Progress in Reproductive Health Research |
Systematic Review of Evidence - The First Step in Research
Systematic review has become widely accepted as a means of evaluating the effectiveness of both medical and nonmedical interventions. Systematic reviews set out to find evidence from scientific studies, and to appraise and synthesize this evidence in order to provide answers to scientific questions. Because it identifies what is known and what is not known, a systematic review is invaluable as a first step before carrying out new research. What is the decision-making process regarding an intervention? First, one needs to establish that the benefits of the intervention outweigh the risks. Next, a variety of factors must be considered - such as differences in the epidemiology of the problem, costs, biological differences, and cultural and ethical issues. After the intervention has been implemented, there needs to be continuous audit of the situation in order to guide further decision-making and to show whether the intervention had the result that was intended.
Systematic reviews often help in selecting research priorities. Recommendations often have to be made without full evidence, so a common approach for assessing recommendations is to grade the evidence and strength of the recommendations. Such a grading system is described below (see box). By using this approach, users can see the quality of the evidence base for each recommendation. The transparency of the approach gives credibility where it is due and helps ensure that resources are not wasted on ineffective recommendations. It can be seen, for instance, that the use of expert panels to make recommendations should be limited.
The main characteristic of a systematic review is the use of an a priori protocol which includes an explicit and comprehensive strategy for searching, identifying and selecting studies for inclusion. The main characteristic of a systematic review is the use of an a priori protocol which includes an explicit and comprehensive strategy for searching, identifying and selecting studies for inclusion. Studies that are identified are then evaluated with regard to their methodological quality before they are included in the review. The Programme not only carries out its own systematic reviews but also commissions them from others. The topics for systematic review are chosen on the basis of three principles:
All protocols of randomized controlled trials supported by the Programme should be backed by a systematic review.
All substantial recommendations in reproductive health (whether medical or nonmedical) should incorporate the strength of evidence behind the recommendation. Recommendations should, wherever possible, be based on evidence synthesized through systematic reviews.
Scientists and health professionals in developing countries are encouraged to propose topics and to participate in the preparation of systematic reviews that contribute to mapping the best reproductive health practices.
As well as preparing systematic reviews, the Programme conducts methodological studies on systematic reviews and meta-analysis in order to assist the development of these research tools.
The Programme's major partner in preparing systematic reviews is Cochrane Collaboration, and its main vehicle for distributing the reviews is RHL.
Grading of evidence I Based on well-designed randomized controlled trials, meta-analyses, or systematic reviews. II Based on well-designed cohort or case-control studies. III Based on uncontrolled studies or external consensus (expert committees). Strength of recommendation A Directly based on Category I evidence. B Directly based on Category II evidence or extrapolated recommendation from Category I evidence. C Directly based on Category III evidence or extrapolated from Category I or II evidence. D Based on the group's clinical opinion.