Pair Each Adverse Event with the Doses of the Medications Given for It

Goal: give one row for each adverse event and each medication of the same subject given for its term: their sequence numbers (AESEQ, CMSEQ), the medication's name (CMTRT) and its dose as a whole number (CMDOSE); and a run report that counts the doses left empty.

Input: one spec.yaml over adverse events, each with its sequence number and reported term, and concomitant medications, each with its sequence number, name, indication and dose as collected text; one dose is written in words.

Variables:

• CMTRT: the name of a medication of the same subject whose indication is the event's term.
• CMDOSE: the medication's dose as a whole number; empty when the collected dose is not a number.

The run report (aecm-report.csv) has one row per item the run handled, each naming the dataset (ARTIFACT), the kind of item (KIND), where the spec states it (SPEC_PATH) and the item's name (NAME): the run and its outcome (OUTCOME) with the rows it built (COUNT); the two collected datasets, each with the records read and its file facts (DETAILS); the pairs that built the rows, out of all pairs (TOTAL); then the doses left empty because they are not numbers, counted once for each pair that holds one. That count gives the rule its empty dose stands in for (RULE) and no output column (TARGET), since each pair's dose is read before any row is built. One row per output column follows, with its missing values and longest text. Nothing is checked, so no row has a label (ID), a SEVERITY, a CONDITION or KEYS.

Note: a subject who reports the same term twice pairs each of its medications with both events, so a dose written in words is left empty in each of those rows, and the run report counts it once for each: twice here.