United States Department of Justice
Interim Report to the State of Oregon

Integration of Community Mental Health and
Compliance with Title II of the Americans with
Disabilities Act
January 2, 2014

 I.

Background

The United States Department of Justice initially opened an investigation of conditions at
the Oregon State Hospital under the Civil Rights of Institutionalized Persons Act in 2006. In late
2010, we expanded our focus to look at Oregon’s broader mental health system pursuant to our
authority under the Americans with Disabilities Act (“ADA”). This expanded investigation
included examining the community services and supports available to persons with mental illness
throughout Oregon.
Shortly after we began our expanded investigation, Oregon announced a transformation
of its healthcare system to include integration of the systems for delivering physical and mental
health care, expanding the number of individuals and services covered under the Oregon Health
Plan, and ensuring improved quality of services through an outcome-driven system. The
transformation is designed to shift financial incentives from acute care to prevention, wellness,
and community-based management of chronic conditions – which is exactly the point of our
ADA investigation, in which we are determining whether Oregon’s mental health care is being
provided in the most integrated setting.
Oregon leadership and the Department agreed that this health transformation process – if
it includes a focus on helping people with serious and persistent mental illness achieve positive
outcomes through the provision of critical community services – provided a unique opportunity,
through a cooperative process, to ensure that Oregon meets that integration mandate of the ADA
as interpreted by Olmstead v. L.C., 527 U.S. 581 (1999).
In order to resolve our investigation, the Department and the State agreed to a process to
ensure that the health transformation initiative brought about changes to the Oregon mental
health system necessary to ensure that people were not placed at a risk of unnecessary
segregation. The agreement included a three-part process. In the first year, largely, although not
exclusively through the managed care provider contracts, the State would collect data that would
permit the mapping of services and outcomes. This was essential to the reform because
statewide data was largely unavailable and the State did not have information essential for the
planning process.
Once the data was collected and the systems understood, the parties agreed to discuss
intermediate metrics to be placed in provider contracts. These would include a minimum array
of services necessary to meet the known needs. We agreed that this was a necessary step to
transition to a system driven solely by outcome measures.
In the final phase, we would work together to develop outcome measures that would be
guided by individual and community health. These would be included in the provider contracts
and would not measure the quantity of each service, but rather set standards to measure whether
the systems are supporting people in the community and preventing unnecessary segregation.
As provided by our November 9, 2012 agreement, Oregon has submitted quarterly data
and information to the Department regarding system development and outcomes regarding
community mental health provisions. We also met with State officials in late March 2013 and
1

 have had various discussions throughout the year regarding the data provided by the State. This
report addresses our initial analysis of that data.
II.

Executive Summary
Several themes emerge from the data provided by the State throughout this past year:
First, and foremost, Oregon appears to have made only limited progress over the past
three quarters in decreasing the use of restrictive institutional settings, decreasing the
rate of readmission to these institutional settings, decreasing the average time
confined in these institutional settings, and decreasing the service dollars spent on
these institutional settings.
Second, while Oregon data reports somewhat of an array of mental health services,
the data reveals a lack of adequate high-intensity services like Assertive Community
Treatment (“ACT”), and critical supports for housing and employment.
Third, the data suggests that the high quantity of services that Oregon is reporting in
some parts of the State do not meet evidence-based models for quality.
Finally, the data is inconclusive whether community services are being appropriately
distributed and allocated statewide, particularly throughout the more heavily
populated centers in the State.

The data we have reviewed clearly shows that Oregon’s community services have not
begun to expand. That has led to a predictable result – Oregon is not yet transitioning to a
community-based system. The State’s data reveals that funding for mental health services is not
yet shifting away from costly restrictive institutional settings to more effective, less costly
community settings.
The State has long been aware of the gaps in its community mental health system. In
2008 and 2009, the Legislature and the Department of Mental Health funded reports from
various workgroups and consultants that document the gaps in the system. 1 For years, the
Department has repeatedly raised this same concern with Oregon leadership. We were assured
that the State was committed to moving to a community-based system of providing mental health
care.

