v9.3 October and v9.4 December 2020

Overview 

  • Conflict of interest dashboard
  • Voting and Delphi process facilitator
  • Make groups for guideline members
  • Make recommendation tags to filter recommendation view
  • More PDF customisation choices
  • More Recommendation strength / type and status options
  • Other things and Fixes

Conflict of interest dashboard


We have introduced a new dashboard to help you manage conflict of interests (CoI) in your guideline panel. 
It includes: 
  • Upload files per panel member or associate authors/ reviewers. It could be  standardised CoI forms, organisation specific forms or other documents relevant to a persons CoI
  • Write a public statement per panel member. It could be made available in a public guideline, or just be available within the panel. 
  • Write an internal comment per panel member that is only visible to the guideline admins. 
  • State potential conflicts per intervention mentioned in a PICO in the guideline. All interventions and comparators are listed, and admins can state level of potential conflicts related to these for each panel member. Admins can also write internal comments for these evaluations, and state whether these potential conflicts would exclude the panel member from voting or other participation in guideline development.
  • The potential conflicts for an intervention will show in the voting feature, so that if you start voting in a recommendation, you will se potential conflicts related to PICOs attached to that recommendation. And similar with voting related to PICOs.
Only admins can edit this dashboard. 

Upcoming features: 
  • Be able to bulk edit potential conflicts related to one intervention (e.g. Low for all panel members), or for one panel member (e.g. Low for all interventions for that member)
  • Show only one instance of re-occuring interventions. Sometimes an intervention is mentioned in multiple PICOs, and thus they appear multiple times in the list of interventions. We will collapse multiple instances so it only shows one time.

______________________________________________________________


______________________________________________________________



______________________________________________________________



______________________________________________________________


______________________________________________________________



_______________________________________________________

Voting and Delphi process facilitator


We have implemented a flexible voting feature. It can also be used to facilitate a Delphi process.

You can vote in relation to a recommendation, a PICO or things related to the guideline as a whole. We are adding more voting types by demand. 

Poll types:
  • Open ended question. Can be used in scoping, to get feedback on ideas, or feedback on phrasing in a recommendation or text block. Participants write their answer in free text.
  • Multiple choice question. Can be used where you give panel members options to choose from, e.g "Should this intervention be included": "Yes, No, Uncertain". Participants vote and can also give a comment in free text.
  • Recommendation Strength (per intervention). In recommendation you can run polls for strength and direction, related to each intervention in question. The poll administrator can include voting related to one or more intervention in the same poll.
  • Direction of effect (per outcome in a PICO). In PICOs you can run polls for direction of effect related to each outcome in question. The poll administrator can include voting related to one or more outcomes in the same poll. You can also use the comment feature to get opinions of magnitude of effect, or use a separate multiple choice poll for that. 
Upcoming poll types:
  • Judgements in evidence to decision factors. In recommendations you can run polls for judgements per evidence to decision factor. The poll administrator can include voting related to one or more factor in the same poll. This will be made to work with both  pairwise and multiple intervention comparisons.
  • Size of effect (per outcome in a PICO). In PICOs you can run polls for size of effect related to each outcome in question. The poll administrator can include voting related to one or more outcomes in the same poll. 
Feature details: 
  • Only admins can start and end polls. Polls can be duplicated and rerunned. 
  • Admins can add any person with permissions to the guideline, as participant in a poll. You can use the group feature (described below), to easier add everyone belonging to a specific panel or guideline group. 
  • If there are potential conflicts of interests registered on a participant, it will be shown to the admin. 
  • Admins do not see who voted what, but see who hasn't voted, and can send a nudge email to the remaining participants
  • Panel members could be given access to see the poll results
  • Admins can write a conclusion of the vote
  • When polls are started, all participants gets an email with instructions and links
  • Admins have access to the full history of polls for each recommendation, PICO and guideline, across guideline versions. 

Upcoming features:
  • Admin can filter results to see only participants without any potential conflicts
______________________________________________________________



______________________________________________________________




______________________________________________________________





______________________________________________________________




______________________________________________________________




______________________________________________________________




______________________________________________________________




______________________________________________________________




______________________________________________________________



______________________________________________________________



______________________________________________________________




______________________________________________________________


______________________________________________

Make groups for guideline members


We have created a group mechanism, where admins can assign guideline members into groups. This makes it easier for admins to get an overview of who in the guideline belong to which group. You can use the group filter when you look at the Conflict of interest dashboard or when you are creating a poll. 

Feature includes:
  • Admins make the group labels
  • Admin can assign any guideline member to one or more groups.
  • Permission level (admin, author, reviewer, viewer) are standard groups that are added to all places where there are group filters 
_______________________________________________________

Make recommendation tags to filter recommendation view


You can now create tags for your recommendations, that allows you to filter the view into only certain recommendations. Examples would be: Children, Pregnancy, Self-care, Community setting... and so on. 

Feature includes:
  • Admins make tags (we recommend the tags to be short)
  • Tags can then be added to any recommendation
  • Once you have tagged recommendations, a filter dropdown will appear on the top of the guideline. 
  • The filter dropdown contains all the used tags, and gives the number of recommendations having that tag
  • If a tag is chosen, only the recommendations with that tag show. Users will see an indicator that the view is filtered.


______________________________________________________________





______________________________________________________________



______________________________________________________________




_______________________________________________________ 

More PDF customisation choices


You can now
  • Choose to include the Evidence profiles in an Annex instead of in the text
  • Choose to not include the evidence profiles in the PDF
  • Choose to include footnotes for the Evidence profiles
_______________________________________________________

More Recommendation strength / type and status options

 
Added recommendation strength / type options
  • Only in research settings
  • Implications for research
Added recommendation status options
  • Possibly outdated
  • Updated evidence, no change in recommendation
  • Reviewed, no new evidence
_______________________________________________________

Other things:


  • Admins can turn off showing banner for explanatory text related to the recommendation strength
  • Widgets now show with practical information open, if chosen in the widget specification.
_______________________________________________________

Planned for the coming months

  • Better track changes and comments in text
  • Multiple comparison tables
  • Better RevMan file import and updating of data
  • Integration with Cochrane's RevMan online

Bigger features on roadmap

  • Implementation of a FHIR API
  • Improved Desktop and Mobile app
  • Diagnostic PICOs
  • Decision aids widgets
  • Better internal linking

Feedback and Knowledge Base