Select derived data element type
There are four types of derived data elements that can serve
as building blocks for creating phenotypes:
- Category data element
A category is a grouping of more than one data element. A patient will match a category data element if any of its component data elements is matched.
- Sequence data element
The sequence data element defines an ordered series of events or observations over time. Data elements may be specified to occur before or after each other, by a minimum or maximum number of days, hours, or minutes. Each data element can be specified to have a minimum or maximum duration. Additionally, a data element can be filtered on property values, if available.
- Frequency data element
Frequencies match at least n or the first n occurrences of a data element. When the data element is a value threshold (see below), an additional option is available for specifying whether or not the occurrences should be consecutive. The occurrences can have a minimum and maximum duration, and they can also be specified to happen with a certain time of each other. The data element can also be filtered on a property value, if available.
- Value threshold data element
Value threshold data elements let you classify observations (such as lab test results, vital sign measurements, etc.) by specifying numerical value thesholds. Lower and upper thresholds may be specified. Thresholds may be defined to apply only within the context of another data element, and that context can be specified to be in effect at least or at most a certain number of days, hours, or minutes before or after the observation. Thresholds for multiple data elements can be specified within a single value thresholds data element, and you can choose whether the data element will be matched if any or all of the thresholds apply.