1

Public Consulting Group, Oregon Department of Human Services Additions and Mental
Health Division, Assessment and Evaluation of the Mental Health Care Delivery System in
Oregon, November 2008; Public Consulting Group, Oregon Department of Human Services
Additions and Mental Health Division, Assessment and evaluation of the adult mental health
system in Oregon, March 12, 2009; Oregon Department of Human Services Community Services
Workgroup Report, A Complement to the Master Plan Phase II Report on the Replacement of the
Oregon State Hospital, March 2009.
2

 Certainly, serving individuals with highly complex mental health needs is best done
through a person-centered approach delivered through an array of effective evidence-based
community services. Furthermore, this approach best preserves State general funds and is more
effective than institution-based services. While State officials have expressed their desire to
create a community-based mental health care delivery system, this is not yet happening. We
look forward to working with the State to now redirect its focus and efforts to make communitybased mental health services a reality in Oregon.
III.

Analysis of Data
A. Current Service Provision of Oregon Community-Based Mental Health Services

Despite the stated commitment to transform to a community-based system, the data
provided demonstrates that there has not been an increase in the provision of community-based
mental health services. As the State develops community services, we would expect to see an
increase in the amount of community-based mental health services (service units) being provided
and a decrease in the amount of institution-based services. This expected increase is not evident.
Indeed, as demonstrated in Chart 1, the quantity provided for most community services has
essentially stayed the same during 2013, and in the case of supported housing, has actually
decreased.
Chart 1 - Service units per adult with a serious and persistent mental illness
provided by Coordinated Care Organizations, per 1,000 clients
8,000
6,888

7,000
6,000

6,568

5,774

5,000
4,000
3,000

2,890
2,514
2,477

2,000

2,729
3,011

2,643

2,161

1,906

1,587

1,000
0

609
228

April 2013 Report

594
294

July 2013 Report

588
333

Oct 2013 Report

Case management

Peer support

Supported employment

Psych-education and living skills training

Supported housing services

Supportive housing services

3

 Similarly, despite Oregon’s stated goal to shift from costly institution-based services to
less costly preventative services, this transformation is not yet occurring. Funding has not yet
shifted away from costly restrictive inpatient settings to less costly, less restrictive community
settings. In fact, the percentage of spending on care for adults with a serious and persistent
mental illness in more restrictive living settings has increased by about 5% over the two quarters
of data provided, as noted in Chart 2. 2 The data provided clearly demonstrates that
approximately 74% of all service dollars spent by Oregon for adults with a serious and persistent
mental illness is going toward restrictive, institution-based settings. Moreover, it is worth noting
that three quarters of the budgeted service dollars are being spent on less than 2% of Oregon
adults with a serious and persistent mental illness.
Chart 2 - Percent of all service dollars for adults with a serious and persistent
mental illness that are used for care, by reported setting 3
35%

31.6%

30.0%

30%

23.6%

25%
20.8%

20%
15%
10%

8.1%
3.6% 3.9%

5%
0%

8.0%

6.6%

0%

1.7%

0%

April 2013 Report

7.0%

3.3% 3.8% 3.9%

Oct 2013 Report

Supported Housing

Supportive Housing

Adult Foster Home

Residential Treatment Home

Residential Treatment Facility

Secure Residential Treatment Facility

Inpatient Hospital

State Hospital

The above care settings are defined herein. 4
2

During calendar year 2013, as evidenced in Chart 2, service dollars spent on care of
adults in institutional settings increased from 69% to74% according to data provided by the State
for System Development Measure 9, items D-H. The reported data demonstrates that the State
continues to spend a vast majority of its service dollars for adults with serious and persistent
mental illness on restrictive settings, including the state hospital and smaller secure settings.
3

Chart 2 includes only spending for people in select settings and, therefore, does not sum
to 100%. The chart does not include spending for people who received services in other settings,
such as those living in their own home, those living with family, those living with friends, and
those who are homeless, for example.
4

