 I'm going to introduce to you today the Prisma 2020 R package for producing flow diagrams that have the possibility to be interactive. So why are flow diagrams important for evidence synthesis? Well they often deal with very complex sets of data, lots of methodological stages, many different numbers from different places, and lots of different records, sometimes tens of thousands and it's important to be able to keep track of all of those records. Flow diagrams do that very nicely. In the case of Prisma, this is the Prisma 2020 pre-print version of the flow diagram, and I'm going to describe the process of how the Prisma 2020 R package and Shiny packages, Shiny apps, work. So the main function is Prisma flow diagram, and it's based on graph biz and uses diagram R. In this code in this language, I use the term nodes for boxes and edges for arrows, and the package automatically connects the boxes with arrows. But for this case in Prisma 2020 I use explicit node locations, so I explicitly say where the boxes should go and another important factor is that the previous and other arms of the Prisma 2020 flow diagram are optional. So if you haven't integrated studies from previous reviews or if you haven't searched other sources of information, those arms can be dropped from the resultant flow diagram. It also has a what's called responsive location of the exclusion reasons node. So the box that has exclusion reasons in where each line is a different exclusion reason can be varying sizes depending on the number of exclusion reasons. And Prisma flow diagram, the function responsively moves the boxes around depending on how big that box is to a certain degree. So this is the flow diagram again, and on the left you can see the optional previous studies arm and on the right the optional other studies arm. So within the R package data can be read in from a template CSV file. Here's a screenshot of the CSV file and the data that can be changed by the user and that is read as different is in this red box. So the other columns are there for the user to understand which box the information relates to. There's also a template schematic that shows for the box column which box is which to make it easier. But users can change the box text. They can change the tool tips which are the text that appears when the mouse hovers over a tool when interactive when the interactive version is used. And the URL is the link that is used when the user clicks on an HTML version if the user specifies interactivity. And then the final column in this case n which was k in the original Prisma 2020 flow diagram is the number of studies or number of records at any step. You'll notice for two of the boxes there are multiple reasons. So where there are multiple reasons each reason is separated from its number by a comma and from the next number by a semicolon and the template shows you how to do that. It won't function if you don't enter the data in that format. So some of the challenges with this are that the vertical aligned text in these blue boxes is incompatible with graphics. So what we needed to do was to append that into the HTML using JavaScript using the packages HTML tools and HTML widgets and it adds text that is rotated through 90 degrees. And this is the function I won't go into detail but there's JavaScript in there. It embeds that JavaScript into the HTML after it's being produced. I also mentioned interactivity. There's a function SR flow interactive and it turns a static Prisma flow diagram into an interactive one. Why is interactivity important? Well it's helps to support full transparency in that it allows you to see the fate of each record. Normal flow diagrams show the number of total records going through each step but they don't show the fate of every single record. And full transparency allows you to see what happens to each record by examining a list of records at each step. It also facilitates data reuse and digital transparency and it makes particularly online communication very simple by explaining methods in as much detail as the user wants to see. So the user can click on a box and either see the methods or the data depending on what you want them to see. And the interactivity again uses JavaScript by appending href links into node elements and those links as I said before are specified in the data entry and tool tips are provided so that when the mouse goes over a bit more information pops up. So there's two levels of interactivity information provided in the tool tips and then information provided in whichever source you link to. There's also a shiny app as I mentioned. The beta testing version is up but it's likely to change over time. You can see that on the left you can specify which arm is included and down below you can enter the text as well for the values of N. At the moment this is all you can change. You can just change the N numbers in the shiny version. So some future developments are to add in the side panels. You'll notice there that the blue side panels aren't working because shiny doesn't support JavaScript. I also want to build interactivity in so that you can click on the nodes and be sent to URLs. We're also trying to build in flexible data entry so that in shiny you can either upload a CSV from the template or you can have manual data entry and to integrate the flow diagram with review management tools so this data automatically is populated from whichever review management tool you use. Thanks very much.