Special study, Department of Health Serivces, Division of Behavioral Health Services, administrative entity system

              SPECIAL STUDY

DEPARTMENT OF HEALTH SERVICES

DIVISION OF BEHAVIORAL HEALTH SERVICES

ADMINISTRATIVE ENTITY SYSTEM

Report to the Arizona Legislature

By the Auditor General

January 1992

92- 1

DOUGLAS R. NORTON, CPA

AUDIT- R GENERAL

STATE OF ARIZONA

OFFICE OF THE

AUDITOR GENERAL

January 27, 1992

Members of the Arizona Legislature

The Honorable F i f e Symington, Governor

Ms. Alethea Caldwell, Director

Department of Heaith Services

Transmitted herewith i s a report of the Auditor General, A Special Study

of the Department of Health Services, Behavioral Health Administrative

Entity System. This report i s in response to a May 8, 1991, resolution

of the Joint Legislative Oversight Committee.

The report contains a limited financial review of three administrative

e n t i t i e s , plus a review of the adequacy of the Department's contracts

with the e n t i t i e s . The report also contains information on a number of

program issues including a c c e s s i b i l i t y o f services, services for special

populations and case management. In addition, the report addresses the

eight questions s p e c i f i c a l l y posed in the Joint Legislative Oversight

Committee resolution.

My s t a f f and I w i l l be pleased to discuss or c l a r i f y items in the report.

This report w i l l be released to the public on Tuesday, January 28, 1992.

DRN : l mn

S- in cerely,

Do glas R. Norton

Au M orGeneral

2700 NORTH CENTRAL AVENUE SUITE 700 . PHOENIX, ARIZONA 85004 ' ( 602) 255- 4385 FAX ( 602) 255- 1 251

SUMMARY

The Office of the Auditor General has completed a special study of the

behavioral health administrative entity system within the Department of

Health Services ( DHS). This study was conducted in response to a

May 8, 1991, resolution of the Joint Legislative Oversight Comnittee.

In directing this study, the Comnittee identified eight areas of

interest, including the use of State behavioral health funds by the

entities and provider agencies, the accessibility of services to the

seriously mental ly i I I ( SMI), the cost efficiency of the entity system,

and the adequacy of management systems. Due to the complexity of the

questions and time limitations, we were unable to thoroughly address a l l

eight areas of interest. Therefore, the scope of the study was limited

to those issues of greatest concern.

Financial Review Of Administrative

Entities ( see pages 11 through 19)

We conducted a limited review of three administrative e n t i t i e s '

expenditures for adult SMI services for fiscal year 1990- 91. These three

entities received approximately $ 35 million in State funding, or 62

percent of a l l adult SMI funds received by the entities from the

Department of Health Services.

While the entities contracted most of the funds they received to provider

agencies, a l l three entities have accumulated significant balances of

unexpended funds that are unrestricted. Unrestricted fund balances for

the three entities we reviewed totaled over $ 10 million as of June 30,

1991. While soma unrestricted funds were expended on behavioral health

programs in fiscal year 1990- 91, these funds were also expended on goods

or services, such as employee bonuses, and food and entertainment, that

are not directly related to the provision of behavioral health services.

In addition, we attempted to estimate entity administrative costs. These

costs are d i f f i c u l t to determine because reporting formats d i f f e r and a

uniform method of classifying administrative costs has not been developed

by Dl%. Nonetheless, entity administrative costa ranged from 10 to 12

percent of the funds expended in fiscal year 1990- 91. We estimate that,

at most, 72 to 79 percent of the funds were expended on direct services

when both entity and provider administrative costs are considered.

Contract Monitorina and

Provisiong ( see pages 21 through 26)

We reviewed the Department's efforts to address deficiencies in contract

moni tor ing and the contract provisions identified in our previous

performance audit of the Department of Health Services, Division of

Behavioral Health ( Report No. 89- 10). The Department's recovery of

monies owed the State by entities continues to be weak, ineffective, and

untimely. In addition, the Department is performing very l i t t l e fiscal

monitoring of the entities at this time.

We also found that the Department's entity contracts for fiscal year

1991- 92 contain many of the same problems we identified in our previous

report. For example, contracts do not specify target populations, and

they lack enforcement provisions. In addition, contracts do not address

other concerns, such as ownership of real property and equipment and the

disposition of interest earnings.

The Behavioral Health Management Information

System ( BHMIS) Has Failed To Meet The Neec&

Of DHS. The Administ mtive Entities. And Service

Providers ( see pages 27 through 34)

Since f i sca l year 1987- 88, DHS has expended over $ 4 mi I l ion design i ng ,

developing, maintaining, and supporting the Behavioral Health Management

Informat ion System ( M I S ) . We found that despite this substantial

comnitrnent of resources, M I S has failed to meet the needs of the

Department and i t s users. The Department intends to use M I S in the

future only for program informational purposes. A separate system w i l l

be deve l oped to hand l e con t rac t payments .

Accessibility And Availabilitv Of

Entitv Servicej ( see paget 45 through 56)

We attempted to determine the length of time required for seriously

mentally ill adults to access the entity system and begin receiving

services. Due to the lack of adequate data, we were able to determine

the length of time for only three of the five enti ties subject to our

study. Although the length of time varies significantly among and within

the three entities, on average clients waited from 21 to 52 days to

receive psychiatric services.

Our analysis of the clients most in need of services indicates that the

entities are improving delivery of services to these clients. For

example, our analysis of SMI adults discharged from the Maricopa County

Jail system revealed that most are either already enrolled or are

successfully accessing community- based services through the entity

system. However, the j a i l population appears especially vulnerable to

becoming " lost" in the referral process and not obtaining the services

they need.

Lack of comnunity services impacts institutional discharges in some areas

of the State. We found that many patients at the Arizona State Hospital

cannot be released when they are c l i n i c a l l y ready for discharge because

there i s no place for them to go.

Our review also found that there are few services specifically targeted

to meet the needs of the homeless SMI population. Outreach services are

limited. Some entities have no outreach programs, and funding for these

programs has decl ined over the past several years. However, DHS and

other comnunity organizations have recently begun to focus planning

efforts on the problems of homeless SMI adults.

Case Manaaement

( see paget 59 thtough 68)

The Arizona comnuni ty- based mental health system focuses on case

management as the mechanism for ensuring clients receive the services

they need, and that services are coordinated and appropriate to the

c l i e n t ' s changing needs over time. We found that although caseload size

varies, overall, caseloads are large, which reduces the case manager's

a b i l i t y to provide adequate individual attention to clients. For the

five ent i ties we reviewed, caseloads averaged 43 c l i ents per case

manager; one case manager had 83 c l ients. Large caseloads prohibit case

managers from spending adequate time with clients. Additional case

managers and fund i ng w i l l be needed to reduce case loads to the l eve Is

that w i l l be required by a court- ordered plan, which DHS is attempting to

implement.

Other needs of the SMI population that are not being adequately met

include a lack of available residential services, dental care, and other

services. Under the court- ordered plan, DHS w i l l have to increase

accessibi l i t y of a l l needed services.

Other States' Proclrams

( see pages 71 through 75)

We were asked to compare Arizona's administrative entity system with

mental health service delivery systems in other states. We found that

Arizona's administrative entity system is unique among the states we

surveyed, and that structures for delivering and paying for services for

the seriously mental ly i l l vary widely from state to state. In addition,

each state is unique in the way i t provides case management services,

targets populations with special needs, and controls the expenditure of

state funds.

TABLE OF CONTEN-CHAPTER

I: FINANCIAL RRllEl OF

AWlNlSTRATlVE ENTITIES. . . . . . . . . . . . . . . . . . .

Methodology . . . . . . . . . . . . . . . . . . . . . . . . .

Results of Financial Review. . . . . . . . . . . . . . . . .

ADAPT, Inc.. . . . . . . . . . . . . . . . . . . . . . . . .

Comnunity Care Network, Inc. . . . . . . . . . . . . . . . .

Questionable Expenditures From

Unrestricted Funds . . . . . . . . . . . . . . . . . . . . .

Administrative Costs May Be Understated. . . . . . . . . . .

Recomnendations. . . . . . . . . . . . . . . . . . . . . . .

CHAPTER I I: CWMCT U) NITORING

AH) PROVlSl O N S . . . . . . . . . . . . . . . . . . . . . . .

Previous Audit Findings. . . . . . . . . . . . . . . . . . .

Monitoring Continues To Be Weak. . . . . . . . . . . . . . .

Contract Provisions Not

Substantial ly Changed. . . . . . . . . . . . . . . . . . . .

Recomnendations. . . . . . . . . . . . . . . . . . . . . . .

CWTER I l l : ME BEHAVlORM HULTH MANAGBENT

INMAYATION S Y m ( HIS) HAS FAILED TO

YET THE NEEDS OF DHS, THE MYlNlSlRATlVE

EWTITIES, UY) SERVICE PROVIDERS. . . . . . . . . . . . . . .

Background . . . . . . . . . . . . . . . . . . . . . . . . .

Despite Comni tment Of Over $ 4 Mi l l ion

BHMlS Does Not Meet The Needs Of The

Behavioral Health System . . . . . . . . . . . . . . . . . .

Several Factors Have Contributed To

W I S 1 Failure To Meet Its Users Needs . . . . . . . . . . .

WS Is In The Process Of Upgrading W I S ;

However , Fundamen ta l Quest i ons Conce r n i ng

The System Still Need To Be Addressed. . . . . . . . . . . .

TABLE OF CONTENTS ( Con' t )

CHAPTER IV: ACCESSIBILITY OF SERVICES. . . . . . . . . . . . . .

Results By Entity. . . . . . . . . . . . . . . . . . . . . .

CODAMA Services . . . . . . . . . . . . . . . . . . . .

Community CareNetwork, Inc.. . . . . . . . . . . . . .

SEABHS . . . . . . . . . . . . . . . . . . . . . . . . .

EVBHA . . . . . . . . . . . . . . . . . . . . . . . . .

ADAPT, Inc. . . . . . . . . . . . . . . . . . . . . . .

CHAPTER V: ACCESSIBILITY OF SEJWICES TO SPECIAL POPULATIONS . . . . . . . . . . . . . . . . . . .

Methodology . . . . . . . . . . . . . . . . . . . . . . . . .

The J a i l SMI Population

And The Entity System. . . . . . . . . . . . . . . . . . . .

The ASH Popu l at i on And The Entity System. . . . . . . . . . . . . . . . . . . .

County Annex Pat i ents AndThe Entity System. . . . . . . . . . . . . . . . . . . .

CHAPTER VI: SERVICES H) R THE HaELES SY1 POPULATION . . . . . . . . . . . . . . . . . . . . . . .

Se rv i ces For The Home l ess SMI Population Are Limited . . . . . . . . . . . . . . . . .

Steps Are Being Taken To

Increase Services. . . . . . . . . . . . . . . . . . . . . .

CHAmER V I I : CASE WNAlrFWr . . . . . . . . . . . . . . . . . .

Case Management Is An Essential Part Of The System . . . . . . . . . . . . . . . .

Large Case loads Restrict

Timewith Clients. . . . . . . . . . . . . . . . . . . . . .

Lower Caseload Ratios W i l l

Require A Substantial Increase

In Funding . . . . . . . . . . . . . . . . . . . . . . . .

TABLE OF CONTENTS ( Con ' t )

CHAPTER V I I I : ARIZONA LACKS SOYE NEEDED

SERVICES FOR THE SERIOUSLY WNTALLY ILL. . . . . . . . . . .

Residential Services Are

Lacking In Arizona . . . . . . . . . . . . . . . . . . . . .

Other Services Are AlsoLacking . . . . . . . . . . . . . . . . . . . . . . . .

CHAPTER I X : OTHER STATES' PROCrWllS FOR THE

SERIOUSLY ENTALLY ILL . . . . . . . . . . . . . . . . . . .

Methodology . . . . . . . . . . . . . . . . . . . . . . . . .

Structure Of StateSystems. . . . . . . . . . . . . . . . . . . . . . . .

Case Management. . . . . . . . . . . . . . . . . . . . . . .

I)' Contracting Practices. . . . . . . . . . . . . . . . . . . .

CHAPTER X: ANSlERS TO LEGISLATIVE WESTIONS. . . . . . . . . . .

ILLUSTRATIONS

FlGLRE 1 Division Of Behavioral Health

Administrative Entities' Service Areas . . . . . . 2

TABLE 1 DEPARTMENT OF HEALTH SERV l CES

Schedule Of Revenues And Expend i tures

For The Seriously Mentally Ill

Fiscal Year 1990- 91 ( unaudited). . . . . . . . . . . 4

TABLE 2 ADAPT, Inc.

Schedule Of Revenues And Expenditures

For The Seriously Mentally Ill

Year Ended June 30, 1991 ( unaud i ted) . . . . . . . 12

TABLE 3 CODAMA Services

Schedule Of Revenues And Expenditures

For The Seriously Mental ly I I I

Year Ended June 30, 1991 ( unaud i ted) . . . . . . . 14

ILLUSTRATIONS ( Con ' t )

Paw

TABLE 4 COMUUN I TY CARE NETWORK

Schedule Of Revenues And Expenditures

For The Seriously Mentally Ill

Year Ended June 30, 1991 ( unaudi ted) . . . . . . .

TABLE 5 Fund Balances Of Entities

June 30, 1991 ( unaud i t ed) . . . . . . . . . . . . . 16

TABLE 6 Comparison Of Services On Registered

Client Forms And BHMlS Data Files

February 1991 - July 1991. . . . . . . . . . . . . 32

TABLE 7 Case Manager To Client Ratio

By Agency As Of September 1991 . . . . . . . . . . 61

INTRODUCTION AND BACKGROUND

The Office of the Auditor General has conducted a special review of the

Department of Health Services ( DHS), Division of Behavioral Health, and

the administrative entity system. This study was conducted in response

to a May 8, 1991, resolution of the Joint Legislative Oversight Committee

and under the authority vested in the Auditor General by Arizona Revised

Statutes 541- 2353.

Structure And Fundina Of

Mental Health Service

Deliverv Svstem

DHS currently provides comnunity- based mental health services through the

administrative entity system. The Division of Behavioral Health is

responsible for providing behavioral health services to those most in

need. To f u l f i l l this mandate, DHS contracts with private, nonprofit

organizations called administrative entities. There are presently eight

administrative entities. Each entity is responsible for administering,

coordinating, and monitoring comnunity- based behavioral health services

in a specific region of the State. ( See Figure 1, page 2 for a map of

the administrative entities and the region served by each entity.) In

turn, the administrative entities contract with other agencies to provide

direct services. Further, each entity is responsible for the ongoing

development and implementation of a case management system.(')

( 1) A case management system consists of a clinical team of psychiatrists, social workers,

case managers, and other professionals. This team i s responsible for developing an

individual treatnent plan for each client i n the administrative entity system and

ensuring continuous client treatment and care.

FIGURE 1

DIVISION OF BEHAVIORAL HEALTH

ADMINISTRATIVE ENTITIES' SERVICE AREAS

Cornnun i ty Organ i zat ion for Drug Abuse, Mental Heal th and

Alcoholism Services, Inc./ Comnunity Care Network/ East

Val ley Behavioral Hea l t h Assoc i at ion ( Mar i copa County )

ADAPT, Inc. ( Pim County)

a Behavioral Health Services of Yuma

( La Paz and Yuma Count i es)

a Northern Arizona Comprehensive Guidance Center ( Apache,

Cocon i no, Uohave , Navajo , and Yavapa i Count i es )

Pinal Gi la Behavioral Health Association, Inc. ( Gi la and

Pinal Counties)

a Southeastern Arizona Behavioral Health Services, Inc.

( Cochise, Graham, Greenlee, and Santa Cruz Counties)

Administrative entities are responsible for five general program areas:

serious mental illness, substance abuse, childrens' services, domestic

violence, and general mental health. This report focuses exclusively on

services for the seriously mentally i l l .

In recent years the court has participated in the development and

oversight of the delivery system for the seriously mentally i l l ( SMI).

In 1981, the Arizona Center for Law in the Public Interest f i l e d suit

( Arnold vs. Sarn) on behalf of five chronically mentally i ll people. The

center sued the Department of Health Services, the Arizona State Hospital

( ASH), and the Maricopa County Board of Supervisors alleging that the

State and County failed to provide these people with adequate c m u n i t y

mental health services. The court ruled in favor of the p l a i n t i f f s and

the decision was appealed to the Arizona Supreme Court. The Arizona

Supreme Court upheld the ruling, stating that both the State and Maricopa

County have mandatory duties to provide the f u l l continuum of services to

a l l seriously mentally ill people who could reasonably benefit from them.

In the spring of 1991, the parties reached an agreement to f u l f i l l the

requirements of the court orders. The implementation provisions of this

agreement are contained in The B l u e ~ r i n t : Implementina Services to thg

Seriouslv Mental lv I II. The purpose of the blueprint i s to ensure that

by September 30, 1995, a comprehensive c m u n i t y mental health system for

the SMI population i s established. The blueprint specifies the types and

number of services that should be made available to comply with the court

order and, therefore, directs the establishment as well as the

continuation of services. The blueprint also calls for a court monitor

to oversee and act as mediator in implementing the terms of the court

order.

Services for the seriously mentally ill population are largely State

funded. Most of the funding appropriated for behavioral health services

for the seriously mentally ill i s passed through to the administrative

en t i t i es to cont ract for comnun i ty- based serv i ces . I n f i sca I year

1990- 91, the Legislature appropriated approximately $ 49 mi l I ion for

services to the SMl population. In addition, DHS received another $ 8

million from other sources. Table 1, page 4 shows the Department's

revenues and expenditures for services for the seriously mentally i l l

during fiscal year 1990- 91.

TABLE 1

DEPARTMENT OF HEALTH SEPVICES

Schedule Of Revenues And Expenditur3s For The

Seriously Mentally Ill

Fiscal Year 1990- 91

( unaudited)

Revenues for SMI

Total appropriations for SMI

Pima County funds

Other funds for SMI

Total funds for SMI $ 57,460,987

Expenditures for outside organizations

ADAPT, Inc. 14,071,885

Comnunity Organization for Drug Abuse,

Mental Health, and Alcoholism 10,898,956

Comnuni ty Care Network 9,829,416

Northern Arizona Community Guidance Center 6,061,133

East Valley Behavioral Health Association 3,911,937

South Eastern Arizona Behavioral Health

Association 2,376,097

Pinal Gila Behavioral Health Association 1 ,926,528

Behavioral Health Services of Yuma 1 ,880,942

Other organizations 5.469.261

Total expenditures

for outside organizations

Administrative costs 459.498

Total expenditures 56,885,653

Excess of revenues over expenditures $ 575.334

( a) Excludes appropriatron for operation of the Arizona State Hospital .

Source: Department of Health Services, Financial On- Line System reports

for the fiscal year ended June 30, 1991.

General Conclusions And

Recommendations Of The Study

The Joint Legislative Oversight Committee ( JLOC) authorized the Auditor

General to perform a review of the administrative entity structure,

including related contract mechanisms and information systems. In i t s

resolution authorizing the study, JLOC outlined eight questions to be

addressed and limited the review to five of the eight entities.(') ( See

Chapter X for a brief response to the eight questions.) To answer as

many of the questions or portions of the questions as possible, audit

work was organized into four general topic areas. The information

compiled in these four areas is included in this report in ten chapters.

Presented be low are the general conclusions and reconmendat ions, i f

applicable.

Cha~ ters I throuah Ill: Financial And DHS Operatlon~

Audit work in these chapters focused on three areas: a limited

review of three administrative e n t i t i e s ' expenditures for adult SMI

services for fiscal year 1990- 91; DHS' e f f o r t s t o address

deficiencies in contract monitoring and contract provisions; and the

Behavioral Health Management lnformat ion System ( BHMIS).

We found that the Department has insufficient control over the use of

contracted monies and the delivery of services. Continued weaknesses

in contract provisions allow the e n t i t i e s t o accumulate unexpended

funds and f a i l t o r e s t r i c t how these funds and the interest earned on

them are to be used. As of June 30, 1991, collectively, ADAPT,

CODAMA, and CCN had over $ 10 mi l l ion i n unexpended funds . A I so, the

Department does not define administrative costs and how these costs

should be classified. Consequently, reported administrative costs

may understate actual costs. We estimate less than 80 percent of

State SM1 monies received by entities are expended on direct services.

The Department's contract monitoring continues to be weak and

ineffective. In addition, problems with M I S l i m i t the use of this

information to assist in monitoring contracts and reconciling

paymen t 8 .

To address these concerns, the Department must strengthen and c l a r i f y

contract provisions to address the use of unexpended funds and define

administrative costs. In addition, DHS needs to strengthen i t s

monitoring of administrative entities. Finally, the purpose of M I S

needs to be determined and efforts made to ensure the quality and

timeliness of data on the system.

( 1) The five enti ties included i n the review are ADAPT, CCN, C O W , EVBHA, and SEABHS.

Cha~ ters IV throush VI: Accessibility Of Services To SMls And SukxK> ulation~

An analysis of services for the SMI population in general and

specific subpopulations indicates that most clients are able to

access services. However, the iength of time i t takes for each

client varies by entity and the severity of the c l i e n t ' s illness.

Some clients must wait over a month to receive i n i t i a l services. in

addition, some SMI persons are lost in the referral process and do

not obtain the services they need.

For the homeless SMI population, i t appears services are particularly

lacking. The present number of residential programs f a l l s far short

of the number needed and the number required by the blueprint. In

addition, funding for outreach services has decreased in recent years.

Efforts are currently underway to improve the a v a i l a b i l i t y and

accessibility of services. Coordination among DHS, the entities,

jai Is, and hospitals is improving and is helping SMI persons to make

a timely transition i n t o comnuni ty- based services. DHS and comnuni ty

service groups have also focused on the needs of homeless SMI

people. However, increased services w i l l likely mean additional

funding.

Cha~ ters VII and VIII: Needed Services

Millions of dollars w i l l be needed to meet the blueprint requirements

for case management and residential services. Currently, case

managers carry an average caseload of 43 clients. The blueprint

l i m i t s caseloads to 25 clients or less per caseworker. Funding for

case management salaries in Maricopa County alone would have to

increase almost $ 8 mi l lion annual ly to provide the estimated number

of case managers needed by 1995.

To meet the blueprint requirements for residential services, several

thousand additional beds w i l l be needed. In addition, several other

types of services, such as day treatment programs, vocational and

supported work programs, and mobi le c r i s i s stabi l ization teams, w i l l

need to be expanded to meet blueprint projections.

Chanters IX & X: Miscellaneous Issues

These chapters present a comparison of other states' programs, and

the answers to the eight questions outlined by the resolution.

Audit

& nd I s t l t l a p

Based on time limitations and legislative interest, the study focuses on

the delivery of services to the seriously mentally ill population. The

fragile nature of this population makes i t particularly sensitive to

problems with the a v a i l a b i l i t y and accessibility of services. Persons

with serious mental illnesses are often unstable; many are

low- functioning and have d i f f i c u l t y locating services. This population

is defined by statute as those who, as the result of a mental disorder,

exhibit emotional or behavioral functioning that is so impaired as to

interfere substantial ly with their capacity to remain in the comnunity

without supportive treatment or services of a long- term or indefinite

duration. Serious mental illnesses include schizophrenia, mood

disorders, and organic and personality disorders.

As noted previously, the scope of our study was established by the May 8,

1991, resolution of the Joint Legislative Oversight Comnittee. This

resolution directed us to address eight specific questions. However, due

to the breadth and complexity of the questions, we informed the

C m i t t e e at the time of the resolution that we would not be able to

thoroughly address a l l eight questions within the time frame provided.

Therefore, we agreed to perform as much work as possible in the time

al lowed.

In addition to the scope limitations imposed by the short time frame and

the breadth and complexity of the questions, persistent problems with

data also restricted our audit work.

Concerns regarding BHMIS data precluded us from relying on i t as a

primary source for service data. ( See Chapter Ill, page 27

regarding BHMIS data.)

Data from client f i l e s proved d i f f i c u l t to use for analysis. Client

f i l e s are not kept in a standard format, nor is a l l client service

information stored in one central location. Furthermore, some f i l e s

lacked adequate documentat ion of serv i ces .

Data is not recorded or maintained in a consistent manner among

entities. Of the five entities from which we requested basic

service information, only two were able to f u l l y comply with our

request. A third provided partial information, and the remaining

two were unable to provide adequate information. ( See Chapter I V ,

page 37.)

Given the time frame and data problems, we were able to compi Ie

information describing the system; however, we did not have time to

obtain sufficient informat ion to assess the relative performance of the

system vis- a- vis the eight questions. Therefore, because we did not form

detai led conclusions and provide the recomnendations normally associated

with a performance audit, we are presenting the results of our work as a

special study.

This study was conducted in accordance w i th government aud i t i ng standards .

The Auditor General and staff express appreciation to the Director and

staff of the Department of Health Services, the Division of Behavioral

Health, and the staff of the administrative entities for their

cooperation and assistance during this study.

SECTION ONE

Chapter I

Chapter I I

Chapter Ill

Financial Review Of

Administrative Entities

Contract Monitoring And

Provisions

Behavioral Health Management

lnformat ion System ( BHMIS)

CHAPTER I

FINANCIAL REVIEW OF ADMINISTRATIVE ENTlTIa

We estimate less than 80 percent of !% I monies received from DHS by

entities are expended on direct services. While most funds for fiscal

year 1990- 91 were contracted to provider agencies, a l l three entities have

accumulated significant balances of State- appropriated behavioral health

funds. Some of these funds, which are considered unrestricted, were

expended on goods or services not directly related to the provision of

behavioral health services. In addition, administrative costs captured

and reported by the e n t i t i e s ' financial accounting systems may

underestimate total administrative costs.

