The Supervision of Persons with Mental Illness onProbation Supervision.

By

Edward J. Latessa, Ph.D.

1996

In Arthur J. Lurigio (Ed.) Community Corrections in America: NewDirections and Sounder Investments for Persons with Mental fllness andCodisorders (1996). Seattle: National Coalition for Mental and SubstanceAbuse Health Care in the Justice System

It is with disease of the mind. as withthose of the body; we are haIf deadbefore we understand our disorder.

and half cured when we do.

-Charles Caleb Colton

40

Dan2erous Comnared to Whom?

The purpose of this chapter is to examine offenders with mental illness under correctionalsupervision in the community. Specifically, I present data that compares offenders with mentalillness to other offender groups including, sex offenders, drug offenders, high risk offenders, andoffenders being supervised on standard probation. Published research on mental illness amongprobation populations is limited, and comparative studies are virtually nonexistent. The datapresented in this chapter are unique because they provide a picture of offenders with mental illnessunder probation supervision and compare them to other offender groups that are commonlysupervised in the community. The primary question that I will address is whether offenders withmental illness on probation pose any more of a risk to the community than do other offendergroups.

Correctional Crisis

-- Across the country. states are experiencing significant increases in their prison populations.Recent figures released by the Department of Justice (1994) indicate that there are over 1.1 millionindividuals incarcerated in prison and nearly 500.<XX> in jail. Despite an unprecedented growth innew prison construction. most states still have shortages of prison space.

The ever-increasing prison population still lags behind the probation population. Probationis the most widely used correctional sanction in the United States. with over 2.8 million adultsunder probation supervision and an additional 700,000 on parole (Department of Justice, 1994).

Many states have turned to probation in attempts to alleviate the prison crowding crisis. Forexample, in 1978, Ohio introduced a probation subsidy program that offered local counties fundingto reduce state incarcerations through pilot community corrections programs. The flfSt county totake advantage of this program was Lucas County, which includes the City of Toledo. The originalgrant funded one intensive probation supervision unit designed to monitor prison-bound offenders.Over the years, Lucas County formed a variety of specialized probation units that target particularoffender types, such as offenders with mental illness.

In addition to Lucas County, there are now over thirty counties taking part in Ohio'sCommunity Corrections Program. Among them is the largest county in Ohio, Cuyahoga, whichincludes the City of Cleveland. In 1985, Cuyahoga County started an intensive supervisionprogram, which operates a Mentally Disordered Offender Unit.

The aim of the Lucas County and Cuyahoga County programs is reducing the countyincarceration rate without seriously increasing risk to the community. These two programs are thesource of the data for this chapter.

Probation :li

Probation agencies have always served a wide range of offender groups. In recent years,however, classification instruments have pemlitted them to use offender characteristics in order tofocus on specific groups of offenders. In many of the larger probation departments, the result hasbeen the development of specialized units, which have allowed probation departments to managecaseloads better and to increase services.

The two most common criteria employed by probation departments to distinguish amongoffenders are risk and needs. Risk classification is commonly based on criminal history and

41

Research Metho~

42

Results

Back&IQund Characteristics

Table I summarizes the background characteristics of the five groups, which were recordedat the time of admission to probation. These data indicate that the drug and mentally disorderedgroups had the highest percentages of females, 29% and 20%, respectively, whereas the sexoffender group reported the lowest, 3%. A majority of probationers in the sex offender and highrisk groups were white, and a majority of the mentally disordered, drug, and regular probationerswere nonwhite. The sex offender group was the oldest whereas the high risk probation group wasthe youngest. The educational levels of the groups were similar, with the mentally disordered andthe sex offender groups the most educated, and the high risk and regular probation groups the leasteducated.

A majority of probationers in all five groups were single; however, the mentally disorderedgroup reponed the highest percentage of single persons, 89%. A majority of probationers in allfive groups, except the mentally disordered, also lived with their families. In terms of length ofresidence, the drug group had the least stable residence whereas the regular probation group hadthe most stable.

The sex offender sample was the only group to report a majority employed at the time ofanest (59%). Not surprising, the mentally disordered group reported the lowest percentageemployed (18%). With regard to presentence investigation recommendations, a majority ofoffender in all five groups, except the drug and high risk groups, were recommended forprobation. The mentally disordered group had the highest percentage recommended for probation(74%). Risk assessment data indicated that the high risk sample was the highest risk group,followed by the mentally disordered group. This finding is particularly interesting given the factthat the majority of the mentally disordered sample were recommended for probation.

