blog

Status updates for research – Michel Counotte and Nicola Low, University Bern, Switzerland, and their group of researchers are the first to publish a living systematic review on F1000Research. They explain why they upgraded from a ‘classic’ to this novel article type and why it is a great step for researchers helping them to keep up to date with the latest evidence on Zika.

During large disease outbreaks, the scientific community focuses its attention and gets to work on research. The Zika virus outbreak in the Americas in 2015-2017 was large and affected many people. It sparked an enormous research effort across disciplines. The pace picked up especially after the World Health Organization declared a public health emergency of international concern on 1^st February, 2016.

Scientists collected evidence from direct observations from infected and uninfected individuals, as well as animal and cell models studied in the lab, and from behind the desk in mathematical and statistical analyses of surveillance data. The output has been immense. Early in 2016, around 300 new publications appeared per month. Even now, at the end of 2019, with the epidemic behind us, still around 200 new publications appear per month (Figure below).

Updating a classic

Typically, people who want to synthesise the evidence across studies will do a ‘classic’ systematic review. They define a research question and document the methods in a protocol, saying how they will search the literature systematically, select relevant publications, extract data, and put together the data. This is what we did early in the Zika outbreak.

A team of nine of us put together all the evidence we could find to see whether Zika was indeed the cause of the outbreaks of microcephaly (a condition where the head is smaller than normal due to damage of the developing brain tissue) and Guillain-Barré syndrome (GBS) (a condition that affects the nerves causing muscle weakness). The search identified 1631 studies and included 106 studies published up to May 2016. The conclusion was that “Zika virus is a cause of congenital abnormalities and is a trigger of GBS”. By the time it was published in September, another 747 studies had been published. How could we keep up?

The answer was a living systematic review. This allows for the continuous incorporation of evidence, while retaining the key features of a ‘classic’ systematic review. By running the searches and all subsequent steps regularly and repeatedly, reviewers can update the content more often to produce a ‘high quality, up-to-date online summary of health research, updated as new research becomes available’.

Where does F1000 Research come in? To update the findings of a review in a publication without causing confusion, you need a system that allows unambiguous tracking of different versions. Julian Elliott and colleagues, primarily pioneered the living systematic review to update systematic reviews on medical interventions, setting up and using the Cochrane Library to publish a new version every time a review is updated. But our questions about Zika aren’t (yet) about interventions, as our reviews are of observational epidemiological studies, laboratory research and modelling studies.

The F1000 Research platform works perfectly for living systematic reviews, which assigns the newly uploaded pre-peer-review manuscript as version 1 and then updates this after authors respond to reviewer comments and revise the manuscript. Rebecca Lawrence, Managing Director at F1000 and Sabina Alam, former Editorial Director of F1000 Platforms, encouraged us early on, but it took a bit of time to make the transition to a living systematic review. After automating our searches, we put together our first update in 2018, which reinforced the original conclusions through a descriptive synthesis.

The first of its kind

We are very proud that our latest update is the first F1000Research living systematic review. We focused our effort this time on studies of epidemiological study design and restricted our inclusion criteria. Here we could quantify the strength of association between Zika virus and its complications (). With every update, we are refining the evidence about the causal relationship, but also highlighting challenges, like trying to explain the large heterogeneity between studies.

In the future, we will carry on incorporating new evidence and we will also continue to improve our workflow. We still have to select studies and extract data manually. But we are exploring new dynamic methods to display information that changes over time, like the map of the distribution of Zika virus cases, which is produced using the shiny app.

Inspiring others to follow suit

We hope that this example inspires others to conduct living systematic reviews. We now maintain a curated database with all Zika virus-related research published since the first description of Zika virus in 1947, indexed and partly annotated. We welcome enquiries from anyone who would like to use this resource to review other aspects of Zika virus systematically, including its biology and immunology. We have also done a living systematic review of the sexual transmission of Zika virus and other flaviviruses.

In a time where ‘status updates’ are all around us, and with the technology available to us, would it not make sense to be able to provide these on our work as well? Living systematic reviews are a great first step.