With respect.

I'm not that professional to answer you very exact, but I'm certain that if you have a very expert and experienced team to search it could be confirmed your work is more reliable. Your search strategy and inclusion criteria you determine in your paper, could be helpful, as well. In addition, every study has unevitable limitations you can not do anything except do your best to minimize them and mention them in your paper. Good luck.

A good way to proceed is :

1) select a relevant database which has a good coverage in your research area

2) have a good research strategy (established generally with a librarian) to be sure you include important papers. There are several strategies (sensitive vs specific) the sensitivity (finding all relevant articles) may come with lack of specificity (a lot of paper to screen). again, this depends on your field of work and if specific filters have been previously validated.

3) a priori define inclusion and exclusion factors which also depends on your question of interest.

For the last part of the question you need to find a way to have an access to paper which are not open access. Open access is quite recent in some scientific field area and therefore only keeping free open access litterature could bias your findings and be considered as an important limitation in the peer-review process...

For the first question, using multiple databases as you suggest usually help to minimize the risk of missing relevant studies. The risk of missing studies by published year not usually an issue, since published year tend to be well documented and indexed.

However, missing studies due to a narrowed scope while searching for literature is always potential risk. If you have an reasonable knowledge of the field you are planning to review, you might have a range of publications already prior to the reviewing process. If you can find all of these publications during your scoping process then your scope is acceptable. If some of these are missing, you might need to extend the scope further.

Another issue that is difficult to account for is the inclusion of journals/publications that are not indexed in the big databases.

For the second question, one option is to contact authors of these subscription-only publications and request a copy. If some publications are essential yet no access can be granted, the last option is to purchase the publication.

1. You can't. The overlap between the different sources you use can help you to guess how much is missed.

2. If your library does not subscribe, inter-library loan, write to authors, pay for, etc.

Look at http://prisma-statement.org/prismastatement/checklist.aspx

Dear @Zahra Hadian, @Sébastien Buczinski, @Carl Alexander Frisk and @Daniel Wright

Thank you very much for your informative answer