ucum-org
Request for: k[arb'U/L
Dear colleagues of the UCUM advisory board,
I have a proposal on behalf of the Dutch Society for Clinical Chemistry and Laboratory Medicine (NVKC) to endorse the use of prefixes with [arb’U]/volume unit expressions.
For many test tests in clinical chemistry it is not feasible to express the results as a mass concentration or substance concentration. Therefore we often use [arb’U]/volume as unit expression. If fact these units are a ratio of a reference measurement of some standard, mostly offered by the manufacturer of the assay, and the measurement of a patient sample. In UCUM rules of use, however, [arb’U] it states this unit can not be used with prefixes like k (of kilo for x 1000) etc., as it is not (always) certain that the atom unit [arb’U] is metric. Although in principle this can be the case, many assays do have a metric response over the valid range of the assay. Investigation of the response of an assay to established the workable range of a test is a vital part of the validation of every assay, as required by i.e. ISO15189 and many other guidelines. In the US many manufacturers and labs present the results in [arb’U]/mL. In Europe we often express the results in [arb’U]/L. The keep in line with the numeric result of such a test we want to use the prefix ‘k’ as [arb’U]/mL = k[arb’U]/L. Because it is/has to be veryfied by each laboratory that the response is metric we propose to endorse the use of prefixes in combination with arbitrary unit concentrations. So we do acknowledge the principle that [arb’U] is not by definition metric, in the practice of tests used in clinical chemistry labs this will be the case.
To repeat our request: we like the LOINC committee to take a stand to endorse the use of prefixes for [arb’U]/volume expressions.
NB In the current LOINC database there already are numerous test expressed as [arb’U]/volume.
Kind regards, Dr. Dirk Bakkeren, Clinical chemist Dutch Labcodeset board, a LOINC subset for The Netherlands
I think I understand your request. I will try to express it in different words: There is a measurement where it is "not feasible to express the results as a mass concentration or substance concentration", so you use [arb'U]. There is however a cross-institutional agreement what [arb'U] means in your context of a specific measurement, maybe because the test and/or the reference sample is provided by the same manufacturer. Maybe there even is a LOINC code assigned to this measurement. All this means, that the unit is not completely "arbitrary" in your case, but it is comparable within a certain weĺl-defined context - even across institutions. Now you're asking to allow prefixes for such a case.
If my understanding is correct, maybe we should consider to add a new code for such "not-quite-arbitrary" units, where it is implied (or required) that details of the measurement are provided in the context of the measurement, e. g. specific LOINC code.
In fact, this is pretty much the way how [IU] works.
My concern is just that we are undermining the purpose of UCUM: To force people around the world to agree on shared standards, and work towards such standards if they are not yet available.
Dear Chrisof,
You rephrased my request perfectly. In fact most of these assays already do have LOINC codes and [arb’U]/mL as example unit. We in ghe Netherlands prefer to express all concentrations as ../L. In order to keep consistency in the numeric results the use of k[arb’U]/L would be greatly instrumental. Anyway this is far better than labs simply using U/L or some other IU/L when this is not appropriate. Indeed these units are not completely ‘arbitrairy’, but at the same time no other unit expression is available in UCUM. Preferably the expression should be as short as possible (< 12 characters), as lab systems often have little positions reserved for unit expressions.
Kind regards, Dr. Dirk Bakkeren Clinical chemist NVKC Dutch Labcodeset board, a LOINC subset.
Verzonden vanuit Outlook voor iOShttps://aka.ms/o0ukef
Van: Christof Gessner @.> Verzonden: zaterdag, maart 29, 2025 11:16 AM Aan: ucum-org/ucum @.> CC: papabak @.>; Author @.> Onderwerp: Re: [ucum-org/ucum] Request for: k[arb'U/L (Issue #392)
[chgessner]chgessner left a comment (ucum-org/ucum#392)https://github.com/ucum-org/ucum/issues/392#issuecomment-2763275528
I think I understand your request. I will try to express it in different words: There is a measurement where it is "not feasible to express the results as a mass concentration or substance concentration", so you use [arb'U]. There is however a cross-institutional agreement what [arb'U] means in your context of a specific measurement, maybe because the test and/or the reference sample is provided by the same manufacturer. Maybe there even is a LOINC code assigned to this measurement. All this means, that the unit is not completely "arbitrary" in your case, but it is comparable within a certain weĺl-defined context - even across institutions. Now you're asking to allow prefixes for such a case.
If my understanding is correct, maybe we should consider to add a new code for such "not-quite-arbitrary" units, where it is implied (or required) that details of the measurement are provides in the context of the measurement, e. g. specific LOINC code.
In fact, this is pretty much the way how [IU] works.
My concern is just that we are undermining the purpose of UCUM: To force people around the world to agree on shared standards, and work towards such standards if they are not yet available.
— Reply to this email directly, view it on GitHubhttps://github.com/ucum-org/ucum/issues/392#issuecomment-2763275528, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AXKSE3YTWO7EIYOY5D3B6UD2WZQFTAVCNFSM6AAAAAB2AMQC7GVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDONRTGI3TKNJSHA. You are receiving this because you authored the thread.