Page History

The IT example below shows 4 types of tobacco listed as tobacco ingredients (STRFID) for the Tobacco Product The following examples illustrate how the list of ingredients of a tobacco product would be represented and quantified using three datasets:

• Tobacco Ingredients (IT) is used for listing the tobacco types and their attributes within the tobacco product under study.
• Non-tobacco Ingredients (IN) is used for listing other ingredients and additives and their attributes as used in the tobacco product under study,
• Quantified Ingredients (IQ) is used for representing the quantity of each ingredient by component of the tobacco product in which they are included. These values represent the design/formulation of the product, not experimentally determined values (as the latter would be represented in the Product Testing (TP) dataset, not shown here).

All three datasets make use of the SPTOBID variable to represent the tobacco product under study, and the STRFID (Substance Reference Identifier) variable to represent the ingredient name. STRFID must be assigned by study sponsors to be unique with regard to the attributes that describe it. If, for instance, the same ingredient varies on one or more attribute (e.g., a different manufacturer supplied a compound that is used in one component of the product than the same ingredient used in another component), STRFID must be unique for each use to reflect this difference, even if they share the same IUPAC name and CAS number. 

IT example shows tobacco filler and the types of tobacco of which it is composed, each listed as tobacco ingredients in the variable IGDCMPID (ingredient or component identifier) for the tobacco product CIG01A (SPTOBID). The variable ITSPECIF shows the specification (/variety ) of the first three 3 tobaccos. ITCAT ITIGDPLX (Ingredient Complexity) is used to indicate that Reconstituted Tobacco is a complex, purchased ingredientthe tobacco filler and the reconstituted tobacco are complex ingredients (i.e., have 2 or more subcomponents). The details of which single tobacco ingredients compose these complex ingredients is made explicit in the IQ dataset example. ITCURMTH indicates the cure method. ITSPFL ITCIGIND (applicant-customized ingredient indicator) is used to indicate whether complex purchased ingredients were made to the applicants applicant's specification. RTOBMF indicates the reference tobacco master file (what is this? why does only one row have this info?)TPMF indicates the Tobacco Product Master File name/number, which should be provided in the data if the component manufacturer has made it available.
Note that there are 2 varieties of burley tobacco in this tobacco product; each is given a unique IGDCMPID value (Burley Tobacco and Burley Tobacco 2).

The IN dataset example shows the list of non-tobacco ingredients used in the tobacco product identified as CIG01A. Each unique ingredient is listed in STRFIDIGDCMPID. Where applicable, IUPAC name and CAS number are listed (IUPACNAM and CASNO variables, respectively). INCAT INIGDPLX is used to indicate whether an ingredient is a single chemical substance ingredient or a complex purchased ingredient. INSPFL indicates  INCIGIND  indicates whether complex purchased ingredients were made to the applicant's specification. Grade, purity, manufacturer name and ID , ingredient unique item number, and the reference tobacco master file Tobacco Product Master File number complete the dataset.

...

Finally, the IQ dataset example shows domain is used to represent the quantity (target, minimum, and maximum) of each ingredient listed identified in STRFID IGDCMPID that the product (SPTOBID) is designed to contain, with the units for each of these values represented in IQUNIT. The function of each ingredient is described in IQFUNCT. Note: Ingredients that appear in more than one 1 component of the tobacco product are quantified separately for each component. (we need examples of this to illustrate. I created one but not sure how valid it is) IQPARENT and IQLEVEL are used to identify the component to which the record applies and and at which level in the overall hierarchy the ingredient appearsIQPARENT identifies a component or complex ingredient composed of the subcomponents/ingredients shown in IDGCMPID. IQLEVEL is used to identify at what level in the overall hierarchy that subcomponent/ingredient appears.

The following implementation rules apply, as described in more detail in Section 2.8.8.6, SDTM Ingredient Quantities by Component (IQ), Assumptions:

• IQPARENT must correspond to a value of STRFID IGDCMPID in the same tabledataset. If  If parent is null, the ingredient is considered a top-level component and is assigned a an IQLEVEL value of "1" in IQLEVEL.
• Records describing subcomponents/ingredients that are sub-components of another ingredient listed compose a parent component listed in the dataset will have IQPARENT set value equal to equal the STRFID IGDCMPID of that ingredientparent component. IQLEVEL will be set to n +1, where n = the parent record's value of IQLEVEL
• Every value of IGDCMPID used in the IQ dataset must have a matching value of IGDCMPID:
  • In the IT dataset (when IQCAT=TOBACCO INGREDIENT), or
  • In the IN dataset (when IQCAT=NON-TOBACCO INGREDIENT)

Therefore, IQCAT is required to relate ingredient quantity records in IQ to the associated ingredient attributes described in either IT or IN.

The concept map below use of IQPARENT and IQLEVEL help make explicit the relationship between complex ingredients or components and the subcomponents/ingredients of which they are composed, and can assist when totaling the quantities of an ingredient that appears in multiple components of a tobacco product.
 
This concept map illustrates the representation of the hierarchical relationships between a complex ingredient component and its subcomponents/ingredients in the IQ domain structure. The hypothetical parent ingredient has two constituent components, A and B. Additionally, B consists of two constituent components, B1 and B2. Therefore, there are a total of 5 ingredient records to be represented in the IQ domain: the parent ingredient, its two constituent components A and B, and B's two constituent components, B1 and B2. 

Image Removed

The IQ dataset example illustrates these principles, using a cigarette tobacco product: