Funded projects

Mental health problems are as common as asthma and, like asthma, are strongly influenced by where we live. One aspect of our living conditions - crime - is an important determinant of both mental and physical health. For example, being afraid of crime can lead to anxiety, depression and insecurity, and lead to reduced social engagement, mistrust of others, and reduced physical activity and other behavioural changes. Moreover risk of crime, and fear of crime are higher in areas of poverty, unemployment and deprivation. On the other hand, it has been suggested that other community and environmental improvements, including some types of policing, can help the development of "mentally healthy neighbourhoods". So it may be possible to improve public mental health through intervening in communities. But how strong is the evidence?

This study aims to find out. In three linked reviews we will bring together the considerable body of research on the topic to allow us to assess the strength of evidence, to determine where and how to intervene to improve mental health. We will start by reviewing existing theories which describe the psychosocial and contextual mechanisms linking crime and health and wellbeing, and their relationship to poverty and income inequality. We will also conduct focus groups in 3 cities to add to the value of the review by helping to refine the review question, understand the pathways between crime and mental health, and aid interpretation of the reviews' findings. This will also allow us to explore community priorities for information on this topic.

Methods: We will start with a review of existing theory describing the relationship between crime, and health and wellbeing outcomes. This will inform the development of a logic model describing the pathways between interventions and outcomes. The use of logic models in the systematic review process has been recommended (see e.g. the recent MEKN report, which emphasises their importance in developing a social determinants approach to tackling health inequalities), but this approach is still uncommon. We will follow standard systematic review methods (CRD, 2009) in doing so. We will then review the evidence on the effectiveness of relevant interventions (as described under "Interventions being evaluated", below). We will also review the qualitative research where it relates to specific interventions. Here again the review will be informed by the new CRD guidance on incorporating qualitative evidence in systematic reviews.

Target population and setting: Community-based studies conducted in any country; English language only; with separate analysis of UK evidence to facilitate local application of findings.

Interventions being evaluated: Any intervention which aims to reduce CFOC and where the intervention involves changes to the built environment or is an intervention delivered at a community or neighbourhood level, with the intention of reducing risk or actual levels of crime or disorder, or fear of crime, and where there is a health-related outcome measure. Area-level programmes to reduce disorder (e.g. diversionary programmes; neighbourhood watch or policing programmes) will be eligible. Individual-level other interventions and measures aimed at offenders will not be included - e.g. sentencing procedures and policies, and programmes aimed at reducing recidivism.

Search strategy: Searches of key databases will be conducted (See the accompanying Detailed Project Description), and we will liaise with the Campbell Collaboration's Crime and Justice group in developing the methods. We already have search strategies which we developed for previous reviews on the urban environment which will be adapted. We have already conducted pilot searches in psychological and crime databases. From our previous experience we know that electronic searches and searches of academic sources will fail to locate non-academic reports. We have also located some of these on websites (e.g. housing associations). As the project develops we will therefore liaise with relevant bodies to identify relevant sources of evidence (e.g., the Department for Communities and Local Government). We will also contact researchers involved in previous evaluations (such as the Crime and Disorder Reduction Partnerships) to identify relevant studies in the published and "grey" literature.

Section G describes the conduct of focus groups with local communities in London/Liverpool/Glasgow at which the scope of the review will be refined, and at which findings will be presented and discussed.

Review process: The identification of the studies (i.e. the initial "sifting" of the results of the searches) will be carried out jointly by the two researchers, with disagreements brought to the wider team for discussion. An overall measurement of agreement (e.g. Kappa statistic) will be calculated. The methodological appraisal of the included studies will be conducted as appropriate to the study design. We have previously used the Effective Public Health Practice (EPHPP) tool for public health systematic reviews, and will pilot it (suitably adapted) for use in this project.

At the Outline stage, the Board asked about the methods for reviewing the qualitative research. The general approach to integrating the qualitative and the quantitative evidence will be based on the new CRD (2009) guidance on this process.

Expected output of research: Some of the subgroups of quantitative studies may be amenable to meta-analysis where heterogeneity permits. Otherwise narrative synthesis will be employed. The review will be structured to make clear a) The theoretical framework, and likely causal pathways between interventions and outcome; b) the strength of evidence for the effects of existing interventions; c) implications for the design of future interventions; d) implications for targeting of interventions; e) implications for inequalities; and f) implications for further research.

Project timetable: The duration is 2 years.

Year 1: Conduct review of theory, develop logic model; obtain ethical permissions and conduct focus groups in London, Glasgow and Liverpool; conduct all electronic and hand searches; conduct review of qualitative data; initiate systematic review of intervention studies;

Year 2: Finalise review of quantitative (intervention) research; final round of focus groups; write and submit methodological paper, write and submit academic papers summarising results of qualitative and quantitative systematic reviews; write non-academic summary report for dissemination via policy and practice networks; revise logic model in light of new data from the reviews.

£391,169