarily capture all of the types of LTSS identified as HCBS for which states participating in the Balancing Incentive Program receive enhanced FMAP. This is particularly true in states using managed LTSS programs. Additionally, depending on the state, expenditures for rehabilitative services may include a significant amount for mental health services. Expenditures for substance use disorders are not separately reported. For a detailed discussion of CMS-64 limitations, see Eiken et al., 2014.

EVALUATION OF THE BALANCING INCENTIVES PROGRAM

This report was prepared under contract #HHSP23320100021WI between the U.S. Department of Health and Human Services (HHS), Office of Disability, Aging and Long-Term Care Policy (DALTCP) and the Research Triangle Institute. For additional information about this subject, you can visit the DALTCP home page at http://aspe.hhs.gov/office-disability-aging-and-long-term-care-policy-daltcp or contact the ASPE Project Officers, Pamela Doty and Jhamirah Howard, at HHS/ASPE/DALTCP, Room 424E, H.H. Humphrey Building, 200 Independence Avenue, S.W., Washington, D.C. 20201; Pamela.Doty@hhs.gov and Jhamirah.Howard@hhs.gov.

Reports Available

CASE STUDIES OF BALANCING INCENTIVE PROGRAM IMPLEMENTATION PROCESS

• HTML:  https://aspe.hhs.gov/basic-report/case-studies-balancing-incentive-program-implementation-process
• PDF:  https://aspe.hhs.gov/pdf-report/case-studies-balancing-incentive-program-implementation-process

DESCRIPTIVE OVERVIEW AND SUMMARY OF BALANCING INCENTIVE PROGRAM PARTICIPATING STATES AT BASELINE

• HTML:  http://aspe.hhs.gov/basic-report/descriptive-overview-and-summary-balancing-incentive-program-participating-states-baseline
• PDF:  http://aspe.hhs.gov/pdf-report/descriptive-overview-and-summary-balancing-incentive-program-participating-states-baseline

FINAL PROCESS EVALUATION OF THE BALANCING INCENTIVE PROGRAM

• HTML:  https://aspe.hhs.gov/basic-report/final-process-evlauation-balancing-incentive-program
• PDF:  https://aspe.hhs.gov/pdf-report/final-process-evlauation-balancing-incentive-program

PRELIMINARY OUTCOME EVALUATION OF THE BALANCING INCENTIVE PROGRAM

• HTML:  https://aspe.hhs.gov/basic-report/preliminary-outcome-evaluation-balancing-incentive-program
• PDF:  https://aspe.hhs.gov/pdf-report/preliminary-outcome-evaluation-balancing-incentive-program

PRELIMINARY PROCESS EVALUATION OF THE BALANCING INCENTIVE PROGRAM

• HTML:  https://aspe.hhs.gov/basic-report/preliminary-process-evaluation-balancing-incentive-program
• PDF:  https://aspe.hhs.gov/pdf-report/preliminary-process-evaluation-balancing-incentive-program