Why conducting one if you can conduct 82 (living) reviews?

Session: 

Oral session: Innovative solutions to challenges of evidence production (1)

Date: 

Sunday 16 September 2018 - 14:20 to 14:40

Location: 

Fintry Auditorium

All authors in correct order:

Rada G1, Bravo G2, Ortiz L2, Pérez-Bracchiglione J3, Madrid E3, Pérez-Cruz P4, Juri C4, Schmidt J4, Arancibia M3, Meza N3, Lobos D2, Zavala C2, Allende F2, Corsi O4, Llovet V4, Morel M2, Guinguis R4, Papuzinski C3, Peña J4
1 Epistemonikos Foundation, Chile
2 Centro Evidencia UC, Pontificia Universidad Católica de Chile, Chile
3 CIESAL, Universidad de Valparaíso, Chile
4 Pontificia Universidad Católica de Chile, Chile
Presenting author and contact person

Presenting author:

Gabriel Rada

Contact person:

Gabriel Rada
Gonzalo A Bravo-Soto
Eva Madrid
Abstract text
Background:
Conducting a systematic review for the first time involves a substantial amount of work for a team. The incremental effort to update or to conduct more reviews in the same topic decreases, since many components of the protocol are reutilised, and there might be gains derived from parallelisation of tasks.

Objectives:
To demonstrate a model of parallel production of multiple living systematic reviews in the same topic.

Methods:
We developed a common protocol for a suite of reviews about medical use of cannabis and cannabinoids, and a map of all the possible PICO questions on that topic based on existing systematic reviews available at Epistemonikos database and other documents.
We run monthly searches in MEDLINE, Cochrane Central Register of Controlled Trials (CENTRAL) and Embase, and search in other sources. Deduplicated records are screened in duplicate for a team of researchers using an online tool developed by our team. Full texts of potentially eligible studies are evaluated in duplicate as well. Discrepancies are solved by a third reviewer.
All of the included trials (i.e. randomised trials on medical use of cannabis and cannabinoids) are allocated to the specific reviews, and then managed by independent teams for each review.

Results:
Based on the revision of 151 existing systematic reviews that include 528 primary studies overall, we developed a map of 82 PICO questions in this topic, namely different conditions or symptoms that might be treated with the intervention.
We have screened 20,118 records, and identified 40 new trials not included in existing reviews. A team of 27 people, composed of methods and content experts is now extracting data for all the reviews in parallel. The completion of all the reviews and submission for publication is expected to be August 2018. All of the reviews will be maintained as living systematic reviews, that is they will be updated as new evidence emerge.

Conclusions:
A production model of multiple living systematic reviews in the same topic is feasible if a large team of researchers is centrally co-ordinated, and the appropriate technological tools to streamline and manage the process are available.

Patient or healthcare consumer involvement:
None.

Relevance to patients and consumers: 

We have developed a production model of multiple living systematic reviews in the same topic by a large team of researchers coordinated centrally, using technological tools to streamline and manage the process. This would allow to produce a substantially larger number of systematic reviews, and keeping them up to date. This approach is a potential solution to the problem of Cochrane reviews having a low coverage of all existing questions, and being frequently out of date.