Dear Researchers,
I am trying to report worldwide prevalence rate of a certain phenomena, i.e. prevalence rate of the phenomena at different countries.
There are not many primary research in the field conducted in the last 20 years and/or most of the researches are with limited scope and hence not suitable for my current research. My queries are,
1. Can i do meta analysis on only 5-6 most suitable studies?
2. This prevalence is an important but only a part of my study. Can i do systematic review with the remaining infirmation, including more studies? ( i.e. may be 5-6 for meta analysis and 12-15 for systematic review on other variables?)
3. If the funnel plot shows 2 out of 6 studies out of plot, can i move ahead with this?
Your kind view ll be of extreme help.
Thanks,
Praheli
Hi Praheli,
A meta-analysis, although often the most definitive way of summarizing findings from similar research, should not always be the ultimate goal of a systematic review.
It would certainly be acceptable to summarize (statistically) only the 5 - 6 studies, assuming minimal heterogeneity, and have a narrative summary of the remaining studies.
Regards,
Hi
To see the world wide prevalence of a public health problem, 5 to 6 studies are not sufficient, I think. But still depends upon your topic of interest or research question.
You can confine your study to a particular region of the world, for initial analysis.
I absolutely agree with the answers by Godfrey Habil Mudhune and Abdul Hamid.
For the question number 3, analysis of publication bias with funnel plot is not recommended with less than 10 studies according to the Cochrane Handbook.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis