Using the Generate Common Drug Name Command
BizInt Smart Charts for Pharmaceuticals Version 2.4 introduces a new command to help work with combined charts: Generate Common Drug Name, found on the Tools menu.
How to use the Generate Common Drug Name command
The Generate Common Drug Name command is used to identify pipeline records which refer to the same compound. This is done by matching names assigned by the various publishers in "drug name", "synonym", "lab code", "trade name", etc. All records which can be grouped by matching at least one name (a transitive operation in which A matches B and B matches C implies A matches C as well), are assigned a common name in a new column.
By sorting on this new column, all records corresponding to a compound (as identified by the match) are brought together.
You should always generate common drug names after combining two charts, even if you have run the command on the two charts separately. The names assigned may change based on the new information added to the chart.
You can force two rows to be assigned the same name by adding the drug name for the first row as a synonym of the second row. You can then run the command again.
How the Generate Common Drug Name Command works
The Generate Common Drug Name command passes through the chart, reading the drug name and synonyms for each row. If one or more of the names matches a name that has already been read in the chart, the row is added to the group of rows corresponding to that name. (The process is somewhat more complicated than this, but for the purposes of understanding the operation, this is accurate).
The particular name assigned to a group of records is taken from the list of names which caused the match. In general this is the most common name in the group of records, but in case of a tie the name is usually the first matched name encountered by the command. Because the drug name is loaded before synonyms, priority is given to the drug name.
Helpful Hints
It is important to keep several things in mind when using this command.
First, the name assigned depends on the order of the records in the chart. If you combine a chart from PJB with a chart from Adis and generate common drug names, the assigned names may be different from those that would be assigned if you combine with the Adis records first.
Second, the name assigned to any group depends on the particular records which match in the chart being examined. That is to say, if you generate common drug names in a chart with data from two databases, and then add data from a third and generate again, you may find that records which were not assigned the same name in the two database case are matched in the three database case (as a result of linkage through a record in the third database).
Third, names are matched without regard to capitalization, without considering the nomenclature markings in IMS, and with much punctuation and space discarded. The assigned name, however, will be an exact copy of name as it appears in one of the databases. The exact text will not necessarily appear in all records. (In fact, as a result of the transitive nature of the matching, the common name may not appear in a record at all!)
You can run the Generate Common Drug Name command repeatedly on a chart. The command will start from scratch, reading the names and synonyms. If you sort between running the command, you may see different names assigned to some groups, but the grouping will not change. You can edit the synonyms (either to remove a value or add a value) and then re-generate common drug names. The common names may change if the new synonyms create new matches or break existing matches.
Finally, if you combine two charts, you should run the command again. This is particularly true if both charts already have common names, because as noted before the particular name assigned to a compound depends on the records which matched and the order in which they appear in a chart.
© 2002 BizInt Solutions