Predicting outcome of assertive outreach across England

BACKGROUND

Assertive community treatment for the severely mentally ill is being implemented increasingly internationally. It is unclear whether recommended characteristics of assertive outreach (AO) teams influence care and outcomes. We hypothesised that recommended characteristics of AO teams such as joint health and social care management would predict reduced hospitalisation in the first year of an AO client programme and related outcomes throughout England.

METHODS

A two-stage design was used: a stratified sample of 100 of the 186 'stand-alone' AO teams in England and a systematic sample of clients from each team with stratification for black and ethnic minority patients. Team characteristics, treatment and outcomes were collected from teams. Analyses took account of patients' histories, clustering and ethnic minority over-sampling.

RESULTS

Under AO the proportion of time spent in hospital following admission decreased. Only 3/1,096 patients went missing in 9 months. Although patient' histories significantly predicted outcomes almost no team characteristics predicted re-admission or other patient outcomes after 1 and 3 years. Ethnic minority clients were more likely to be on compulsory orders only on jointly managed teams (P = 0.030). Multidisciplinary teams and teams not working out of hours significantly predicted that patients received psychological interventions, but only 17% of sampled patients received such treatments.

CONCLUSIONS

Characteristics of AO teams do not explain long-term patient outcomes. Since recommended team characteristics are not effective new models of care should be developed and the process of care tested. Managing teams to implement evidence-based psychological interventions might improve outcomes.

Lawfulness of a randomised trial of the new community treatment order regime for England and Wales

This article debates and defends the lawfulness of a randomised controlled trial of compulsory outpatient treatment under the mental health legislation for England and Wales. The trial is designed to compare the outcomes for patients of their treatment under the new Community Treatment Order (CTO) regime with their treatment under the older leave scheme - the two main forms of compulsory care in the community now authorised by the revised Mental Health Act 1983. The methods for the trial involve the random allocation of some patients between the two schemes, when they are considered by their Responsible Clinicians to be eligible for some form of compulsory outpatient care. The main question we consider is the lawfulness of that aspect of the methods. Can a carefully selected group of patients be allocated at random between the two regimes to permit an evaluation to proceed? Or would that involve some departure from the decision-making criteria specified by law? We argue that a group of patients can be identified who meet - simultaneously - the tests for treatment under both the CTO and the leave schemes. Those patients could then be allocated lawfully to treatment under either scheme. This opens the door, we argue, to their random allocation between the two schemes for the purposes of the research. In reaching this conclusion, we explain the methods and aims of the trial and closely compare the respective features of the leave and CTO regimes.

Measuring patients' views: a bifactor model of distinct patient-reported outcomes in psychosis

BACKGROUND

Patient-reported outcomes (PROs) are widely used for evaluating the care of patients with psychosis. Previous studies have reported a considerable overlap in the information captured by measures designed to assess different outcomes. This may impair the validity of PROs and makes an a priori choice of the most appropriate measure difficult when assessing treatment benefits for patients. We aimed to investigate the extent to which four widely established PROs [subjective quality of life (SQOL), needs for care, treatment satisfaction and the therapeutic relationship] provide distinct information independent from this overlap.

METHOD

Analyses, based on item response modelling, were conducted on measures of SQOL, needs for care, treatment satisfaction and the therapeutic relationship in two large samples of patients with psychosis.

RESULTS

In both samples, a bifactor model matched the data best, suggesting sufficiently strong concept factors to allow for four distinct PRO scales. These were independent from overlap across measures due to a general appraisal tendency of patients for positive or negative ratings and shared domain content. The overlap partially impaired the ability of items to discriminate precisely between patients from lower and higher PRO levels. We found that widely used sum scores were strongly affected by the general appraisal tendency.

CONCLUSIONS

Four widely established PROs can provide distinct information independent from overlap across measures. The findings may inform the use and further development of PROs in the evaluation of treatments for psychosis.

Community treatment orders: how ethical without experimental evidence?

Compulsory community treatment orders are a feature of most advanced mental health systems without convincing experimental evidence of benefits. Is it ethical to continue without such evidence ? This paper argues that the responsibility for ensuring it is collected lies as much with Parliament as with researchers.

Developing a user-generated measure of continuity of care: brief report

OBJECTIVE

This paper describes a measure of continuity of care, establishes its reliability and tests it in a field trial sample for evidence of its validity. In contrast to others, this measure has been generated from the perspectives of service users. As continuity of care is a concern particularly for those with severe mental illness, we have confined our work to this population group.

METHOD

Service users in focus groups and expert panels generated the measure. The researchers were themselves service users. Test-retest reliability was assessed with an independent sample. The measure was administered to a final independent field trial sample to determine their experiences of continuity of care and for further psychometric testing.

RESULTS

The measure generated by service users has satisfactory psychometric properties. Service users in the field trial sample were more satisfied when continuity, as assessed by this measure, was in place.

CONCLUSION

It is possible and valid to construct outcome measures in mental health entirely from the user perspective. This has not been done before.

Continuity of care in mental health: understanding and measuring a complex phenomenon

BACKGROUND

Continuity of care is considered by patients and clinicians an essential feature of good quality care in long-term disorders, yet there is general agreement that it is a complex concept. Most policies emphasize it and encourage systems to promote it. Despite this, there is no accepted definition or measure against which to test policies or interventions designed to improve continuity. We aimed to operationalize a multi-axial model of continuity of care and to use factor analysis to determine its validity for severe mental illness.

METHOD

A multi-axial model of continuity of care comprising eight facets was operationalized for quantitative data collection from mental health service users using 32 variables. Of these variables, 22 were subsequently entered into a factor analysis as independent components, using data from a clinical population considered to require long-term consistent care.

RESULTS

Factor analysis produced seven independent continuity factors accounting for 62.5% of the total variance. These factors, Experience and Relationship, Regularity, Meeting Needs, Consolidation, Managed Transitions, Care Coordination and Supported Living, were close but not identical to the original theoretical model.

CONCLUSIONS

We confirmed that continuity of care is multi-factorial. Our seven factors are intuitively meaningful and appear to work in mental health. These factors should be used as a starting-point in research into the determinants and outcomes of continuity of care in long-term disorders.

Assessing Social Functioning in Schizophrenia

Aims:

Social functioning has increasingly become recognised as an important outcome in schizophrenia. While measures of symptom status are highly developed and widely used there has been only limited work on developing instruments for measuring social functioning in a reliable and consistent manner. We aimed to review the schizophrenia literature to identify the structured social functioning measures that have been used with any frequency and compare their features (1).

Method:

A detailed electronic literature search (1990 - 2006) using the key words schizophrenia and social function was carried out and those papers containing details of any structured assessment of social function were used. The most frequently used instruments were identified. A search was also conducted for the use of social function measures in trials of antipsychotics in schizophrenia.

Results:

301 articles employed 87 potential social function measures and of these only 20 were used 3 or more times. Only 14 RCTs of antipsychotics employed them.

Conclusions:

There is limited consensus on the definition and measurement of social functioning but two or three scales show promise for regular usage.

Essential Features in Outpatient Care of Severely Mentally Individuals

Aims:

Most community mental health services have evolved in response to the downsizing and closing of mental hospitals. Their form varies in different health care settings and their evolution was rarely subjected to formal evaluation. The introduction of Assertive Community Treatment in 1980 lead to a flurry of research activity which yielded conflicting results. Examination of these results provides the opportunity to distinguish essential from redundant components of care.

Methods:

Two separate studies were conducted - a systematic review followed by cluster and regression analysis of reported components of care and a secondly a meta-regression analysis of published studies.

Results:

The meta-regression demonstrated that most of the variation in hospitalization outcome was due to differences in baseline hospital usage. Staffing levels (e.g. caseloads) were not found to be crucial but multi-disciplinary working and integrated psychiatrists were. The systematic review indicated the importance of integration of health and social care and outreach.

Conclusions:

Multidisciplinary teams which include both social workers and psychiatrists and which focus on visiting patients at home are highly successful in managing severely ill patients in the community. Many of the detailed requirements proposed for ACT teams are probably superfluous.

Objective social outcomes index (SIX): a method to summarise objective indicators of social outcomes in mental health care

OBJECTIVE

No method has yet been established to summarise different objective indicators of social outcomes into one score. The study aimed to develop and test a simple, brief and meaningful index of social outcomes that can be used across a broad range of mental health studies and is potentially obtainable from routine documentation.

METHOD

An index capturing employment, accommodation and living situation (range from 0 to 6; SIX) was devised and tested in three longitudinal datasets of patients with severe mental illness from Germany, Sweden and the UK.

RESULTS

SIX showed an almost normal distribution cross-sectionally in all three datasets. Between 58 and 78% of patients changed scores over time. Change scores were weakly associated with change scores of symptoms, quality of life, global functioning and disability.

CONCLUSION

A simple method to summarise different indicators of social outcomes provides meaningful results. It can be widely used in research and routine care.

Prevalence of Chlamydia trachomatis and associated risk factors in women inmates admitted to a youth offenders institute in the UK

The aim of the study was to estimate the prevalence of Chlamydia trachomatis genital infections in young women inmates aged 17-21 located on the Youth Offenders Institute of HMP Holloway. The study consisted of a questionnaire and a urine sample for analysis. A health advisor administered a questionnaire to establish sociodemographic profile, sexual history and illicit drug use. C. trachomatis was diagnosed using a nucleic acid amplification test. Overall, the prevalence of C. trachomatis was 13.2%. Drug use was common with 41.2% using crack cocaine and 17.3% using intravenous heroin. A majority (80.2%) reported not using condoms during the last sexual encounter and 62.6% had a history of a previous pregnancy. Over half (53.3%) had a previous sexually transmitted infection (STI) screening and 74.7% had never been previously diagnosed with a STI. Sexual health programmes in prisons should include screening as well as education about STI risk reduction and drug prevention.

A review of instruments developed to measure outcomes for carers of people with mental health problems

OBJECTIVE

Community-based care for mental disorders places considerable burden on families and carers. Measuring their experiences has become a priority, but there is no consensus on appropriate instruments. We aimed to review instruments carers consider relevant to their needs and assess evidence for their use.

METHOD

A literature search was conducted for outcome measures used with mental health carers. Identified instruments were assessed for their relevance to the outcomes identified by carers and their psychometric properties.

RESULTS

Three hundred and ninety two published articles referring to 241 outcome measures were identified, 64 of which were eligible for review (used in three or more studies). Twenty-six instruments had good psychometric properties; they measured (i) carers' well-being, (ii) the experience of caregiving and (iii) carers' needs for professional support.

CONCLUSION

Measures exist which have been used to assess the most salient aspects of carers outcome in mental health. All require further work to establish their psychometric properties fully.

Social functioning as an outcome measure in schizophrenia studies

OBJECTIVE

Deficits in social functioning are a core feature of schizophrenia.

METHOD

A literature search of English language articles published between January 1990 and December 2006 was undertaken to identify: i) scales used most frequently to assess social functioning in schizophrenia; and ii) the most frequently used social functioning scales in randomized, controlled trials of antipsychotics. A further search (without time limits) examined their psychometric properties.

RESULTS

A total of 301 articles employed social functioning scales in the assessment of schizophrenia. These contained 87 potentially relevant measures. Only 14 randomized, controlled studies of antipsychotic agents were identified that examined social functioning. Scales varied greatly in terms of measurement approach, number and types of domains covered and scoring systems. A striking lack of data on psychometric properties was observed.

CONCLUSION

Limited consensus on the definition and measurement of social functioning exists. The Personal and Social Performance Scale is proposed as a useful tool in future research.

Predictors of violent victimization amongst those with psychosis

OBJECTIVE

The aim of this study was to establish the prevalence and predictors of violent victimization amongst a community-dwelling sample of individuals with psychosis.

METHOD

The 2-year prevalence of self-reported violent victimization was estimated for a sample of 708 individuals with chronic psychosis living in the community in four urban UK centres. Baseline socio-demographic and clinical factors were examined as possible risk factors for victimization over the 2-year follow-up period.

RESULTS

The 2-year prevalence of violent victimization in the sample was 23%. Four factors were found to be independently predictive of victimization - history of victimization, less than daily family contact, young age at illness onset and the presence of co-morbid Cluster B personality disorder.

CONCLUSION

Those with psychotic illnesses are at elevated risk of being assaulted. Given the likely adverse health implications, clinicians should routinely enquire about victimization in their assessments of those with psychotic disorders particularly amongst those who are socially isolated, with a younger age of illness onset and in those with co-morbid personality disorder.

Day centres for severe mental illness

BACKGROUND

The number of people with severe mental illness who receive treatment whilst living at home has increased greatly over the last 30 years. Day centres and day hospitals frequently supplement this treatment.

OBJECTIVES

To determine the effects of non-medical day centre care for people with severe mental illness.

SEARCH STRATEGY

We updated our search in September 2005. All databases and searches are detailed in the body of the text.

SELECTION CRITERIA

We would have included all randomised controlled trials where seriously mentally ill people were allocated to non-medical day centre care.

DATA COLLECTION AND ANALYSIS

We reliably selected studies, quality rated them and extracted data. For dichotomous data, it had been hoped to estimate the fixed effects Relative Risk (OR) with 95% confidence intervals (CI) and the number needed to treat statistic (NNT). Analysis was to have been by intention-to-treat. Normal continuous data were to have been summated using the weighted mean difference (WMD) and scale data presented only for those tools that had attained pre-specified levels of quality.

MAIN RESULTS

Electronic searches identified over 300 citations but none were relevant to this review. We found no trials of non-medical day centres.

AUTHORS' CONCLUSIONS

We feel that the inclusion of any studies less rigorous than randomised trials would result in misleading findings and that it is not unreasonable to expect well designed, conducted and reported randomised controlled trials of day centre care. More precise nomenclature would greatly help identify relevant work. At present non-randomised comparative studies give conflicting messages about the roles provided by day centres and the clinical and social needs they are able to meet. It is therefore probably best that people with serious mental illness and their carers, if given the choice, take a pragmatic decision on which type of unit best meets their needs. There is a clear need for randomised controlled trials of day centre care compared to other forms of day care, and when resources are limited, day centre care within the context of a pragmatic randomised trial may be the only way of ensuring equity of provision.

Do care patterns change over time in a newly established mental health service? A report from the UK700 trial

PURPOSE

Data on the process of mental health care is scant. Most studies focus on services at their inception when activity may be atypical and then usually present data only mean values for the reported variables over the whole study period. We aimed to test whether care delivery changes over time, and to describe any changes at the individual patient and team levels.

METHODS

Process data on 272 patients in three new intensive case management (ICM) teams were collected over 2 years. Interventions were prospectively recorded using clinician-derived categories. Changes over time are described at both patient and team level.

RESULTS

The number of contacts and the proportion of face-to-face activity were remarkably constant after the first month at the patient level. The proportion of 'psychiatric' interventions (main focus on medication or a specific 'mental health' intervention performed) increased greatly after the first 6 months. The care activity received by individual patients varied considerably. Overall, teams varied significantly in the extent to which their activity rates were sustained over time.

CONCLUSIONS

New ICM teams deliver highly individualised care with more marked differences in treatment patterns between patients in the same team than mean differences between teams. The early 'engagement' period is marked by a greater focus on social care. There is evidence of differences in sustainability of the services by site.

The impact of intellectual functioning on symptoms and service use in schizophrenia

BACKGROUND

This study aims to evaluate differences in the clinical profiles and use of psychiatric services by people with schizophrenia with and without borderline intellectual functioning. Both groups in this study were receiving standard community psychiatric care.

METHODS

A naturalistic sample of 372 people with schizophrenia completed the National Adult Reading Test. Data were collected prospectively over 18 months on psychiatric symptoms and service use. Three hundred and thirteen had normal intellectual functioning (mean age 43, range 20-76 years) and 59 had borderline or lower intellectual functioning (mean age 45, range 21-81 years). This was defined by a National Adult Reading Test error score of more than 40.

RESULTS

People with borderline or lower intellectual functioning had a lower quality of life, more severe psychotic symptoms, reduced functioning and fewer antidepressant prescriptions. There were no significant differences in service use including hospital admission.

CONCLUSIONS

People with schizophrenia and borderline or lower intellectual functioning are a more disabled group within general adult psychiatric services who should be the focus of initiatives for improved service delivery.

The cardiovascular and respiratory health of people with schizophrenia

OBJECTIVE

To examine the cardiovascular and respiratory health of people with severe mental illness (SMI) and compare findings with the Health Surveys for England.

METHOD

A prospective, multi-centre observational prevalence study of 602 patients with schizophrenia-related psychoses carried out in six locations across the UK over 24 months.

RESULTS

Compared with general population subjects, people with SMI reported higher rates of angina and respiratory symptoms and had poor lung function. Much of this increased risk could be explained by lifestyle risk factors; there were increased levels of obesity among younger people with SMI.

CONCLUSION

Key indicators of the cardiovascular and respiratory health of people with SMI are poor compared with those of the general population. Care plans should prioritize interventions to attenuate lifestyle risk factors. Evidence of increasing obesity in younger patients is of particular concern, predicting even greater health needs in the future.

Maintenance antipsychotic medication patterns in outpatient schizophrenia patients: a naturalistic cohort study

OBJECTIVE

Newer antipsychotics are increasingly used in schizophrenia maintenance. The UK change has been slow with little known on switching patterns. We aimed to investigate antipsychotic prescribing patterns in schizophrenia patients.

METHOD

A naturalistic six-site cohort sample of 600 patients were interviewed by researchers at 6-monthly intervals for 2 years to record their clinical and social functioning; use of services and medication for the preceding 6 months was obtained by structured extraction from clinical case notes.

RESULTS

Alterations in antipsychotic medication were frequent in this group, mainly during periods of inpatient care. Atypical prescribing increased steadily, though slowly, across the period. Polypharmacy was less than anticipated.

CONCLUSION

Inpatient care remains the main forum for switching of antipsychotics. The UK maintains a slow shift to atypical antipsychotics.