Hello! I'm working on a systematic review (without meta-analysis) and when evaluating the quality of the papers only one presented low quality. I did not find guidelines in the instrument itself for making such a decision and would like help. My research is exploratory and the subject is statistical anxiety.
I appreciate any help!
Thanks!
I think it depends on how many studies there are in this field of research. If you only have ten studies I would also included those of low quality, as they may still contribute with some knowledge . If there are 50 studies, I would probably make an exclusion criteria for those studies of low quality, as they contribute less to the overall knowledge in this field of research.
You always have the possibility to exclude low-quality papers from your review, but you have to mention that in your methodology protocol explicitly.
Actually no need to exclude the low-quality studies in the first place as long as the studies meet your inclusion criteria.
If you want to integrate meta-analysis, you can later conduct a sensitivity analysis excluding the low-quality studies to estimate whether excluding the low-quality studies have any effect on your overall results.
Good luck!
Dr Asiful
The role of a systematic review with or without meta-analysis is to appraise the quality of studies not to exclude them!!! Don't forget the small study effects. I strongly recommend that you read the AMSTAR2 criteria for good systematic reviews.
You can mention that in the discussion/ limitation section rather than exclude those articles.
As Md Asiful Islam mentioned, I think you could conduct a (qualitative) assessment with and without that particular study (IE what would be analogous to a sensitivity analysis if you were conducting a meta) and then comment on whether the conclusion you arrive at is significantly different.
If, for instance, you find that the (qualitative) results are no longer as impressive after excluding the high-bias study, then you'd obviously need to mention that in the limitations (IE that we have a greater degree of uncertainty in our conclusions due to the results/conclusions being no longer as clear-cut after excluding the high-bias study by X et al.).
Also, note that differences in the results between study results may be attributable to factors ASIDE from bias! (for instance, X et al. may have been analyzing a different population). So it is important to thoroughly inspect the difference in results and seek out the different possible explanations that could explain this discordance.
Hello Dr, the concern might be arisen especially if the low-quality paper has conclusions grossly different from the other papers. One possibility is to perform a sensitivity analysis with and without that paper, to understand how it impact on the final results.
- Badges
- Science topic
- Similar topics
- Papers