How many more systematic reviews do we need - an estimation on the proportion of possible interventions covered by The Cochrane Library

Article type
Authors
Kamprath S, Timmer A
Abstract
Background: The number of systematic reviews prepared by The Cochrane Collaboration has shown an impressive increase since The Cochrane Library was first published. However, there are no data estimating to what extent relevant interventions have been covered.

Objectives: By the example of Type II diabetes, we aimed to estimate the number of relevant interventions amenable to systematic reviews of effectiveness and contrast this with the scope of topics evaluated.

Methods: Based on published clinical guidelines, relevant interventions for the therapy of Type II diabetes were identified. In addition, patient subgroups were defined. Outcomes were classified as follows: A - major (e.g. mortality, stroke, myocardial infarction) B - surrogate (e.g. glycaemic control, blood pressure), and C - socioeconomic (e.g. quality of life, days off work). The Cochrane Library (2006, Issue 1) was searched for diabet* (record title, abstract, keywords).

Results: Overall 153 diabetes therapy related interventions and 58 systematic reviews (17 protocols) were identified (see Table).

[table excluded]

Conclusions: Less than a third of all interventions identified as potentially relevant for the treatment of diabetes mellitus have been covered by Cochrane systematic reviews. Of note, quality of life and socioeconomic issues have not been considered in any review examining specific types of insulin therapy. Our data will be further explored by using an expert opinion based grading system of relevance for therapeutic interventions. A search for controlled clinical trials will also be performed.