At DistillerSR, we believe in the power of evidence-based research. As a software company, we believe in harnessing the power of technology to make the lives of researchers easier, allowing them to work more efficiently and effectively.
There’s a lot of hot tech emerging with the potential to cause a paradigm shift in the systematic review community. So how is technology solving today’s biggest systematic review challenges? Let’s take a look:
1. Comprehensive Tools
Tools that are specific to singular systematic review workflow tasks and stages have existed for some time. Think reference management software, data repositories, screening software — these are all common tools used by researchers today. The issue with these tools is that they only help with one part of the systematic review workflow, and researchers often run into compatibility issues. For example, if something in the review changes, it’s much harder to go back into each different program and make changes to all the reports, outcomes, files etc. This creates a risk of inconsistencies or errors in the review.
Comprehensive tools such as DistillerSR are designed to address a broader section of the review lifecycle. A comprehensive tool is one that performs a multitude of systematic review tasks across the entire workflow. From importing and storing references, screening, data extraction, and creating reports–a comprehensive tool can reduce friction, save time, help ensure adherence to protocols, and lower the risk of inaccuracies and errors.
However, it’s important to also note that not all tools are designed equally, and might not meet the very specific needs of the user. If you are currently in the market for systematic review software, it’s important to research your options to find the program that best suits your unique needs.
Automation has already been one of the biggest game changers in terms of improving the speed of the systematic review workflow and streamlining simplistic tasks for researchers. As researchers struggle to keep up with the demand for quality evidence, there is enormous potential for innovative approaches to automation to help them work faster and more accurately.
Although some might wish it so, automation doesn’t mean pressing a button and having the review done for you. Today’s automation focuses on addressing the repetitive, mechanistic tasks in the systematic review workflow. Think of it this way: if you feel like a computer when you’re doing the task, a computer can probably do it.
Repetitive tasks that do not require the analytical skills of a human brain are particularly amenable to automation. For example, full-text retrieval, AI-driven screening prioritization, deduplication, and NLP-driven screening using classifiers are all tasks that can be completed quickly and easily with the help of automation today.
The future of automation is boundless at this point. As the technology becomes more refined, we will see further developments in NLP-guided data extraction and other automations to support research teams.
Interoperability means technology used to connect external systematic review resources as well as other platforms and formats. In DistillerSR, this looks like integrated features that allow users to import references directly from PubMed, import hierarchical datasets as JSON files, reuse files, and more. With new integrations becoming available all the time (the newest DistillerSR update included some pretty exciting integrations to help make full-text procurement faster and easier), we will see interoperability as an increasingly integral feature that will ultimately save significant time for researchers during their review.
Why researchers need this technology today
Evidence-based researchers can’t be replaced for tasks such as developing protocols, defining research questions, and assessing the validity of individual sources of evidence and the applicability of the evidence. By saving time on the repetitive, mechanistic tasks involved in a systematic review, researchers will be able to optimize their time and focus on what really matters.
Organizations such as the International Collaboration for the Automation of Systematic Reviews (ICASR) underline the importance of creating tools to help automation in the name of giving researchers more opportunities to do the meaningful work they’ve been trained for.
Upon its inception in 2015, ICASR’s mandate was centered on the development of tools to automate certain SR tasks, but today, the mandate has shifted to identifying what tools currently exist and how to increase the uptake of this type of technology in systematic reviews.
Our colleagues at ICASR believe that automation in systematic reviews is the present and the future–and we agree. By making more comprehensive tools with interoperability features and automation, researchers could save more time on their work with little to no risk. Although it must be said that many tools are currently in their infancy and must be used with caution, currently there is no limit on the development of such tools. As always, a pragmatic approach is best when integrating new technologies, but there is no reason why today’s researcher can’t start reaping the benefits of all the new “hot tech” in systematic reviews today.