Report to the President and Congress
Medicaid Home and Community-Based Alternatives to
Psychiatric Residential Treatment Facilities Demonstration
As Required by the
Deficit Reduction Act of 2005 (P.L. 109-171)
From the
Department of Health and Human Services
Office of the Secretary

Kathleen Sebelius
Secretary of Health and Human Services
July 2013

 Executive Summary
In 2005, Congress authorized a five-year Demonstration project to test whether children and
youth who meet the requirements to be served in a psychiatric residential treatment facility
(PRTF) could successfully and cost effectively be served in the community.
The Centers for Medicare & Medicaid Services (CMS) selected ten states to compare effective
ways of providing home and community-based services (HCBS) as an alternative to care in
PRTFs for children and youth with serious emotional disturbances enrolled in the states’
Medicaid program. The Demonstration served children and youth who were either “diverted”
from being served in a PRTF, or were “transitioned” from a PRTF into the community, often
earlier than would have been possible without the Demonstration. By the fourth year of the
Demonstration, the states provided community-based alternatives to institutional treatment to
over 4,000 children and youth. Using a “systems of care” approach, the Demonstration
successfully enabled participants to either improve or maintain their functioning status at less
than a third of the cost of serving them in an institution.
The functional status for the children and youth, diverted or transitioned, was measured in the
domains of mental health, juvenile justice, school functioning, alcohol and other drug use and
social support. All participants improved or maintained their functional status in these domains.
However, the children and youth with the highest levels of need consistently showed improved
mental health status, less frequent interaction with law enforcement, better performance in
school, reductions in substance abuse and better relationships with peers and family throughout
periods measured during the project.
Background
Over the last decade, PRTFs have become a major provider for children and youth with mental
illness and serious emotional disorders requiring an institutional level of care. PRTFs are not
recognized as hospitals, nursing facilities, or intermediate care facilities for individuals with
intellectual disabilities under the Medicaid statute. Therefore, states have been unable to use the
1915(c) waiver authority to provide home and community-based alternatives to institutional care,
which would keep children and youth in their homes and with their families or in the community.
In July 2003, the New Freedom Commission on Mental Health released a report, Achieving the
Promise: Transforming Mental HealthCare in America, which outlined significant barriers to
providing community-based services for children and youth with serious emotional disturbances
as an alternative to placing them in PRTFs. Children, youth, and families typically have little
influence over decisions affecting service delivery, planning, and the use of financing to deliver
care. When comprehensive community-based options are unavailable, some children and youth
may end up incarcerated in the juvenile justice system, institutionalized for long periods, or in
the care of the child welfare system. To address this problem, the Commission recommended
that CMS conduct a Medicaid Demonstration project to test community-based alternatives to
institutional care.

Report to Congress on Community-Based Alternatives
to Psychiatric Residential Treatment Facilities Demonstration

Page 1

 In response to the statutory barriers to providing Medicaid HCBS to children and youth with
serious emotional disturbances, the Medicaid Home and Community-Based Alternatives to
Psychiatric Residential Treatment Facilities Medicaid Demonstration project was authorized by
the Congress in section 6063 of the Deficit Reduction Act of 2005 (DRA, P.L. 109-171). The
Demonstration project allowed up to ten states to compare effective ways of providing care for
children and youth enrolled in the state’s Medicaid program. Specifically, the demonstration
tested the effectiveness of HCBS as an alternative to care in PRTFs. To eliminate the statutory
barrier to providing HCBS to children and youth who meet a PRTF institutional level of care, for
purposes of the Demonstration, PRTFs were deemed facilities as specified in section 1915(c) of
the Social Security Act so participating states could receive federal Medicaid matching funds for
these services.
The project targeted children and youth who might not have otherwise been eligible for
Medicaid-funded, intensive community-based services and supports available through a 1915(c)
waiver authority. CMS awarded $217 million to ten states, with each state receiving between
$15 million and $50 million each over the grant period (FY 2007 through FY 2011) and the
remaining $1 million was awarded for the evaluation of the demonstration. Each participating
state was required to provide non-federal Medicaid matching funds. One of the ten states,
Florida, did not continue in the Demonstration after the first year due to difficulty securing the
non-federal matching funds necessary to implement the program. The nine fully participating
states were Alaska, Georgia, Indiana, Kansas, Maryland, Mississippi, Montana, South Carolina,
and Virginia. The development of home and community-based services in these states was built
on historical funding to create systems of care from the Substance Abuse and Mental Health
Services Administration (SAMHSA) Children’s Mental Health Initiative (CMHI) grant program.
An independent national evaluation of the Medicaid Home and Community-Based Alternatives
to Psychiatric Residential Treatment Facilities Demonstration was conducted by IMPAQ
International and submitted to CMS in May of 2012. Data and lessons learned from this
Demonstration and a national evaluation of SAMHSA’s Systems of Care grant program were
used to issue the recent SAMHSA-CMS Informational Bulletin, “Coverage of Behavioral Health
Services for Children, Youth and Young Adults with Significant Mental Health Conditions.”

Design of the Evaluation
As part of the Demonstration, Section 6063 of the DRA required an evaluation. For purposes of
this evaluation, CMS constructed two specific questions:
Question One: Did the Demonstration services result in the maintenance of, or improvement in,
a child’s or youth’s functional status?
Question Two: Was it cost-effective to provide coverage of home and community-based services
as an alternative to psychiatric residential treatment for children and youth enrolled in the
Demonstration?

Report to Congress on Community-Based Alternatives
to Psychiatric Residential Treatment Facilities Demonstration

Page 2

 The evaluation addressed these two questions, as well as additional questions introduced by
CMS to help identify successful strategies and the subpopulations for which those strategies are
most effective. The evaluation was designed to develop reliable cost and utilization data to
evaluate the effectiveness of community-based service-delivery models.
Summary Findings
The evidence from the evaluation yields the following results:
Question One: Did the Demonstration services result in the maintenance of, or improvement in,
a child’s or youth’s functional status?
Findings: Overall, the project successfully enabled children and youth to either maintain or
improve their functional status. The common theme across all states is that children and youth
with the highest level of need at baseline benefited the most from participating in the
Demonstration. These participants showed the most
improvement in the following areas: decreased juvenile
justice involvement, increased school functioning,
decreased alcohol and other drug use and increased social
Major Findings
support, over the most follow-up periods. The findings
1. Children and youth
also indicate that children and youth that were transitioned
generally maintained or
out of PRTFs had better outcomes on average than
improved their functional
children who were diverted from PRTFs.
status when receiving
services in the community.
Question Two: Was it cost effective to provide coverage
of home and community-based services as an alternative
2. Waiver services cost
to psychiatric residential treatment for children and youth
about a third (32 percent)
enrolled in the Demonstration?
of comparable services
provided in PRTFs.
Findings: For all nine states over the first three
Demonstration years for which cost data was available to
be collected, there was an average savings of 68 percent.
In other words, the waiver services cost only 32 percent of
comparable services provided in PRTFs.
The Demonstration proved cost effective and consistently maintained or improved functional
status on average for all enrolled children and youth. As discovered through satisfaction surveys,
it is encouraging that enrollees and their families liked the outcomes of the Demonstration and
their involvement in the treatment, as well as other aspects of the Demonstration.
State Demonstration Enrollment Profiles
The Demonstration enabled states to use the 1915(c) waiver program for the first time and
receive Medicaid reimbursement to serve a population of children and youth with serious
emotional disturbances in their homes and communities, rather than in PRTFs. With any

Report to Congress on Community-Based Alternatives
to Psychiatric Residential Treatment Facilities Demonstration

Page 3

 Demonstration, challenges are expected. The nine states, to varying degrees, experienced initial
delays in enrollment due to factors such as:
•
•
•
•
•
•

Recruiting providers of home and community based services;
Training staff to address the complex needs of the children who would have typically
been served in an institutional setting;
Educating families to assure them that their children could safely be served in the
community with the appropriate planning and intensive services;
Developing partnerships among the child-serving systems that may not have been
accustomed to working together;
Addressing the requirements associated with receiving Medicaid reimbursement
including programming for Medicaid Management Information Systems (MMIS); and,
Implementing state regulatory changes as needed.

The initial slow enrollment patterns changed and actual enrollment grew over the four waiver
years included in this report. The growth of state provider networks and geographic expansion of
the project helped spur enrollment.
Demonstration Project Growth
Year
2008
Children and
253
Youth Served

2009

2010

2011

978

1300

4000

Mississippi and Indiana had the largest number of children enrolled in the project since its
inception, together accounting for over half of the total children and youth enrolled in the
Demonstration. Some of Mississippi’s early success lies in its education and outreach initiatives.
Indiana’s success can be attributed both to an existing workforce trained in the wraparound
practice model within a system of care framework and a team of statewide staff dedicated to
working only on the Demonstration project.
Male children were more likely to be enrolled in the Demonstration than female children. The
project served children across different age groups, the highest percent being youth between the
ages of 15 and 18.
Final Analytical Sample by Gender and Age -All States
Key Individual
Gender
Characteristics
N
%

Age

Male

Female

<6

6 to 11

12 to 14

15 to 18

2036
63.7

1162
36.3

26
0.8

904
28.3

996
31.1

1206
37.7

Report to Congress on Community-Based Alternatives
to Psychiatric Residential Treatment Facilities Demonstration

>18 Unknown

21
0.7

45
1.4

Page 4

 Findings on Assessment-Based on Functional Outcomes
One of the key evaluation questions was whether there was an improvement or maintenance of
the children’s functional status. To address this issue, it was important to assess changes in the
level of functioning of the children and youth in the Demonstration by focusing on several
outcomes that are common across grantees. All the common outcome measures reflect changes
in five selected domains: mental health, juvenile justice, school functioning, alcohol and other
drug use, and social support.
The grantees used one of three instruments to assess changes in children’s behavioral and mental
health functional outcomes: the Child and Adolescent Needs and Strengths (CANS), the Child &
Adolescent Functional Assessment Scale (CAFAS) and the Child Behavioral Checklist (CBCL).
Detailed information about each of the instruments can be found in Appendix A. Each of these
instruments collects a set of outcome measures that relates to the functional domains under
review. Functional outcomes data from these instruments and other Demonstration waiverspecific measures were collected to form a Minimum Data Set (MDS). These data were then
used to generate an overall picture of children and youth’s functional outcome changes after
project participation. The MDS collected data at 6-month intervals and disenrollment, which
enabled the evaluation – at each follow-up point – to analyze a child or youth’s change in
functioning from baseline.
Although the findings vary by domain, most children showed improvements for most domains
and most follow-up periods. The functional improvements, which reflect changes in level of
need (LON) rather than simple changes in the absolute score for a particular instrument, indicate
a substantial likelihood of improvement on average for children at most LON and regardless of
their admission status (diversion or transition). These findings are quite positive and reflect the
need for a more permanent format for HCBS mental health programs for children and youth.
The findings also provide rich information on what domains are more susceptible to change and
which groups of children are likely to see the most positive changes in functioning as a result of
the Demonstration. More detailed findings indicate that the Demonstration had particularly
positive effects on mental health, family functioning, and alcohol and other drug use. The
common theme across all state grantees is that children and youth with the highest LON at
baseline benefited the most from participating in the Demonstration. These children showed the
most improvement, across the most domains, and over the most follow-up periods.
Children and Youth with Highest Level of Need (LON)
Duration in Demonstration Waiver
Functional Status
Domains
6 Months
12 Months
18 Months
Maintained
Maintained
Improved
Mental Health
Improved
Improved
Improved
Juvenile Justice
Maintained
Improved
Improved
School Functioning
Improved
Improved
Improved
Alcohol & Other Drug Use
Improved
Improved
Maintained
Social Support
Findings Related to Cost Effectiveness
Report to Congress on Community-Based Alternatives
to Psychiatric Residential Treatment Facilities Demonstration

Page 5

 In response to question two, as a result of the Demonstration, all participating states have seen
significant savings in the costs of caring for children and youth with severe emotional disorders,
although the extent of the savings varies by state. The three states with the largest number of
participants (Indiana, Kansas, and Mississippi) had average savings of $20,000 to $30,000, close
to 45 percent savings over comparable PRTF services. Initial outreach/education in Mississippi
and access to a trained and existing provider networks in Indiana and Kansas may explain why
these states experience higher cost than the 9 states on average. The ability to provide and
expand services rapidly to meet the needs of the participants in a more comprehensive manner
could explain overall high community based costs.
Cost effectiveness of the Demonstration was evaluated by comparing each state’s fiscal year’s
expenditures on HCBS services provided under the Demonstration to data on PRTF expenditures
that states submitted annually. This information provided data on expenditures by participant, by
service, per waiver year. To calculate the average per capita cost, the data submission included
the number of users per service and the total number of unduplicated waiver participants for
which claims were paid. These calculations were made by using the three years of cost data
available for the evaluation.
Across the first year of the Demonstration and through year three of the evaluation, all states
taken together had Demonstration costs around 32 percent of the average per capita total
Medicaid costs for services in PRTFs, an average per capita savings of $36,500 to $40,000
across the states. Demonstration costs increased as a proportion of the average per capita total
Medicaid costs for services in institutions; however, all the Demonstration projects remained
cost effective. The rising trend may be due to more Demonstration waiver services being added
as states solidified their project.

Waiver Year

Average Per Person
1915(c) Cost

Average Per Person
PRTF Cost

1915(c) Cost as a
Percentage of
PRTF Cost

WY 1

$9,792

$42,343

22%

WY 2

$12,244

$55,783

28%

WY 3

$23,122

$79,452

32%

Average Years
$15,869
$55,107
32%
1 to 3 (1)
Source: Department of Health and Human Services, Centers for Medicare & Medicaid Services: Medical Assistance
Expenditures by Type of Service for 1915(c) HCBS Waiver, CMS MOD-PRTF DEMO 372 report: (1) Waiver years
(WY) 1 to 3 include all expenditures for an estimated 2,820 children, however, expenditures for the average years
row, excludes Waiver and Medicaid costs for Grantees with less than 5 children per waiver year because of the
likelihood of a biased estimate due to the small number of observations. Estimates for this row are based on 2,808
children.

Conclusion

Report to Congress on Community-Based Alternatives
to Psychiatric Residential Treatment Facilities Demonstration

Page 6

 The Demonstration has easily met cost effectiveness tests and on average has consistently
maintained or improved functional status for all children and youth. In addition, as discovered
through satisfaction surveys, families and children liked the outcomes of the Demonstration and
their involvement in the treatment, as well as other Demonstration aspects.
States have reacted positively to the improved outcomes seen for children and youth
participating in the Demonstration and seven states have received approval for “bridge” 1915(c)
waivers. These bridge waivers permit states to continue services to the children and youth who
started receiving services under the Demonstration but not to additional children or youth
because of the statutory barrier to providing 1915(c) waiver services to children and youth
meeting a PRTF level of care. The seven states operating bridge waivers have indicated that they
would be interested in applying for a 1915(c) waiver if the statute permitted a waiver for this
purpose. Three states are using or considering the use of the 1915(i) state plan authority which
allows children and youth to receive services in the community without the institutional level of
care as long as the participants are at or below 150 percent of the Federal Poverty Level (FPL).
However under this authority, the only way to increase the income limits would be for the
children or youth to meet the same institutional level of care required for 1915(c) waivers.
Income level restrictions, statewide requirements and the inability under the authority to regulate
the number of participants in the program make this alternative less attractive to states.
The findings highlight the positive benefits of the project and the desire of states to sustain the
waiver beyond the Demonstration period. The improved outcomes and positive reactions to the
Demonstration may have increased the involvement of the participating children, youth and
families, which is likely to have made the project even more successful in program adherence
and behavior modification. In order for the objectives of the Demonstration to continue, the
statutory barrier to providing HCBS to children and youth who meet a PRTF institutional level
of care criteria would have to be eliminated by identifying PRTFs as a designated institution
under section 1915(c) of the Social Security Act so states could receive federal Medicaid
matching funds for waiver services.
For more detailed information regarding findings described in this report, please see “The
National Evaluation of the Medicaid Demonstration Waiver Home- and Community-based
Alternatives to Psychiatric Residential Treatment Facilities Final Evaluation Report” that can be
found on the Medicaid.gov website: http://www.medicaid.gov/Medicaid-CHIP-ProgramInformation/By-Topics/Delivery-Systems/Institutional-Care/Alternatives-to-PsychiatricResidential-Treatment-Facilities-Demonstration-PRTF.html .

