Coercion in Outpatients under Community Treatment Orders: A Matched Comparison Study

Abstract

Objective:

Since the deinstitutionalization of psychiatric services around the world, the scope of outpatient psychiatric care has also increased to better support treatment access and adherence. For those with serious mental illness who may lack insight into their own illness, available interventions include coercive community practices such as mandated community treatment orders (CTOs). This paper examines the perceptions of coercion among service users treated with a CTO.

Method:

We used a cross-sectional comparative design where service users treated under a CTO were matched to a comparison group of voluntary psychiatric outpatients. Both groups were receiving intensive community mental health services (n = 69 in each group). Participants were interviewed using a series of questionnaires aimed at evaluating their perceptions of coercion and other aspects of the psychiatric treatment.

Results:

The level of coercion reported by service users treated under a CTO was significantly higher than that in the comparison group. However, in adjusted analyses, service users’ perception of coercion, irrespective of their CTO status, was directly correlated with their previous experience with probation and inversely correlated with the sense of procedural justice in their treatment.

Conclusions:

Evaluation of psychiatric service users’ experiences of coercion should consider their past and current involvement with other types of coercive measures, particularly history of probation. Clinicians may be able to minimize these experiences of coercion by incorporating procedural justice principles into their practice.

Introduction

Since the deinstitutionalization of individuals with severe mental illness around the world in the 1960s, many mental health care interventions have moved to outpatient settings. This in turn led to the incorporation of a continuum of coercive measures in the community psychiatric treatment of these individuals, including compulsory Community Treatment Orders (CTOs). CTOs aim to ensure treatment compliance in service users who may otherwise refuse treatment. These individuals may be frequently involuntarily hospitalized, and at times be considered at risk of harming themselves or the public. CTOs have been in use in various jurisdictions since the 1960s. The efficacy and ethics of CTOs has been a topic of intense debate. Two randomized control trials comparing CTOs and standard care^1,2 and another comparing CTOs with extended leave,³ have not managed to increase consensus on this debate, in part due to their significant methodological shortcomings.^4–7 CTOs are complex interventions involving legal, medical, and social services systems that are unique to each jurisdiction.^8,9 Hence, conducting and interpreting CTO studies are beset by difficulties. Furthermore, the suitability of randomised controlled trials (RCTs) in evaluating complex interventions, such as CTOs, has been questioned.¹⁰ Bearing this in mind, there is some evidence for the positive effects of CTOs on frequency or length of psychiatric hospitalisations, and improved treatment adherence in several pre-post and large-scale longitudinal cohort studies.^11–13

There has never been a RCT of CTOs in Canada. All 4 Canadian quantitative studies using pre-post or case-controlled designs support a positive effect for CTOs, including a reduction in the number of hospitalisations.^14–17 However, these types of studies have their own methodological limitations.¹⁸

Outcome studies often narrowly focus on involuntary re-hospitalisations and length of stay in hospital as the most relevant indicators for measuring the effectiveness of CTOs, but research on recovery points to a more complex view of treatment success, which incorporates issues of quality of life, self-worth, and agency.¹⁹ Hence, understanding service users’ experiences of their treatment under a CTO is important. The current literature indicates that CTOs tend to contribute to an increased sense of coercion;⁹ although, there are significant variations between study findings. For example, some studies do not report a statistically significant difference in perceptions of coercion among participants with or without a CTO,^1,26 with both CTO and their comparison group counterparts reporting high levels of perceived coercion. On the other hand, some studies report statistically significantly higher levels of perceived coercion in the CTO v. the comparison group,²⁷ but with a low overall level of perceived coercion in both groups. One possible explanation for these differences may be that CTO legislation and enforcement as well as health care, legal, and social service systems vary greatly across different jurisdictions. The concurrent use of additional types of coercive measures, such as financial guardianship, may also play an important role in service users’ overall perception of coercion.²⁰ Additionally, there are some indications that procedural justice, or the process of ‘including people in just and fair decision making’²⁷ is inversely correlated to perceived coercion among service users treated with CTOs.²⁹

Qualitative studies in this field are limited in Canada.^21,22 However, similar to other jurisdictions, they show that service users tend to find the orders coercive or may view them as an obstacle to their recovery^21–25 but they are preferred to psychiatric hospitalization and sometimes seen as beneficial; for example, in connecting individuals to necessary services.^21,23–25

