Labelling: A Missed Step Between Record Screening and Report Screening of Systematic Reviews

Farhad Shokraneh
3 min readMar 8, 2022

--

Increasing Value and Reducing Waste during Systematic Reviews

Sorry for keeping you a cold turkey for a few weeks! For those of you who asked, hell is a really busy place to be! Thank you for your great wishes to send me there.

I knew you already knew about it! Apparently, no one has taken labelling seriously so far, so I thought to bring this to everyone’s attention.

Labelling is important and can save your XXX; by XXX, of course, I mean time and resources!

In the good old days, when we were screening in EndNote, we used to create groups, sub-groups, smart groups to assign the records to them. Age of boredom and wisdom!

At the age of Rayyan and Covidence and the big family of semi-automation programmes, you can undoubtedly do better with more features and optional accessories!

What is labelling?

Labelling or tagging is the practice of assigning a keyword or a group of keywords to some or all of the records during screening. Then, the system allows you to choose one of these keywords or tags and see all the assigned records to that keyword. A good example is hashtags on Twitter.

What’s the point of labelling?

While many might see it as an extra step, and indeed, it might be irrelevant to many systematic reviews, you will find many benefits in labelling [at least] the included records.

  1. Classification: Librarians can remember that when social media were finding their dance feet, some started talking about hashtags and folksonomies (in relation to taxonomies) as classification done by folks! Many of us love classification simply because it feels good, and we control the information. Although it is not a controlled vocabulary, it has the same function and helps you in information [record] retrieval.
  2. Map or Overview or Scoping: Labelling your included records allows you to have a classification system giving you a map in the jungle of included records. Believe me; you would appreciate such a map. It can also show biases in the topic or search or journals’ scopes.
  3. Plan, Narrow Down, Broaden Up: such an overview can help you narrow down, broaden up, or plan ahead for the rest of your review. For example, if you have a tag such as ‘Non-English’ assigned to 38 records, you need to plan to find translators or money for translation ASAP. If the sub-topics look like a manageable sub-systematic review, why not ask one of your enthusiast students to dig it!
  4. Changing the Protocol! You can change the protocol for your systematic review, including your eligibility criteria, after seeing the distribution of the labels. I know it might look like an odd or unprofessional suggestion, but it is based on realism. We don’t always have a good idea about the monstrosity of the review at the start.

Systematic reviews — in most cases — are research type without the pre-specifiable sample size (number of included studies). Pragmatic decisions are usually required to adapt the size of the review to the available resources. Any changes from the pre-specified protocol should be documented and reported with reason. Changing the protocol is not the forbidden fruit.

5. An Extra Scoping Review: not many say no to an extra publication! It can be a rapid bonus for your team who struggle to recycle data from rubbish (published studies) for months. Maybe you are not ambitious, and a conference abstract would satisfy you before you throw away your screening project (archive it). It can also be part of your report to your funder to keep them happy while you are scuba diving into the full texts. Look at it as a side product.

Conclusion

It might become routine for many systematic reviews to have an extra step of labelling between the record screening (title-abstract) and report screening (full-text). Labelling can benefit you in several ways; however, you should not dive deeper than needed because it might become time-consuming to return to the surface. Remember, the primary purpose behind labelling is to increase value, reduce waste, and not waste your limited time and resources.

If you liked this blog post, please support me by pressing the green Follow button and signing up so I can write more. Email Subscription Dysfunctions. Thank you :D

--

--

Farhad Shokraneh
Farhad Shokraneh

Written by Farhad Shokraneh

Evidence Synthesis Manager, Oxford Uni Post-Doc Research Associate, Cambridge Uni Senior Research Associate, Bristol Uni Director, Systematic Review Consultants

No responses yet