Towards a Cochrane Information Retrieval Methods Group: a progress report

The international Cochrane Collaboration, a spearhead of the evidence-based healthcare movement, was established in 1992 in response to Archie Cochrane's challenging assertion that many healthcare decisions were not well-informed, were not based on reliable evidence and reflected a widespread collective ignorance about the effects of healthcare. Cochrane (1909–88), a British epidemiologist, published his ideas in an influential work ‘Effectiveness and efficiency: random reflections on health services’.1 Subsequently, he summarized his credo ‘it is surely a great criticism of our profession that we have not organized a critical summary, by specialty or sub-specialty, adapted periodically, of all relevant randomised controlled trials’.2 Consequently, information retrieval lies at the heart of an evidence-based healthcare movement that is committed to enabling patient care decisions to be well informed and based on up-to-date, reliable information and knowledge. For a review to be unbiased it is essential that a broad and sensitive literature search is conducted to retrieve the maximum number of randomised controlled trials and other relevant published and unpublished studies. However, until recently, information professionals contributing to the Collaboration have lacked a forum within which they could systematize their provision of expertise, advice and support, conduct research and facilitate information exchange. Cardiff's links with evidence-based practice date from Archie Cochrane's appointment in 1948 to the MRC Pneumoconiosis Research Unit based at Llandough Hospital. Subsequently, they were strengthened with his tenure from 1960 as Professor of Tuberculosis and Chest Diseases in the Welsh National School of Medicine (now University of Wales College of Medicine—UWCM). It was during this time that Cochrane, together with colleagues, developed his seminal concepts. The Cardiff connection was further boosted during Doctor (now Sir) Iain Chalmers’ time at UWCM during the 1970s when he picked up the evidence-based torch from the retired Archie Cochrane. Chalmers went on to Oxford to establish the first-ever Cochrane Centre. During the 1990s UWCM Library identified support for evidence-based approaches to healthcare as part of its core business and named its site at Llandough Hospital ‘The Archie Cochrane Library’. This mission is now led by the Library's Support Unit for Research Evidence (SURE), currently working with systematic review teams and other projects and initiatives evaluating the evidence-base of healthcare. The first glimmering of a proposal to establish an Information Retrieval Methods Group within the international Cochrane Collaboration stemmed from a discussion with the UWCM Vice-Chancellor about the library's wish to build on existing support for evidence-based approaches. Membership of the proposed Group will be international in scope, drawing from those already contributing to the Collaboration as well as some first-time contributors to the formal Cochrane agenda. It will have five principal functions Providing policy advice on information retrieval issues to the Collaboration world wide and liaising with publishers, database producers, database suppliers and with the Cochrane Library Users Group to see how the Group's findings might be relevant to the Cochrane Library. Providing training and support in effective information retrieval skills for those undertaking relevant searching activities and for those involved in training others—including workshops at the international Cochrane Colloquia. A Web resource will be developed, identifying and evaluating databases of potential value to those preparing Cochrane Reviews. Current and completed information retrieval research will contribute to the Cochrane Methodology Register. Conducting empirical research, including systematic reviews into information retrieval methods and strategies as appropriate to the developing remit of the Collaboration. Helping to monitor the quality of systematic reviews including liaison with the Cochrane Quality Advisory Group and others with regard to possible improvement. The Group is expected to have a role in advising on good practice, in reporting search methods and working with the Quality Advisory Group, the Handbook Advisory Group and others. Serving as a forum for discussion and exchange of views regarding the contribution of information retrieval methods to the goals of the Collaboration. This will involve active liaison and co-operation with Advisory Groups within the Collaboration, the Cochrane Library Users’ Group as well as other relevant national and international Groupings. Finally, the work of the Group will be disseminated as widely as possible to encourage further discussion, to stimulate the development of its agenda and to recruit additional collaborators. Despite UWCM's long-standing commitment to furthering the evaluative approach to healthcare, it had not previously contributed formally to the Cochrane Collaboration. At an early stage, therefore, it was important to test the initial (and rather vague) concept of the proposed Methods Group with staff of the UK Cochrane Centre. An initial draft of ‘A Proposal to Establish a Cochrane Methods in Information Retrieval Group’ was submitted to Sir Iain Chalmers and his colleagues. With their advice and encouragement, a second draft was developed and entered into the three-stage process required before a possible submission for registration and adoption of the Group by the Cochrane Collaboration namely: a pre-exploratory national meeting; an exploratory international meeting; submission to the Cochrane Collaboration of an application for registration. Several key figures from the UK were invited to the pre-exploratory meeting in Cardiff on the 31 May 2002 and endorsed the general direction indicated in the second draft of the document. Then the text was scrutinized word by word. Nuances of phrasing and terminology were debated. Areas of potential collaboration, or conflict, with other initiatives and programmes were identified. Changes were made to the text in the interests of clarity, accuracy and to ensure close alignment with the requirements and structures of the Cochrane Collaboration and, particularly, of its other Methods Groups. Following further consultation via e-mail, an amended and revised document was adopted as draft three. With support assured from the UK, the next stage was to seek substantial international commitment at the full exploratory meeting held in central London on 2nd December 2002 to coincide with the International Online conference. Eighteen colleagues from the global information community, within and outwith the Cochrane structures, from USA, France and the UK, attended in person or by video link. Apologies for absence and expressions of support were received from a further 15 figures from UK, Eire, USA, Australia and New Zealand. The meeting unanimously endorsed the principles behind the proposed Group and agreed that it should concentrate on providing practical support for the development of information retrieval techniques and facilities for information researchers. Those present volunteered to contribute to various areas of the proposed programme of work. The draft proposal document had been recast, as required, into a ‘Module for Publication in the Cochrane Library’ and, following some modifications, was accepted for submission for registration by the Collaboration. An informal supper followed the meeting—offering further opportunities to pick over the bones. By the time this paper appears in print it is expected that the Group will have submitted itself for formal recognition by the Cochrane Collaboration. If this succeeds, the programme of work outlined in the section above entitled ‘What will the Group do?’ will be rolled out as an action plan with agreed responsibilities, targets and time scales. The central significance of information retrieval to the Collaboration and the necessity of ensuring that the Group works effectively—and with a measurable impact—means that additional members from anywhere in the world who wish to make a contribution will always be welcome. Further details and information are readily available from the Convenors (Carole Lefebvre—clefebvre@cochrane.co.uk, Steve Pritchard—pritchard@cf.ac.uk) or the co-ordinator (Alison Weightman, weightmanal@cf.ac.uk). Professor Stephen Tomlinson struck the initial spark which Sir Iain Chalmers fanned into a flame. Subsequently the fire was maintained by the ready and congenial support, advice and guidance that was always available from staff of the UK Cochrane Centre—in particular Mike Clarke, Carol Lefebvre, Jon Deeks and Sally Hopewell. Grateful thanks are extended also to all colleagues and protocollaborators who have contributed their effort and support.