CTO legislation was introduced in Ontario, Canada, in 2000 with the objective of ensuring public safety as well as providing care to individuals with severe mental illness.²⁸ The Ontario CTOs are valid for up to 6 months at a time and are initiated by a physician. Explicit criteria must be met before a CTO can be initiated including a minimum of 2 psychiatric admissions or a psychiatric admission lasting for at least 30 days within the past 3 years. Additionally, CTOs can be issued for anyone who has been on a CTO in the previous 3 years. A key element of the Ontario CTO legislation is consent. A physician cannot impose a CTO unilaterally: the individual or the designated substitute decision maker (SDM) must provide informed consent. Moreover, there are various mechanisms in place in Ontario for individuals to contest a finding of incapacity for psychiatric treatment and/or their CTO(s).

We designed a cross-sectional comparative study to assess the perception of coercion among service users treated under a CTO compared to a matched comparison group of voluntary psychiatric outpatients and examined the potential predictors of perceived coercion. Both groups received either intensive case management (ICM) or assertive community treatment (ACT) services.

Method

Sample

The inclusion criteria included: age (18 to 65 years), ability to speak English and give informed consent for participating in a survey. Participants in the CTO arm were subjected to a CTO within the previous 18 months and participants in the control arm had never been placed on a CTO. CTO participants were matched to controls based on gender, primary diagnosis, and age (±8 years). Efforts were made to match on self-identified racial/ethnic identity; however, only 73.9% of the final sample was matched on this variable. Primary diagnosis was self-reported, and consent was also provided by 125 (81.5%) participants to cross-validate their self-reported diagnosis with their chart diagnosis. There were discrepancies in 13 cases (across both groups). Where discrepancies between self-report and the chart review occurred, the diagnosis from the chart review was used.

Participants were drawn from 3 mental health centres in inner city Toronto. Individuals on CTOs were identified by an administrative coordinator who maintains a database of CTO service users. Potential participants were informed by their community mental health teams of the research and offered the contact information of the research coordinator for additional information or, alternatively, participants gave permission to have their contact information shared with the research coordinator. Participants in the comparison group were selected from the same ICM or ACT service providers based on the matching criteria upon referral by their case managers. Interviews were conducted by an experienced research assistant, who obtained the participants written informed consent before the interview. Research ethics approval was obtained from the research ethics boards of St Michael’s Hospital and the Centre for Addiction and Mental Health.

During this study, case managers referred 189 individuals who met participant criteria to the research team. From this group, 13 individuals declined to participate and 3 could not be reached. From the remaining 166 potential participants, 14 did not meet the inclusion criteria. Finally, 152 face-to-face structured interviews were conducted (79 in the CTO group and 73 in the comparison group). Of these, 14 participants were excluded due to withdrawal from the study (n = 2), incorrect identification as a match by their case managers (n = 8), and being matched to participants who withdrew or did not meet study criteria (n = 4). From this group, 138 participants, 69 in each arm, were included in the final analysis.

Measures

Colorado symptom index (CSI)

CSI was used to query the presence of symptoms of mental illness. Possible scores ranged from 14 to 70, with higher scores representing greater frequency of symptoms.³⁰ A score of 30 has been proposed as the “clinical cut off” for CSI, where it can identify individuals who are likely in need of mental health services.⁴⁰

Quality-of-life scale

A 20-item measure, developed by Lehman and colleagues,^31,32 was used to measure quality-of-life for people with psychiatric disabilities.

Procedural Justice and Perceived Coercion

Questions and instruments were adapted from a study by Lidz and collegues.³³

Procedural justice

Four questions related to various aspects of the construct of procedural justice in relation to the service user’s primary case manager/treating team were asked:

1. Motivation of service provider (‘To what extent did s/he do what s/he did out of concern for you?’),

2. Respect (‘How much respect did s/he treat you with?’);

3. Validation (‘How seriously did s/he consider what you had to say?’), and

4. Fairness (‘How fairly did s/he treat you?’).

Response options for these items ranged between 1 = very much/ fairly and 4 = not at all, with lower scores representing greater procedural justice. We also used a single global measure of Voice (‘How much of a chance did you have to say what you wanted to about your mental health treatment?’), with a 4-point Likert scale ranging from 1 = very to 4 = none. The sum of these five items was used to create an index of procedural justice with possible scores ranging from 5 to 20. Due to the skewed nature of the distribution, this variable was transformed in analysis and the scoring reversed so higher scores represent greater levels of perceived procedural justice. The construct of deception was also included and measured by a yes/no question (‘Did anyone try to trick you, lie to you or fool you into receiving treatment?’).³³ Participants were also asked about different ‘negative pressures’ that may have been used by others in order to influence their treatment including persuasion, inducement, threats, or force.³³