 Significantly, as the State itself recognizes, the movement from an institution-based
system and a subsequent reinvestment into a community-based system will result in major
savings. This is because the cost of providing care at the state hospital is significantly higher
than the cost of providing individualized services in the community. For example, with an
estimated cost of care at Oregon State Hospital of $945 per day, 5 the annual cost of a state
hospital stay is approximately $344,925 per person. Given that care provided by the state
hospital is generally not reimbursable through Medicaid or other insurance, these costs have
largely been borne by the State of Oregon.
In contrast, providing services to individuals in the community would not only lower
cost, but would draw down federal dollars through Medicaid or other programs. For example, a
national estimate of the average annual total cost of effectively supporting an individual in the
community with Assertive Community Treatment services is about $15,000. 6 Taking into
account that ACT services are reimbursable through Medicaid and draw federal dollars, the costs
to Oregon are only approximately $5,634 per person annually. The table below summarizes
annual projections comparing inpatient hospitalization with intensive community services.

4

“Supported housing” as defined in the November 9, 2012 agreement, refers to permanent,
scattered-site housing for no more than two people in a given apartment or house where housing,
support, and individualized care services are provided as an integrated package. “Supportive
housing” as defined in the November 9, 2012 agreement, refers to permanent housing in which
tenants who live together in a single building or complex of buildings receive support services.
“Residential Treatment Facility” is a state-licensed live-in facility that provides services on a 24hour basis for six or more residents with serious and persistent mental illness. “Secure
Residential Treatment Facility” is a state-licensed locked facility that provides services on a 24hour basis for six or more residents with serious and persistent mental illness. By regulation, a
resident is restricted from exiting the facility or its grounds through the use of approved locking
devices on resident exit doors, gates, or other closures for the stated purpose of protecting the
public. “Residential Treatment Home” is a group home for five or fewer residents with serious
and persistent mental illness where services are provided on a 24-hour basis. “Adult Foster
Home” is a state-licensed group home in which residential care is provided to five or fewer
adults. By regulation, Adult Foster Homes are required to have written posted house rules
“regarding hours, visitors, use of tobacco and alcohol, meal times, use of telephones and kitchen,
monthly charges and services to be provided and policies on refunds in case of departure,
hospitalization or death.”
5

Oregon Health Authority, Oregon State Hospital Cost of Care,
http://www.oregon.gov/oha/amh/osh/Pages/cost-of-care.aspx.
6

National Alliance on Mental Illness, Assertive Community Treatment: Investment Yield
Outcomes, September 2007.
5

 Chart 3 – Institutional costs compared to community services
Annual Cost Basis

Institutional

Hospital Days:
Oregon State Hospital
($945 daily rate)

Total Annual Cost
Average
stay (256
days)

Full year
cost

$241,920

$344,925

Federal
Medicaid

$0

Net State Cost
Average
stay (256
days)

Full year
cost

$241,920

$344,925

Community
ACT @ $15,000

$15,000

$9,366

$5,634
$236,286 to $339,361

Range of Annual Savings/Person:

B. Consumer Outcomes
The State’s data raises concerns that Oregon continues to rely heavily on institutional
settings for persons with mental illness and is not yet providing an adequate array or volume of
services in the community. Assessing the data Oregon provided shows a trend of relatively
unchanged use of restrictive living placements, relatively unchanged lengths of stay in these
settings, and relatively unchanged rates of readmission to institutional settings. These “flat line”
data trends are shown in Charts 4-7 below, and all are poor consumer outcomes.
Chart 4 - Number of adults with an identified serious and persistent mental illness
in restrictive settings
1000
900

817

825

825

788

771

762

497

496

474

265

257

257

193

191

208

800
700
600
500
400
300
200
100
0
April 2013 Report
State Hospital
Residential Treatment Facility
Adult Foster Home

July 2013 Report
Oct 2013 Report
Secure Residential Treatment Facility
Residential Treatment Home

6

 Chart 5 - Average admission rate for adults with serious and
persistent mental illness, per 1,000 clients

10.00
9.00

8.6

8.0

8.00

8.0

7.00
6.00
5.00
4.00
3.00
2.00

2.0 2.0
1.2

1.8

2.0

1.5

1.7

1.2

1.5 1.7 1.4 1.7

1.00
0.00
April 2013 Report

July 2013 Report

State Hospital
Residential Treatment Facility
Adult Foster Home

Oct 2013 Report

Secure Residential Treatment Facility
Residential Treatment Home

Chart 6 - Average length of stay for adults with serious and persistent
mental illness, by setting
685.1