Report to Congress on Community-Based Alternatives
to Psychiatric Residential Treatment Facilities Demonstration

Page 7

 Appendix A:
Functional Assessment Instruments
The CANS assessment refers to a group of outcome management tools developed by John Lyons
(Lyons, 2009) together with many stakeholders across multiple states. The CANS instrument is
used in Indiana, Maryland, Mississippi, and Virginia, which together covered more children in
the Demonstration waiver than were covered by the remaining states using the other two
instruments.
The CANS was developed to assess the strengths and needs of children and youth who have
emotional and behavioral disorders, and to aid in the development of treatment plans to guide
service delivery. The core domains of the CANS Comprehensive Multisystem Assessment are
life functioning, child strengths, acculturation, caregiver strengths, caregiver needs, child
behavioral/emotional needs, and child risk behaviors. Extension modules are triggered by core
questions and include developmental disability, health, sexuality, adoption, trauma, substance
use, violence, juvenile justice, fire setting, and psychotropic medication. Specific items or
questions are the same across all versions.
Each CANS item has four levels of assessment and each level translates into separate needs and
strengths assessments. The basic scoring metric for CANS items is 0 through 3. In the case of
needs assessment, a score of 0 indicates no evidence of need, while a score of 3 indicates that
immediate/intensive action is required. In the case of strength assessments, 0 reflects a
centerpiece strength while 3 shows no strength identified.
The CBCL (Achenbach & Rescorla, 2001) is an extensively used parent-report questionnaire that
allows clinicians and researchers to assess a wide range of behavior problems and competencies
in children and youth. The CBCL functional assessment instrument is used in Kansas, Montana,
and South Carolina.
The CBCL uses T scores to sort subjects into three groups: in the normal range, on the border
line, or in the clinical range. These clinical categories would have been ideal for developing the
low, middle, and high needs categories (as done in the CANS), while offering enhanced clinical
implications of the results. However, due to the lack of T scores in the Demonstration waiver’s
MDS, in determining children and youth’s baseline needs categories the cut-off points were
based on the raw score. In particular, the profile of competence/syndrome score sheet in the
CBCL Manual was used to identify the cut-off points closest to those by T scores to approximate
the clinical categorization. This enhances the clinical implications of cut-off points by raw scores
and, thus, analysis results.
The CAFAS (Hodges 1990, 1994) is an inventory for measuring functional impairment in
children and adolescents originally designed for use in a mental health policy research project.
The CAFAS inventory used in the Demonstration waiver consists of five child scales: Role
Performance, Thinking, Behavior toward Self and Others, Mood/Emotions, and Substance
Abuse, as well as two child caregiver scales: Basic Needs, and Family Social Support. The
CAFAS scale is used in Alaska, Georgia, and Kansas.

Report to Congress on Community-Based Alternatives
to Psychiatric Residential Treatment Facilities Demonstration

Page 8