Perceived Coercion

Perceived coercion was measured using 4 items:

1. Influence (‘What has more influence on your mental health treatment: what you want or others want?’),

2. Control (‘How much control do you have over your current mental health treatment?’),

3. Choice (‘How much is your current mental health treatment based on your choice?’), and

4. Freedom (‘How free did you feel to do what you wanted regarding decisions about your current community mental health treatment?’)

These questions were measured individually on a Likert scale ranging from 1 to 5, with higher values indicating more perceived coercion. The items were summed to create an index, with scores ranging from 5 to 20.³³

Additional measures

Participants were also asked about their involvement with the legal system, diagnosis, current treatment, perceived need for treatment, substance use, housing situation and past history of hospitalization.

Statistical Analysis

Pearson’s Chi-square test was used to assess for differences between groups on categorical variables and independent t-tests for normally distributed continuous variables. The Mann–Whitney U test was used to compare ordinal variables and variables with distributions deviating from normality. Continuous and dichotomous variables were correlated with scores of participants’ perceptions of coercion using Pearson’s product moment correlation. Variables significant at P < 0.05 were included in a multiple linear regression model.

An a priori power analysis was conducted in G*Power. Using an alpha of 0.05, a power of 0.80 and a medium effect size, 64 subjects were needed in each group to test mean differences and a sample of 139 was need for a multiple regression model with 15 predictors.³⁵

Before the analysis, variables were examined for missing values and fit between their distributions and the assumptions for multiple linear regression. Five participants (3.6%) had missing values on the procedural justice scale and 7 participants (5.1%) had missing values on the CSI. The median for all cases was used to replace missing values on these scales. A logarithmic transformation was used on the CSI and a reflected logarithmic transformation was used on the procedural justice scale to reduce the number of outliers and to improve normality, linearity, and homoscedasticity. However, even after transformation, the procedural justice measure remained highly skewed, resembling a J-shaped distribution. It was dichotomized at the median, which is at the inflection point of the distribution, and analyses were run on both the skewed and dichotomized procedural justice variables. The median for the untransformed procedural justice scale was 7, which equates to a value of 1.18 in the reflected log₁₀ transformed procedural justice measure. Multiple regression analyses were run separately using the transformed and dichotomized versions of this measure to examine whether skewness of the variable affected the parameter estimates within the regression model. The correlations, multiple R and P values did not substantially change when the dichotomized procedural justice variable was substituted for the skewed variable. Consequently, the skewed, transformed continuous procedural justice variable was included with the final regression analysis. All analyses were conducted using SPSS version 24 (IBM Corp; 2015).

Results

Table 1 presents the socio-demographic and other characteristics of the sample. A high percentage of participants in both groups reported belonging to non-white/racialised groups. This is reflective of Toronto as a diverse city with a large immigrant population.

Table 1.

Demographic and Characteristics of the Study Sample.

	Control Group		CTO Group
	n = 69	%	n = 69	%	P
Demographics
Male	43	62.3%	43	62.3%	1.000
Age, Mean (SD)	39.4 (10.9)		38.4 (11.1)		0.611
Any post-secondary education	33	47.8%	36	52.2%	0.610
Single, never married	57	82.6%	48	69.6%	0.072
Receives disability/social assistance	62	89.9%	63	91.3%	0.771
Moved 2 times or more in the last 6 months	20	29.0%	26	37.7%	0.279
Race
Caucasian	33	47.8%	35	50.7%	0.788
Asian	8	11.6%	9	13.0%
Black	20	29.0%	15	21.7%
Other	8	11.6%	10	14.5%
Diagnosis
Psychotic disorder (as per chart)	58	84.1%	58	84.1%	1.000
Was told has a Psychotic disorder (self-report)	54	78.3%	50	75.8%	0.434
Believes has a mental illness	62	89.9%	48	69.6%	<0.001
Believes has a psychotic disorder	46	66.7%	18	27.3%	0.003
Believes has a substance use problem	22	31.9%	24	34.8%	0.718
Psychiatric Hospital Admission
Was admitted to a psychiatric unit last year	9	13.0%	53	76.8%	<0.001
Has had an involuntary psychiatric admission	2	2.9%	46	66.7%	<0.001
Has had a voluntary psychiatric admission	7	10.1%	15	21.7%	0.063
Criminal Justice Involvement
Previous history of imprisonment	20	29.0%	30	43.5%	0.077
Has been on probation in the past	22	31.9%	29	42.0%	0.217
Any past legal involvement (arrested, charged, etc.)	34	49.3%	43	62.3%	0.123
Treatment
Believes needs treatment	60	87.0%	48	69.6%	0.013
Receives an intramuscular injection medication	17	24.6%	46	66.7%	<0.001
Believes current treatment is effective	54	78.3%	45	65.2%	0.089
Has a substitute decision maker for treatment	5	7.2%	38	55.1%	<0.001
Negative Pressures and Procedural Justice in Treatment
Felt persuaded into taking treatment	27	31.9%	39	56.5%	0.041
Felt induced into taking treatment	11	15.9%	15	21.7%	0.384
Felt threatened into taking treatment	2	2.9%	21	30.4%	<0.001
Felt forced into taking treatment	14	20.3%	39	56.5%	<0.001
Procedural justicea (Mean [SD])	1.2 [0.19]		1.1 [0.21]		0.004
Felt deceived in treatment process	11	15.9%	18	26.2%	0.144

^aTo reduce extreme skewness, a reflected log₁₀ transformation was used on the measure of procedural justice.

Sixty-seven of 69 (97%) participants in the CTO group were under a CTO at the time of their interviews, and 40 participants (58%) in this group had experienced 2 or more CTOs.

A significantly smaller number of participants in the CTO group believed that they needed treatment or that their current treatment was effective as compared with the control group. In addition, a significantly greater number of the CTO group also reported receiving intramuscular antipsychotic injections compared with their counterparts. CSI showed that participants in the CTO group endorsed to be significantly less symptomatic than the comparison group (mean score for the CTO group, 27.8 [SD, 10.2] and control group 33.2 [12.3]. With the logarithmic transformation, this is 1.49 [0.161] in CTO group v. 1.42 [0.152] in the comparison group, P = 0.010). There was no difference on the global measure of Lehman’s quality-of-life scale (70.15 in the comparison group v. 68.85 in the CTO group; P = 0.845). There were differences between the groups in relation to procedural justice, with participants in the comparison group showing a higher sense of procedural justice in their treatment (7.4 [2.90]) than the CTO group (9.3 [4.21]; with the reflected log₁₀ transformation, this is 1.22 [0.192] in the comparison group v. 1.12 [0.210] in the CTO group, P = 0.004). Regarding experiencing negative pressures in their mental health treatment—across 4 categories of feeling persuaded, induced, threatened and forced into taking treatment—the CTO group scored higher, with statistically significant differences between the two groups noted for persuasion, threat, and force.

Table 2 compares the responses between the 2 groups on the dimensions of perceived coercion experienced during mental health treatment. Across all 4 dimensions of perceived coercion (i.e., having influence, control, choice, and freedom with respect to their mental health treatment), the CTO group reported significantly more perceived coercion than the comparison group. Although some participants in the CTO group did not feel coerced, several participants in the comparison group felt quite coerced with regard to their treatment.

Table 2.

Perceived Coercion.

	Control Group	CTO Group	Statistics	p	Effect Size
Perceived Coercion
Influence on mental health treatmenta	62.3	76.7	U = 1,883.5	0.029	r=0.186
Control over mental health treatmenta	57.33	81.67	U = 1,540.5	<0.001	r=0.305
Choice in mental health treatmenta	57.31	81.69	U = 1,539.5	<0.001	r=0.314
Freedom in decisions on mental health treatmenta	59.46	79.54	U = 1,687.5	<0.001	r=0.258
Perceived Coercion Scale  mean (SD)	10.8 (4.26)	13.8 (5.14)	t = 3.82	<0.001	d=0.673

