Effectiveness of community treatment orders

for treatment of schizophrenia with oral or

depot antipsychotic medication: changes in
problem behaviours and social functioning

Graham Ingram, David Muirhead, Carol Harvey

Objective: Involuntary outpatient commitment (IOC) has been in use in various countries
for a number of years and has recently been implemented (in the form of supervised
community treatment) in England and Wales. Several studies indicate that IOC reduces
relapse and readmission rates and decreases length of stay on inpatient units in patients
diagnosed with schizophrenia. The aim of the present study was to examine whether the
use of IOC in the Australian context, in the form of community treatment orders (CTOs),
may be associated with a reduction in problem behaviours and improved social
functioning.
Method: A naturalistic retrospective mirror image study of case notes, with each case
serving as its own control, was used. Behavioural and social outcomes were examined:
episodes of aggression and suicidal and self-harming behaviour, episodes of homelessness,
frequency of contact with family members and overall quality of relationship between family
and patient, and employment status.
Results: Ninety-four sets of case notes were identified as meeting the criteria for inclusion.
The number of episodes of aggression was found to be halved from the year before the
CTO to the subsequent year (p  0.0001). Significant reductions in the number of episodes
of homelessness were experienced by patients (p  0.05) when the pre-CTO year was
compared with the CTO year.
Conclusion: A CTO may contribute to improved outcomes related to patient quality of life.
This may be seen to mitigate concerns about infringement of civil rights.
Key words: community care, community mental health, involuntary outpatient commitment,
mental health service, schizophrenia.

Australian and New Zealand Journal of Psychiatry 2009; 43:10771083

We previously reported the clinical effectiveness of com-

Graham Ingram, Consultant Psychiatrist (Correspondence)
St Georges Hospital, St Georges Park, Morpeth, Northumberland NE61
2Nu, UK. Email: graham.ingram@ntw.nhs.uk
David Muirhead, Director of Clinical Services
North West Area Mental Health Service, Melbourne, Victoria, Australia
Carol Harvey, Associate Professor in Psychiatry, University of Melbourne,
Melbourne, Victoria, Australia; Consultant Psychiatrist, North West Area
Mental Health Service, Melbourne, Victoria, Australia
Received 30 September 2008; accepted 19 May 2009.
munity treatment orders (CTOs) in terms of reduced
relapses and readmission rates, decreased length of stay,
and increased contact with Community and Crisis Teams
in two groups of patients with schizophrenia [1]. These
groups consisted of patients on depot or oral antipsychotic
medication. One factor contributing to poor outcome in
community settings is lack of adherence to prescribed
treatment. A response to non-adherence is the use
of legislation-based involuntary outpatient commitment

2009 The Royal Australian and New Zealand College of Psychiatrists

1078 INVOLUNTARY OUTPATIENT COMMITMENT
[IOC) [24]. In England and Wales an amendment to the
Mental Health Act (1983) was made and in October 2008
an IOC in the form of supervised community treatment
(SCT) was introduced into clinical practice [5]. Despite
such increasing use of IOC, outcomes that may be associ-
ated with improved quality of life such as a reduction in
problem behaviours and improved social functioning in
patients subject to CTOs are rarely examined.
Behavioural change and quality of life
Increased suicide rates among patients with schizo-
phrenia have been a consistent nding in the literature.
A meta-analysis by Palmer et al. in 2005 estimated
that 4.9% of people with schizophrenia commit suicide
during their lifetime, usually near the illness onset [6].
Further, deliberate self-harm is a strong predictor of
suicide among patients with schizophrenia [6].
Schizophrenia is now well established as a risk factor
for violence. Walsh et al. concluded that 10% of societal
violence may be attributable to schizophrenia [7]. The
prevalence of community violence in discharged patients
has been estimated to be 9% [8]. Swanson et al. found a
reduced incidence of violent behaviour among patients
subject to IOC [9].
Reduction in episodes of deliberate self-harm and suicide
and of violent or aggressive behaviours among people with
schizophrenia would be important indicators of improved
outcomes of patients with schizophrenia subject to CTOs.
Social functioning and quality of life
Studies of rates of schizophrenia among homeless per-
sons give ranges between 15% and 46% [10,11]. Schizo-
phrenia was found to be a signicant risk factor for
homelessness by Folsom et al. [12]. Compton et al., in a
sample of 204 patients with serious mental illness, found
a signicant short-term decrease in the risk of homeless-
ness in patients subject to IOC versus those who were not
[13]. Compared with homeless people with mental ill-
ness, those who have stable accommodation were not
only more satised with their living arrangements, they
were also more satised with family relations, nances
and daily activities [14]. In a study by the same group,
persons with severe mental illness who were treated
assertively spent 31% more days in stable housing com-
pared with those receiving care by usual services [15].
Based on these studies, the authors would hypothesize
that a CTO may enable more assertive treatment for
patients who would otherwise refuse treatment and this
may positively impact on their accommodation status.
Family relationships may be regarded as an important
outcome indicator in patients treated with IOC, because
many patients reside with their families. It is plausible that,
when CTOs are clinically effective, this may enable family
relationships to be repaired. Mullen et al. in qualitative inter-
views with 27 families of patients with serious mental ill-
ness subject to compulsory treatment orders, found that
families were generally in favour of compulsory community
treatment due to perceived positive inuence on their rela-
tive, on themselves, on family relationships, and on relations
with the clinical team [16]. The authors were unable, how-
ever, to identify studies that examine family relationships
among involuntary patients in the community.
High unemployment among persons with severe men-
tal illness is well recognized. An unemployment rate
of  75% in a sample of 1664-year-old Australians with
psychosis has been reported [17], suggesting that better
occupational and vocational services may benet this
group. The value of employment in terms of improved
self-esteem and quality of life among people with severe
mental illness is increasingly recognized [18].
In this paper we aim to examine behavioural and
social outcomes of a group of patients with schizophre-
nia subject to a CTO. This sample was described in a
previous publication [1]. Information was gathered from
case notes that might provide external indicators of the
social functioning and problem behaviours of patients
with schizophrenia before and after the instigation of
a CTO. These indicators included recorded episodes of
aggression and suicide attempts as well as documented
periods of homelessness and employment and docu-
mented comments regarding frequency and quality of
interaction between the patient and family members.
We hypothesized that CTOs reduce problem behaviours
such as self-harm and attempted suicide and violence,
and that CTOs would improve social functioning through
reduced periods of homelessness, higher employment
and improved quality of family relationships.
Methods
This was a naturalistic retrospective mirror image study of case notes
with each case serving as its own control. Behavioural and social out-
comes of patients diagnosed with schizophrenia and treated under
a CTO were examined. The study was conducted in three mental
health services covering urban and suburban areas in metropolitan
Melbourne. Eligible patients attended any of these services, were
diagnosed with schizophrenia, and placed on a CTO during the study
period. Cases were identied by a manual search of Mental Health Act
reports of current and past patients placed on CTOs. The sample was
divided into two subgroups consisting of patients treated with either
oral or depot antipsychotic medication. A more detailed description of
the method is provided by Muirhead et al. [1], including inclusion and
exclusion criteria.
 G. INGRAM, D. MUIRHEAD, C. HARVEY
Data collection
Data were collected by retrospective medical record review for the
period of 12 months before the implementation of a CTO (the pre-CTO
year) and for the period of 12 months following the implementation of
the CTO (the CTO year). A data collection instrument was developed
specically for this study. Variables were selected that would give the
most reliable indicators of outcome considering the potential unreli-
ability of medical records, and including those deemed to be important
in outcome research in related studies. Variables included number
of episodes of aggression and suicidal and self-harming behaviour,
number of episodes of homelessness, frequency of contact with family
members and the overall quality of relationship between family and
patient, as judged using all available information, and nally, employ-
ment status. Data were extracted from the notes guided by the following
parameters: (i) episodes of aggressive behaviour included physical
force used against others or property, intimidating behaviour and verbal
aggression sufcient to cause others to be fearful of their safety; (ii)
suicidal or self-harming behaviour was recorded but not suicidal ide-
ation due to perceived lack of reliability of recording of this latter infor-
mation in the medical records; (iii) homeless episodes were recorded
when there was a basic lack of shelter or patients were living in tem-
porary or emergency accommodation; (iv) the overall quality of the
relationship between the patient and their family was assessed, making
use of documented observations such as whether the patient was par-
ticipating in, or withdrawing from, family interaction and activities; (v)
frequency of contact with family was estimated from reports in the
clinical le; and (vi) best employment status whether unemployed,
voluntary, part time or full time work was recorded.
Interrater reliability
Two researchers were involved in collecting the data for this study
(GI, DM). Data collection from case note studies can be notoriously
unreliable. In order to estimate the reliability of the data that were col-
lected, the researchers both independently assessed a randomly chosen
10% of notes previously assessed by the other. These two sets of data
were used to calculate interrater reliability scores (expressed as either
Spearmans or Kendalls ).
Statistical analysis
All data were analysed using SPSS version 14.0 (SPSS, Chicago, IL,
USA). For all results p  0.05 was considered statistically signicant.
Repeated-measures tests were used to compare data from the pre-CTO
year with the data from the CTO year for the total group, the rst
experimental subgroup (oral medication) and the second experimental
subgroup (depot or combination medication). The distribution of the
bulk of the ordinal measures was highly skewed and the Wilcoxon
signed rank test was therefore used to determine statistical signicance.
The feasibility of performing a repeated-measures ANOVA with time
(pre-CTO vs CTO year) as a within-subject variable, and group (depot
or not) as a between-subject factor, was explored. Because the data were
1079
highly skewed, an attempt was made to transform the relevant variables.
Only one of these six variables (episodes of aggression in pre-CTO
year) could be transformed by any method (log, square root and inverse
of zero methods). Because we were unable to adequately full the
underlying assumptions, we did not proceed with this analysis. Sig-
nicant changes in categorical repeated measures were investigated
using the marginal homogeneity test because several cells had low num-
bers.
Results
Of the sampling frame of 212 patients placed on a CTO for at least
10 months, 94 were identied as meeting the inclusion criteria. The
majority of patients were excluded because they did not have a diag-
nosis of schizophrenia [1].
An overview of the characteristics of the entire patient group and the
two subgroups is documented in our previous publication [1]. In brief,
the mean age of the total sample was 39 years (range  1866 years).
Most were male, almost two-thirds had never married, and just over
half lived with families. Mean length of illness for the total sample was
13 years (range  139 years). Secondary diagnoses within the sample
were relatively rare; for example, 16 (17.0%) had a secondary diagno-
sis of alcohol misuse and 29 (30.1%) had a diagnosis of substance
misuse. Comparisons between the two subgroups (oral and depot) on
demographic and clinical variables showed no statistically signicant
differences, except for a signicantly higher antipsychotic dose pre-
scribed to the oral subgroup in the CTO year (Mann Whitney Z  2.608,
p 0.009) [1].
When comparing the ratings made by the two researchers to assess
interrater reliability, the following correlation coefcients were
obtained: for aggressive episodes in the pre-CTO year,  0.71 and in
the CTO year,  0.81; for suicide attempts in both the pre-CTO and
CTO years,  1; for episodes of homelessness in the pre-CTO
year,  0.8, and in the CTO year,  1. The interrater reliabilities for
the remaining measures were: frequency of family contact in the pre-
CTO year and the CTO year, 0.96; quality of family contact in the
pre-CTO year,  0.05, and in the CTO year,  0.32; and employment
status in the pre-CTO year,  1, and in the CTO year,  0.66. Qual-
ity of family contact was excluded from further consideration due to
poor interrater reliability.
There was a signicant reduction in the number of episodes of
aggression (by approx. half in each instance) when comparing the pre-
CTO and CTO year in the total patient group (Z  3.494, p  0.0001),
the oral group (Z  2.157, p  0.05) and the depot group (Z  2.707,
p 0.01; Table 1). The number of suicide attempts made during the year
of CTO was not reduced in any of the three groups. There was a sig-
nicant reduction in the number of episodes of homelessness in the
total patient group (Z  2.111, p  0.05), but signicance was not
achieved for the oral (Z  1.732, p  0.083) and depot subgroups
(Z 1.414, p  0.157) alone.
There was no improvement in the frequency of contact between
patient and family for the total, oral or the depot group (Table 2).
1080 INVOLUNTARY OUTPATIENT COMMITMENT

Table 1. Aggression, suicidal behaviour and homelessness before and during the CTO year

Wilcoxon signed rank test

Pre-CTO year CTO year Z P

Total sample
No. violent aggressive episodes
Mean 0.98 0.50 3.494 0.0001
n 94 94
Median 1 0
No. suicide or self-harm attempts
Mean 0.26 0.16 0.963 0.336
n 94 94
Median 0 0
No. episodes of homelessness
Mean 0.14 0 2.111 0.035
n 94 94
Median 0 0
First subgroup (oral antipsychotics)
No. violent aggressive episodes
Mean 1.23 0.65 2.157 0.031
n 31 31
Median 1 0
No. suicide or self-harm attempts
Mean 0.19 0.006 1.414 0.157
n 31 31
Median 0 0
No. episodes of homelessness
Mean 0.13 0.003 1.732 0.083
n 31 31
Median 0 0
Second subgroup (depot antipsychotics)
No. violent aggressive episodes
Mean 0.86 0.43 2.707 0.007
n 63 63
Median 0 0
No. suicide or self-harm attempts
Mean 0.29 0.21 0.403 0.687
n 63 63
Median 0 0
No. episodes of homelessness
Mean 0.14 0.008 1.414 0.157
n 63 63
Median 0 0
CTO, community treatment order.

Finally, there was little change in employment status for all three
groups, although there was a trend towards fewer patients having full-
time work in the total and depot subgroups during the year of operation
of the CTO (Table 3).
Discussion
Strengths and limitations of the study
The strengths of this study include that it is a real-
world study of rarely examined outcomes that may be
associated with the use of CTOs. The study was con-
ducted in a setting in which the use of CTOs is well-
established. Further, there were few study exclusions.
Limitations include the mirror image, retrospective
design using relatively low numbers of case notes to
obtain data. Also, some of these data are open to assessor
bias. Interrater reliability for some outcomes is poor
(notably quality of family relationships). We also note
the relatively low rate of comorbid conditions (notably
alcohol and drug abuse); there may be a problem in iden-
tifying comorbid conditions retrospectively from case
notes due to poor documentation. There may be problems
 G. INGRAM, D. MUIRHEAD, C. HARVEY 1081

Table 2. Frequency of contact with family before and during the CTO year

Pre-CTO year CTO year MH statistic p

Total sample
Living with family 50 51 1.291 0.154
More than weekly 10 13
Every 14 weeks 16 17
Less than monthly 9 6
None 5 6
Unknown 4 1
Total 94 94
First subgroup (oral antipsychotics)
Living with family 16 16 1.000 0.313
More than weekly 4 5
Every 14 weeks 6 6
Less than monthly 3 2
None 2 2
Unknown 0 0
Total 31 31
Second subgroup (depot antipsychotics)
Living with family 34 35 0.905 0.281
More than weekly 6 8
Every 14 weeks 10 11
Less than monthly 6 4
None 3 4
Unknown 4 1
Total 63 63
CTO, community treatment order; MH, marginal homogeneity.

of regression to the mean affecting the outcomes, as
outlined by Churchill et al. [19].
We have previously reported that CTOs may be effec-
tive in reducing relapse rates and time spent in hospital
for selected persons with schizophrenia [1]. These
results might be seen as suggestive of improved quality
of life but provide only indirect support for such a
notion. The present study examines whether there is
improvement in measures that might provide prelimi-
nary evidence of enhanced social outcome and a reduc-
tion in problem behaviours (which are suggestive of
enhanced quality of life). The present ndings suggest
that reduced periods of homelessness and reduced epi-
sodes of violence occur during treatment with a CTO
compared with the preceding year without CTO treat-
ment. Past justication for the use of CTOs has been
sought on the grounds of decreasing risk (2022),
reducing relapses and reducing time spent in hospital,
but these ndings provide an argument for their poten-
tial value in improving social outcomes as well as
reducing problem behaviours.
The present study showed a signicant reduction in
violence in all three patient subgroups during the CTO
year. This replicates ndings in other studies (9,23).
Further, the New York State Ofce of Mental Health audit
found a reduction in harmful behaviours in the rst
6 months of a CTO [24].
Patients with schizophrenia had a signicant reduction
in homelessness in the CTO year. This is consistent with
the ndings of the New York State Ofce of Mental
Health audit [24]. Homelessness is quite often associated
with unemployment, poverty and social isolation, and par-
ticularly estrangement from family. It should be noted that
the majority of patients in the present study lived with
their families. This may also explain the lack of signi-
cant increase in family contact during the CTO year.
As we have previously reported [1], this postulated treat-
ment effect of using a CTO might be mediated by increased
clinical contact with the patient as well as increased medi-
cation adherence. We suggest that this might facilitate
earlier intervention in the event of relapse, enabling thera-
peutic intervention before violence and other adverse
events have occurred, including homelessness.
Episodes of self-harm and suicidal behaviours and
changes in employment showed no improvement when
comparing the pre- and post-CTO year. One explanation
may be the low incidence of such events and a sample
size that was insufcient to demonstrate any positive ben-
et. The slight worsening of overall employment status
among the patients in the present study during the CTO
year requires further investigation. It is plausible that re-
gaining employment requires interventions in addition to
the treatments that are most closely linked with the use
of a CTO.
1082 INVOLUNTARY OUTPATIENT COMMITMENT

