Re: [Gnumed-devel] More lab test result considerations: groupings

[James Busser]

On 26-Feb-08, at 3:14 AM, Karsten Hilbert wrote:

> On Sun, Feb 10, 2008 at 10:53:09AM -0800, James Busser wrote:
>
> > The lab results in BC CA include the HL7 message segment OBR 004
> > "Universal Service ID" of the form "Test code^Test name" for example
> > "LYTE^Electrolytes".
> >
> > The thing to note is that a *single test code* can represent  
> > *multiple
> > results* that will be received.
>
> I think we are misinterpreting the LOINC code.

Not me :-)

> The LOINCs
> are simply other labels for "things". In a sense they are
> like Esperanto. What we call Natrium and you call Sodium is
> both LOINC-2951-2 such that we can know what each other
> means w/o having to know that Sodium == Natrium in an
> English to German sense.

LOINC entities are characterized by:

"LOINC_NUM"	"COMPONENT"	"PROPERTY"	"TIME_ASPCT"	"SYSTEM"	"SCALE_TYP"	"ME 
THOD_TYP"	"RELAT_NMS"	"CLASS"	"SOURCE"	"DT_LAST_CH"	"CHNG_TYPE"	"COMMENT 
S"	"ANSWERLIST"	"STATUS"	"MAP_TO"	"SCOPE"	"NORM_RANGE"	"IPCC_UNITS"	"REF 
ERENCE"	"EXACT_CMP_SY"	"MOLAR_MASS"	"CLASSTYPE"	"FORMULA"	"SPECIES"	"EXM 
PL_ANSWERS"	"ACSSYM"	"BASE_NAME"	"FINAL"	"NAACCR_ID"	"CODE_TABLE"	"SetRo 
ot"	"PanelElements"	"SURVEY_QUEST_TEXT"	"SURVEY_QUEST_SRC"	"UnitsRequire 
d"	"SUBMITTED_UNITS"	"RelatedNames2"	"SHORTNAME"	"ORDER_OBS"	"CDISC_COMM 
ON_TESTS"	"HL7_FIELD_SUBFIELD_ID"	"EXTERNAL_COPYRIGHT_NOTICE"

In your example above the record is
"2951-2"	"Sodium"	"SCnc"	"Pt"	"Ser/ 
Plas"	"Qn"		"NA;"	"CHEM"	"AF"	"20041102"	"MIN"											 
1				"NATRIUM,"		"Y"			0				"Y"		"Na; Substance concentration; Level;  
Point in time; Random; SerPl; SerPlas; SerP; Serum; SR; Plasma; Pl;  
Plsm; Quantitative; QNT; Quant; Quan; Chemistry"	"Sodium SerPl- 
sCnc"	"BOTH"	"Y"

So there seems to already be some support in LOINC to recognize the  
same "thing", represented by any one LOINC_NUM, by more than one  
"word" e.g. component "Sodium" (component) aka "NATRIUM" (ACSSYM)

ACSSYM
        see http://www.nlm.nih.gov/research/umls/archive/2004AA/METAB2.html
	see http://www.dimdi.de/dynamic/en/ehealth/loinc/downloads/loinc/ 
release_notes.html
	LOINC chemical name synonyms, alternative names and chemical  
formulae from the Chemical Abstract Society.

... there is also some work on internationalizing LOINC:
        http://www.regenstrief.org/medinformatics/loinc/international

> The LOINC code for an LYTEs OBR means just that: this is a
> LYTEs OBR.
> It does NOT mean: a LYTEs OBR includes Sodium and Potassium.

I did not misunderstand, and would in fact point out that there *is  
no* LOINC for an LYTES OBR, only a test code as defined by that  
particular lab, and having (usually) no reliable relationship to any  
identically-named test code by some other lab. The OBR refers to the  
test *codes* IOW the test *panel* as arbitrarily defined by the  
laboratories without any standardized naming (though multiple  
Canadian labs are starting to standardize, at least informally.) The  
LOINC is only for each of the individual results (one or multiple)  
that come from any test code / panels.