You are viewing an old version of this page. View the current version.

Compare with Current View Page History

Version 1 Next »

Body:


Attributes:

OIDoid

Namename

DataType(text | integer | float | date | time | datetime | string | boolean | double | hexBinary | base64Binary | hexFloat | base64Float | partialDate | partialTime | partialDatetime | durationDatetime | intervalDatetime | incompleteDatetime | incompleteDate | incompleteTime | URI )

LengthpositiveInteger(optional)
SignificantDigitsnonNegativeInteger(optional)
SASFieldNamesasName(optional)
SDSVarNamesasName(optional)
Origintext(optional)
Commenttext(optional)


Contained in:


An ItemDef describes a type of item that can occur within a study. Item properties include name, datatype, measurement units, range or codelist restrictions, and several other properties.

The DataType attribute specifies how the corresponding value elements are to be interpreted for comparison and storage. The Length attribute is required when DataType is text or string, optional when DataType is integer or float, and should not be given for the other datatypes. The SignificantDigits attribute is optional when DataType is float, and should not be given for the other datatypes. When DataType is float both Length and SignificantDigits must be given or both be absent.

Note: In version 1.3.0 the usage of Length and SignificantDigits was ambiguous. The above text clarifies the intention of these two attributes.

If DataType=integer, Length=N is a requirement that the receiving system be able to process and store all whole number values of magnitude less than 10N. Larger values may be rejected.

If DataType=float, Length=N and SignificantDigits=S is a requirement that the receiving system be able to process and store all numeric values of magnitude less than 10N-S that are multiples of 10-S. Larger values may be rejected. Intermediate values may be rounded to the nearest multiple of 10-S.

If DataType=text, Length=N is a requirement that the receiving system be able to process and store all text string values of length less than or equal to N. All characters are allowed in text string values. Data of type Text should be transmitted in an ItemDataString element.

Note: Length and SignificantDigits are statements about an item's data values, not the number of characters used to represent these values in value elements. For example, the character "<" might be represented as "&lt;".

Note: Data characters that are not included in the encoding character set for a particular ODM file must be represented using XML entities or character references. For example, Æ could be represented as "&#198;".

The SDSVarName, Origin, and Comment attributes carry submission information as described in the latest version of CDISC SDTM .



Note: In the ODM model, all internal keys are assumed to be unchangeable. This was done to make the audit trail issues work: if the SubjectKey in the model were the actual external subject identifier (or randomization ID) of a patient, and that value is sent incorrectly in one ODM file, there would be no way to correct the mistake in a followup file. In doing this, we intend that the external subject keys (and other externally visible key variables) should be defined as Items in the metadata. Thus they can be modified through normal modify/audit mechanism. While this solves the problem of supporting modification of study keys, it leaves the user without a way to identify which ItemDefs have special meaning or what the meaning is. The most obvious place where this is a problem is in matching up patients when loading data from an external source. If you can't find the patient ID how do you do the matching?

The answer is to use the SDSVarName attribute of ItemDef. SDSVarName is an optional attribute which can be used to tag the Item with a business meaning. Rather than try to enumerate all possible meanings in the ODM model, the ODM working group thought it best to rely on the set of variable names defined in the CDISC SDTM, since this list covers the core variables used in managing clinical data. Software that is processing an ODM-compliant XML instance can therefore use specific values of the SDSVarName attribute to identify standard, frequently used variables. The use of this attribute is restricted to variables defined in the SDTM model. In tagging a variable, you are identifying it as matching the SDTM definition for that variable. A partial list of commonly used values includes:

  • STUDYID (Study Identifier Unique within a Submission)
  • USUBJID (Study Identifier Unique within a Submission),
  • SUBJID (Subject Identifier Unique within a Study),
  • SITEID (Unique Identifier for a study Site)
  • SEX (Sex or Gender, coded value),
  • VISITNUM (Clinical encounter Number)
  • VISIT (Protocol-defined description of clinical encounter),
  • VISITDY (Planned study day of VISIT)

See the SDTM Specification and Implementation Guide for more information about SDTM variables.


The Question element contains the text shown to a human user when prompted to provide data for this Item. The ExternalQuestion element does the same but refers to an externally defined question. If both are present, they should be consistent.

The MeasurementUnitRefs list the acceptable measurement units for this type of item. Only numeric items can have measurement units. If only one MeasurementUnitRef is present, all items of this type carry this measurement unit by default, i.e. if a MeasurementUnitRef is defined on the ItemDef, and no MeasurementUnitRef is given on the corresponding ItemData, the value given by the ItemData has the units given by the ItemDef-MeasurementUnitRef.
If no MeasurementUnitRef is present on the definition of a numeric Item, the Item's value is scalar (i.e., a pure number).

The RangeChecks constrain the acceptable values for items of this type.

The CodeListRef (if present) constrains the acceptable values for items of this type to be members of the referenced codelist.

Note: Items do not repeat within an item group.


  • No labels