Hand searching

Electronic searches miss many reports of trials.12 This may be because the articles were not written with a clear indication in the title or abstract that the contents describe a clinical trial, or because the journal has not been entered in a database. As there are well over 200 international dermatology publications, many of which are not on databases, electronic searching alone cannot identify trials that may have been published in these journals. The aim of the Collaboration is to identify reports of trials back to 1948. Much of this work can be done only by searching for trials by hand - "hand searching".

Hand searching refers to the task of physically turning each page of a journal looking for indications that an article reports a clinical trial. This huge task is being undertaken by the Collaboration as a "one-off" exercise to trawl systematically the medical literature worldwide for reports of trials. Each review group is responsible for searching for trials in journals relevant to their scope. This exercise includes searching journals that are listed on databases such as Medline. The advantage of hand searching is that as each article is examined page by page, references are found to trials that are mentioned only in the "Methods" section of a paper or in the correspondence pages. An example within our own review group shows the benefit of hand searching. When all the issues of Archives of Dermatology between 1976 and 1998 were searched, 270 trials were found. Of these 103 had not been identified by Medline because they had been classed as "journal article" or "letter", with no indication they were trial reports. In addition, there were 16 articles that were categorised as "randomised controlled trial", "controlled clinical trial" or "clinical trial" when in fact on closer inspection they were not trials at all. The Cochrane Collaboration and the National Library of Medicine are working together on the "Medline Re-Tagging" project to rectify these errors.

To prevent duplication of the considerable effort involved in hand searching, the Collaboration holds a Master List of the journals that are being searched. The Trials Search Coordinator of each review group is responsible for registering journals and for training the hand searchers.

A number of criteria must be met for an article to be a report of a clinical trial.

• A trial must be planned in advance and be a comparison of two or more interventions (where one may be a placebo).

• A trial must involve a single original population of human beings, or groups of human beings or parts of their body.

• If the allocation of the interventions to the single population is randomised and explicitly stated to be so this is classified as an RCT.

However, in many reports the method of allocation of interventions is not stated very clearly. These reports are classified as CCTs. There is much variation in the quality of the random allocation even among RCTs, so inevitably many of the CCTs turn out to be unreliable sources of evidence for the systematic reviewer. However, the hand searcher's task is to find reports of trials, not to make value judgements about the trials identified.

Was this article helpful?

0 0
Natural Treatments For Psoriasis

Natural Treatments For Psoriasis

Do You Suffer From the Itching and Scaling of Psoriasis? Or the Chronic Agony of Psoriatic Arthritis? If so you are not ALONE! A whopping three percent of the world’s populations suffer from either condition! An incredible 56 million working hours are lost every year by psoriasis sufferers according to the National Psoriasis Foundation.

Get My Free Ebook

Post a comment