Linked PICO- connecting Systematic reviews, recommendations and patients
An increasing number of collaborators (see below) share a common interest in harmonizing data models/ representation of PICOs and common content in single studies,
systematic reviews, guidelines and decision support in order to improve patient care. We want to achieve easier export/ import between the various tools out there, better flow and accessibility of information, getting closer to Big Data through identifying what we call core elements, that are centered around PICO questions.
To get a picture of how easy it will be to harmonize core PICO elements we have started out by getting an overview over what is out there and where the overlaps are.
We would like to see the development of search engines that can pull data from various sources and show it to the end user. It's perfectly fine to keep your Guidelines, Systematic reviews or Evidence profiles on your local server as long as you make their data searchable, findable and/or extractable for others to see and use.
Existing data elements
We have started a Survey of datamodels/ data elements in existing tools for authoring/ searching/ reviewing/linking/ assessing or publishing guidelines/ systematic reveiws/ single studies/evidence profiles or decision support.
It's started it together with people from:
Duodecim/EBMeds, Cochrane, Epistemonikos, DECIDE, GRADE, Clinical Decision Support Consortium (CDSC) and Health eDecisions
But has expanded through collaborative work to include people from: Covidence, EROS, Dynamed, Dr.Evidence, DIPS ...
To look into how the flow of data and knowledge can help guideline developers, systematic reviewers and developers of decision support systems to continuously update their data, we have initiated a series of explorations of data transfer and started sketching out possible solutions and frameworks.
We would want people from any related tool, software, framework or organization out there to join this work:
Reference managers, Electronic Health records,, Knowledge repositories, Search engines, Guideline developer tools, Decision support systems, tools for Quality assessment and production of Systematic reviews, Research assisting tools and Medical journals.
We have set up an API for our MAGICapp and are currently working with other software systems to test strategies of proper flow of data and knowledge
Our APIs currenly using RESTful JSON, but we are looking at a move towards JSON for linked data: http://json-ld.org/
Contact Linn Brandt if you want to join us, or if you would like more information.