Met hodo l oqy

To determine the proportion of funds expended providing direct services

in relation to administrative costs, we selected the three entities that

received the most State funding for adult SMl services for fiscal year

1990- 91. The entities were ADAPT, Inc.; Community Organization for Drug

Abuse, Mental Health and Alcoholism Services, Inc. ( CODAMA Services); and

Cmunity Care Network, Inc. ( CCN). Combined funding to these three

entities represented 62 percent of a l l adult SMI funds received from the

Department of Health Services in fiscal year 1990- 91. A t each of the

three en t i t i es , we rev i ewed f i nanc i a l records document i ng how adu l t SM I

funds were expended. In addition, we reviewed financial records at the

two largest provider agencies under contract with each entity to further

determine how funds were expended at the provider level.

Results Of

Financial Review

In fiscal year 1990- 91, DHS expended $ 56,885,653 for adult SMI services.

Of this amount, $ 56,426,155 was contracted to outside organizations. The

three entities we reviewed received $ 34,800,257 of this amount; the

remainder went to the five other entities and to other organizations.

ADAPT, Inc.

As shown in Table 2, ADAPT. received $ 14,071,885 from DHS for adult SMI

services in fiscal year 1990- 91. Almost $ 13 million of this amount was

paid to provider agencies. The Arizona Center for Clinical

TABLE 2

ADAPT, INC.

Schedule Of Revenues And Expenditures

For The Seriously Mentally Ill

Year Ended June 30,1991

( unaudited)

SMI revenue from DtiS

Expenditures

To providers:

Arizona Center for

Clinical Management

La Frontera Center

Southern Arizona

Mental Health Center

Kino Hospital ( overflow

providers)

Comnuni ty Organization

for Personal Enrichment

Intermountain Centers

P r i mvera Founda t i on

Other providers

Total to providers

Direct services ( Arizona Hotel)

Medication

Administ rat ive

Total Expenditures

Excess of revenue over expend i t u r es

Management ( ACCM) received the largest share of ADAPT's contracted

funds. Fifty- one percent of ADAPT'S funding was contracted to ACCL( to

provide case management services for a l l SMI clients in ADAPT'S service

area, and to contract with provider agencies for other direct services.

ACCM's financial records indicate that approximately $ 3.5 million was

expended by the agency on case management, and $ 2.7 m i l I ion was

contracted to other providers. ADAPT is the only entity currently

con t rac t i ng the case management f unc t ion w i th another agency.

ADAPT reports spending $ 523,009 on administrative costs in fiscal year

1990- 91. This represents 3.8 percent of i t s total expenditures, which is

low when compared to the other administrative entities that do not

contract for case management services. However, i f ACCM's administrative

costs for case management of $ 869,187 are included, the percentage of

administrative costs rises to 10 percent. Approximately $ 176,000 of the

SMI funds received remained unexpended at the end of the fiscal year.

CODAMA Services

As indicated in Table 3, page 14, CODAMA expended almost $ 9.5 million of

i t s adult SMI funding in fiscal year 1990- 91. Of this amount, $ 5.8

mi l l ion was contracted to other providers. Unl ike ADAPT, COOAMA provides

case management services directly rather than contracting this function

to another agency. CODAMA'S financial records indicate that

approximately $ 2.3 mi l lion was spent on case management.

CODAMA reports spending $ 962,664 on administrative costs in fiscal year

1990- 91. This represents over 10 percent of i t s total expenditures.

Approximately $ 1.4 mi I I ion of the SMI funds CODAMA received remained

unexpended at the end of the fiscal year.

TABLE 3

CODAMA Services

Schedule Of Revenues And Expenditures

For The Seriously Mentally Ill

Year Ended June 30, 1991

( unaudited)

SMI revenue from DHS

Expenditures

To providers:

Phoenix South Comuni ty Mental

Health Center

Triple R Foundation

Maricopa County Health Services

Toby House

New Ar i zona Fam i l y

AHCCCS

Project Arts

Survivors on Our Own

Behavioral Health Services

Total to providers

Case management

Administrative

Medication

Other operating

Total expenditures

Excess of revenue over expenditures

Communitv Care Network. Inc.

Comnuni ty Care Network contracted with providers for almost $ 5.2 mi 1 1 ion

of the $ 9.6 million it expended on adult SMI services in fiscal year

1990- 91. A8 indicated in Table 4, page 15, CCN spent $ 844,115 on case

management. In addition, CCN contracted $ 959,871 to providers for case

management services.

TABLE 4

COMMUNITY CARE NETWORK, lnc.

Schedule Of Revenues And Expenditures

For The Seriously Mentally Ill

Year Ended June 30,1991

( unaudited)

SMI revenue from DHS $ 9,829,415

Expend i t u res

To providers:

Ter ros

Wayland Family Centers

Toby House I I

Good Samaritan Regional Medical

Center

Comnunity Behavioral Health

Presbyter i an Service Agency

Phoenix Interfaith Counseling

Service

Survivors United

Jewish Family and Chi Id Services

Other organizations

Total to providers

Pi lot program

Case management

Med i ca l supp l i es

Payments to psychiatrists

Administrative

Total expenditures

Excess of revenue over expenditures

Of the three entities, CCN reported the highest administrative costs.

However, $ 528,727 of the administrative costs CCN reported were for the

pi lot costs that were not incurred at ADAPT or CODAMA. CCN

reports spending $ 1,162,979 on administrative costs in fiscal year

1990- 91. This represents almost 12 percent of i t s SMI expenditures.

Approximately $ 225,000 of the SMI funds received remained unexpended at

the end of the fiscal year.

Our review raised some concerns about the e n t i t i e s ' use of unrestricted

funds, a portion of which are SMI funds received by the entities. There

( 1) Pilot programs were established i n 1986 to test alternative delivery systems such as

capi tated systems.

are no restrictions in the contracts between DHS and the administrative

entities regarding the use of unexpended funds or the interest earned on

these funds. Therefore, these monies can be expended by the entities at

their discretion. A l l three entities have accumulated significant

balances of unrestricted funds. Based on our review of expenditures from

these unrestricted funds, we noted some unrestricted funds were expended

on behavioral health programs. However we also identified some

expenditures that were not directly related to providing behavioral

health services.

Fund balances - We found that the three e n t i t i e s ' fund balances,

consisting of cash and other assets, totaled over $ 10 million at June 30,

1991. These fund balances were not only from SMI funds. However, ADAPT,

CODAMA, and CCN received 95, 96, and 100 percent, respectively, of their

revenue from State funds. Accordingly, these cumulative fund balances

consisted mainly of State funds that remained unexpended at the end of

each fiscal year, and interest earnings on these monies. The interest

earned on these funds was approximately $ 690,000 for fiscal year

1990- 91. Periodically the entities had large receivable balances from

DHS and AHCCCS which required them to use part of the fund balances to

pay providers. Because DHS contract provisions do not speci fy how these

monies are to be used or disposed of, the entities may retain these funds

or their interest earnings and expend them at their discretion ( see

Chapter I I , page 25).

Table 5 reports the June 30, 1991, unaudited fund balances for the

administrative entities we reviewed.

TABLE 5

Fund Balances Of Administrative Entities

June 30, 1991

( unaudi ted)

Adinistrative Entity Fund Balance

ADAPT, Inc.

CODAMA Services

CCN

TOTAL

There are no restrictions in the DHS contracts regarding the use of these

funds. However, both the Legislature and the Department of Health

Services originally intended that these monies be used to provide

behavioral health services. They were not intended to provide the

entities with discretionary funds. Left unrestricted, these funds may

not be used for the purpose intended or in the best interest of the State.

Questionable exoenditures - As part of our review, we examined check

registers, vendor f i l e s , and a limited number of specific expenditures

for each entity to determine whether entity expenditures appeared

appropriate and reasonable. We identified a number of expenditures from

the unrestricted and SMI funds that demonstrate how unexpended behavioral

health monies can be spent in subsequent fiscal years i f they remain

unrestricted.

We found that entities spent these funds on employee bonuses, food and

entertainment, retreats and conferences, and other items. For example:

a ADAPT distributed $ 102,960 in bonuses, primarily to executive staff

between f i scat years 1990- 91 and 1991- 92. One- ha I f of the bonus poo l

was distributed in January 1991, and one- ha1 f in July 1991. Twelve

emp loyees rece i ved bonuses rang i ng f rom $ 1,560 to $ 26,154.

ADAPT spent over $ 1 ,800 for food and accomnodat i ons for meet i ngs , and

$ 800 for flowers for various occasions.

Community Care Network spent $ 5,300 on a retreat for directors of

provider agencies, $ 5,200 for i t s annual board meetings, $ 4,600 for

an annual board retreat, and $ 1,200 for a provider picnic.

0 CODAMA spent $ 2,500 for i t s annual board retreat, $ 700 for i t s

Christmas party, and almost $ 500 for flowers.

To prevent behavioral health monies from being expended for unintended

purposes, UiS should r e s t r i c t the use of State monies and disallow costs

and expenditures that are not related to contract provisions. Unlike the

bidding process for most State contracts, entities have experienced no

competition in obtaining behavioral health contracts. DHS received only

one proposal for each of the administrative entity areas. Consequently,

a1 l applicants for the contract were awarded the designation of

administrative entity. Due to the ci rcumstances of this award process,

the administrative entities should be treated as i f they were grantees

rather than vendors. Accordingly, DHS and the administrative entities

should work together to ensure that fund balances and related interest

earnings are expended in the best interest of the State, as was

originally intended. DHS should negotiate the disposition and use of the

unrestricted fund balances accumulated from prior- year contracts. In

future SMI contracts, DHS should restrict the use of unexpended funds and

interest earnings to ensure they are spent only for the purpose

originally intended.

Administrative Costs

Mav Be Understated

Whi le entity administrative costs are d i f f i c u l t to determine accurately

due to inadequate and inconsistent DHS reporting requirements, our review

of expenditures suggests administrative costs reported may understate

actual costs. For example, some administrative costs, such as those

associated with case management, are categorized as direct services.

Entitv costa - Because DHS has not defined administrative costs and

specifically directed how costs should be classified ( see Chapter II,

page 25), we were unable to determine or compare the administrative costs

of the entities. The entity tables shown earlier present administrative

costs as reported to DHS. These costs were calculated based on

methodology provided by DHS; however, this methodology determines direct

and indirect costs, not administrative costs.

Further, the reporting formats required by DHS and used by the

administrative entities and the providers were inadequate to determine

amounts for administrative or direct service expenditures. However, our

review of the entities1 and providerst expenditures indicated that the

administrative expenditures reported in Tables 2, 3, and 4 are low. DHS

allows the e n t i t i e s to treat case management as a direct service for

purposes of determining administrative costs. A t the administrative

en t i t i es we rev i ewed , many costs we re comb i ned under the head i ng of case

management. For example, COOAMA recorded a l l of the operational costs of

the mental health c l i n i c s as case management costs. Also included in the

case management costs were expend i tures for employee benef i ts, travel ,

advertising, telephones, and supplies. We consider these expenditures

administrative.

Provider costa - Entity administrative costs did not include provider

administrative costs, which can be significant. As part of our review,

we examined the expenditures of the two providers that received the most

funding from each entity. Providers examined included the Arizona Center

for Clinical Management and the La Frontera Center funded by ADAPT,

Phoenix South Cornunity Mental Health Center and the Triple R Foundation

funded by CODAMA, and Ter ros and Way I and Fam i I y Centers funded by

Cmuni ty Care Network.

To determine the percentage of total SMI funding spent on direct

services, we considered both entity and provider administrative costs.

We estimated the percentage of total SMI funding expended on direct

services for fiscal year 1990- 91 by combining entity and provider

adm i n i st rat i ve costs , and then deduct i ng these amounts and unexpended

funds from the total revenues received. For the providers examined in

the calculation, we used the administrative expenditures they reported.

These expenditures ranged from 7 to 17 percent of total SMI

expenditures. For the providers not examined, we assumed administrative

costs of 10 percent.

Our calculations estimate that 72 to 79 percent of revenues received are

being expended on direct services. This estimate may be high since no

adjustments were made for administrative costs that may have been

classified by the entities or providers as direct services.

DHS should develop uniform accounting and reporting guidelines that would

require more detailed reporting of program expenditures. This would also

ensure consistent reporting of expenditures between the e n t i t i e s and

providers, and assist DHS in mni toring SMI expenditures.

1. DHS and the administrative enti ties should work together to ensure the

expenditure of fund balances is in the best interest of the State, as

was originally intended. In future SMI contracts, DHS should restrict

the use of expended funds to ensure they are spent for the purpose

original ly intended.

2. DHS should develop uniform accounting and reporting guidelines that

would require more detailed reporting of program expenditures. This

would also ensure consistent reporting of expenditures between the

entities and prov'ders and aid DHS in monitoring the !% I expenditures.

CHAPTER II

CONTRACT MONITORING AND PROVISIONS

The Department continues to have deficiencies in contract monitoring and

contract provisions that we identified in our previous audit. We found

that weak and ineffective contract monitoring persists. In addition,

contract provisions continue to exhibit the same weaknesses previously

identified, although the Department plans to overhaul i t s entity

contracts for fiscal year 1992- 93.

