SQL delete unique row in table

Question

I've got the following table:

PatientID | DiagID

...where PatientID is a unique identifier and DiagID is a diagnosis ID, which is not unique. How do I delete the patient from table who's got unique DiagID (the only in the table, who's got this specific DiagID)?

Answer 1

Presumably you want to delete any patients that only have a single instance of a diagnosis ID. In which case the SQL would be something like:

DELETE Patient
WHERE DiagID in
(SELECT DiagID FROM Patient GROUP BY DiagID HAVING COUNT(*) = 1)

Answer 2

A delete where clause doesn't have to use the primary key, example:

Delete From Table
Where DiagID = 5

If you mean you want to delete all uniques, then something like:

Delete From Table
Where DiagID In (Select DiagID
                 From Table
                 Group By DiagID
                 HAVING count(*) = 1)

Answer 3

You say there's only one patient with this specific diagnosis id?

delete
from YourTable
where DiagID = @diagId

Answer 4

DELETE
    D
FROM
    DiagTable D
    JOIN
    (SELECT DiagID FROM DiagTable D2 GROUP BY DiagID HAVING COUNT(*) = 1) foo ON D.DiagID = foo.DiagID

The subquery find unique (ie used once) DiagIDs. This is used to JOIN bak to delete rows for these unique DiagIDs

Answer 5

Are you trying to delete patients that only have one DiagID?

DELETE FROM table
    WHERE PatientID IN 
    ( SELECT DiagID FROM table
    GROUP BY DiagID 
    HAVING COUNT(*) = 1)

Answer 6

If you know it is the only one then there is nothing wrong with this

DELETE FROM Patient 
WHERE DiagID = @inDiagID

If you want to be safe, something like this might be wise:

IF NOT EXISTS(SELECT DiagID FROM Patient WHERE DiagID = @inDiagID GROUP BY DiagID HAVING COUNT(DiagID) = 1)
BEGIN
      DELETE FROM Patient 
      WHERE DiagID = @inDiagID
END