Criminal Histo~

The criminal histories of the five groups are presented in Table 2. The drug, high risk, andregular probation groups had a more extensive criminal history than the other two groups,especially the sex offender group. The high risk group had the highest percentage of persons withprior state commitments (33%), followed by the sex offender group (23%).

S~ial Problems

I examined the special problems of the five groups at program entry. This is an importantconsideration because these areas are likely to be used to select offenders into one of the units.These data are presented in Figure 1. As expected, the vast majority (89%) of the mentallydisordered group had a psychiatric history. In contrast, the sex offender group was second withonly 25%. Similarly, 47% of the mentally disordered group reponed a prior suicide attempt versus24%, 12%, and 11% of the high risk, drug, and sex offender groups, respectively.

A majority of offenders in all five groups had a history of drug use, ranging from a high of100% in the drug group to a low of 59% in the sex offender group. A history of alcohol abuse waspresent in a majority of the drug, sex offender, and high risk groups (88% for drug and sexoffender groups, and 84% for the high risk group), with the mentally disordered group reportingonly 36%.

43

frobation StatYS

44

drug group was the least likely (38%). The drug group was more likely to be revoked (32%) andto abscond (24%). The mentally disordered group reported 21% revoked and 16% absconded.

Probation status data were collapsed into two categories (success and failure) and arepresented in Figure 9. A majority of three groups were classified as successful: sex offender(78%), mentally disordered (63%), and regular probationers (61 %). The drug and high riskgroups reported more failures than successes.

Probationers with Mental lllness

Although predominately male, the mentally disordered group had a relatively higherpercentage of females, who typically are in their mid-30s with a high school education or better.They are single and, compared to the other groups, less likely to live with their families and to beemployed at admission. They are likely to be recommended for probation and are generallyclassified as high to medium risk. This group is likely to have been on probation previously, and43% had prior felony convictions. They were selected for specialized supervision on the basis oftheir psychiatric histories (e.g., 47% had attempted suicide). A majority had a history of drug use.If arrested. this group was likely to commit a felony; however, overall, they were no more likely tobe arrested or convicted than the regular probation group and were less likely to be arrested orconvicted than the drug or sex offender groups. A high percentage had a technical violation. butthis was also true for every group except the sex offenders. Despite the technical violations for thisgroup, the majority were considered successes (i.e., they were not likely to abscond or berevoked).

Caveats

A number of caveats should be noted concerning these data. First, they were drawn fromtwo probation departments in one state and are not necessarily representative of probationdepanments elsewhere. Thus, the generalizability of these results is limited. Second, thecombination of cases from two probation departments also poses several problems. The selectioncriteria used by these two departments varies and so do their policies and supervision practices.These differences can affect technical violations, absconder rates, and probation outcomes. Size isanother factor that has to be considered. One department services a very large urban area whereasthe other operates in a midsize county. In addition, the treatment and service programs offered inthe counties also differ greatly, which can affect outcomes.

Another imponant limitation is the length of time under supervision for each group. The-mentally disordered group was under supervision for just over one year compared to sixteenmonths for the regular probationer group. Time at risk can be an imponant factor in follow-upstudies. There are some groups, such as sex offenders (Furby, Weinron. & Blackshaw, 1989) andhabitual drunk drivers (Langworthy & Latessa, 1993), that require a considerably longer follow-upperiod to adequately gauge recidivism.6 Whether this is the case for offenders with mental illness isnot known at this time. Finally, the data related to arrests and convictions were limited to officialcrime statistics. Self-repon data were not available. 7

Conclusions

Despite these limitations, the data presented here can help to address a crucial question:Does the supervision of offenders with mental illness in the community pose a greater risk than thesupervision of other offender groups? The answer appears to be, no. These data indicate thatoffenders with mental illness are being supervised in the community without increasing risk to

45

public safety. Offenders with mental illness perfonn as well and often better than drug offenderssex offenders, high risk offenders, and regular probationers. Given these findings, what are theimplications for supervising offenders with mental illness in the community?

As mentioned previously. these two probation departments are not necessarilyrepresentative of other depanments. On the other hand. there is no reason to believe that they areany different from the majority of midsize to large probation agencies across the country. Hence,the following implications for probation practices need to be considered.

Developing a specialized unit for offenders with mental illness on probation necessitatesmore than just the differential classification of offenders. It involves a philosophical shift from ageneralist approach, which requires officers to handle a cross-section of cases irrespective of theirspecial characteristics, to a specialized model, which requires officers to handle one type ofoffender in a specialized caseload.