Previous Audit Findin-

In our previous audit report dated November 1989 ( Performance Audit

Report No. 89- 10), we noted several deficiencies in the Department's

monitoring of administrative entity contracts and contract provisions:

Limited and inconsistent monitoring of entity performance, and weak

and superficial follow up on problems resulted in the failure to

f u l l y address or correct problems.

Staff lacked definite direction from management on monitoring duties

and responsibilities.

The DHS Policies and Procedures manual for behavioral health was

outdated.

In addition to monitoring deficiencies, we also identified the following

weaknesses with the Department's contract provisions:

The Department's contract contained few specific definitions of who

was to receive behavioral health services, and did not contractually

establish target populations to be serviced by the administrative

entities and providers. Because the Department based contract

compliance on units of service provided, the administrative entities

may not have been providing services to those most in need.

Contracts did not contain provisions establishing penalties for

f a i l u r e t o perform, or for failure to submit timely financial reports.

Contracts did not have provisions requiring the administrative

ent i t i es to conduct qua l i ty assurance.

An examination of the steps taken by the Department to implement the

recomnendat ions made i n our November 1989 aud i t f o l lows .

Monitorina Continues

To Be Weak

The Department has made l i t t l e improvement in i t s monitoring of the

administrative e n t i t i e s ' performance. The Department's recovery of

unallowable costs is weak, and the discontinuation of fiscal monitoring

by program representatives raises concerns. In addition, l i t t l e progress

has been made in addressing other deficiencies noted previously.

Ineffective recovery of funds - The Department's recovery of unallowable

costs continues to be weak, ineffective, and untimely. In the 1989

audit, we reported that a Department review of an e n t i t y ' s c l i e n t service

records revealed that the entity may have overcharged the State more than

$ 150,000. A t the time of the previous audit, the entity had not repaid

these funds and, as of this review, the Department s t i l l has not

collected these monies. Instead, the Department appears to have let the

issue lapse before f i n a l l y submitting i t to the Audit Disposition

Comnittee ( ADC),(') on September 10, 1991, for resolution. The entity

has since requested a formal administrative hearing, which was scheduled

for December 6, 1991 -- more than two years after we original ly reported

the issue.

The Department appears to continue to be lax in recovering funds. For

example, we reviewed a May 6, 1991, finding for another entity that

i den t i f i ed $ 7,932 i n una l l owab l e costs and requested a response from the

entity by June 6, 1991. When no response was received, Department audit

staff submitted the finding to the ADC for resolution. According to

Department audit s t a f f , the committee decided at their September 10,

1991, meeting to return the finding to the entity for a response before

making a decision.

Discontinuation of fiscal nonitorinq - In addition to the Department's

weak cost recovery efforts, the discontinuation of fiscal monitoring by

program representatives raises concerns. A t the time of our last audit,

( 1) The Audit Disposition Comittee was established by the Department to serve as the

fi rst step for resolving Department audi t finding and recornendation conflicts between

the Departnent audit staff and the adnini strative enti ties.

program representatives were responsible for fiscal reviews of entities

to ensure that the entities had actually provided the services for which

they were paid. However, due to a reorganization of OCW, program

representatives no longer perform this function.

As a result of this change and fragmentation of responsibilities, i t

appears the Department is doing very l i t t l e fiscal monitoring. Personnel

from Provider Services, the office responsible for receiving the

e n t i t i e s ' invoices and authorizing payment for services, indicate that

they rely on the program representatives to ensure that the services the

entities claim they are providing have actually been provided. However,

as previously noted, program representatives are not performing this

function. In addition, Department audit staff say that they rely on the

annual independent audits of the e n t i t i e s t o verify that the services the

entities are reporting and being paid for have actually been provided.

However, when we interviewed an audit firm, we found that although this

firm claimed to verify services, they did not conduct a sampling or

review any client f i l e s . Instead, they sent a l e t t e r t o the Department

asking DHS to verify i f the service units their client claimed agreed

with units the Department had purchased.

Lack of Progress in other areas - We found the Department has made l i t t l e

progress in addressing three other deficiencies: a lack of consistency in

monitoring, a lack of timeliness, and policies and procedures that have

not been updated. A review of the six " draft" s i t e v i s i t reports we

received revealed that staff focused on different aspects of entity

performance and used different methods to report their findings.

In addition, while we found s i t e v i s i t s for a l l eight administrative

entities were conducted by the Division for fiscal year 1990- 91, timely

completion of reports and annual verification of data continue to be

problems. Site v i s i t s for fiscal year 1990- 91 were conducted during

April, May, and June 1991; however, as of October 7, 1991, a l l six

reports we had received were s t i l l in " draftft form, which is contrary to

Division policy. Furthermore, the s i t e v i s i t report for CODAMA notes that

data verification, which was last performed in April/ May 1990, w i l l be

conducted again in September 1991. However, as of November 1991, this

verification had not been done. In addition, ADAPT'S site v i s i t report

states that a complete data verification w i I l be performed in the future,

when sufficient data is available.

Finally, the Department has not made significant and necessary changes to

the OCBH Policies and Procedures manual. The manual provides guidelines

to be used by OCBH s t a f f i n administering behavio 7ealth contracts. A

majority of the outdated policies we found duri. wr previous audit

remain in effect.

Contract Provisions Not

Substantiallv ChanggL!

Behavioral health services contracts have not changed substantially since

our last audit, and the problems previously identified continue. Some

additional concerns have also been identified since our previous audit;

However, the Department anticipates making changes to the entity

contracts in the future.

Previous orobleas remain - We reviewed contract provisions for fiscal

year 1991- 92 contracts, as we1 l as providing copies of these contracts to

the State Purchasing Office for their review. We found that fiscal year

1991- 92 contracts s t i l l contain few specific definitions of who is to

receive behavioral health services. Contracts also lack penalties for

noncompliance. Furthermore, the Department's standard contract is s t i l l

based on a unit of service approach rather than performance contracting.

As mentioned in the previous audit, without contractually defined

populations to be served, the entities determine who w i l l receive

ava i lab l e serv i ce , and because the Department bases con t rac t comp l i ance

on units of aervice provided, the enti ties may not be providing services

to those most in need. This system results in providing services based

on the units of service rather than based on the needs of a targeted

population. In Chapter IX, we found some other states' contracts are

more specific regarding the population to be served.

Additional concer. denti if id - Several additional concerns with the

Department's contract provisions have been identified since our previous

audit. F i r s t , the fiscal year 1991- 92 contract f a i l s to address real

property and equipment ownership or interest earnings. In February 1991,

our Office conducted a special financial review of ADAPT'S purchase of a

building. Our review revealed that the contracts are not specific as to

whether administrative entities may purchase fixed assets. The lack of

contract language that delineates the purchase with State funds and the

ownership of real property and equipment is a matter that needs to be

addressed. Although contract provisions restrict the spending of certain

Federal funds ( e. g., they cannot be used for inpatient services, the

purchase of land, buildings, or major medical equipment), there are no

such restrictions on State funds. Interest earnings present another

concern. We identified an entity that had accumulated large amounts of

interest on State funds; however, the Department's contract f a i l s to

address this issue.

In addition, entity contracts may not provide, either directly or by

reference to Department policy, a sufficient and appropriate definition

of administrative costs. While the fiscal year 1991- 92 contract

stipulates an administrative cost l i m i t up to 8 pepcent of the total

contract amount, it does not specify which costs . on be included as

administrative costs. Without a definition of administrative costs,

entities may misinterpret and inconsistently classify these costs. In

addition, contract language is unclear whether the 8 percent

administrative cost ceiling includes administrative costs at the provider

level. Consequently, this lack of c l a r i t y in the contract language does

not enable accurate calculation of the percentage of administrative costs

allowed by the entities and their providers. This concern i s addressed

in greater detail in Chapter I, the financial review of administrative

entities.

Contract - 8 D- - The newly appointed Director of the Department

of Health Services has indicated that changes are planned for DHS

contracts and the 1992- 93 Request for Proposals. These changes w i l l

include developing performance contracts, requesting a service plan and a

quality assurance plan from the entities, as well as requesting

reimbursement from entities i f services are not provided as planned. The

Department w i l l use Florida's behavioral health contracts, which are

based on performance contracting, as a reference point for improvements

in the new DHS contracts.

1. DHS should recover unallowable costs in a more timely manner.

2. DHS should strengthen fiscal monitoring of administrative entity

contracts.

3. DHS should improve the consistency and timeliness of i t s program

monitoring efforts.

4. In revising entity contracts, DHS should consider adding provisions

to target service popuiations, c l a r i f y the definition of

administrative costs, address ownership of real property and

equipment purchased with State funds, and provide for the disposition

of interest earnings.

CHAPTER Ill

THE BEHAVIORAL HEALTH MANAGEMENT INFORMATION SYSTEM ( BHMIS)

HAS FAILED TO MEET THE NEEDS OF DHS,

THE ADMINISTRATIVE ENTITIES, AND SERVICE PROVIDERS

Although DHS has comnitted over $ 4 million to the Behavioral Health

Management Information System ( BHMIS) since fiscal year 1986- 87, BHMIS

f a i l s to adequately meet the needs of DHS, the administrative entities

( AEs) , and service providers. BHMlS has been plagued by operat ional and

data problems since i t was brought on- line. Although the Department i s

currently in the process of upgrading the system in an attempt to better

address users' needs, some problems remain and fundamental decisions need

to be made.

In 1986, the Legislature required the Division of Behavioral Health to

" contract for the design and development of a computer system to track

and monitor chronical ly mental ly i l l cl ients and to provide the division

with information on a l l behavioral health programs." The Legislature

ultimately required that the system be on- line by January 1, 1990.( l)

Consequently, DHS designed and implemented BHMIS, which went on- line in

July 1989.

The Department designed M I S to provide the information and management

tools necessary to plan, operate, monitor, and evaluate behavioral health

services throughout Arizona. Specifically, BHMIS was to provide

information for a c t i v i t i e s such as case management; client tracking;

contract compliance; program monitoring; and client, program, and

resource assessment.

p p - . . IS ~ r o b l - In our 1989 report on the

Division of Behavioral Health ( Performance Audit Report No. 89- 10), we

identified problems with M I S . We observed that the selection of the

hardware and software for M I S may have been premature, given that the

( 1) Legislation called for the system to be implemented on a Statewide basis no l a t e r than

July 1, 1987. The Legislature extended the deadline to July 1, 1988, and then to

January 1, 1990.

27

final systems design was not completed u n t i l after those components were

purchased. We also reported that DHS was having serious problems with

system performance , and i dent i f i ed weaknesses i n the eva l ua t i on component

of BHMIS. We warned that i f improvements were not made, DHS would have

to consider various options, including reduced data collection,

restricted access to on- line reporting, and changes or reductions in

information reporting.

DHS responded to these suggested opt ions by saying that BHMlS was on- l ine

and operating according to expectations. They also said that any

start- up problems had been identified and were being resolved. The

Department claimed that the program evaluation concerns expressed in our

1989 report were unfounded. However, our current review indicates that

the BHMlS problems identified in our 1989 report have not been resolved

and have affected the a b i l i t y of the system to adequately meet the needs

of the Department.

Des~ ite Commitment Of Over $ 4 Million BHMlS D m

Not Meet The Needs Of The Behavioral Health Svsterq

Since fiscal year 1986- 87, DHS has expended over $ 4 m i l l i o n i n designing,

developing, maintaining, and supporting M I S . Despite this huge

comnitment, BHMlS has failed to adequately meet the needs of the Division

of Behavioral Health, the administrative entities, and direct service

providers.

mer survev indicates BHIIS is not meetina the needs of thq

a i n i s t r a t i v e e n t i t i e s and service ~ r o v i d e r s - A survey of M I S users

conducted by our 0ffi. a found that M I S is not meeting the needs for

which i t was designea. We surveyed administrative entity and service

provider amnaganent staff to try to determine the extent to which BHUlS

meets thei r needs . ( I )

( 1) We nailed 225 surveys to administrative entity and service provider agencies, which

were i d e n t i f i e d i n -)) IS reports, by administrative entity staff, and i n other l i s t s .

Thirteen surveys that we were unable to find a forwarding address for were returned as

undeliverable. Thi rteen survey respondents indicated that they were no 1 onger funded

by DHS or did not have a current contract with m ahrinistrative entity. Six

respondents indicated that they were responding for note than one subcontractor. For

the reruining 193 surveys, wo received 118 responses ( 61 . I percent). Sone of those

who did not respond indicated that they did not feel that the survey was relevant to

t h e i r agency md declined to respond.

According to our survey, BHMlS has failed to meet the needs of and

provide information useful to administrative entity and service provider

s t a f f . For example:

Less than 17 percent of the respondents f e l t that M I S was

successful in meeting the needs of their organization. When asked to

rate the extent to which BHMlS assists them in the functions for

which i t was designed, users overwhelmingly indicated that BHMlS was

not very useful.

Only 36 percent of the respondents indicated that they use M I S

reports on a regular ( at least monthly) basis. Only 26 percent of

respondents f e l t that i t was easy to obtain information from M I S .

