This chapter has been prepared by the Non-Randomised Studies Methods Group (NRSMG) of The Cochrane Collaboration (see Box 13.8.a). It is intended to support review authors who are considering including non-randomized studies in Cochrane reviews. Non-randomized studies (NRS) are defined here as any quantitative study estimating the effectiveness of an intervention (harm or benefit) that does not use randomization to allocate units to comparison groups. This includes studies where allocation occurs in the course of usual treatment decisions or peoples’ choices, i.e. studies usually called observational. There are many types of non-randomized intervention study, including cohort studies, case-control studies, controlled before-and-after studies, interrupted-time-series studies and controlled trials that use inappropriate randomization strategies (sometimes called quasi-randomized studies). Box 13.1.a summarizes some commonly-used study design labels for non-randomized studies. We explain in Section 13.5.1 why we do not necessarily advise that these labels are used in Cochrane reviews.
This chapter aims to describe the particular challenges that arise if NRS are included in a Cochrane review, and is informed by theoretical or epidemiological considerations, empirical research, and discussions among members of the NRSMG. The chapter makes recommendations about what to do when it is possible to support the recommendations on the basis of evidence or established theory. When it is not possible to make any recommendations, the chapter aims to set out the pros and cons of alternative actions and to identify questions for further methodological research.
Review authors who are considering including NRS in a Cochrane review should not start with this chapter unless they are already familiar with the process of preparing a systematic review of randomized trials. The format and basic steps of a Cochrane review should be the same whether it includes only randomized trials or includes NRS. The reader is referred to Part 2 of the Handbook for a detailed description of these steps. Every step in carrying out a systematic review is more difficult when NRS are included and a review author should seek to include expert epidemiologists and methodologists in the review team. As an example of such collaboration, a review of NRS included nine authors, five of whom were methodologists (Siegfried 2003).