Question

I have a table that reuses a foreign key and I’d like to join a different value for each case. The first table describes a drug dosing activity. It has ID’s that point to drugs, routes, vehicles, etc… The confusing thing is that it reuses the Unit ID which is a foreign key in a single description table.

GROUPS (Simplified)

GROUP_NO    DRUG      DRUG_AMOUNT     DRUG_UNIT     CHECK_UNIT
   1        568           5               7              5
   2        689           1               7              5
   2        568           5               7              5
   3        19            0.5            10             
   4        984           10             10              5

UNITS (Simplified)

UNIT_ID      UNIT_DESCR
   5         kg
   7         mg
   10        mL

I’d like to generate a query that returns a row for each drug dose for all groups. I can do everything but the units. I’d like to use a CASE statement to display the dosing units. The select statement would look something like this:

'DOSE UNITS' =
CASE
WHEN CHECK_UNIT IS NULL THEN DRUG_UNIT_DESCR
ELSE CONCAT(DRUG_UNIT_DESCR+'/'+CHECK_UNIT_DESCR)
END

I’m trying to get the results to look like such for this example:

RESULT

GROUP_NO      DRUG          DRUG_AMOUNT     'DOSE UNITS'
    1       HelpsAlot           5              mg/kg
    2       HelpsMore           1              mg/kg
    2       HelpsAlot           5              mg/kg
    3       DoesNothing         0.5            mL
    4       WhoKnows            10             mL/kg

Thanks for any help.

Answer 1

You need two joins, one for each key:

select g.*,
       'DOSE UNITS' = (CASE WHEN CHECK_UNIT IS NULL THEN du.UNIT_DESCR
                            ELSE CONCAT(du.UNIT_DESCR+'/'+cu.UNIT_DESCR)
                       END) 
from groups g left outer join
     units cu
     on g.check_unit_id = cu.unit_id left outer join
     units du
     on g.drug_unit_id = du.unit_id