Some users claimed they get nothing from M I S .

Finally, a majority of M I S users f e l t that BHMlS data was not

timely or reliable. One- half of those who responded to a question

asking them to rank the timeliness of BHMlS data indicated that i t

was not timely. Less than one- half f e l t that BHMlS data was accurate

i e . , reliable, correct), and only 36 percent f e l t that it was

complete i e . , a l l records that should be in BHMlS are in the

system).

M I S also does not meet the needs of DHS - M I S has also not

successfully met the needs of DHS. Although much e f f o r t has been

expended on the system, l i t t l e use is being made of M I S data for

operational concerns. For example:

a DHS SMI program representatives responsible for monitoring compliance

with behavioral health contracts indicated that, due to l imitations

with W l S data, it is of l i t t l e use to them. Several program

representatives also indicated that i f they need information, instead

of going to W I S , they w i l l often request i t from the administrative

entities.

M I S data has not been useful in contract reconciliations performed

by the Department. Provider Services within the Division of

Behavioral Health is responsible for authorizing payments to

administrative entities for services provided as part of their

contract with DHS. One auditor in Provider Services presented us

with work sheets he had prepared when reviewing administrative entity

contracts. He found that informat ion presented by the administrat i

entity and information in M I S varied considerably in s m

instances. As a result, he determined that M I S information could

not be reliably used for performing his review and had to rely on

informat ion prepared by the administrative entities.

Additionally, the M I S evaluation component also appears to be of

limited use. A recent series of studies performed for the

Legislature by Clegg and Associates cited limitations due to BHMlS

data problems. Among other problems, Clegg found discrepancies

between service data reports produced by administrative entities and

those produced by M I S .

29

Sever Factors Have Contributed TQ

- BHM* Failure To Meet Its Users Needg

A number of factors have contributed to M I S ' failure to adequately meet

the needs of i t s users. BWlS has been plagued by operational and data

problems since i t was brought on- line. The computer hardware and

software used by BHMlS proved inadequate for managing the volume and type

of work. Changes in DHS operations have also affected BHMIS. In

addition, M I S data quality and timeliness have been susceptible to

problems with systems maintained by other organizations.

Hardware and software robl lens - DHS purchased the original computer

hardware and so f tware based on recomnendat i ons made by a consu l t i ng f i rm

hired by the Department. However, this equipment was purchased prior to

the completion of the f u l l BHMlS design. This violates standard

principles of systems development. Consequently, inadequate hardware and

software have severely impacted BHMIS' effectiveness.(') For example, at

the time of our audit:

Software limitations restricted the number of users that could access

BHMlS data at one time. Although 31 work stations are tied into

M I S , the system could accomnodate only four users on- line at one

time. I f a f i f t h user attempted to obtain access, system performance

and response time were severely affected. DHS had to " bump" the

f i f t h user o f f the system.

Due to the length of time i t tames to process some reports, most

report production had to be done during off- peak hours when users

were not on the system. This limits the avai labi l i ty of timely

reports.

BWlS users could not access information on- line about a l l services

received by clients. Instead, users had to request reports to obtain

this informat ion.

- Operational changes within the Division of

Behavioral Health have also affected W I S ' effectiveness. For example,

DHS has changed the way i t contracts for behavioral health services. In

( 1) DHS claims that limitations placed on then by the Department of Administration in

regard to the kind of machine and operating systm they could obtain has also been a

factor.

the past, administrative entities were required to adhere to the service

units negotiated in the contracts. Currently, administrative enti ties

are ailowed more f l e x i b i l i t y i n the way they meet their contract

requirements. Changes like these have an affect on BHMIS. Changes in

contract information must be incorporated into BHMlS to keep data current

and meaningful for contract compliance and program monitoring. A t the

present time, there is no mechanism in place to ensure that such changes

are transmitted to BHMlS s t a f f .

Administrative entity iumact on timeliness and data auality - The method

by which information is transmitted to BHMlS also affects the system's

abi l i ty to meet the needs of i t s users. Most data ( 90 percent or more)

is transmitted to DHS from the administrative enti t i e s ' computer

systems. This reporting arrangement makes BHMlS susceptible to problems

with the systems maintained by other organizations. For example, our

survey revealed that six administrative entities have backlogs of one

month or longer. One administrative entity with internal system problems

did not report data to BHMlS for approximately nine months. As of

October 1991, BHMlS s t i l l did not contain a l l the data for services

provided in fiscal year 1990- 91.

There are also problems with data accuracy. Our own data testing showed

problems with missing and potentially inaccurate data.(') We tested a

sample of services to registered clients for the five administrative

entities included in the scope of this audit. We compared hard- copy

records for a six- month period to data in M I S f i l e s . We found that

over one- fourth of the records we reviewed were not in BHMlS data f i les.

In addition, we found discrepancies'between . re information on the

hard- copy forms and the informat ion in the BHMlS data f i les for almost 11

percent of the records we reviewed. Table 6 ( see page 32) presents a

sumnary of our findings.

( 1) For the entities selected, we collected Services to Registered Client forms from 13

subcontractors for February 1991 through July 1991. We checked records against M I S

f i l e s containing service data from January 1991 to mid- September 1991. Although we

did not attempt to draw a s t a t i s t i c a l l y significant sample, we did review over 1,800

records.

We chose the Services to Registered Client form for testing because i t was reasonably

consistent among a l l subcontractors, was feasible i n the audit time frame, and would

s t i l l allow us to identify timeliness and potential accuracy problems with the data.

3 1

TABLE 6

Month

February

March

Apr i l

June

July

Comparison Of Services On Registered Client Forms

And BHMIS Data Files

February 1991 - July 1991

Percent Percent Pe rcen t

Uatched Did Not k t c h Unable to Locate

Overal l 62.3% 10.9% 26.8%

Source: Office of the Auditor General, staff analysis of a sample of

Services to Registered Client forms compared against M I S

data f i l e s .

We also found additional problems with BHMlS data. For example:

one administrative entity consistently entered ndumny" dates for

client records into the BHMlS database, and

some administrative entities used inconsistent codes to designate

certain functions ( e. g., intake screenings were coded as 23 by one

administrative entity and 03 by another administrative entity.)

Lack of accurate and timely data affects BHMIS' abi l i t y to perform any of

the functions for which i t was designed. Under the current reporting

arrangement, these problems are not likely to be t o t a l l y resolved.

Qradina BHMIS;

However. Fundamental Questions Concerning

Still Need To Be Addressed

Although DHS is endeavoring to upgrade EJtiMIS, problems w i l l remain, and

fundamental questions concerning the system w i l l s t i l l need to be

addressed.

IS wrrentlv w a d n a- - DHS i s in the process of upgrading

M I S . The Department recently replaced the original computer with an

upgraded model and, at the time of our study, was in the process of

converting the operating system and database software on which W l S is

based. This conversion is expected to be f u l l y completed in early 1992.

The new hardware and software should increase system performance and

allow more users to access the system on- line at the same time. DHS

obtained nearly $ 600,000 in thi rd- party financing(') for this new

computer hardware and software.

Some fundamental arobleins with W l S s t i l l need to be addressed

Although the Department's e f f o r t s t o upgrade BHMlS may address some of

the problems with the system, other problems remain and the following

fundamental problems with M I S s t i l l need to be addressed:

Controls over data reporting should be established. As presently

constituted, M I S is highly dependent on systems maintained by other

organizations, and DHS does not have a procedure for enforcing i t s

reporting requirements. M I S data w i l l continue to be only as

accurate and timely as the systems from which it receives data.

Questions related to what, to whom, and how information should be

reported need to be considered.

There has been some confusion as to what information should be

collected. Although most survey respondents indicated that they

attempt to report to BHMlS a l l the behavioral health services they

provide, over 20 percent said that they did not. Many providers

explained they report only services funded by DHS. Others appear to

report a l l services, regardless of whether the service is funded by

DHS or another source.

Once DHS has determined what the primary purpose of BHMlS should be, the

Department should then consider whether the current system can be adapted

to meet those needs. The new DHS director has indicated that the

Department plans to l i m i t the use of EHMlS in the future to program

informational purposes and w i l l develop a separate system to handle

contract payments. I f the role of BtiMlS goes beyond that, the current

software used to support BHIIlS might need to be replaced. Often, when

systems in other states or agencies are determined to be effective, they

( 1 1 DHS obtained this funding through Chrysler First Financial Services Corporation and

makes quarterly payments on the loan from their general appropriations. According to

a DOA official, it i s not unconnon for state agencies to seek third party financing

arrangements.

can be adapted to meet other needs. For example, the Department of

Economic Security ( DES) obtained a system from the State of Utah that

allows DES to do client tracking for their developmentally disabled

clients, Utah's system has a mental health component that DES did not

convert. DHS should attempt to identify other systems that might meet

their needs and consider whether i t would be more efficient and effective

to adapt another system or modify the current system with the resources

at hand.

1. DHS should establish controls over data reporting. These controls

should address questions related to what, to whom, and how

information should be reported to BHMIS.

2. DHS should determine what the primary purpose of BMlS should be, and

then consider whether the current system can be adapted to meet those

needs.

3. DHS should attempt to identify other systems that might meet their

needs and consider whether i t would be more efficient and effective

to adapt another system or modify the current system with the

resources at hand.

SECTION TWO

Chapter IV Accessibi l i ty Of Services

Chapter V

Chapter VI

Accessibility Of Services

To Special Populations

Services For The Homeless

SMI Population

CHAPTER IV

ACCESSIBILITY QF SERVICES

The length of time required for an adult SMI client to access the

comnunity behavioral health entity system varies greatly. We reviewed

the extent to which behavioral health entity services were accessible;

however, our efforts were hindered due to the inadequacy of data provided

by two of the five entities. Analysis of service data from the three

remaining entities revealed some clients obtained services the day of

referral while others waited two months or longer.

To address the question of service accessibility, our analysis focused on

identifying how quickly clients receive i n i t i a l services. Specifical ly,

we compared the following:

referral date to psychiatric evaluation date, and

referral date to f i r s t service date other than meeting with a case

manager . ( I )

We requested the entities to provide basic service information on a l l new

referrals received in the last quarter of fiscal year 1990- 91. Of the

f i v e e n t i t i e s within the audit scope, only two, CODAMA and SEAMS, were

able to f u l l y comply with our request for information; a third, CCN,

provided adequate information about psychiatric evaluations but not about

f i rst services. ADAPT and EVBHA were unable to ident i fy the basic

information needed to document the length of time new el ients waited for

service.

In evaluating the data, it should be noted that the entities are not

always responsible for the delays cl ients experience. For instance,

( 1) The analysis focused on accessibility at the i n i t i a l psychiatric evaluation and f i r s t

service other than case management because professionals working i n the system

identified these areas as bottlenecks for new referrals entering the system. In

addition, case management services nay have been provided prior to the service dates

analyzed.

although clients in the county jai l and inpatients at both the Maricopa

Medical Center and Arizona State Hospital can be referred to the entity

and case managed prior to discharge, clients cannot be evaluated and

services cannot be provided unt i l they are released. Also, some delays

are caused by client behavior. According to one intake specialist,

unstable clients are prone to reschedule or postpone appointments.

Results Bv Entity

We found accessibility of service varies among entities. CODAMA clients

i n c r i s i s appear to access services quickly; however, clients not in

c r i s i s may wait. Although CCN did not provide the data necessary to

determine the length of time clients waited for services, it did provide

enough information about psychiatric evaluations to indicate that CCN

clients appear to wait longer for evaluation than CODAMA clients. SEAMS

is general ly able to serve clients more quickly than urban entities

because the intake process is streamlined in i t s rural comnunities.

CODAMA reported receipt of 245 referrals for the last quarter of fiscal

year 1990- 91, but not a l l of the referrals were included in this

analysis. For example, clients who refused services or failed to show up

for appointments, or for whom data was unavailable, were eliminated from

the population.(') The average length of time cI ients waited for

psychiatric evaluation was determined for 127 referrals. F i r s t service

times were determined for 115 referrals. Analysis of these referrals

revealed that many clients wait over one month to receive a psychiatric

eva l uat i on and serv i ces other than case management .

(' I I n the psychiatric evaluation analysis, certain referrals had to be dropped for the

following reasons: some refused services ( 34); some were referred to another agency

( 25); some were determined not to be SHI ( 4); some had an i l l o g i c a l date sequence of

service p r i o r to referral ( 4); and for others, no date of service or evaluation was

reported ( 39). The l a t t e r might happen for several reasons. Service might not yet

have been received or possibly service was provided but not yet logged. Twelve

additional new clients were known to have been i n j a i l or the Arizona State Hospital.

which vould l i m i t t h e i r receipt of service. While not exactly the same distribution,

referrals were dropped from the service date analysis for similar reasons.

Psychiatric evaluation - Our review of the timeliness of psychiatric

evaluations revealed the following:

Twenty- five percent of the clients were evaluated by a psychiatrist

within one week of referral.

F i f t y percent of the clients waited longer than one month for an

evaluation.

Some evaluations were performed the day of the referral; however, one

client waited 119 days for a psychiatric evaluation.

The average delay between referral and psychiatric evaluation was 35

days.

The results of our analysis appear to reflect CODAMA'S intake policy.

CODAMA p r i o r i t i z e s clients and tries to provide services most quickly for

those in c r i s i s or running out of medication. Those clients that appear

stable and have sufficient medication wait significantly longer. This

philosophy may explain why a large number of clients were seen within the

f i r s t week after referral, although one- half waited more than a month.

F i r s t service - Our analysis of the time between the referral date and

the date of f i r s t service revealed the following:

Thirty- one percent of CODAMA1s clients received services within one

week of referral.

Forty- five percent waitedmore than onemonth to receive services.

Some clients received services on the day of referral; however, one

client in the study group did not receive services u n t i l 135 days

after referral.

The average delay between referral and f i rst service was 32 days.

The average time to obtain i n i t i a l service was s l i g h t l y less than the

average time to obtain evaluation. The lower average wait for service

may result from clients receiving medication or c r i s i s services p r i o r t o

their scheduled evaluation.

Community Care Network. Inc.

CCN reported receipt of 323 referrals for the last quarter of fiscal year

1990- 91; however, not a l l of the referrals were included in this

analysis.(') The average length of time cl ients waited for psychiatric

evaluation was determined for 187 referrals. Analysis of these referrals

revealed that most clients wait over one month to receive a psychiatric

evaluation. CCN did not provide the data necessary to determine the

average length of time clients waited for services other than case

management.

k y c h i a t r i c evaluation - Our review of the timeliness of psychiatric

evaluation revealed the following:

Two percent of the clients were evaluated by a psychiatrist within

one week of referral.

Seventy percent of the clients waited longer than one month for an

evaluation.

a One evaluation was performed the day of the referral, but one client

waited 167 days for an evaluation.

The average delay between referral and psychiatric evaluation was 52

days.

During the period of our review, on average, CCN clients waited at least

two weeks longer than CODAMA clients to receive a psychiatric

evaluation. According to CCN o f f i c i a l s , several factors may account for

( 1) Again, certain referrals had to be dropped for the following reasons: solre refused

services or did not keep their appointments ( 39); some were referred to another agency

( 3); two were detemined not to be MI; some reported an i l l o g i c a l date sequence of

psychiatric evaluation p r i o r to referral ( 4); some were i n ASH or i n j a i l ( 4); and for

others, no date of evaluation was reported ( 91). As with C O W clients, the CCN

c l i e n t evaluation dates my not have been reported f o r several reasons.

the longer wait. For example, although CCN now has responsibility for

conducting psychiatric evaluations, during the three- month period of our

review, providers performed this function. Thus, CCN had l i t t l e control

over the process. CCN staff also attribute some of the delay to large

caseloads. According to one CCN o f f i c i a l , large caseioads impact the

amount of time case managers can devote to proactive a c t i v i t i e s , such as

scheduling a client for a psychiatric evaluation.

F i r s t service - CCN did not provide the information needed to determine

the average length of time clients waited before receiving i n i t i a l

services.

SEABHS

Thirty of the t h i r t y - f i v e new SEABHS referrals reported for the quarter

were used in our analysis. Four clients were dropped from the analysis

because no date of service was provided, and one client refused service.

The review indicates that many clients were evaluated and received

service within a week of referral.

P ~ v c h i a t r i c evaluation - We noted the following about the timeliness of

SEABHS evaluations:

Forty- three percent of the evaluations were performed within one week

of referral.

Twenty- three percent of the clients waited longer than one month for

evaluation.

@ While several clients were evaluated the day of referral, one client

wa i t ed 77 days.

@ On average, referred clients waited 21 days for a psychiatric

evaluation.

F i r s t s e r v i a - Our review of the time between the referral date and the

date of f i r s t service revealed the fol lowing:

* Sixty- three percent of SEABHS clients received a service within one

week of referral.

a Thirteen percent of SEABHS clients waited longer than one month.

The longest wait for service was 58 days.

On average, SEABHS clients waited 10 days before receiving their

f i r s t service.

In general, SEABHS is able to serve clients sooner than CODAMA or CCN.

SEABHS' shorter time frames are not surprising because SEABHS' intake

process d i f f e r s from the metropolitan entities in that those seeking

assistance can v i s i t SEABHS service providers ( which have a high prof i le

in their smaller c m u n i ties) directly. SEAMS' providers are also able

to i n i t i a t e services for the individual clients through SEABHS without

the client needing to v i s i t SEABHS.

We were unable to analyze EVBHA1s performance because of problems with

EVBHAts data. EVBHA did not provide us with data on c r i s i s clients, who

are handled more quickly than other cases. Consequently, any analysis

would be skewed. We also found other problems with EVBHAts data. For

example:

a EVBHA could not provide a l i s t of a l l new clients referred to i t

during the review period.

a For those clients EVBHA did identify, the information was

inadequate. For some clients, EVBHA often provided only the date of

the c l i e n t ' s f i r s t service with a referring agency p r i o r t o referral

to EVBHA. Because that date is not indicative of EVBHA1s response,

the length of time before receiving i n i t i a l service could not be

determined.(') For others, no date of service was provided.

ADAPT. la^,

Due to lack of appropriate data, we were unable to analyze services to

clients in Pim County. Although ADAPT did make several attempts to

provide us with information, there were several problems w h the data

that prevented us from conducting any meaningful analysis.

( 1) Although NBHA subsequently offered to provide the f i r s t service date a f t e r referral

to NW, sufficient : Em was not available to conduct the analysis.

ADAPT could not provide a l i s t of a l l new clients referred to i t

during the review period.

Of the 80 new clients that were identified, evaluation dates were

provided for only 32, and service dates were provided for only 23.

Therefore, no meaningful analysis was possible.

Some of the problems with ADAPT data were the result of organizational

changes taking place in Pima County at that time. During our study

period, an ADAPT assessment team was reevaluating a l l SMI adults enrolled

with ADAPT, in addition to processing new referrals and then passing both

groups on to the newly created case management contractor, Arizona Center

for Clinical Management. Thus, separating clients actually new to the

ADAPT system and those being reass i gned was prob l emat i c.

This chapter provides information only, therefore no recomnendations are

presented.

CHAPTER V

ACCESSIBILITY OF SERVICES

TO SPECIAL POPULATIONS

Our analysis indicates that most people with serious mental illness who

are released from j a i l or discharged from the Arizona State Hospital

( ASH) or the Maricopa County Psychiatric Annex do enroll in the entity

system and receive cmunity- based services. In fact, many appear to

have already been enrolled in the system before their arrest or

hospitalization. However, some of these people, especially those in the

j a i l population, are lost in the referral process and do not get the

services they need. Coordination is improving between the entities and

the j a i l s and hospitals.

We were dhected by the Joint Legislative Oversight C m it tee to examine

the exterrr to which behavioral health services are accessible to

populations of the greatest need, including the j a i l population. We

expanded the scope of our review to include patients at the Arizona State

Hospi t a l ( ASH) and the Mar icopa County Psychiatric Annex, because they

too are among those with the greatest need for mental health services

when they are discharged into the comnuni ty.

To determine whether those in these special populations made a successful

and timely transition into the entity system, we asked the j a i l s , ASH,

and the Uaricopa County Psychiatric Annex to provide l i s t s of persons

they referred to the entities during April, May, and June 1991. A t the

same ti-, we asked the entities to provide l i s t s of the referrals they

received during the same period. We then compared the l i s t s , and

attempted to resolve any discrepancies by reviewing M I S data and

interviewing s t a f f a t the entities and institutions.

Our analysis was limited to institutions in Maricopa County because we

were unable to obtain reliable information about referrals in Pima

County. Pima County Jai l staff directed our inquiries to ADAPT, which,

as discussed in Chapter I V , could not provide the information.

The Jail SMI Pooulation

And The Entitv Svstem

Our review indicates that most people with serious mental illness who are

released from jai l make a transition into the enti ties' comnuni ty- based

care system. Some do not, partly due to factors outside the control of

the entities. The entities have taken steps to ensure that fewer people

are lost through the cracks of the system. According to j a i l o f f i c i a l s ,

these efforts are making a positive impact.

Most of the SYI releases we reviewed were enrolled bv the entitieg - We

attempted to determine whether 74 SMI persons released from Maricopa

County j a i l s during April, May, and June 1991 received entity services.

Most are now enrolled in the entity system. However, some refused

services, or their service histories could not be determined.

Service histories for most of the SMI releases could be determined using

records provided by the jai Is and the entities.

Forty- six people ( roughly 60 percent of the seventy- four releases)

were enrol led in the entity system. Twenty- seven were already

enrolled in the entity system before arrest; twelve more enrolled

during their incarceration or upon release; and seven were referred

to service providers where they could obtain services.

Thirteen people ( about 18 percent) refused services.

Five people ( or 7 percent) including two known only as Jane and John

Doe, could not be tracked because the jai Is could not provide enough

information.

Ten people ( about 14 percent) appear to have Iffallen through the

cracks." Three were not referred because they were in jai l for only

a short time. The other seven were referred to the entities,

according to j a i l s t a f f ; however, the entities and the M I S system

have no record of them and ent i ty staff have no know ledge of these

individuals. The current whereabouts and mental health status of

these people is unknown.

Factors outside the e n t i t i e s ' control may contribute to the problem of

getting released MI individuals enrol led in entity services. According

to Maricopa County o f f i c i a l s , sane SMI inmates are released from j a i l too

quickly for the entity to enroll them in the system. About one- half of

those jailed are released in 24 hours, and the jai 1 cannot hold anyone i f

the person is released by the court. Even i f someone is identified as

SMI, these hurried releases can lead to failure to notify the entities.

m r d i n a t i o n is improving - According to j a i l o f f i c i a l s , coordination

between the entities and the jai Is is improving. Some enti ties have a

structure to identify and offer services to the SMI j a i l population. For

example, one Mar icopa County ent i ty has designated a case manager to act

as j a i l coordinator and v i s i t both county j a i l s one day a week to

establish contact with prospective clients and maintain contact with

ongoing clients. Another entity has designated three case managers to

handle referrals from the j a i l s . In Pima County, one case manager has an

o f f i c e at the j a i l . J a i l o f f i c i a l s report that these efforts, which have

been in development for as long as seven years in some places, are

beginning to have an effect on improving the transition from j a i l to

conrnunity services.

In addition to the e n t i t i e s ' e f f o r t s , DHS has taken action. WS staff

have been meeting with jai l o f f i c i a l s for over a year, weekly at f i r s t

and now monthly, to learn which inmates are in psychiatric units and

determine how and where to place them in the comnunity- based system. DHS

has assigned a program representative to work with each county j a i l and

the entities to develop a coordinated process for connecting those in

j a i l with the entities. To comply with blueprint requirements for a

written plan, due February 1, 1992, which w i l l ensure that each SMI

person in the j a i l population receives appropriate services, the

Department has assigned a staff member to work with j a i l and Department

of Corrections personnel in developing a draft. Finally, DHS plans to

use a l l of the available new funding for fiscal year 1991- 92 to develop

alternative housing for specified groups, including clients released from

jai I.

The ASH Pwulation

And The Entitv Svstem

Our analysis indicates that most patients discharged from ASH make a

successful transition into the entity system. However, although 137

patients discharged during a recent three- month period were either

already enrolled in the entity system or had made the transition into i t ,

the status of 16 other patients could not be determined. In addition,

some patients do not make the t., nsition into the entity system

successfully. ASH identified several patients with poor outcomes,

including readmission to ASH. The hospital also has several patients who

are c l inical ly ready for discharge but cannot be discharged because there

are no suitable beds available for them in community- based f a c i l i t i e s .

Many ASH patients are d i f f i c u l t to place in other programs. Some have a

history of mental illness spanning 10 years or more, and have received

most of their treatment as inpatients. These patients are extremely

dependent on the hospital and consider i t their home. Elderly patients

who, in addition to psychiatric care, need medical care face an

additional problem because nursing homes are reluctant to accept

psychiatric patients and other SMI residential f a c i l i t i e s are unable to

care for their medical needs.

Most of the ~ a t i e n t sw e reviewed make a transition - ASH staff identified

a total of 153 adult SMI clients discharged during April, May, and June

1991 with referrals to the entities in the audit study group. As in the

j a i l population, we were able to determine the outcome of most of these

patients.

One hundred thirty- seven patients ( almost 90 percent) are enrolled in

the entity system. One hundred twenty were already enrolled,

thirteen new clients enrol led upon referral from ASH, and four more

were enrolled with the entity system after their discharge.

Sixteen patients ( a l i t t l e over 10 percent) could not be found in

entity records. ASH reported these people were referred to the

entities, but the entities and the ' BHMIS system have no record of

receiving the referrals. The current status of these people is

unknown.

S# le natients had to ba rehosoitalized - Of the 153 patients discharged

between April and June, 35 were readmitted to ASH by mid- September.

According to ASH'S Director, this 23 percent readmission rate is

acceptable for this type of patient. However, ASH staff provided a

synopsis of the reasons for readmission in each case, and in 7 of the 35

cases, there were problems with case management, coordination, and

service provision. For example, one patient was placed in a supervisory

care home where the amount of supervision was inadequate for the

patient's needs. Another patient was in an entity's p i l o t program u n t i l

his court- ordered treatment expired; however, when the p i l o t program

stopped treating him, the entity did not transfer him to i t s regular case

managers.

Availabilitv of services i s a problem - Another side to the accessibility

issue is the number of patients who cannot be discharged when they are

c l i n i c a l l y ready because there is no place for them to go. Availability

, f appropriate residential services is an especially d i f f i c u l t problem

for ASH patients because most of them require residential f a c i l i t i e s upon

discharge. Such faci li ties are limited in some areas of the State. A t

any g i ven t i me, ASH may have dozens of pat i en ts who have reached the

maximum benefit from their hospital izat ion but must wait for an avai lable

bed. Some patients can wait years for an appropriate bed. For example,

ASH has an 81- year- old patient who has been c l i n i c a l l y ready for

discharge for about six years. However, because she is periodically

disruptive ( she screams), nursing homes are unwilling to take her, and

SMI f a c i l i t i e s are not set up to take care of a person her age.

The Department of Health Services i s attempting to increase the number of

available beds by developing alternative housing for special groups. The

Department has also opened new re- ent ry faci l i t ies ( REFS) in the 1990- 91

fiscal year to provide a residential transition for persons recently

discharged from ASH. Despite these steps forward, creating the needed

number of residential beds in Arizona w i l l be challenging ( see Chapter

V l l l of this report).

According to ASH, 76 patients were ready for discharge as of

October 9, 1991 . ( I ) Analysis of the barriers to discharge shows that 26

patients ( over one- thi rd of those ready for discharge) were waiting

because no appropriate bed was available for them. Some of these

pat i en t s had spec i a I needs as a r esu I t of cond i t i ons such as pregnancy or

incontinence, or required a f a c i l i t y for dually diagnosed clients ( such

as those w i th mental i l l ness and chemi ca l dependency) , wh i ch made them

harder to place. Five more patients remained at ASH due to

administrative problems such as delays in the referral process. For the

remaining 45 patients, there was either no reason for not discharging

them or they were not discharged for reasons that were unrelated to the

entities or the a v a i l a b i l i t y of services.

The problem with a lack of community services for former ASH patients

w i l l become more crucial in the future. According to the blueprint, only

those with documented medical necessity may be admitted to ASH, and they

must be discharged as soon as hospitalization is no longer necessary.

The blueprint requires an evaluation of a l l long- term ASH patients, and

sets time limits for evaluating new clients. These evaluations were

being conducted during our study. A placement schedule must be

established for patients who, according to the evaluation, no longer need

hospitalization. ASH already meets to plan placements for patients ready

for discharge, but a lack of avai table faci l i t i e s hampers i t s abi I i t y to

schedu I e p I acemen t s .

Countv Annex Patlenb

And The Entitv System

In the past, some patients who received services at the County Annex have

had p r o b l w making the transition into the entity system. However,

Annex s t a f f report that coordination of services between the Annex and

the e n t i t i e s i s improving.

( 1) This i s a comprehensive l i s t for ASH, so sona patients my not be - 1. However. the

1 i s t indicates the barriers to discharge that 941 patients can encounter.

The Annex has 82 inpatient beds and provides c r i s i s services for about

800 patients at a time in Maricopa County. Many patients under court

order for psychiatric treatment receive i t at the Annex. The more

serious involuntary cases transfer to ASH for long- term care, but most

Annex patients are referred out to the administrative entities for

comnuni ty- based services. During Apri I, May, and June 1991, the Annex

provided services for a total of 1,756 patients . ( ' I

Annex ~ a t i e n t s have had ~ roblems: however. services are imroving

Delays in connecting patients with entity services have been a persistent

problem according to Annex s t a f f . This is particularly true for patients

whose court- ordered treatment i s expiring. Annex staff attribute much of

this problem to Arizona's lack of sufficient residential beds. Annex

staff also expressed concern about the large caseloads of entity case

managers ( see Chapter V I I, page 59). Annex staff said they sometimes

delay referring patients from their own case managers, who have much

smaller caseloads, because some patients need more intensive case

management than entity staff may be able to provide.

According to Annex s t a f f , the entities and the Annex work together to

coordinate the transition for Annex patients. A CODAMA staff member

v i s i t s the Annex regularly to f a c i l i t a t e patient transitions, and CODAMA

staff cooperate with the Annex in expediting referrals for patients with

special needs. In July 1991, the Annex set up a box for each entity, and

the entities now stop in to collect referral documents in their box. In

addition to these steps, the Annex tries to refer patients before they

are ready for discharge, to allow the entities enough lead time to pick

up the referral.

This chapter provides information only, therefore no recomnendations are

presented.

( 1) We were unable to evaluate Annex referrals by the same method used for the j a i l and

ASH populations. Annex records are not ccmputeri zed, and a manual search of the high

volume of patient f i l e s was not practical. Annex staff used a randon sampling method

to create a representative l i s t of referrals for the purpose of testing entity records

for v a l i d i t y and completeness; however, the small nunber of referrals l i s t e d was

insufficient for drawing conclusions about the Annex population.

CHAPTER VI

SERVlCES FOR THE HOMELESS

$ MI POPULATION

Services currently available to homeless % I adults are limited.

However, in the last year, DHS and comuni ty organizations have begun to

focus planning efforts on the problems facing homeless SMI people in

order to increase services.

No substantive data about the number of homeless SMI people in Arizona is

available, and estimates vary significantly. There is no consensus about

the size of the homeless population in general or what percentage of the

homeless are seriously mentally ill. Estimates of the homeless SMI

population in Maricopa County range from 2,400 to 4,400.(') One service

provider in Pima County estimates that there are 750 to 1,500 SMI

homeless persons in that county. Although homeless SMI people are not

s t r i c t l y an urban problem, there are fewer homeless persons with serious

mental illness in the State's more rural counties; therefore, our audit

work focused primarily on Maricopa and Pima counties.

Services For The Homeless

SMI Pooulation Are Limitad

There are few services targeted specifically for the homeless SMI

population. Outreach services for homeless SMI people are limited and

vary by entity. In addition, the need for residential f a c i l i t i e s greatly

exceeds the avai lab i I i ty .

Qutreach sarvicttr, - Currently, outreach services for the homeless SMI

population are limited and vary by entity. There are only a few outreach

programs that serve a small segment of this population. Because homeless

SMI persons are typically the most treatment- resistant, outreach programs

( 1) Based on research for the Maricopa Association of Governments ( MAG) Homeless Task

Force. June 1991 report on the homeless, a task force spokesperson estimates the

homeless population i n the county to be 7,251 to 13,415. Of that group she also

estimates approximately 33 percent are SMI.

are v i t a l in serving this group. Outreach workers must locate these

people by v i s i t i n g places where they congregate, such as shelters, soup

kitchens, or beneath bridges at river bottoms. Outreach workers

typically contact homeless SMI persons repeatedly and attempt to gain

their confidence by providing them with minor necessities. For example,

workers in one program we contacted give out water during the summer.

With repeated contact over time, outreach workers may be able to convince

a homeless % I person to accept treatment.

Funding for outreach programs serving the homeless SMI population has

decreased over the past several years.( 1) DHS has targeted over $ 470,000

to serve homeless SMI individuals in fiscal year 1991- 92 compared to over

$ 540,000 in fiscal year 1988- 89. Although funding for these programs is

largely Federal money, some State funds are allocated for these

services. Our review of services indicates the extent of outreach

a c t i v i t i e s varies considerably among the f i v e e n t i t i e s we reviewed.

CODAMA Services - For fiscal year 1991- 92, CODAMA is receiving

$ 207,297 for homeless- related outreach programs through a State

contract. Seventy- five percent of this figure is Federally funded

with the other twenty- five percent coming from the State. CODAMA

contracts with Phoenix South Mental Health Center to provide outreach

services for homeless SMI people.( 2) In the p r i o r f i s c a l year, the

Phoenix South Psychiatric Outreach Project was the only homeless SMI

outreach program funded by DHS in Maricopa County. The program

served 881 homeless SMI persons -- 20 to 30 percent of those

estimated to need services.

RIBHA'S outreach services for homeless SMI persons have fluctuated

over the last several years. Outreach services were provided by two

subcontractor agencies in fiscal years 1988- 89 and 1989- 90. Funds

for contracted service were not available in fiscal year 1990- 91

because EVBHA lost t h i s Fede ra I fund i ng . Consequent I y , EVBHA

assigned several case managers to provide l i m i ted out reach services

in fiscal year 1990- 91. However, these case managers stopped

providing outreach services in April 1991 due to large caseloads.

The State has funded an EVBHA in- house team to provide

( 1) Projects for Assistance i n Transition for Honolessness ( PATH) grants provide most of

the funding for DHS1s honeless 941 programs. However, this funding has been

decreasing .

( 2) Mricopa County also has a " drop- in" center ( CHAPS) that p ovides case management and

psychiatric services to honeless SMI people. However, this program i s not an outreach

program because it does not attempt to locate clients and i s not supported by funding

for the SnI honeless.

outreach services in fiscal year 1991- 92. Again, 75 percent of the

State's $ 81,440 contract is Federal ly funded. EVWA is in the

process of f i l l ing the two positions slotted to provide this service.

0 ADAPT, Inc. - For fiscal year 1991- 92, ADAPT is receiving $ 181,077

through a State contract. Again, 75 percent is Federally funded, and

25 percent is State funded. ADAPT has contracted with La Frontera

for the Readily Accessible People Program to provide homeless SMI

outreach for the past several years. This program served 384 clients

in fiscal year 1990- 91, only about one- quarter to one- half of the

population estimated to be in need.

CCN - CCN does not receive monies specifically designated for

home less SMI out reach programs and has no programs spec i f i ca l ly

targeted for such services. According to a DHS o f f i c i a l , CCN did not

request funding to provide homeless -% I services during fiscal year

1991- 92.

SEABHS does not have a homeless SMI outreach program either;

however, according to the entity director, this i s not an area of

real need in southeastern Arizona. He estimates 90 percent of a l l

homeless SMI people in the four SEABHS counties are identified

through contacts with other local agencies.

Residential services - Upon consenting to treatment, homeless SMI people

are in particular need of residential services. According to outreach

program s t a f f , homeless SMI people need residential services tai lored to

meet their specific needs. Many SMI residential f a c i l i t i e s are

unequipped to deal with this group. SMI individuals are often d i f f i c u l t

clients, and homeless shelters are often forced to turn these people away

because they " act out" in the shelter and become disruptive.

Furthermore, some homeless SMI persons coming o f f the street are unable

to conform to the structure required by many SMI faci l i t i e s . According

to one program manager, some seriously mental ly i l l people are unable to

deal with treatment programming because for them just getting out of bed

each day i 8 an achievement .

There are only a few specialized residential services, although many are

needed. The June 1991 Mar icopa Association of Governments' ( MAG) study

identified only 49 beds to serve between 2,400 and 4,400 homeless SMI

people in Maricopa County. Recently, an 18 bed f a c i l i t y was opened in

Maricopa County to serve the homeless SMI. Currently, Pima County has no

beds specifically earmarked to serve i t s roughly 750 to 1,500 homeless

SMI population.

Steos Are Beina Taken TQ

Increase Services

In the last year, DHS and comuni ty service groups have focused on the

needs of the homeless SMI population. A recently established DHS

Homeless SMI Task Force is currently working on a plan required by the

Arnold v. Sarns court order blueprint .( I) The blueprint requires that

the plan consider the number and types of services that need to be

developed, including specialized services for homeless SMI persons.

After review by the lawsuit's p l a i n t i f f s and the court monitor, a final

plan w i l l be developed. A l l parties must agree on the plan and i t must

be implemented by September 30, 1995. However, additional funding w i l l

probably be needed to implement the plan.

Although the blueprint has stimulated action to address the needs of the

homeless SMI population, i t s restrictions have also hindered the

development of services. For example, recent efforts to obtain

add i t i ona l beds for home l ess SM I peop l e have encountered prob l ems. MAG'S

Homeless Task Force and the City of Phoenix have been working together to

provide additional beds for the homeless SMI population in Maricopa

County. HUD money was obtained to fund three new residential f a c i l i t i e s

to house 52 beds.( 2) However, only 24 beds could be developed due to the

court- imposed restriction of 8 beds per f a c i l i t y . Thus, 28 of the 52

planned beds cou Id not be developed, and some of the HUD funding had to

be returned. According to a staff person for MAG'S Homeless Task Force,

HUD expressed disappointment at the loss of the 28 beds, given the need

for the beds and the a v a i l a b i l i t y of funding to provide them.

This chaptrr provides information only, therefore no recomnendations are

presented.

( 1) The court order called for the homeless plan to be completed by October 1, 1991;

however, an extension until January 1, 1992, has been given. As of January 10, a plan

had not been final i zed.

( 2) Two of the foci 1 i ties were to house 20 beds each, and one was to house 12 beds.