Table 3. Best employment status before and during the CTO year

Pre-CTO year CTO year MH statistic p

Total sample
Full-time work 10 5 1.410 0.098
Part-time work 10 11
Voluntary work 1 1
Unemployed 73 77
Total 94 94
First subgroup (oral antipsychotics)
Full-time work 3 2 0.000 0.594
Part-time work 3 4
Voluntary work 0 1
Unemployed 25 24
Total 31 31
Second subgroup (depot antipsychotics)
Full-time work 7 3 1.588 0.073
Part-time work 7 7
Voluntary work 1 0
Unemployed 48 53
Total 63 63
CTO, community treatment order; MH, marginal homogeneity.

Ethical controversy has prevented the introduction of
CTOs in Europe, although their introduction in England and
Wales has occurred following their introduction in Scotland
in 2005. A review for the UK Department of Health of
international experience with CTOs highlighted that evi-
dence of a positive treatment effect, for example reduction
in length of stay in hospital, may not provide sufcient jus-
tication for CTO use without an additional appraisal of the
associated benets and harms [19]. The lack of robust ran-
domized controlled trials as highlighted by Kisely et al. and
Kisely and Campbell emphasizes that good-quality studies
assessing efcacy of CTOs are lacking [25,26]. But as
OReilly pointed out, this attitude may be slightly nave in
that assessment of CTOs hardly compares with assessment
of a drug effect and is a more complex process requiring a
more creative and exible approach to methodology [27]. A
further problem complicating comparison of studies is that
the legislative framework supporting CTOs and the powers
to intervene conferred on mental health services vary sig-
nicantly between the different countries or jurisdictions
where they have been introduced. Additionally there is sub-
stantial variation in the clinical procedures involved with
how CTOs are implemented in clinical practice. The authors
suggest that the evidence within the present study might be
seen to mitigate concerns about the infringement of civil
rights involved in the implementation of involuntary treat-
ment in the community [24].
Conclusion
The present study reports on the reduced homelessness
and reduced violent episodes that may be associated
with the use of a CTO. The argument that CTOs may be
effective is reinforced and could be seen to counterbal-
ance concerns about infringement of civil rights. Further
research into the impact of involuntary community treat-
ment on quality of life is required, making use of larger
samples and, where practicable, prospective and random-
ized designs.
References
1. Muirhead D, Harvey C, Ingram G. Effectiveness of community
treatment orders for treatment of schizophrenia with oral or depot
antipsychotic medication: clinical outcomes. Aust N Z J Psychiatry
2006; 40:596605.
2. McIvor R. The community treatment order: clinical and ethical
issues. Aust N Z J Psychiatry 1998; 32:22332228.
3. Brooks AD. Outpatient commitment for the chronically mentally
ill: law and policy. New Directions Ment Health Serv 1987; 36:
117128.
4. Mulvey E, Geller JL, Roth LH. The promise and peril of involun-
tary outpatient commitment. Am Psychol 1987;42:571584.
5. Department of Health. Mental Health Bill: Amending the Mental
Health Act 1983. UK: Department of Health, 2007.
6. Palmer BA, Pankratz VS, Bostwick JM. The lifetime risk of suicide
in schizophrenia: a re-examination. Arch Gen Psychiatry 2005;
62:247253.
7. Walsh E, Buchanan A, Fahy T. Violence and schizophrenia:
examining the evidence. Br J Psychiatry 2002; 180:490495.
8. Monahan J, Appelbaum P. Effective prevention of crime and vio-
lence among the mentally ill. In: Hodgins S, ed. Reducing violence
risk: diagnostically based clues from the MacArthur Violence Risk
Assessment Study. Dordrecht: Kluwer Academic Publishers, 2000;
1934.
9. Swanson JW, Swartz MS, Borum R, Hiday VA, Wagner HR, Burns
BJ. Involuntary out-patient commitment and reduction of violent
 G. INGRAM, D. MUIRHEAD, C. HARVEY
behaviour in persons with severe mental illness. Br J Psychiatry.
2000; 176:324331.
10. Kermode M, Crofts N, Miller P, Speed B, Streeton J. Health indi-
cators and risks among people experiencing homelessness in Mel-
bourne, 19951996. Aust N Z J Public Health 22;1998:464470.
11. Teesson M, Hodder T, Buhrich N. Psychiatric disorders in home-
less men and women in inner Sydney. Aust N Z J Psychiatry 2004;
38:162168.
12. Folsom DP, Hawthorne W, Lindamer L et al. Prevalence and
risk factors for homelessness and utilization of mental health
services among 10,340 patients with serious mental illness in a
large public mental health system. Am J Psychiatry 2005; 162:
370376.
13. Compton SN, Swanson JW, Wagner HR, Swartz MS, Burns BJ,
Elbogen EB. Involuntary outpatient commitment and homelessness
in persons with severe mental illness. Ment Health Serv Res 2003;
5:2738.
14. Lehman AF, Kernan E, DeForge BR, Dixon L. Effects of homeless-
ness on the quality of life of persons with severe mental illness.
Psychiatr Serv 1995; 46:922926.
15. Lehman AF, Dixon L, Hoch JS, Deforge B, Kernan E, Frank R.
Cost-effectiveness of assertive community treatment for homeless
persons with severe mental illness. Br J Psychiatry 1999;174:
346352.
16. Mullen R, Gibbs A, Dawson J. Family perspective on community
treatment orders: a New Zealand study. Int J Soc Psychiatry 2006;
52:469478.
17. Waghorn G, Chant D, White P, Whiteford H. Delineating disa-
bility, labour force participation and employment restrictions
among persons with psychosis. Acta Psychiatr Scand 2004;
109:279288.
1083
18. Mueser KT, Becker DR, Torrey WC et al. Work and nonvocational
domains of functioning in persons with severe mental illness: a
longitudinal analysis. J Nerv Ment Dis 1997; 185:419426.
19. Churchill R, Owen G, Hotop M, Singh S. International experiences
of using community treatment orders. London: Department of
Health, Institute of Psychiatry, Kings College, 2007.
20. Swartz MS, Burns BJ, George LK et al. The ethical challenges of
a randomized controlled trial of involuntary outpatient commit-
ment. J Ment Health Admin 1997; 24:3543.
21. Swartz MS, Swanson JW, Wagner HR, Burns BJ, Hiday VA. Effects
of involuntary outpatient commitment and depot antipsychotics on
treatment adherence in persons with severe mental illness. J Nerv
Ment Dis 2001; 189:583592.
22. Fernandez GA, Nygard S. Impact of involuntary outpatient
commitment on the revolving-door syndrome in North Carolina.
Hosp Community Psychiatry 1990; 41:10011004.
23. Power PJR. Outpatient commitment: is it effective? Melbourne:
University of Melbourne, 1999.
24. New York State Ofce of Mental Health. Kendras law: nal report
on the status of assisted outpatient treatment. New York State:
Ofce of Mental Health, 2005.
25. Kisely S, Campbell LA, Preston N. Compulsory community
and involuntary outpatient treatment for people with severe
mental disorders. Cochrane Database Syst Rev 2005; (3):
CD004408.
26. Kisely S, Campbell LA. Community treatment orders for psychi-
atric patients: the emperor with no clothes. Can J Psychiatry 2006;
51:683685; discussion 691.
27. OReilly RL. Community treatment orders: an essential therapeu-
tic tool in the face of continuing deinstitutionalization. Can J
Psychiatry 2006; 51:6868; discussion 689690.
Copyright of Australian & New Zealand Journal of Psychiatry is the property of Taylor & Francis Ltd and its
content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for individual use.