Bookshelf

Search strategy

All databases were searched using an adaptation of the economics search filters developed by the NHS Centre for Reviews and Dissemination combined with the search terms used in the effectiveness literature search strategy. The overall search strategy drew on work on retrieving cost information previously conducted at the NHS Centre for Reviews and Dissemination107 and used the terms ‘cost benefit’, ‘cost effectiveness’, ‘cost utility’, ‘cost consequences’, ‘cost minimisation’, ‘economic evaluation’, ‘quality of life’, ‘utility’, ‘incremental cost effectiveness analysis’, ‘incremental cost effectiveness ratio’, ‘net present value’ and ‘incremental net benefit’ combined with the search terms used in the clinical literature search strategy. Sensitive searching [e.g. economics (ec) as a floating subheading] was used. Mirroring the effectiveness review, terms were developed with input from experts in the field.

In April and May 2012 a search was conducted in the following databases for studies of peer interventions in prisons that included economic evaluations, health utilities or other cost information: ASSIA, The Campbell Collaboration Library, Cost-effectiveness Analysis Registry, CINAHL, Conference Papers Index, Dissertations & Theses, EMBASE Classic, EMBASE, Research Papers in Economics (RePEc) (IDEAS), IBSS, MEDLINE, MEDLINE In-Process & Other Non-Indexed Citations, National Criminal Justice Reference Service Abstracts, NHS Economic Evaluation Database (NHS EED), PsycINFO, Social Services Abstracts, Sociological Abstracts, Science Citation Index Expanded, Social Sciences Citation Index Expanded, Conference Proceedings Citation Index- Science, Conference Proceedings Citation Index – Social Science & Humanities. Full details of database hosts and database search dates are available in Appendix 8. The searches identified 1347 references, which were reduced to 1157 following deduplication.

Databases were selected that were likely to contain economic evaluations and cost and health utility studies, along with a range of health and social databases that include prison- and young offender-based studies. Grey (unpublished) literature was sought from conference, dissertation and working paper databases. The selection of databases was consistent with and complementary to those used for the effectiveness review searches. The search strategies were devised in collaboration with the effectiveness review information specialist. The searches were constructed around three concepts: prisoners, peer interventions and cost/health utility studies. The effectiveness review project team supplied potential search terms for ‘peer interventions for prisoners’. A draft MEDLINE search strategy for the effectiveness review was checked by project team members and the search modified accordingly. The strategies comprised both text word searches and subject heading searches. For pragmatic reasons proximity operators (adjacency) were used between prison and peer intervention text word searches to improve the search precision. Some databases had the subject heading ‘peer intervention’ but it was discovered that this did not appear in some known relevant references. Therefore, a number of subject headings (e.g. ‘focus groups’, ‘social support’) were included in our search to retrieve references with relevant peer intervention content but without the ‘peer intervention’ subject heading. Searches were not limited by language or publication date.

The final MEDLINE ‘peer interventions for prisoners’ search was translated as closely as possible to all other databases. The cost search for most databases consisted of the ‘peer interventions for prisoners’ search plus a search for cost or health utility studies. We used previously designed searches to identify cost and health utility studies, which were adaptations of the NHS EED search strategies [see https://sites.google.com/a/york.ac.uk/issg-search-filters-resource/filters-to-find-i (accessed June 2014)]. Searches of economic databases were developed separately from but consistently with the effectiveness review searches. Full details of all search strategies are provided in Appendix 8.

Bibliographies of articles selected for inclusion were hand searched for relevant studies, and the wider research group was also contacted for advice, for peer review and to identify additional published and unpublished references. Any additional papers were subjected to the abstract review process before inclusion in or exclusion from the quality assessment process.