SECTION THREE

Chapter VII Case Management

Chapter Vlll Other Needed Services

CHAPTER Vli

CASE MANAGEMENT

The Arizona comnunity- based mental health system relies on case

management to ensure that clients receive the services they need, and

that services are coordinated and appropriate. Although we found that

caseloads vary greatly among case managers, overall, caseloads are

large. As a result, case managers cannot devote adequate time to

individual clients. Limiting caseloads to comply with recently mandated

standards will require many additional case managers and millions of

do1 lars.

Case Manaaement Is An

Essential Part Of The System

Case management is an essential part of Arizona's community- based mental

health system. Case management services in Arizona are provided by

administrative entities through cl inical teams.(') These cl inical teams

are responsible for developing an individual treatment plan ( ITP) for

each client.( 2) As part of the clinical team, the case manager has

primary responsibi I i ty for identifying and obtaining services in the

client's ITP. In addition, the case manager has the ongoing

responsibility of monitoring the services provided to the client and

assessing the client's progress in achieving his or her goals.

To understand the case management function and workloads, we collected

caseload data from the five entities. We also observed ten case managers

performing a variety of functions, including visiting clients in various

( 1) The c l i n i c a l t e a may consist of a nurse, social worker, vocational therapist,

psychiatrist, and care mangers.

( 2) The Individual Treatment Plan ( ITP) i s a written document describing services to

assist the client i n meeting identified needs, and the objectives and long tam goals

to be achieved.

settings, such as the county j a i l , Arizona State Hospital, treatment

programs, private homes, supervisory care homes, etc. In addition to

v i s i t i n g clients, we also observed case managers speaking with clients by

telephone, providing them with transportation to appointments, consulting

with other professionals about their cases, assisting clients in securing

basic needs, such as food stamps and lodging, and helping them in times

of c r i s i s .

Larae Caseloads Restrict

Time With Clients

Although caseload size varies, overall, caseloads appear large, which

reduces the case manager's a b i l i t y to provide clients with adequate

attention. For the f i v e e n t i t i e s we reviewed, caseload size averaged 43

clients per case manager; one case manager had a caseload of 83 clients.

Our observations confirmed case manager's comments that the large number

of cases prohibits them from spending adequate time with each client.

Caseload size - Using information provided by the four administrative

entities and the Arizona Center for Clinical Management ( AcCLO(') for

September 1991, we found that 187 case managers handled 8,053 clients, an

average of 43 clients per case manager. Caseloads ranged from 10 to 83

clients. Table 7, page 61 shows case manager t o c l i e n t ratio by

agency .( 2)

( 1) The administrative entity i n Tucson, ADAPT, Inc., contracts with the Arizona Center

for Clinical Management ( ACCM) to provide case management.

( 2) These nunbers are not static; they change as new clients enter the system and others

leave, and as a result of case manager vacancies.

TABLE 7

Case Manager To Client Ratio

By Agency As Of September 1991

Nunber Of Case Total Nmber Caseload Range Of

A~ ency Uanaclersf a) gf Cl ientstbl Rat iotcy Case I oads

ACCM 61

EVBHA 22

CODAMA 34

CCN 59

SEABHS 11

( a) Number does not include supervisors who may have caseloads, or new case managers with

1 imi ted caseloads.

( b) Number excludes 202 c l i e n t s served by supervisors and new case managers.

( c) Numbers are rounded.

( d) Includes p i l o t project case managers. When these case managers are excluded, case

manager t o c l i e n t r a t i o s average 1 : 58.

Source: Office of the Auditor General, staff review of data provided

by ACCM, EVBHA, CODAMA, CCN, and SEAMS.

There are several reasons for the wide range in caseload size. For

example, CCN has a pi lot project that funds smaller caseloads. CCN

caseloads ranged from 10 to 28 clients for the 11 case managers in the

p i l o t program during September 1991. In addition, ACCM has several case

managers with smaller caseloads because they are assigned clients who are

more d i f f i c u l t to manage or because clients l i v e in rural areas.

Case managers are l i k e l y to continue to carry large caseloads. In fact,

their caseloads may increase. Currently, only about one- half of the

estimated SYI population in Maricopa County is receiving case management

services through the entity system. Once a l l other clients are brought

into the system, they w i l l require case management services. In

addition, the " checklistN of e l i g i b i l i t y was recently broadened to

include personality and organic brain disorders. Thus, an even greater

number of people w i l I quai i fy for case management services. However, no

add i t i ona I fund i ng has been des i gnat ed for case management serv i ces in

fiscal year 1991- 92. Consequently, the client to case manager ratio may

increase.

Larae caseloads l i m i t time available for clients - Large caseloads, as

we1 I as the wide range of services case managers are expected to provide

impact the amount of time case managers have to spend with their

clients. In addition to identifying and obtaining services for each

client and monitoring the services provided, case managers are

responsible for evaluating each client every 90 days. Therefore, they

need to spend enough t i me w i t h each c I i en t to adequate l y assess the

c l i e n t ' s level of functioning. However, a l l of the case managers we

observed indicated their caseloads are too large and they are unable to

spend enough time with their clients. Some case managers complained that

they must spend at least 30 percent of their time on paperwork. Two of

the ten case managers we observed stated that they have not been able to

meet some of their clients because they have not had the time. One of

these case managers ( with a caseload of 71 clients) told us that he is

only able to process new intakes and provide services for those who " make

enough noise to get his attention." The following case example

i l l u s t r a t e s the variety of functions case . anagers perform during the

course of a day and the limited amount of time they have to spend with

thei r c l ients.

We spent the day w i th one case manager who had a case l oad of 50

c l i en ts . The pace a I l day was f rant i c even though the case manager

had scheduled and inned the day efficient ly. He began the day

early by transport a client to an appointment and then rushing to

attend a training session on housing for the seriously mentally i l l .

Back in the office, the case manager fielded phone c a l l s and did

paperwork simultaneously. One of the calls was from a client who

wanted to see a doctor because she was very depressed. The case

manager imnediately discussed the client with a psychiatrist and was

able to schedule an appointment for this client later in the day. He

then transported another client to an appointment, and went back to

the o f f i c e where he returned three phone calls from other clients who

had called about their medications while he was out. The case

manager also spent about five minutes each with two additional

clients. The case manager explained that he normally tries to spend

about 30 minutes with each c l i e n t ; however, on this particular day,

he was already running behind schedule and could not spare the time.

In fact, the case manager did not break for lunch.

In the afternoon the case manager visited several clients. One of

them l i v ~ i n an apartment with only mattresses on a concrete floor.

The case anager wanted t o t a l k with the landlord about this client's

l i v i n g conditions; however, the client feared being evicted i f he

c r i t i c i z e d the situation. After v i s i t i n g with this client for about

20 mlnutss, the case manager drove to v i s i t another client to ensure

t h i s c l i e n t was taking his medication. This client is s t i l l paranoid

and believes his medication is k i l l i n g him. The client does not want

to participate in a treatment program.

The case manager next attended a meeting at a treatment f a c i l i t y .

One client involved in the program had been living in a supervisory

care home, and the discussion focused on whether t h i s c l i e n t was

ready for independent l i v i n g and what problems t h i s c l i e n t might face

in an independent living situation. Following the meeting, the case

manager returned b r i e f l y to his office and attempted to c a l l two more

clients; however, he was unable to reach either of them. He then

went to v i s i t another client who is living independently and has a

job. The case manager spent about 10 minutes talking with the

client about how he was doing and whether he was having any

problems. The client appeared to be doing we1 I. After this v i s i t ,

the case manager planned to pick up a prescription for a client and

v i s i t another client at home.

The case manager told us that this was a typical day for him. He had a

number of unplanned calls to which he had to respond and more work than

he could r e a l i s t i c a l l y handle. He also noted that he s t i l l had to

complete paperwork about the day's a c t i v i t i e s , and this most l i k e l y would

be done after hours.

Lower Caseload Ratios Will

R e i r e A Substantial Increase In Fundinq

By 1995, additional case managers as well as additional funding w i l l be

needed to reduce the size of caseloads to the levels required by the

blueprint. By September 1995, most caseloads w i l l be limited to 25

clients or fewer. Caseloads with intensive clients, ( those more

d i f f i c u l t to treat), w i l l be limited to 10 clients. However, as

illustrated in Table 7, page 61, none of the five agencies we reviewed

are close to achieving these caseload ratios. We found caseloads average

almost double the number specified by the blueprint, and almost 90

percent of the case managers have caseloads of more than 25 clients.

Consequently, many additional case managers w i l l be needed. For example,

according to the blueprint, Maricopa County w i l l need approximately 459

case managers to serve an estimated population of 11,589 SMI people in

1995. As of September 1991, there were only 115 case managers in

Maricopa County. As a result, funding for case management salaries

in Maricopa County alone would have to increase almost $ 8 m i l l i o n

annual ly to provide the estimated number of case managers needed by

1995.(') DHS believes that some efficiencies in cc s management may be

possible through better interagency coordination or other means. I f

efficiencies are possible, the cost of meeting the blueprint caseloads

standards may be s i g n i f i c a n t l y less.

RECOMMENDATIONS

WS should study whether efficiencies can be realized in case management

through interagency coordination.

( 1) The current average mid- point salary of a Maricopa County case manager ($ 23,022,

excluding employee- related expenses) was multiplied by 344 ( the estimated nmber of

additional case managers needed by 1995) to arrive at the $ 8 million amount.

CHAPTER Vlll

ARIZONA LACKS SOME NEEDED SERVlCES FOR

THE SERIOUSLY MENTALLY ILL

In addition to the need for more case management services described in

Chapter V I I , we also identified other services that seem to be lacking in

Arizona. Residential beds appear to be the area of greatest need. Other

services, including dental care, are also needed.

Residential Services Arg

Lackina In Arizona

The lack of residential services was a recurring theme throughout the

review. Case managers and other professionals reiterated the need for

more residential services. The blueprint requires that a variety of

housing and residential options be provided for SMI persons. The State

does not have a sufficient number of residential beds for the number of

clients who need them, and some clients are currently l i v i n g in settings

that w i l l not be available in the future due to blueprint restrictions.

Blueprint reauires residential services - The blueprint requires DHS to

plan for, develop, and maintain a variety of housing and residential

options. By 1995, a l l clients are to be receiving the housing and

residential services that can reasonably benefit them. Clients are to be

integrated into the c m u n i t y in residential programs of no more than

eight people, or in apartment settings where no more than 25 percent of

the apartments are occupied by clients. Clients currently l i v i n g in

supervisory care or boarding homes are to be evaluated and moved to

alternative housing i f appropriate.(') The blueprint defines several

( 1) The blueprint limits each residential f a c i l i t y to eight or fewer clients, and almost

a l l supervisory care h m s currently have more than eight residents. According to the

blueprint, approximately 900 SMI people l i v e in supervisory care settings in Haricopa

County. To canply with the blueprint, alternative housing w i l l have to be found for

many of these clients by 1995.

different types of residential programs. For example, intensive

residential programs are staffed on a 24- hour basis with a high staff to

client ratio, and provide vocational and other support services;

semi- supervised group l iving arrangements are minimal l y staffed, and

allow clients to function as part of a household and develop their

independence; and supported l i v i n g provides support services to clients

who live on their own. The blueprint directs DHS to develop housing that

w i l l be flexible enough to meet each c l i e n t ' s needs as those needs

change.

Residential f a c i l i t i e s are limited - Arizona has an insufficient number

of residential beds. The blueprint contains estimates of available beds

at the end of fiscal year 1990- 91 and projected needs for each type of

bed. According to the blueprint, at the present time, Arizona has only

235 intensive 24- hour beds, although i t w i l l need 1,145 beds Statewide by

1995. Similarly, the State has only 350 semi- supervised beds, compared

to a need for 4,905 semi- supervised beds by 1995. In addition, the State

has no supported- living beds, and 2,943 w i l l be needed.

se aanaaers rewrted their frustrations - Our discussions with case

managers and other professionals revealed frustrations in connecting SMI

persons with residential services. We met with case managers and other

e n t i t y s t a f f , State hospital and County Annex psychiatric s t a f f , and

patient advocates. The insufficient number of residential beds compared

to the number of patients who need them was mentioned repeatedly in these

interviews. They told us that some residential programs are unwilling to

accept clients with special needs. These programs refuse clients who

might be disruptive, such as those coming out of inpatient

hospi t a l i r a t ion and those w i th substance- abuse problems . Also, some

clients arc, placed in less restrictive environments than are c l i n i c a l l y

recomnended because not enough 24- hou r supe rv i sed beds are ava i l ab l e i n

some areas.

Case managers and others also cited the following specific problems

related to residential placements.

F a c i l i t i e s for the dually diagnosed SMl/ substance abuser are in short

supply. In k r i c o p a County, the only option avai IabIe for

detoxification is LARC, the Local Ambulatory Reception Center, which

is not appropriate for some SMI clients.

The Department of Economic Security's program for the mentally

retarded has a statutory provision that i t can accept people only to

the limits of i t s resources. Thus, some SMI individuals who are also

mentally retarded may wait as long as four years to transfer from ASH

to a faci l i ty