Neil Preston

Assessing the outcome of compulsory psychiatric treatment in the community: epidemiological study in Western Australia

Abstract Objective: To examine whether community treatment orders for psychiatric patients reduce subsequent use of health services in comparison with control patients not placed on an order. Design: Epidemiological study with a before and after, two stage design of matching and multivariate analysis, controlling for sociodemographic variables, clinical features, and psychiatric history. Setting: All community based and inpatient psychiatric services in Western Australia, covering a population of 1.7 million people. Participants: 228 subjects placed on a community treatment order, matched with an equal number of controls to give a total of 456 patients. Main outcome measures: Inpatient admissions, bed days, and outpatient contacts one year after subjects were placed on a community treatment order or the index date of matched controls. Results: Both subjects and their matched controls had reduced inpatient admissions and bed days in hospital. Subjects had significantly more outpatient contacts. Multivariate analysis indicated that being placed on a community treatment order was associated with increased outpatient contacts in the subsequent year compared with the control group. Otherwise, orders did not affect subsequent use of health services. Other factors associated with increased use of health services were age and inpatient admissions, bed days, and outpatient contacts before the order or index date. No covariates were shown to be associated with changes in within pair differences in inpatient admissions or bed days. Conclusions: The introduction of compulsory treatment in the community does not lead to reduced use of health services. What is already known on this topic Various forms of compulsory treatment in the community have been suggested as being effective in reducing use of services by patients with mental health disorders Studies have often lacked epidemiological sampling frames and control for possible confounding factors What this study adds Patients placed on community treatment orders and those not on such orders had reduced hospital admissions and bed days one year later Placement of an order did not predict subsequent use of services Community treatment orders may not be an effective alternative to assertive community treatment programmes

Gender differences in premorbid adjustment of patients with first episode psychosis.

Gender differences in premorbid adjustment, clinical presentation, and longitudinal course have been considered increasingly in explanatory models of psychotic disorders, such as the schizophrenias. Indeed, findings of a male propensity to poor premorbid adjustment, negative and non-affective symptoms, and poor outcome relative to their female counterparts, has led to suggestions that males are more prone to an early-onset dementia praecox type of schizophrenic disorder. The current study investigated a sample of 38 male and 20 female patients presenting with their first episode of psychosis (broadly defined, but excluding obvious drug-induced disorders) from a defined catchment area population, which had been systematically ascertained without prejudice to diagnostic subtype or illness duration. The study investigated gender, diagnosis and interaction of gender and diagnosis on differences within the three developmental age categories of childhood, early adolescence and late adolescence, to identify where, within these age categories, differences lie. The second part of the study was to investigate the relationship between premorbid adjustment, gender, and psychopathology as measured by the PANSS and SCL-90. General linear modelling revealed that males were reported to have had poorer premorbid adjustment in late adolescence when compared to females, notably in items examining school performance, adaptation to school, social interests and sociosexual development. Males were observed to have higher levels of negative symptoms but not for positive or general symptoms on the PANSS. This finding is independent from the effect of diagnosis or of the interaction effect between gender and diagnosis on premorbid adjustment. There were no gender effects for the self reported global indices on the SCL-90. The results suggest that in comparison with their female counterparts, males who develop a psychotic illness have significantly poorer premorbid adjustment at the late adolescent stage and that this may contribute to higher levels of negative symptoms.

Randomized and non-randomized evidence for the effect of compulsory community and involuntary out-patient treatment on health service use: systematic review and meta-analysis.

BACKGROUND There is limited randomized controlled trial (RCT) evidence for compulsory community treatment. Other study methods may clarify their effectiveness. We reviewed RCT and non-RCT evidence for the effect of compulsory community treatment on hospital admissions, bed-days, compliance and out-patient contacts. METHOD A systematic review of RCTs, controlled before-and-after (CBA) studies, and interrupted time series (ITS) analyses. Meta-analysis of RCTs. RESULTS Eight papers covering five studies (two RCTs and three CBAs) met inclusion criteria (total n=1108). There was no statistical difference in 12-month admission rates between subjects on involuntary out-patient treatment and controls. Survival analyses of time to admission were equivocal. All five studies reported decreases in the number of bed-days following involuntary out-patient treatment but this only reached statistical significance in one situation; patients receiving the intervention were less likely to have admissions of over 100 days. There was no difference in treatment adherence between the intervention and control groups in either RCT or two of the CBA studies. However, the third CBA study reported a statistically significant increase of nearly five visits in the mean number of overall contacts in the involuntary out-patient treatment group. CONCLUSIONS The evidence for involuntary out-patient treatment in reducing either admissions or bed-days is very limited. It therefore cannot be seen as a less restrictive alternative to admission. Other effects are uncertain. Evaluation of a wide range of outcomes should be included if this type of legislation is introduced.

Reducing all-cause mortality among patients with psychiatric disorders: a population-based study

Background: Among patients with psychiatric disorders, there are 10 times as many preventable deaths from physical disorders as there are from suicide. We investigated whether compulsory community treatment, such as community treatment orders, could reduce all-cause mortality among patients with psychiatric disorders. Methods: We conducted a population-based survival analysis of an inception cohort using record linking. The study period extended from November 1997 to December 2008. The cohort included patients from all community-based and inpatient psychiatric services in Western Australia (state population 1.8 million). We used a 2-stage design of matching and Cox regression to adjust for demographic characteristics, previous use of health services, diagnosis and length of psychiatric history. We collected data on successive cohorts for each year for which community treatment orders were used to measure changes in numbers of patients, their characteristics and outcomes. Our primary outcome was 2-year all-cause mortality. Our secondary outcomes were 1-and 3-year all-cause mortality. Results: The study population included 2958 patients with community treatment orders (cases) and 2958 matched controls (i.e., patients with psychiatric disorders who had not received a community treatment order). The average age for cases and controls was 36.7 years, and 63.7% (3771) of participants were men. Schizophrenia and other nonaffective psychoses were the most common diagnoses (73.4%) among participants. A total of 492 patients (8.3%) died during the study. Cox regression showed that, compared with controls, patients with community treatment orders had significantly lower all-cause mortality at 1, 2 and 3 years, with an adjusted hazard ratio of 0.62 (95% confidence interval 0.45–0.86) at 2 years. The greatest effect was on death from physical illnesses such as cancer, cardiovascular disease or diseases of the central nervous system. This association disappeared when we adjusted for increased outpatient and community contacts with psychiatric services. Interpretation: Community treatment orders might reduce mortality among patients with psychiatric disorders. This may be partly explained by increased contact with health services in the community. However, the effects of uncontrolled confounders cannot be excluded.

The brief symptom inventory and the positive and negative syndrome scale: discriminate validity between a self-reported and observational measure of psychopathology.

Concern within the literature has emerged from time to time arguing the poor validity of self-reported measures in psychopathology, namely, the reporting of psychotic experience. Although it is commonly believed that patients who have had a psychotic episode cannot accurately self-report their experience, very few studies have been conducted to measure the concordance between self-reported and observational measures of psychopathology using multivariate statistical techniques. Sixty-nine patients presenting their first psychotic episode were interviewed and assessed on the Positive and Negative Syndrome Scale (PANSS) and were asked to complete the Brief Symptom Inventory (BSI). By clustering symptom dimensions from the BSI into discriminate functions, the research demonstrated that these symptom dimensions could adequately classify high versus low scores on the PANSS subscales and total score. When the same clusters were entered into multivariate analysis of variance (MANOVA) models, they also demonstrated significant differences between high versus low observed symptomatology on the PANSS Positive and General Subscale Groups and Total Score Groups. The current findings shed some doubt on the supposition that those who experience psychosis are unable to report symptom dimensions that concord with those who observe the psychosis. It appears that models, operational definitions, and the language used in measuring psychopathology may differ significantly from those who experience the psychotic experience and those who observe it. Techniques such as multitrait multimethod are discussed as ways of overcoming these concerns.

The Health of the Nation Outcome Scales: validating factorial structure and invariance across two health services

Objective: The Health of the Nation Outcome Scale (HoNOS) was developed in the mid-1990s as an inclusive and comprehensive instrument to measure patient outcomes in four main factors: behaviour, impairment, symptoms and social problems. This paper attempts to investigate whether similar health services rate the HoNOS with equivalent psychometric calibration. Method: The purpose of this study was to test for invariant construct interpretation of the instrument across two equivalent health services, using simultaneous confirmatory factor analysis. Results: Although the four-factor model of the HoNOS was confirmed, structural non-invariance occurred, casting doubt on the equivalent interpretability and generalisability of the instrument across similar heath services. Conclusions: Over-inclusiveness, lack of specificity and questionable independence of observations may have contributed to the non-invariant factor structure between the two health services. Such results shed some doubt on the ability of the HoNOS and like observational instruments to provide equivalent comparisons between health services.

