Hi to all!
I have another query on this topic, as discussed many times on research gate that less number of studies included in meta analysis results in low power prediction but at present I have different case in my hand . have a look and give your esteemed suggestions:
I carried out meta analysis of 2 studies with overall population size 86. I got I2= 50% which means I have low heterogeneity with in the data.
It seems reasonable to me to combine them, so as to give generalization of results available till now.
Kindly share your views
Regards
Hitesh
If it is possible to get the raw data from the original investigators, and the studies are of similar design, it may be possible to do a pooled data analysis.
Meta analysis itself is a statistical combination of results from two or more separate studies. As long as these two studies meet certain requirement: homogeneous, not at risk of bias, Not serious publication bias/ or reporting bias. Meta analysis is possible. the idea behind meta- analysis is to increase power, improve precision...
I agree with Marcel. To reinforce the point, see the SPSS syntax linked below. What it shows is that the pooled estimate of the odds ratio that one gets via the Cochran-Mantel-Haenszel (CMH) method is virtually identical to the pooled estimate of the OR via standard meta-analysis. And both of them are virtually identical to the OR from a logistic regression model that includes the explanatory variable of main interest, the stratum variable, but no interaction term. (The interaction term, when you include it, is virtually identical to the "test of heterogeneity" you get with the CMH method.)
HTH.
http://www.angelfire.com/wv/bwhomedir/spss/pooling_odds_ratios.txt
While the point estimate has no issue, the problem is precision - the meta-analysis of two studies results in spurious confidence intervals that are too narrow. Based on simulation studies, the minimum is five studies to get the correct estimate of precision
Hi Suhail. Maybe it's just too late in the day, but I'm struggling to understand how correctness of precision is determined. Are the simulation studies you mention published? If so, please provide links or citations. Thank you.
Hi All, The first steps of your systematic review are to outline your specific aim and refine your search strategy. We tell our students that, in many cases, the systematic review is used to identify gaps in the literature. As such, in the event that not much has been published on a given topic, your systematic review could be very valuable in identifying potential directions for future research. On the other hand, I believe that failing to include all relevant publications on a given topic could potentially bias your results. Traditionally we recommend an electronic database search to identify potentially relevant articles. The references below provide a little more detail. In addition, I would also recommend contact authors or leading experts in the field in case they may know of other articles not returned during the database search. Finally, I would recommend manually searching the reference lists of previous systematic reviews, meta-analyses, and original research papers for additional articles. Whiting P, Westwood M, Burke M, Sterne J, Glanville J. Systematic reviews of test accuracy should search a range of databases to identify primary studies. Journal of clinical epidemiology. 2008;61(4):357-64. Papaioannou D, Sutton A, Carroll C, Booth A, Wong R. Literature searching for social science systematic reviews: consideration of a range of search techniques. Health information and libraries journal. 2010;27(2):114-22.
Similar question was asked before, and the link below provides access to that thread as well:
https://www.researchgate.net/post/What_is_the_minimum_number_of_papers_that_one_can_use_for_a_systematic_review#view=5a60005ecbd5c2d1e73918b9
I'm not sure what simulations were run Suhail but if the meta-analysis needs 5 studies in order for the CI to correctly cover at the stated percentage then that just means the CI is calculated incorrectly.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis