On Mon, 24 Mar 2008 20:25:15 -0700 (PDT), --CELKO-- wrote:
>>> "have you been in the United Stated during the last 90 days"? and two check boxes conveniently marked yes and no. How would you model that? <<
>
>They don't care about your trips to AIDS infected Africa? Avian flu
>in China? Mad Cow Disease in the UK?
Hi Joe,
They don't care about trips to Africa, since AIDS only infects in
specific ways. I do have to fill in some Y/N checkboxes with questions
about my sexual activities (I recall questions about sex with males, for
money, or with drug addicts of the top of my head; there are probably
more), use of non-sterile needles, or reception of blood products.
I don't recall questions about China (it's been a while since my last
donation), but they do care about mad cow disease, because I have to
declare not having visited the UK during a specific time period (in this
case, it's an absolute time period, not relative to the moment of the
blood donation).
> I think I would go with three
>tables -- Donors for your general demographics, Donations for a
>history of donations and a Quarantine list with countries and dates
>that would exclude a donation at this time.
I don't know about blood centres in the US, but here in the Netherlands
the questionnaires for blood donors are designed based on health
inspection and legal requirements, not by some database guy.
The database guy gets to design the database. The input for that
database has to come from the questionnaire.
>I would first check the date of the last donation to be sure that you
>are not giving too much and too often. In the US you get recognition
>for donating over a gallon, etc. We want a history in the DB.
The Dutch blood collecting organisation maintains records of all my
donations. They don't even invite me or allow me to make an appointment
when insufficient time has passed since the last time. However, that is
unrelated to the question I asked you.
>Then I would ask "what countries have you visited since << insert
>date>>?" One of the options could include the Netherlands, altho the
>front end might present it as "None" or something similar to the
>donor. When the situation changes, I change the Quarantine list.
Again, the question list (that does indeed change from time to time as
new risc factors are discovered) is not designed by the DB guy. So I
repeat my question - how would your table design look, given that the
input comes mainly from check boxes ticked by the donor?
--
Hugo Kornelis, SQL Server MVP
My SQL Server blog:
http://sqlblog.com/blogs/hugo_kornelis