5 Tips for starting a systematic review

A systematic review is type of literature review summarising and critically analysing the results of available studies. Compared to a literature review, a systematic review follows a standard set of (systematic) stages to gather all the relevant literature for the research question.

My systematic review

My review is a qualitative systematic review, also called a meta-synthesis. I am exploring the barriers and facilitators of adults using an app or wearable for monitoring physical activity and/or sedentary behaviour.
My review is currently ongoing, and I am at the analysis stage. Therefore, my tips below will mostly relate to getting started with a systematic review.
My top 5 tips.

1. Do your research first.

This first one might be obvious, but before you start you need to know what has already been done. What research there is in your field and would a systematic review would be useful. Has a systematic review already been conducted? If there has, does it warrant an update?
You also need to decide on your research question, depending on what literature is out there, your question may be very specific or broad. Having a good research question and objective will help with the development of your protocol, search strategy and inclusion/exclusion criteria.

2. Write a protocol.

You need to know what you’re doing before you start your review. Developing a protocol will get you to think about why you are doing a systematic review, where you are going to search, what you might include or exclude, what data you will extract, and how you will analyse and report the results.
For my protocol, I used the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) 2015 checklist. This checklist was really helpful to make sure I had considered and thought about all the different parts of the systematic review before I started. I found my protocol extremely helpful to refer to between screening to ensure that the papers I was including were relevant and going to help me answer my research question.
Once we had a good draft of the systematic review, we registered the review on PROSPERO, the International Prospective Register of Systematic Reviews. This is a really good thing to do as it avoids the duplication of reviews, and improves transparency and rigor of reviews.
You could also think about publishing your systematic review protocol. My protocol has just been accepted for publication in Digital Health and the whole process was really useful, including the reviewers comments!

3. Think about having a team of double screeners.

My fantastic team of screeners have been so helpful! When I was doing my screening I missed a few relevant papers and excluded them when they were actually included. This is likely to happen with anyone, as screening can take a long time and can be mind-numbing at times. Luckily, as we are doing 100% double screening the second screener spotted them and included them. I found if I was not concentrating on it and giving it 100% of my attention, it was so easy to miss some. I think it is great to have that second screener there to just check it. Having second screeners has also been really useful to get a second set of eyes and a different perspective on the papers.
I did 100% double screening but I have a great team behind me willing to help and take on the extra work. Talking to fellow PhD students, most doing a systematic review have done 10 or 20% double screening, which is absolutely fine. Doing 100% double screening has really added to the rigor of the review and would recommend if you can find just one or two people willing to help with screening it’s a really useful thing to do.

4. Trial your search strategy.

Well, the search strategy took months! We searched, searched and searched again. Each time we learnt something new about the search, took terms out and put terms in. Even though it took a long time, it was worth it. Eventually we got to a really comprehensive search strategy that we were all happy with and it hopefully identified as many relevant papers as it could.
I also got to know the databases I was using. Adding relevant MESH or thesaurus term and using speech marks to search for phrases, and an asterix (*) for different variations of the word. I used 6 databases and each had its own hidden quirks. Therefore I would also recommend chatting with your subject librarian and getting some advice on your search strategy.
It was hard to find that point where I said ‘that’s good enough’ as I am such a perfectionist. We made so many changes and went back and forth on terms. It is inevitable that something is still missing but that’s why I am checking the reference lists and citations of included papers.

5. Get organised.

I cannot stress this enough! Get organised, have a spreadsheet that you can export all your references into from your searches. Add a column for your include/exclude decision, a notes column, plus columns for your second screeners decisions and their notes. I kept a copy of the main file for me and created second copies for my double screeners with just the papers they need to screen or a selection of papers. This way they didn’t have to deal with my huge spreadsheet of columns and codes that only make sense to me, and I could still work on my document at the same time as they were screening. I then just copy and pasted over their screening decisions when they send it back to me. I also had a lot of filters on my columns so I could easily select included or excluded rows papers or ones yet to screen.

Thank you for reading and I hope my tips are helpful.

If you are also doing a systematic review, I would love to hear about your topic and experiences. Please get in touch if you have any questions too!

Sign up for updates on new blog posts by entering your email address!