Many probation administrators committed to the generalist philosophy claim that it issimply too expensive to start and operate specialized units, given the wide availability ofcommunity services. They also claim that, because of widespread staff shonages, it is moreefficient to use generalized supervision strategies and to rely on available community resources toprovide for offenders' special needs. They argue that to have community resources available, andnot to use them, would be inefficient and ill-advised.

46

with mental illness can be undertaken without increasing the risk to public safety, at least whencompared to other offender groups who are being routinely released into the community.

I For an excellent review of the risk factors the mentally ill. particularly as they relate to violent behavior,

see part two of Hodgins. S. (Ed.). (1993). Mental di§Ord~r and crim~. Newbury Park. CA: Sage.

2Those offenders still under supervision were classified as successes since they were meeting all theconditions of their probation at the time the data were collected.

~e contact data does not include group sessions or contacts by external treatment staff.

41t should be noted that each probation agency folIows different guidelines for the filing of technicalviolations. Since data from two probation departments are being utilized it can explain some of the variation in theserates.

S Absconders are those offenders whose whereabouts are unknown. The "Other" category includes those

offenders who transferred to another jurisdiction. dies while under supervision. and incarcerated on a previous charge.

6 All of these groups are still under study and will be followed indefinitely.

'For some offender groups. such as sex offenders. official arrest and conviction data are generally consideredinadequate measures of criminal behavior.

47

References

Brennan, P. A., Mednick, B. R., & Mednick, S. A. (1993). Parental psychopathology, congenitalfactors, and violence. In S. Hodgins (Ed.), Mental disorder and crime (pp. 244-262).Newbury Park, CA: Sage.

Furby, L., Weinrott, M. R., & Blackshaw, L. (1989). Sex offender recidivism: A review.Ps~cholo&ical Bulletin. 105,3-30.

(1993). Mental disorder and crime. Newbury Park. CA: Sage.Hodgins, S. (Ed.

Langworthy, R., & Latessa, E. J. (1993). Treatment of chronic drunk drivers'project. Journal of Criminal Justice, 21, 265-276.

The turning point

Latessa, E. J., Parks, E., Allen, H. E., & Carlson, E. W. (1979). Specialized supervision inprobation: Implications, research, and issues. Prison Journal,.52, 27-35.

Link, B. G., Andrews, H., & Cullen, F. T. (1992). The violent and illegal behavior of mentalpatients reconsidered. American Sociolo~ical Review,.1], 275-292.

Maughan, B. (1993). Childhood precursors of aggressive offending in personality-disorderedadults. In S. Hodgins (Ed.), Mental disorders and crime (pp. 119-139). Newbury Park,CA: Sage. .

Moffitt, T. E. (1984). Genetic influences of garenta1l2s~chiatric illness of vi~lenl and ~_i.~ivis.ticcriminal behavior. Unpublished doctoral dissertation, University of Southern California.

Monahan, J. (1993). Mental disorder and violence: Another look. In S. Hodgins (Ed.), Mentaldisorder and crime (pp. 287-302). Newbury Park. CA: Sage.

National Mental Health Association. (1987). Shima: A lack of awareness and understandinK.Alexandria. VA: National Mental Health Association.

Robins, L. N. (1993). Childhood conduct problems, adult psychopathology. and crime. In SHodgins (Ed.), Mental disorder and crime (pp. 173-193). Newbury Park, CA: Sage.

Taylor, P. J. (1993). Schizophrenia and crime: Distinctive patterns in association. In S. Hodgins(Ed.), Mental disorder and crime (pp. 63-85). Newbury Park, CA: Sage.

U.S. Department of Justice, Bureau of Justice Statistics. (1994, September 11). Probation andgarole gQ~ulatiQns reach new hi&hs.

48

- - - - -

Table 1Background Characteristics of the Drug, Sex.

