Hi,
I am planing to write systematic review in rare disease. My inclusion and exclusion criteria meets 35 research articles. However, most of the studies are retrospective and only limited studies have case-control and clinical trials.
Is it appropriate to include these studies together in systematic review (retrospective studies without control)?
If the retrospectives studies fullfill your critera for inclusion, yes. Although for having more impact for your review, you should include only studies with a very solid method.
If you are plaining to do stadistic analysis ( Meta-analysis ) you also need to calculate the sensibility analysis and selection bias, wich means that you redo the analysis but in every attempt you exclude one of the studies that was part of your sample.
Thanks Bruno Oliveira.
I have a bit confusion regarding retrospective studies as some already published review includes these studies along with case-control and clinical trials.
I will shortout soon.
Go to the following link
http://joannabriggs.org/assets/docs/sumari/Reviewers-Manual_Methodology-for-JBI-Scoping-Reviews_2015_v2.pdf
It certainly is appropriate to include all kinds of studies in a systematic review. Each type of study of course has its own strengths and weaknesses. But it is good to have a look at a problem from multiple perspectives, different kinds of populations being studied, and using different statistical designs. And contrary to popular belief, randomized controlled clinical trials aren't necessarily the best type of study: for drug studies they are necessary, but other types of studies (for example retrospective analyses) often look at more real-world data and outcomes in a non-artificial environment, and often with larger numbers of subjects, than in clinical trials with carefully-chosen subjects and limited outcomes under study.
But of course some types of study like correlational studies can not prove cause and effect (but may be suggestive or hypothesis-generating) and are therefore weaker forms of evidence. As you go through the literature, you may find a progression of evidence and strength of findings over time, from weak correlational studies, to stronger hypothesis-driven studies, to larger well-powered studies that build on previous work and control for confounding variables.
NO, APPLES SHOULD NOT BE MIXED WITH ORANGES. Case students comes from real case with real data; this is called empirical data. Experimental data does not have the same status; it may come from simulation as part of experimental design.
META ANALYSIS is what you are asking when you say "systematic analysis." See details on meta analysis and how samples studies are selected and include in the meta sample. See link:
https://en.wikipedia.org/wiki/Meta-analysis
Hi Nirmal, How are you getting on with this systematic review? Did you include case studies and case control studies?
I can imagine there would be quite a mixture of literature types on your topic.
This may be helpful.
https://www.bmj.com/content/368/bmj.l6890
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis