Systematic review (SR) is one of main research area in health c are research, Contrary to narrative review, SR can provide more reliable conc lusion by comprehensive literature searching, critical appraisal of literatu res, synthesis and summarizing results such as meta analysis, From its beginning, SR has focused on randomized controlled studies (RCTs) which are considered standard methods to assess efficacy of health care intervention. As SR has been used widely in recent years, there has been increased need to include non randomized studies (NRSs). This need is more evident in situation where RCTa are not feasible or difficult to conduct such as pub lic health fields. Also as RCTs are done to selected population group and more ide al situation, studies based on more broad range of population or conducted in more ‘real’ world such as community setting or population based studies. Al l this factors raised increase need for inclusion of NRSs particularly in publ ic health field or health policy issue In this article, some aspects of conducting SR when including N RSs are discussed;
1. In which study areas, SR on NRS are more feasible?
2. Which characteristics of NRS are to be considered in research conduct?
3. What is the main differences between NRS and RCTs in interpreting results?
4. What is the limitations in interpreting results and drawing conclusion from SR on NRS?
NRSs are widely used in situation where RCTs are not feasible due to ethical problems, long follow up period, studies on rare disease or studies on side effect. Meanwhile, as NRS are more high risk of biases such as confounders or selection
bias, careful interpretation of study results are essentially needed. Several aspect of SR on NRS including study planning, study selection, critical appraisal, and interpretation, heterogeneity or publication biases are discussed in this article.
|