700.0
650.0

605.7

597.6

600.0
550.0
Days

500.0
450.0

512.4

506.9

472.2
440.4
423.9

400.0
350.0
300.0

300.3

262.5

250.0
200.0

308.1

263.6
258.6

256.0

July 2013 Report

Oct 2013 Report

234.0

April 2013 Report
State Hospital

Secure Residential Treatment Facility

Residential Treatment Facility

Residential Treatment Home

Adult Foster Home

7

 Chart 7 - Readmission rates for inpatient psychiatric care for adults
with serious and persistent mental illness
35.0%

30.6%

29.6%

30.0%
25.0%
20.0%

19.2%

18.1%
15.4%

17.7%

15.0%

30-day
180-day

10.0%
5.0%
0.0%
April 2013 Report

July 2013 Report

Oct 2013 Report

Charts 4-7 reflect that mental health services in Oregon have not shifted away from
restrictive settings to less restrictive settings in accordance with the integration mandate of the
Americans with Disabilities Act, as interpreted by the Supreme Court in Olmstead v. L.C., 527
U.S. 581 (1999). Moreover, the data provided by the State reflects that services in the
community are not sufficient to keep individuals out of restrictive settings.
In addition, the State’s data (Chart 8) shows only a minimal decline in the number of
consumers seeking emergency room services. Again, if the State were following through on its
stated goal of investing in preventative community-based services, we would expect to see the
number of emergency room visits for individuals with mental illness start trending significantly
downward. As evidenced by the data, this has not happened. Furthermore, while the number of
total emergency room visits may appear to have decreased somewhat, the number of consumers
using community-based crisis stabilization services (such as mobile crisis teams and walk-in
crisis centers) has not increased as shown by the State’s data for System Development Measure
3, evidencing a lack of intended outcomes.
Chart 8 - Number of emergency room visits for adults with a
with a mental health diagnosis
3,000
2,500

2,572

2,562

April 2013 Report

July 2013 Report

2,000

2,428

1,500
1,000
500
-

8

Oct 2013 Report

 IV.

Utilization of Community-Based Mental Health Services by Adults with a Serious
and Persistent Mental Illness
A. Data Integrity

Chart 9, below, summarizes the key community-based services that Oregon reports
providing to people with serious and persistent mental illness, either through its counties or
through managed care organizations – predominantly the fledgling Coordinated Care
Organizations, but to a lesser extent Mental Health Organizations and fee-for-service providers.
An analysis of the data provided calls into question both the accuracy of the data and suggests
that the high quantity of services that Oregon is reporting in some parts of the state do not meet
evidence-based models for quality.
Chart 9 - Oregon community-based mental health services: Number of adults with a
serious and persistent mental illness served and number of contracted providers

Communitybased mental
health
service
Crisis hotline
Mobile crisis
teams
Walkin/drop-off
crisis centers
Crisis
apartments /
respite
Short-term
crisis
stabilization
units

# of adults
with a
serious
and
persistent
mental
illness
served by
counties

# of adults
with a serious
and persistent
mental illness
served by
# of managed
# of county
managed care
care
subcontractors organizations
subcontractors
Community-Based Crisis Services 7

Total # of
adults
with a
serious
and
persistent
mental
illness
served

Total # of
subcontractors

37,931

47

37,931

47

5,579

45

5,579

45

3,803

38

3,803

38

150

27

150

27

574

25

574

25

CCOs partner with the counties,
who typically run the crisis
system, according to Oregon’s
“Data Table.”