Mentally Disordered, High Risk, and Regular Probation Groups

DRUGN %

SEX MENTALN %

REGULARN %

HIGH RISKN %

FAcrORN %

SEX973

19147

8020

719158

8218

57571

8911

Male 86Female 35'

RACEWhite 45 37

Non-white 76 63

AGE1~21 6 522-28 41 3429-35 50 4136+ 24 20

EDUCA 110N

< than H.S. 55 46H.S. Grad 53 44Post H.S. 13 10

MARrr AL ST A 11.15

Single 97 80Married 23 19

LIVE WITH FAMILY

Yes 88 73No 33 27

LENGn! OF RESmENCE

< than 1 yr 80 66

1 to 2 yr 13 11>than2yr 28 23

EMPLOYED AT TIME OF ARREST

Yes 31 26No 90 74

PSI RECOMMENDA 110N

Probation 26 22Incarceration 94 78

RISK CLASSIFICA 110N

High 46Medium 61Low 12

71

29

622

83158

34

66

220578

2872

350296

54

464618

7228

2181925

328

3039

21588375

9243532

105

318

240

206

12372824

119245

145133

18382321

483715

490

31569

56368

392

21439

61336

27278

434314

1199237

42

226634

21026

8911

686

1758020

517125

8119

75

2S

4716

7525

118120

4951

637220

7426

484161

561628

3S7

21

561133

1264159

581824

291142367

3618

46

362

100

183

5941

42196

18

82331497

40

60188457

29

713726

178439

2971

3230

5248

16258

74

26491311

6139

395110

203013

324821

117

71

47

503020

254469134

305515

619214

963

>1

49

Table .1

so

~

::::::::::::"'..' 1 (t)

I:::::::::::. :::::::::::,.. 1 e:::::::::::

~:::::::::::E:::::::::::

~(tJ

~..c(J')-(.)~

""""C Q)0 c.>

CW')()":EW~Oa:..c:c:J"'Eo~o~-~~u..~..c

'+-00 ...

c..

cnt-o~z0. 0

-J~'0t-

~.

~

xwfI)

~~\,..~..

<D~i...

a>..cE:J

Z,- .10

53

7"7 j',':Ic

'c".

(X)

(X)~

CI:W0CI:0C/)0-J~ZW~

0

'-

0croQ)

EQ)

"'CC/).-~

ro

"CQ)-().->c:0U

~

(f)or-

~

~1.qowa::::>~-U-

~

(')~

(t)(t)

"CGJ-(/)GJ'-'-

-<

~

~0 ~I

~' -"", -'" ~ -

0(\J

10~

0'p-

LO 00.,.

10C")

0C")

10(\I

S4

~

'+-

0

-cQ)(.).->c:0(.)

-0Q)(/)Q)~~

-<

. . . .I -- - - - ~~ :::::::::::::::::::f

~=>a:0

~C/)

a:~~-~

0 :::::::::::::::::::::::. ~ :::::::::::::::::::::::I :::::::::::::::::::::::

. " " :::::::::::::::::::::::

a:w0a:0Cf)-

0

-J~Zw2

0

~~~

c:0-Q)

u...-'

~

"CQ)-Co)

.-

>

c

0

(,)

';:{

LO

W(I:~~-L.L.

0

~(f)po

,'.. 10 , ... I :::::::::::::::::::::::::::::::::::::

~ , , :::::::::::::::::::::::::::::::::::::

I::::::::::::::::::::::::::::::::::::, .. .. ... ...(\I(\I

"CQ)-cnQ)'-'-

-<

0),...

CSI~0

;" ~;...;:.,

NI

I1,,- I"," I ' I r "

10 0 10 0 10 0C\I ~ ... ~

55

\t-

o

"'CQ)

(.).->c0

()

"'CQ)cnQ)~~

<1:

xwrn

~;ja:c

~<..J~~w~

~cn

a:

:%:~

:%:

"

to... ::::::::::::::::::::~ ,

a:wca:0a)C-J~ZW~

D

K'c~Cc.o(\I

"CQ)-0.->~0C.)

(C

w(I:::)~-LL.

~IId.J

~I

~

~I~<-J~~w~

..

.. :.::::::::::::::::..

)(wCI)

~

10C\J

~ -

~0-~...

.2

~I"CQ)...U)Q)~~

<

~II

J ~cn-cc::r:~%

~~a:0

I

~I""'I I' 1 ~ I '

0 0 0 0 0 0U) (f) ,N 9118

56

7",~. -==-~-_.~'" .~

-cQ)

~0.->c:0

U

-cQ)

~(/)Q)~~

~

100

%C

harged

45

FIG

UR

E 7

~

SE

X

23Tech V

iolation

57

44/:r;

43CJ

ME

NT

AL D

ISO

RD

ER

t.. --

, ,

:::::::::::::::: /

£. Violation

41

..~

. -

.c.ti

, ,

jt~ tI

DR

UG

with

a Technical

ngm R

EG

ULA

RH

IGH

R

ISK