Cochrane library

In the first article of Family Medicine's new series 'Evidence in practice', Ruth Frankish discusses the Cochrane Library, an electronic bible of systematic reviews

To provide effective health care to patients, decision makers about the provision of health care should be aware of all the available evidence on what does and does not work and on what provides the best value for money. However, keeping up to date with the best research evidence is a problem that confronts all health professionals. Very often, it is an impossible task because you need to be able to get access to the information in the first place, let alone have the time to search for papers, to sit and read them, to interpret, appraise and draw conclusions from them. In fact just to have reading time during the working day is a luxury that many health professionals do not have.

Reviews of research evidence can help increase access to this knowledge base by ordering and evaluating the evidence available. Systematic reviews of research are now widely accepted as providing the most reliable sources of information about effectiveness. They adhere to a strict methodology which aims to be explicit and objective in the way that various research results on a given topic are assessed and summarised. As comprehensive summaries of the available evidence, they are valuable sources of information for decision makers.

The Cochrane Library is an electronic publication providing regularly updated systematic reviews of the effects of health care. It is now regarded as the best source of information on the effectiveness of healthcare interventions. From its origins in the field of pregnancy and childbirth, it now covers many areas of healthcare, many of relevance to primary health care. The information on the Cochrane Library reflects the hierarchy of evidence, in that it contains full text of systematic reviews from meta-analysis of randomised controlled trials (RCTs), assessments of other systematic reviews and references to RCTs.

Four databases

The Cochrane Library consists of four separate databases, plus other information.

The Cochrane Database of Systematic Reviews (CDSR) contains the full text of systematic reviews carried out by people in the Cochrane Collaboration. This is split between on-going reviews and those that have been completed. An important aspect of Cochrane reviews is that they are kept up-to-date, in that any new, relevant research, identified via regular literature searches, will be incorporated into the review.

The Database of Abstracts of Reviews of Effectiveness (DARE), contains assessments of systematic reviews which have been published elsewhere in the medical literature. The abstracts comment on the quality of the methods used in the review and only those meeting certain quality criteria will be included in the quality section of the database. Those reviews which pass the evaluation process receive detailed structured abstracts that explain the aims of the reviews, their main clinical results or conclusions, together with a critical commentary of the methods used. The critical commentary outlines any shortcomings or areas of caution about how the review was carried out, which might need to be taken into account by those reading the review.

The Cochrane Controlled Trials Register contains the bibliographic references to all the randomised controlled trials identified by those in the Collaboration, through systematic hand-searching and database searching (of Medline, and more recently Embase). It now contains almost 200,000 references to RCTs and is now regarded as the best resource for finding RCTs. No other type of primary research is included.

The Cochrane Review Methodology Database (CRMD) contains references to articles and book chapters that deal with the methods needed to carry out a systematic review. It does not contain information about effectiveness, but is used as a resource for researchers carrying out systematic reviews, when they need help or suggestions as to how to carry out the various stages of the review.

Other information

As well as the four databases, there is a host of other information; contact details and information about the various entities within the Cochrane Collaboration, including the 'Cochrane Field in Primary Healthcare', which aims to ensure that there is a focus on primary care in systematic reviews produced by those in the Collaboration. You might want to look at this section to get an idea of the areas they are covering at present or intend to cover in the future. There is also a section containing a list of sources of effectiveness information on the web, a glossary containing definitions of some of the commonly used terms contained in systematic reviews and also the Cochrane Handbook, which takes readers through the various stages of producing a Cochrane Review.

Searches

Searching the Cochrane Library is via a search interface that allows you to either enter a few keywords to search with, or to construct more sophisticated searches. It is important to note that not all the records on the Cochrane Library are indexed using subject headings. Some are indexed using MeSH, as with Medline, but not all.

To do a complete search of the database, therefore, it is necessary to do a search that combines MeSH searches and text-word searches. For example it would be necessary to search for the MeSH term AGED to find references dealing with the old, but to find all references you might also do a search for OLD, ELDERLY or GERIATRIC as text words.

More detailed help in searching the library can be found in the help screens or on the guide which can be downloaded from the CD-Rom.

The next article in Family Medicine's series 'Evidence in practice' will take a look at how to assess the quality of systematic reviews.

Ruth Frankish is Cochrane Library Trainer, Centre for Reviews and Dissemination, University of York