7

It is also likely the data includes individuals who are not diagnosed with serious and
persistent mental illness, given that the counties do not appear to have a system for filtering who
qualifies as serious and persistent mental illness versus someone diagnosed with serious mental
illness or someone without mental illness. Additionally, the State is unable to provide
unduplicated counts for these services.
9

 Communitybased mental
health service
Assertive
Community
Treatment
Intensive Case
Management
Case
Management
Peer Support
Supported
employment
Psych-education
and living skills
training
Assessment
Supported
housing services
(DOJ definition)
Supportive
housing services
Own Home
Total Adults
Served

Community-Based Mental Health Services and Supports
# of adults
with a
serious and
persistent
# of adults
mental
with a serious
illness
and persistent
served by
mental illness
managed
# of managed
served by
# of county
care
care
counties
subcontractors organizations subcontractors
169

Total # of
adults
with a
serious
and
persistent
mental
illness
served

Total # of
subcontractors

17

349

24

518

41

41

310

289

2,920

330

39

5,553
404

289
62

5,553
813

289
101

248

27

534

36

779

63

692

35

2,366

62

3,058

97

564

10,113

636

2,610
Data not provided for
counties
409

3,330

72
6,783
Community-Based Housing

460

0

460

873

0

873
21,419
22,752

In conversations with Oregon since this data was provided, State officials were confident
that the information provided is accurate. The data regarding Intensive Case Management
illustrates why some of the data may be flawed. If the information provided is accurate, Oregon
provides this service to almost 3,000 individuals – the vast majority of whom receive Intensive
Case Management through the counties. The counties are reported to have 41 subcontractors –
which the State asserted represents 41 teams – providing services to 2,610 individuals. The
managed care organizations are reported to have 289 subcontractors or 289 teams, serving 310
individuals. Thus, the counties have an average case load of 64:1 which is more than 60 times
greater than the caseload provided by the managed care organizations for the exact same service.
Although it is likely the discrepancy arises from how managed care organizations count the
number of subcontractors providing ICM, the data shows an obvious discrepancy.

10

 In addition, the data reflects the inadequacy of the totality of services currently being
provided. Given the total number of adults with a serious and persistent mental illness being
served in the community, the level of services that they are receiving is extremely low. For
example, we would expect that most of these individuals would receive some type of case
management or care coordination. The data reveals, however, that only 41% of these individuals
received case management or care coordination services.
B. Current Service Caseloads
Setting aside the accuracy issue, the current service population data illustrates a critical
point: many service teams in Oregon report caseloads dramatically lower than what we have
seen in states using evidence-based models.
The Intensive Case Management data also highlights this point, as shown in Chart 10,
below. Looking at just the county-level data and excluding the managed care data, Oregon
county Intensive Case Management teams serve caseloads of roughly 64 people per team with
caseloads varying from 0 people per team to 527 people per team.
Chart 10 - Comparison of Oregon Intensive Case Management team caseloads,
by county

County
Gilliam
Grant
Harney
Morrow
Wheeler
Clatsop
Columbia
Lake
Wallowa
Clackamas
Union
Coos
Sherman
Tillamook
Yamhill
Deschutes
Hood River
Multnomah
Josephine
Jackson

Number of Intensive Case
Management teams
1
1
1
1
1
1
1
1
1
3
1
1
1
1
1
1
1
4
1
1

Unique number of people with serious
and persistent mental illness served
0
0
0
0
0
1
1
3
3
12
4
5
5
5
5
7
12
55
18
20
11

Average
caseload per
team
0.00
0.00
0.00
0.00
0.00
1.00
1.00
3.00
3.00
4.00
4.00
5.00
5.00
5.00
5.00
7.00
12.00
13.75
18.00
20.00

 Number of Intensive Case
County
Management teams
Douglas
1
Baker
1
Lane
2
Wasco
1
Benton
1
Linn
1
Jefferson
1
Polk
2
Washington
3
Lincoln
1
Marion
2

Unique number of people with serious
and persistent mental illness served
29
33
73
51
65
86
135
274
440
214
1,054

Average
caseload per
team
29.00
33.00
36.50
51.00
65.00
86.00
135.00
137.00
146.67
214.00
527.00

As shown in Chart 11, below, the ACT team caseloads are similarly low and reflect
substantial variation among ACT providers. These caseload numbers are not consistent with
full-fidelity ACT services, which typically serve a caseload of 75 to 100 people per ACT team.
In Oregon, ACT teams serve, on average, 10 people. 8 When broken down by provider, though,
the caseloads vary wildly. The data also raises the question of whether there are sufficient ACT
services in the State. For instance, Multnomah County ACT team serves only 25 of the roughly
13,173 estimated adults with a serious and persistent mental illness in the county. 9 By any
measure, this is woefully inadequate as researchers have estimated that ACT teams should be
provided to between 20-40% of those with serious mental illness. 10 Given that consumers with
serious and persistent mental illness present greater needs than people with serious mental illness
do, that number may need to be even higher, particularly when there has previously been a high
rate of institutionalization.

