gnumed-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnumed-devel] type of search pattern for demographics


From: J Busser
Subject: Re: [Gnumed-devel] type of search pattern for demographics
Date: Sun, 22 Aug 2004 11:01:32 -0700

At 11:02 PM +0200 8/21/04, Karsten Hilbert wrote:
 >> O"Malley
We don't currently support that to the best of my knowledge.
Can you give a few examples and a bit of explanation ?

 > 2) spaces
 > de Groot
It is rather tricky to get the semantics exactly right.
What *are* those name parts ? Should we ignore them ? Are
there others in the same category (eg. van, von, of, ...) ? Do
we need a separate backend field for that name part ?

As stated by someone earlier, it is true in the example like deGroot vs. de Groot and maybe even da Groot (Spanish form???) one could instring search just on Groot.

For the case of apostrophe, I expect the Irish O' and French d' may be commonest. It is hard to know whether office staff and doctors would reliably enter the apostrophe or whether they might (though I doubt) use a quotation (") or, for those with much font knowledge and keyboard savvy, a curly apostrophe.

For the issue of capitalization and spaces we have
de (French, Dutch)
la (French)
le (French)
von (German)
van (Dutch)
Al and Al- (Arabic)
Au and Au- (Oriental e.g. Au-Yeung)

There are also combination van de (van De) and van der (van Der) as combining forms.

Then of course there are people with two-word last names, which may (or ambiguously, may not) be hyphenated.

You have already brought up the umlaut and then of course there are the french accents acute, grave and circumflex (sorry Dominique, I forget how to spell the french forms!).

This is further complicated because in some forms of identification and in some registries, there are can be limitations or errors on whether such things as punctuation/apostrophes, hyphens, accents or even correct spelling is carried through (e.g. in the printing of plastic cards). You may see a patient for the first time in the emergency room and take to the office information derived from the plastic cards used by the hospital to emboss paper records or may carry forward to your office a hospital error. Your patient at your next contact may provide you with their correct spelling. They may be unaware of or may have given up trying to get an insurer or government agency to fix incorrect spelling and may at times have to present an incorrect name or date of birth because it may be the one needed for eligibility and payment purposes.

And then of course we may have mis-spellings eg character reversal upon keyboard input in our own practices.

Sounds like Tim has been all through this.

There may be others but maybe it is not necessary to catalog every possible occurrence?

We might be most of the way there, for Western names anyway, for the search to:
- disregard case (iLike?)
- disregard accents (plain letters would need to be substituted for accented ones)
- disregard spaces (drop them; would cover  two and three part surnames)
- disregard punctuation (e.g. apostrophes and hyphens)

We could in addition optionally offer soundex searching per the url given by Syan which specifies dropping to vowels in order to extend the matches, this would cover the van/von and also the Mc/Mac challenge.

Tim's method sounds like it would extend to include letter reversals?

One question that I have is whether it is efficient to have all of this factored into a search function, or whether this will be slow and if it would be faster, at the time of creating or editing a patient name, for Gnumed to store a converted form of the names for search purposes. Surely it is faster to search a key or index file than it is to have to locate on the physical hard drive every name in order to evaluate? If the user were permitted to see how the name is proposed to be converted and stored for search purposes, people might wish to edit it to capture the misspelling used by official agencies or insurers in dealing with the patients (although if this essentially becomes an "alternate name" it may be better served through an "alias" function wherein we permit each patient to have multiple names as can happen with marriage/divorce as well as in other circumstances.




reply via email to

[Prev in Thread] Current Thread [Next in Thread]