An eleven-year evaluation of the effect of community treatment orders on changes in mental health service use

Many studies of compulsory community treatment have assessed their effect early on after the implementation of legislation. Although compulsory community treatment may not prevent readmission to hospital, there is evidence of an effect on length of stay before and after the intervention when compared to controls. This paper examines whether outcomes change as clinicians gain experience in the use of community treatment orders (CTOs). Cases and controls from three linked Western Australian databases were matched on age, sex, diagnosis and time of hospital discharge or community placement. We compared changes in bed-days and outpatient visits of CTO cases and controls using multivariate analyses to further control for confounders. We identified 2958 CTO cases and controls from November 1997 to December 2008 (total n = 5916). The average age was 37 years and 64% were male. Schizophrenia and other non-affective psychoses were the commonest diagnoses (73%). CTO placement was associated with a mean decrease of 5 bed-days from before the order when compared to controls (B = -5.23, s.e. = 1.60, t = -3.26, p < 0.001). There was an increase of 8 days in outpatient contacts (B = 8.31, s.e. = 1.17, t = 7.11, p < 0.001). There was little change in CTO use and outcomes over the 11 years. Compared to controls, CTOs may therefore reduce lengths of stay from before placement on the order. They also increase outpatient contacts. This study illustrates the importance of selecting an outcome that directly addresses the objective of the intervention.

A comparison of health service use in two jurisdictions with and without compulsory community treatment

BACKGROUND This study examines whether community treatment orders (CTOs) reduce psychiatric admission rates or bed-days for patients from Western Australia compared to control patients from a jurisdiction without this legislation (Nova Scotia). METHOD A population-based record linkage analysis of an inception cohort using a two-stage design of matching and multivariate analyses to control for sociodemographics, clinical features and psychiatric history. All discharges from in-patient psychiatric services in Western Australia and Nova Scotia were included covering a population of 2.6 million people. Patients on CTOs in the first year of implementation in Western Australia were compared with controls from Nova Scotia matched on date of discharge from in-patient care, demographics, diagnosis and past in-patient psychiatric history. We analysed time to admission using Cox regression analyses and number of bed-days using logistic regression. RESULTS We matched 196 CTO cases with an equal number of controls. On survival analyses, CTO cases had a significantly greater readmission rate. Co-morbid personality disorder and previous psychiatric history were also associated with readmission. However, on logistic regression, patients on CTOs spent less time in hospital in the following year, with reduced in-patient stays of over 100 days. CONCLUSIONS Although compulsory community treatment does not reduce hospital admission rates, increased surveillance of patients on CTOs may lead to earlier intervention such as admission, so reducing length of hospital stay. However, we do not know if it is the intensity of treatment, or its compulsory nature, that effects outcome.

Conditional Release in Western Australia: Effect on Hospital Length of Stay

OBJECTIVE The goal of this study was to determine whether the introduction of community treatment orders, which allow for conditional release from a psychiatric hospital, reduced inpatient episode durations in Western Australia by providing an alternative to extended inpatient stays. METHODS The design compared 129 persons given community treatment orders and 117 matched control patients without such orders-all of whom were hospitalized during the same period both before and after the introduction of the community treatment order law that allows for conditional release. A multivariate analysis of covariance was used to evaluate the impact of community treatment orders on change in inpatient episode duration. RESULTS The model showed a significant effect on inpatient episode duration (R(2)=.23, adjusted R(2)=.17, N=243, F=3.99, df=17 and 226, p<.001), indicating that community treatment orders (after taking all control factors into account) enabled a 19.16-day reduction per episode of inpatient care (t=2.13, df=1, p=.034) for persons given conditional release. Community-initiated treatment orders intended to prevent hospitalization, yet failing to do so, were associated with increased duration of subsequent hospitalizations (35.18 days; t=-3.36, df=1, p<.001). CONCLUSIONS Community treatment orders can be a useful tool for some but not necessarily all objectives. In the form of conditional release, orders reduce the likelihood of extended hospital stays. As a means to prevent hospitalization, the utility of community treatment orders is more complex, being dependent on services provided and on the judicious selection of persons for these orders.

What Determines Compulsory Community Treatment? A Logistic Regression Analysis Using Linked Mental Health and Offender Databases

OBJECTIVE Western Australia has one of the highest published rates of the use of compulsory treatment orders in the English-speaking world. Differences in patient characteristics, legislation and service setting may explain variations in the reported efficacy of compulsory community treatment. Our objective is to investigate predictors of Community Treatment Orders (CTO) placement in the first year of implementation in Western Australia and see if there were any differences in the type of patients placed on these orders compared to other studies. METHOD A population-based record linkage study of Mental Health and Offender Databases comparing 265 patients on CTOs with a consecutive control group (CCG) of equal number matched on date of discharge from inpatient care or CTO placement. RESULTS Previous health service use, after-care placement, mental disorder history including schizophrenic history, a positive forensic history of violence to others as well as patients marital status were the significant predictors of CTO placement. CONCLUSIONS Studies of compulsory community treatment appear to be of similar populations. In spite of the comparatively high rate of use, psychiatrists in Western Australia do not appear to be applying community treatment orders to different types of patient compared to elsewhere. We need further research to establish the relative contribution of patient characteristics, legislation and service setting toward the use and outcome of compulsory community treatment.