We require a method that describes the universe or inclusion/exclusion criteria from different data sources, but especially the development and or validation data.
There are several concepts of the universe that are potentially important to describe:
- Study base. This is the broadest intended use of the algorithm. Potential examples include population-based, community-dwelling, home-care recipients, hospital patients.
- Development or validation population(s). These are the population(s) where the algorithm has been developed and/or validated. Information in this file describes the inclusion and exclusion criteria, in machine-actionable format for generating the population. For example, exclude people under age 20 or over 100 years, people who are pregnant, people whose sex is unknown.
- Implementation population(s). This may be the same as (2), (but flagging in case it is not).
Related, we require the instructions how to transform the starting point of data or exposures to generate the starting variables in variables.csv. Examples of these instructions include recodeFlow files or Lime survey file. Limesurvey transforms data from web-based questions to variables described in variables.csv.
We require a method that describes the universe or inclusion/exclusion criteria from different data sources, but especially the development and or validation data.
There are several concepts of the universe that are potentially important to describe:
Related, we require the instructions how to transform the starting point of data or exposures to generate the starting variables in variables.csv. Examples of these instructions include recodeFlow files or Lime survey file. Limesurvey transforms data from web-based questions to variables described in variables.csv.