Question

I got a requirement where the user enters test data
I have fields:

example

 Blood   Pulse 
 ----    -----
 12      13

There is requirement to add another another field we don’t know what the test will be for.
I was thinking about have a field called Misc1Label, Misc1Value

Say if Misc1Label is Oral and Misc1Value is 88

Later I can use Pivot if we need to capture value of Misc1Label and use Pivot so that I can have something like

 Blood  Pulse  Oral
 -----  ----   ----
   12    13     88

I was thinking if there is any other best practice of handling this.

Thanks in advance

Answer 1

Do you really need to keep Blood and Pulse as separate fields? You could consider Blood and Pulse as different Attributes as well. Sounds like something like this would work and make it easy to query against without having to use a PIVOT command.

PersonAttribute
    PersonId int (I'm only presuming here)
    AttributeId int

Attribute
    AttributeId int
    AttributeTypeId int
    AttibuteValue varchar(100)

AttributeType
    AttributeTypeId int
    AttributeType varchar(100)

You could then store Blood, Pulse, Weight, etc. in your AttributeType table and use the PersonAttribute table as a 1-N approach with your main table. Just a thought.

SELECT Distinct PA.PersonId
FROM PersonAttribute PA
   INNER JOIN Attribute A ON PA.AttributeId = A.AttributeId 
   INNER JOIN AttributeType AT ON A.AttributeTypeId = AT.AttributeTypeId 
WHERE AT.AttributeType = 'Blood'

Of course the same model could be applied leaving Blood and Pulse in your main table if desired.

Good luck.