8

Assuming full- fidelity, these lower caseloads would result in extremely high costs per
individual – way beyond what we would expect for full-fidelity, evidence-based ACT teams.
9

This includes only individuals served by the county ACT team, not those served by
Coordinated Care Organization ACT teams covering Multnomah County. Two Coordinated
Care Organizations – Family Care and Health Share – cover Multnomah County. These
Coordinated Care Organizations reported having one ACT team and three ACT teams,
respectively. These teams – which also partially cover Clackamas, Marion, and Washington
Counties – serve 78 people with ACT. This means even if all of the ACT services provided by
these Coordinated Care Organizations were actually provided in Multnomah County, at most
only 103 people would be receiving ACT services in the Portland area.
10

Gary R. Bond, et al., Assertive Community Treatment for People with Severe Mental
Illness: Critical Ingredients and Impact on Patients, Dis Manage Health Outcomes, 141-159
(2001); Gary S. Cuddeback, et al., How Many Assertive Community Treatment Teams Do We
Need?, Psychiatric Services, 1803-1806 (December 2006).
12

 Chart 11 - Comparison of Oregon ACT team caseloads, by provider

County or CCO
County: Columbia
County: Douglas
County: Gilliam
County: Morrow
County: Wheeler
County: Union
County: Wallowa
CCO: FamilyCare
CCO: Western Oregon Advanced
Health
County: Lane
CCO: PrimaryHealth Josephine
County: Josephine
CCO: Columbia Pacific
CCO: Yamhill County Care
County: Hood River
CCO: Intercommunity Health
Network
CCO: Eastern OR
County: Benton
CCO: PacificSource
CCO: Trillium
County: Deschutes
CCO: All Care
County: Washington
County: Multnomah
CCO: Health Share
County: Wasco
County: Yamhill

Number of
ACT teams
1
1
1
1
1
1
1
1

Unique number of people
with a serious and persistent
mental illness served
0
0
0
0
0
1
1
1

Average caseload
per ACT team
0.00
0.00
0.00
0.00
0.00
1.00
1.00
1.00

1
2
2
1
3
2
1

1
5
8
5
18
12
8

1.00
2.50
4.00
5.00
6.00
6.00
8.00

2
3
1
2
4
1
1
1
1
3
1
1

16
33
12
30
71
19
20
22
25
77
27
44

8.00
11.00
12.00
15.00
17.75
19.00
20.00
22.00
25.00
25.67
27.00
44.00

C. Geographic Distribution of Current Services
Given the data provided by the State to date, we do not have enough information to map
precisely where services are concentrated. While we have information about what the counties
provide, the complication arises for the services provided by Coordinated Care Organizations.
Several Coordinated Care Organizations cover multiple counties, and several counties are
covered by multiple Coordinated Care Organizations. In places where overlap exists, we do not

13

 have a way to accurately overlay where services are available. 11 The State should amend its data
collection process to permit this analysis to be done. It will be essential to its ability to address
gaps in services and successfully move to an outcome-based system.
V.

Conclusion

Although the accuracy of the data provided by Oregon over the past year is concerning,
our review of intended outcomes does show substantial gaps in the community mental health
services being provided. Despite our expectation that changes in the service system would help
reduce Oregon’s reliance on institutionalization, that key outcome for consumers has not taken
hold. Likewise, key consumer outcomes in general have not appeared to improve over the past
three quarters. Further, despite data reporting a high quantity of some services, the data suggests
those services likely do not meet the quality of evidence-based models for care. Oregon is far
behind where it needs to be in providing the high-intensity community services that are most
necessary for serving people with serious mental illness in the most appropriate integrated
settings.

11

We have reached out to the State about this issue. Oregon informed us that it expected to
provide us this breakdown by January 1, 2014, but we have not yet received this information.
Receiving this information will assist in conducting a mapping of services throughout Oregon.
14