Effects of organisational interventions for mental health services
Systematic review
|Updated
The purpose of this report was to provide an overview of systematic reviews on the effects of interventions to organize services for children, youth and adults with mental disorders.
Key message
The purpose of this report was to provide an overview of systematic reviews on the effects of interventions to organize services for children, youth and adults with mental disorders.
We identified 17 systematic reviews of interventions to organize services for people with mental illness. In terms of content the various organizational interventions complemented each other by having different purposes, covering topics such as multidisciplinary teams, treatment setting, integrated care and continuity of care.
• Intensive coordination of psychosocial work compared with standard therapy possibly led to slightly less number of days in hospital, some more users who maintained contact with mental health care, higher functional score, a few more living independently and possibly slightly higher user satisfaction.
• Use of community multidisciplinary teams versus hospital-based
services may lead to slightly fewer hospitalizations
• Day hospitals versus hospitalisation probably led to longer duration for index admission but also probably a slightly better score for social function.
• For all the other interventions with more restricted purposes, the quality of the documentation was either too low, the results had wide confidence intervals or showed inconsistent effects across studies, making it difficult to draw meaningful conclusions about efficacy.
Of the identified organizational interventions, intensive case management compared with standard therapy appears to be the intervention with positive effects on more outcomes. The quality of the evidence for these results was low. Overall, the scientific evidence for the effect of organisational interventions is generally of low and very low quality. The main part of the studies was of an earlier date. The scientific evidence appears to be a rather inadequate basis for decisions on future organisation of mental health care.
Summary
Background
Particularly since the 1970s a de-institutionalization and decentralization of mental health care in large parts of the western world have taken place. More and more services and tasks have been moved from central institutions to the local community. In Norway the National Programme for Mental Health 1999 to 2006 established that the goal of this reorganization was to improve the quality and quantity of services, ensure a comprehensive planning and to improve cooperation between municipalities and counties with a focus on user needs. The purpose of this report was to provide a systematic overview of the effect of various interventions for organising services for children, youth and adults in this relatively new model for mental health care.
Methods
We searched for systematic reviews of interventions for the organisation of mental health services in the Cochrane Library (SR), MEDLINE, EMBASE, Database of Abstracts of Reviews of Effects (DARE) and Health Technology Assessments (HTA) and ISI Web of Science in June 2011. All titles and abstracts were screened with respect to whether references appeared to meet inclusion criteria. Potentially relevant publications were ordered in full text and assessed for inclusion or exclusion. All reviews that could be included were critically assessed according to the Knowledge Centre's checklist for assessing quality of systematic reviews. Only reviews of high quality were included. All pre-selected outcomes in the final included reviews were critically assessed for the risk of systematic bias. This was done according to an internationally recognized checklist for this purpose. The quality of the total evidence for each presented outcome was assessed using the GRADE tool for grading of evidence. The reviews were classified according to an adapted version of the EPOC (Cochrane Effective Practice and Organisation of Care Review Group) taxonomy of organisational interventions. Screening, inclusion or exclusion decisions and critical assessments were done by two persons independently. Grading was done by one person while another person checked the data extracted and the grading of the evidence.
Results
We identified a total of 17 systematic reviews of high quality. Three of these reviews did not reveal any relevant studies from the literature search. Most reviews were about interventions of a more limited character. Of these, one review was classified to the category multidisciplinary team s, ten reviews to setting for treatment , five reviews to integrated care services . Only one review was classified to the category continuity of care , which involves a more comprehensive perspective on user needs, extent and continuity of care. For the categories changing professional roles and remote communication , we identified no reviews of high quality. In terms of content, the organisational interventions in each category complemented each other by having different purposes. However, the overlapping of studies in some reviews suggests some conceptual overlap between intervention contents.
We found that the intervention that was about continuity of care, intensive case management, possibly resulted in some fewer number of days in hospital, some more users who maintained contact with mental health care, a higher functional score, a few more who were living independently and possibly a slightly higher user satisfaction compared to standard care. The use of community multidisciplinary teams versus hospital-based services may lead to some fewer hospital admissions. Day hospital as compared to hospitalisation probably resulted in longer duration of the index admission but also probably a slightly better score for social function. For the other measured outcomes of these two comparisons, the results were either too uncertain due to wide confidence intervals or the quality of the documentation was too low for us to draw meaningful conclusions. For the rest of the other interventions with more restricted purposes, the quality of the documentation was too low, the confidence intervals for the results were too wide or the results were inconsistent across studies, making it difficult to draw conclusions about the effect.
Of the interventions that were compared with hospitalisation, such as community multidisciplinary teams, twenty-four hour care in residential units, day hospital, different types of home treatments and treatment in outpatient clinics, statistical significant differences between treating patients within or outside hospital, were not demonstrated for most outcomes. In such cases, the confidence interval for the effect estimate indicates that the intervention may have a better, equal to or less effect than the comparison intervention on the measured outcomes. This means that one cannot draw any conclusions about the true effect on the outcomes of the intervention.
For the interventions that we identified the evidence of effect was generally inadequate and contained inconclusive results. The quality of the evidence base was mainly graded down because of the risk of systematic bias in primary studies - unclear descriptions of how the randomisation sequence was generated, whether allocation had been concealed and inadequate treatment of dropout of patients - and because of the lack of precision of results.
Discussion
The detailed treatment content of the various organizational interventions was somewhat unclear, because overall this was only briefly described. If the interventions are to be implemented, the content will be of importance, not just the setting or structure. The use of ‘standard treatment’ as comparison group was inadequately described, making it difficult to assess what the organisational intervention actually was compared to. There is also a challenge that the terms used to describe the various organisational interventions are broad, general and vague, with the consequence that the boundaries between them are often unclear. Organisational interventions consist of complex services that are not possible to standardise. They will never be identically implemented from setting to setting. Interactions between all contextual factors are complex and difficult to predict. Consequently, trials of the same type of intervention sometimes demonstrated significant effects in one place but not in another. Because we do not know the context well enough we cannot explain these variations in effects.
The changes and reorganisations of mental health care that have been implemented in many parts of the western world has primarily occurred through health policy decisions based on general attitude changes in society and social, economic and ethical considerations, rather than scientific evidence.
Conclusion
Of the identified organizational interventions, intensive case management compared with standard therapy appears to be the intervention with positive effects on more outcomes. The quality of the evidence for these results was low. Overall, the scientific evidence for the effect of organisational interventions is generally of low and very low quality. The main part of the studies was of an earlier date. The scientific evidence appears to be a rather inadequate basis for decisions on future organisation of mental health care.