Units of measure mismatches

[Jongmassey]

When calculating the total amount of usage of a given VMP there are two points at which mismatches of units may occur:

1. scmd.unit_of_measure_identifier <=>vpi.strnt_nmrtr_uom or vpi.strnt_dnmtr_uom
2. ddd.DDD_UOMCD <=> vpi.strnt_nmrtr_uom or vpi.strnt_dnmtr_uom

Some of these mismatches are simple order of magnitude differences (e.g. mg-g) and may be resolved via a lookup table such as https://docs.google.com/spreadsheets/d/1YiTznOoOH0Jzt9nwvVAf14JsSgPUGUxiSiWLoPYpL6c/edit#gid=0

Some of these mismatches are less trivially resolvable such as IU<->mg, mmol<->mg and require external lookup data such as molecular weights or IU quantities. An example found by @milanwiedemann is Sodium acetate 3mmol/50ml / Sodium oxybate 2mmol/50ml / Sodium propionate 1.5mmol/50ml enema (28897511000001100)

[Jongmassey]

The uom mapping google sheet has been imported into bigquery as jonm.uom_relations but is as yet unvalidated

[milanwiedemann]

@brianmackenna, this describes a problem I ran into with the narcolepsy study, for example for "sodium oxybate"

• Strength value numerator unit: 500mg (see here)
• ddd unit is: 7.5 g (see here)

our current conversion of ddd volume doesn't check for this so we're getting wrong answers when these measurement units dont match up