^aMean Rank

Bivariate correlation analyses were conducted to examine the relationship between perceived coercion and various potential predictors, including CTO status. Table 3 summarizes the unadjusted analysis results. Being on a CTO; having intramuscular antipsychotic injections; previously being on probation; having any hospital admission or involuntary inpatient admission in the previous year; having a substitute decision maker; endorsing negative pressures of persuasion, threats and force; and reporting deception were all positively correlated with coercion. Conversely, believing that one has a psychotic disorder, endorsing a need for treatment, believing that treatment received is effective, and endorsing more procedural justice were negatively correlated with perceived coercion. All of these variables were included in a regression model, except for having any hospital admission, because it was highly correlated with having an involuntary inpatient admission (r = 0.81); this suggests collinearity, which can distort the interpretation of the multiple regression results.³⁶ Table 4 displays findings from a multiple linear regression model examining the correlates of perceived coercion. Whereas being on probation was positively correlated with perceived coercion, reporting greater procedural justice was negatively correlated with perceived coercion. No other variables were significantly associated with perceived coercion, including being on a CTO.

Table 3.

Unadjusted Regression Analysis of Perceived Coercion.

	r	P
Demographics/Characteristics
Male	−0.06	0.498
Age	0.11	0.218
Racialised group	0.02	0.856
Any post-secondary education	0.00	0.973
Single, never married	0.02	0.778
Receives disability/social assistance	0.04	0.684
Moved two times or more in the last 6 months	0.11	0.216
Diagnosis
Psychotic disorder (as per chart)	0.11	0.194
Was told has a Psychotic disorder (self-report)	0.07	0.419
Believes has a mental illness	−0.15	0.083
Believes has a psychotic disorder	−0.22	0.009
Believes has a substance use problem	−0.10	0.244
Psychiatric Hospital Admission
Was admitted to a psychiatric unit last year	0.16	0.056
Has had an involuntary psychiatric admission	0.28	0.001
Has had a voluntary psychiatric admission	−0.07	0.408
Treatment
CSI measurea	0.01	0.929
Treated under a CTO	0.31	<0.001
Believes needs treatment	−0.25	0.003
Receives an intramuscular injection medication	0.21	0.013
Believes current treatment is effective	−0.30	<0.001
Has a substitute decision maker for treatment	0.26	0.002
Criminal Justice Involvement
Previous history of imprisonment	0.13	0.126
Has been on probation in the past	0.22	0.008
Any past legal involvement (arrested, charged, etc.)	0.11	0.194
Negative Pressures and Procedural Justice in Treatment
Felt persuaded into taking treatment	0.20	0.021
Felt induced into taking treatment	0.08	0.356
Felt threatened into taking treatment	0.27	0.001
Felt forced into taking treatment	0.40	<0.001
Procedural Justiceb	−0.51	<0.001
Felt deceived in treatment process	0.31	<0.001

^aTo reduce skewness, a log₁₀ transformation was used for the Colorado Symptom Index.

^bTo reduce extreme skewness, a reflected log₁₀ transformation was used on the measure of procedural justice.

Table 4.

Perceived Coercion Regression Model.

Variable	B	SE	β	t	P
(Constant)	21.924	2.750		7.973	<0.001
Treated under a CTO	−0.205	1.165	−0.021	−0.176	0.861
Believes has psychosis	−0.708	0.816	−0.072	−0.868	0.387
Endorses need for treatment	−0.562	1.007	−0.047	−0.558	0.578
Has IM treatment	0.634	0.811	0.064	0.782	0.435
Believes treatment is effective	−1.102	0.913	−0.101	−1.207	0.230
Any involuntary admission	0.811	1.021	0.078	0.795	0.428
Previously on probation	1.728	0.750	0.169	2.303	0.023
Felt persuaded into taking treatment	0.312	0.779	0.032	0.401	0.689
Felt threatened into taking treatment	0.010	1.124	0.001	0.009	0.993
Felt forced into taking treatment	0.954	0.924	0.094	1.032	0.304
Has an SDM	0.955	0.922	0.090	1.036	0.302
Procedural justice scalea	−8.640	2.020	−0.361	−4.277	<0.001
Deception	0.540	1.031	0.045	0.524	0.601

^aTo reduce extreme skewness, a reflected log 10 transformation was used on the measure of procedural justice.

F(13,124) = 6.052, P < 0.0001, adj. R² = 0.324.

Discussion

Using a cross-sectional study design, we compared perceptions of coercion between a group of service users on CTOs and a matched comparison group of voluntary service users receiving similar intensive psychiatric care in the community. The analysis of the socio-demographic data and other characteristics indicates that both groups were facing difficult psychosocial challenges, including substance use issues, poverty, and a history of criminal justice involvement.

Our data show that a significantly larger proportion of the participants in the CTO group did not consider themselves to have a mental illness or require treatment compared to the control group. Hence, many had a substitute decision maker and were being treated with injectable antipsychotics. Not surprisingly, given the CTO criteria, a significantly higher number of participants in the CTO group were involuntary hospitalized over the previous year, potentially indicating the recent severity of their illness. However, at the time of this study, the participants in the CTO group reported to be significantly less symptomatic than the comparison group. There are several possible explanations for this finding. It is possible that CTO group participants under-reported symptoms or did not believe that their experiences were symptoms of illness. However, it is also possible that CTO group participants were experiencing fewer symptoms due to their mandatory community treatment and better compliance with their care (for example, by receiving intramuscular injections).

In this study, being on a CTO was only associated with perceptions of coercion in the bivariate analysis and not in the linear regression model, whereas procedural justice explained a large portion of the variance in coercion scores. This contrasts with findings in a study of CTOs in New Zealand, which found being on a CTO was associated with perceived coercion in a linear regression model whereas procedural justice was not.²⁷ These differences in study findings may be an artifact of differences in the administrative system by which compulsory outpatient treatment ensues; in the criteria and procedures applied for the rendering of such treatment; and in the powers conferred to treatment providers.³⁷ For example, in Ontario, a CTO cannot be made without the consent of the service user or their substitute decision maker if an individual lacks the capacity to consent to treatment. In Australasia, by comparison, no determination of incapacity is required to place individuals on a CTO.³⁷ The inverse relationship we found between procedural justice and perceived coercion is consistent with findings in numerous other studies,²⁹ especially in inpatient settings.^38,39 Our data also show that participants’ perception of coercion is positively correlated with their past experience with probation. About half (49%) of the comparison group and nearly two-thirds (62%) of the CTO group reported criminal justice involvement in their lives.

Many psychiatric service users are subjected to formal and informal pressures and different types of leverage in the community to adhere to treatment, including management of their finances and their housing.³⁴ The combination of these measures may impact perceptions of their overall treatment. For example, a secondary analysis of a North Carolina RCT showed that the application of CTOs and financial guardianship together was correlated with significantly higher levels of perceived coercion, whereas neither CTOs nor financial guardianship was independently correlated. These findings point to a possible cumulative effect of multiple types of pressure in the treatment of psychiatric service users.²⁰ Combining this with the struggles that many mental health service users have to access care, housing and income support may explain to some extent the high level of perceived coercion across the whole study group irrespective of CTO status.

Current psychiatric care includes several interventions that can be experienced as coercive by service users, including involuntary psychiatric admissions and the use of CTOs. As the debate about the use of coercive measures such as CTOs continue, it is important for clinicians to develop strategies that could reduce service users’ perceptions of coercion in their care. Utilizing and operationalizing the tenants of procedural justice may be a viable option to achieve this objective when such interventions are deemed necessary.

With regards to the limitations of this study, our sample size was relatively small and limited to service users treated in a large Canadian urban center under Ontario’s medical, social, and legal system. Our findings may have also been influenced by selection bias as well as issues related to appropriately matching our comparison group. Additionally, the cross-sectional design precludes any inferences about causality. Moreover, data on the number and characteristics of service users who were presented with information about the study from their case manager but who declined to learn more from the research team were not collected. Despite these limitations, this is the largest study in Canada examining perceptions of coercion and procedural justice among patients of community mental health teams and highlights important areas for further research.

Conclusions

In this study, participants’ perception of coercion, irrespective of their CTO status, was directly correlated with their previous experience with probation and inversely correlated with their sense of procedural justice in their treatment. Service users’ perceptions of coercion may be worsened by other types of leverage and coercive measures experienced in their day-to-day lives. However, their perception of coercion may be reduced by better incorporating procedural justice principles into their care.

Abbreviation List

CTO

Community Treatment Orders

ICM

Intensive Case Management

ACT

Assertive Community Treatment

CSI

Colorado Symptom Index

SDM

Substitute Decision Maker