Funded projects

This project will review existing research about the effectiveness and cost-effectiveness of a range of health service changes all of which aim to reduce the need for inpatient hospital care. The main purpose of ...

This project will review existing research about the effectiveness and cost-effectiveness of a range of health service changes all of which aim to reduce the need for inpatient hospital care. The main purpose of the review is to develop a conceptual framework that will allow the variety of schemes and their implementation to be better described, and also provide a better basis for explaining variations in their effectiveness and cost-effectiveness. We will use an approach to systematic review called realist review which specifically aims to develop and test ideas about how such services are meant to be more effective than the hospital alternative. Such reviews produce explanations of how programmes work - and why they work better or not in different circumstances - by progressively specifying the underlying mechanisms which produce the programmes outcomes, and also collating empirical data about the specific contexts which help or hinder the operation of those mechanisms.The review will be conducted by a core team based in an established centre of excellence in evidence synthesis and systematic review methods (PenTAG, Peninsula Medical School). The Chief Investigator (Dr Anderson) and lead reviewer (Dr Pearson) have been producing high quality policy-relevant systematic reviews together for NICE on public health topics since 2008, and both have an established interest in extending the methods of realist review to a wider range of health policies and issues (e.g. incorporating economic evidence). They will work closely with a topic expert (Dr Sasha Shepperd, author of two of the relevant Cochrane reviews in 2008 and 2009) and a methodological expert (Prof Ray Pawson, who developed and has previously applied the approach of realist review).PenTAG is also closely linked with the Peninsula Centre for Leadership in Applied Health research and Care (PenCLARHC), which will enable formal input to the review process from NHS managers and other stakeholders in the South West through a Project Reference Group.The success the review will be measured by: the degree to which the resulting conceptual framework makes sense and is used by the main target audience (NHS managers and decision makers); the degree to which it can help explain the already documented variations the effectiveness and cost-effectiveness of such programmes; and, the degree to which future evaluations and systematic reviews in this area make use of the conceptual framework, and the key intervention mechanisms and contexts that it will describe, in designing data collection and analyses.We do not believe that this review will raise any significant ethical issues.

The evidence synthesis will be conducted using the approach of realist review. This approach to systematic review is inherently explanatory in aim, and in particular aims to specify the transferrable underlying mechanisms by which ...

The evidence synthesis will be conducted using the approach of realist review. This approach to systematic review is inherently explanatory in aim, and in particular aims to specify the transferrable underlying mechanisms by which a service or programme produces their intended outcomes, and also the important contexts which either help or hinder the operation of these mechanisms. The last phase of the review will involve 'testing' selected 'context-mechanism-outcome' configurations using a selection of empirical studies and previous systematic reviews.Realist reviews are also systematic reviews, in the sense that they have explicit research questions and clearly described methods for identifying, selecting, appraising relevant studies, papers or reports and also clear methods of evidence synthesis. however, another feature of our proposed review method will be the involvement of a Project Reference Group of NHS managers (from Devon and Cornwall) and academics with experience in evaluating or designing such services.Search strategy:Locating relevant evidence for a realist review entails using a different logic to that conventionally used in a systematic review. The search for evidence in a realist review is iterative, being progressively extended and re-focused (based on the identified sources) as the review progresses and greater insight is attained into the issues concerned. This means that searching may be needed at multiple points in a review as more data is needed to further understanding. The initial representative range of papers will be identified using the database search strategy used in the two Cochrane effectiveness reviews on the topic. However the search will be broadened. The study design methodological filter will be removed from this search so that the broader range of papers appropriate for a realist review will be identified. Reviews, commentaries and editorials will initially be included.Further searches and synthesisNext, in order to create a 'map' of underlying programme theories, a search to locate the range of underlying theories that underpinned the design and implementation of schemes will be conducted. Finally, a search for relevant empirical studies will be conducted, to identify a purposive sample of studies which reflect a wide range of the possible mechanisms, contexts and outcomes which underly the effectiveness of sich programmes. In between these search phases (see flow diagram) there will be an iterative process of note taking, annotation and conceptualisation to identify and then refine the posited context-mechanism-outcome configurations. Relevant data/information from included studies and reports will be extracted to a standard data extraction form. Evidence synthesisThe process of evidence synthesis in realist review is achieved by:a) Juxtaposition of sources of evidence - for example, where evidence about implementation in one source enables insights into evidence about outcomes in another source.b) Reconciling of sources of evidence - where results differ in apparently similar circumstances, further investigation is appropriate in order to find explanations for why these different results occurred.c) Adjudication of sources of evidence - on the basis of methodological strengths or weaknesses.d) Consolidation of sources of evidence - where evidence about mechanisms and outcomes is complementary and enables a multi-faceted explanation to be built.e) Situating sources of evidence - where outcomes differ in particular contexts, an explanation can be constructed of how and why these outcomes occur differently.The transparency of a synthesis in a realist review is achieved by documenting these reasoning processes, the way in which they are grounded in the empirical evidence.The findings of the review will be summarised through narrative synthesis, using text, summary tables and where appropriate graphics to summarise individual papers/reports and draw insights across papers/reports. Given the overall purpose and methodological approach of the review, a meta-analysis will not be conducted.