I currently have a number of Postgraduates working on systematic literature reviews and we have reached the 'Search Strategy' stage. Our reviews focus on stroke rehabilitation, treatment of stuttering/stammering in adults and lymphoedema post breast cancer. We only have access to the free database Pubmed, and to Science Direct in our academic institution as we do not have a School of Medicine.
Before we go looking for collaborators with access to CINAHL, EMBASE, OVID MEDLINE etc, (and I notice that most published systematic literature reviews search 3-4 different databases) when can I be confident that I have searched sufficiently to not have missed any important RCT's?
Thanks Ken
Hi Kenneth Monaghan ,
In my opinion, we don't have/we don't know the exact number of "how many database should I use in my SR-MA?". Many materials in this topic talk in use a minimum of 3 (correct) database. It is important to do an extensive search and then select the most relevant database in your topic.
Good luck!
:)
Hi, exactly there is no specific number of a database in SLR. It depends on a research question, that is related to single disciplinary or multidisciplinary.
Good luck
I recommend following Cochrane's guidelines (https://training.cochrane.org/handbook/current). You cannot go wrong following them. You have to use MEDLINE, CENTRAL, and Embase. Read Chapter 4 in the handbook for additional recommendations more specific to your research question.
In general, the more databases you search, the better it is to improve the "systematicity" of your review.
As stated in the Cochrane handbook, "the three bibliographic databases generally considered to be the most important sources to search for reports of trials are CENTRAL, MEDLINE and Embase" (chapter 4).
Additionally, in an exploratory study published a few years ago, it was reported that "optimal searches in systematic reviews should search at least Embase, MEDLINE, Web of Science, and Google Scholar as a minimum requirement to guarantee adequate and efficient coverage".
Article Optimal database combinations for literature searches in sys...
In particular, including specific sources like Google Scholar along with more traditional scientific databases (such as MEDLINE) may help you to screen the gray literature as well.
Finally, if your topic regards rehabilitation, you may consider searching even PEDro, a free database entirely dedicated to studies about Evidence-Based Physiotherapy (https://www.pedro.org.au/).
Best wishes for your research!
Maria, thank you for taking time out to answer my question. I hope all well in Montevideo, Uruguay. I think the most obvious reply to your suggestion is to possibly know if I have published a good quality RCT in a peer reviewed high quality journal, how many of those databases will pick it up? It would be helpful if we had a league of databases like the pyramid of strength of evidence that told us the number 1,2,3, etc databases to search. I do see that 3 databases at least are chosen in most decent systematic literature reviews I have seen published.
Thank you Ken.
Mohammed, thank you for taking time out to answer my question. I hope all well in Baghdad, Iraq. The topic involved does certainly seem to be important and for example if the topic has a Nursing subject matter then CINAHL would be suggested as a sensible database to search. Ultimately, Gordon L Warren has the answer when he suggests reading the Cochrane guidelines but I still would like to know what the minimum number of database searches I could get away with to save time and energy. Thank you Ken
Gordon, thank you for taking time out to answer my question. I hope all well in Atlanta. You have a huge ResearchGate score so I will very much consider your advice. I have published a Cochrane protocol in the past but never got to finish the full review because my topic was so huge and for many reasons but I always regret that I could not add that to my CV. I agree with your suggestion that the Cochrane Handbook is the place to go but because we have very little access to databases in our Institution I wanted to try and reduce the workload for the Postgraduates. As I suggested to Maria above it would be interesting to know the similarity from database searches. Thank you Ken
Michele, thank you for taking time out to answer my question. I hope all well in Parma. This was a brilliant answer and is probably the best summary of everything that was contributed before. Thank you so much for sending the reference for the Bramer 2017 paper. I will send a copy to all my Postgraduates to read because it explains the importance of those databases so well.
We will also take your Pedro suggestion. Good luck with your medical work. I would imagine that in your specialty you have been extremely busy over the last few months due to COVID etc. Thank you Ken
Kenneth Monaghan, thanks for your reply: I am glad to prove helpful with my hints!
If you need additional material to provide your postgraduate students with a brief methodological guide about how to properly understand reviews of the scientific literature, then take a look at our latest article on the topic:
Article Reading and interpreting reviews for health professionals: a...
This guide was specifically designed for healthcare professionals with limited or no research background. Furthermore, it includes both explanatory examples and easy-to-understand suggestions to translate theory into practice, so it could be precious for junior researchers.
Best wishes for your work!
Michele, congratulations on writing this excellent paper and I have forwarded it to my 'Neuroplasticity Research Group' team and advised them to read it. Thank you so much for your assistance and good luck with your work. If you ever are travelling to Ireland then we would love to buy you a coffee or maybe a pint of Guinness to thank you!
- Badges
- Science topic
- Similar topics
- Human-Computer Interaction
- Collaboration