How to represent 'use as directed' [was RE: MEDICATION SECTION EVIL_TS VOCABUALRY set url]
Tom de Jong
Hello Kumara,
Responding to your question below (with a new subject heading that fits the topic)…
Actually the answer is much the same as the one you got from Lloyd about specifying ‘sufficient amount of cream’ (or any non-quantified dose). When “use as directed” is all you have, this could apply to both the frequency in <effectiveTime> and the dose in <doseQuantity>. Instead of trying to fiddle this into one or both of these elements, it’s better to just capture in the <text> element of the <substanceAdministration> as a whole.
As Lloyd pointed out, CDA R3 (i.e. data types R2) will allow you to tie a textual form to any data element, although you argue that “use as directed” would still fit best in <substanceAdministration><text>, since it refers to the administration as a whole and not just the frequency or the dose.
I think the pool of C-CDA examples should certainly contain this use case, since it’s quite common in many jurisdictions (although not best practice).
Best,
Tom
Van: Kumara Prathipati [mailto:kuma...@yahoo.com]
Verzonden: vrijdag 12 juli 2013 23:00
Aan: Kumara Prathipati; Tom de Jong; 'Benjamin Flessner'; 'Boone, Keith W (GE Healthcare)'; 'Brian Weiss'; ccda_s...@googlegroups.com
CC: 'Structured Documents WG'; phar...@lists.hl7.org
Onderwerp: Re: MEDICATION SECTION EVIL_TS VOCABUALRY set url
Sometimes we say
XYZ tablet 10 mg PO , use as directed No 30 quantity
"use as directed" is used in lieu of frequency
Not a common usage but will happen.
This cann't go into 2nd <effectivetime> element in medication section which requires only EIVL_TS or PIVL_TS.
Can we add this in <entryrelationship> under criterion using <originaltext>?
Or is there some other way to represent this
Kumara
From: Kumara Prathipati <kuma...@yahoo.com>
To: Tom de Jong <t...@nova-pro.nl>; 'Benjamin Flessner' <Benj...@epic.com>; "'Boone, Keith W (GE Healthcare)'" <keith...@ge.com>; 'Brian Weiss' <brian...@cdapro.com>; "ccda_s...@googlegroups.com" <ccda_s...@googlegroups.com>
Cc: 'Structured Documents WG' <stru...@lists.hl7.org>; "phar...@lists.hl7.org" <phar...@lists.hl7.org>
Sent: Friday, July 12, 2013 11:13 AM
Subject: Re: MEDICATION SECTION EVIL_TS VOCABUALRY set url
Tom,
Thanks for the help with EIVL_TS data type vocabulary value set.
Are these 2 below examples accurate or did I make any mistakes?
Kumara
//lunch time//
<effectiveTime xsi:type="EIVL_TS" operator="A">
<event code="CD"/>
</effectiveTime>
//1 HR BEFORE LUNCH//
<effectiveTime xsi:type="EIVL_TS" operator="A">
<event code="CD"/>
<offset>
<low value='-1' unit='h'/>
</offset>
</effectiveTime>
From: Tom de Jong <t...@nova-pro.nl>
To: 'Kumara Prathipati' <kuma...@yahoo.com>; 'Benjamin Flessner' <Benj...@epic.com>; "'Boone, Keith W (GE Healthcare)'" <keith...@ge.com>; 'Brian Weiss' <brian...@cdapro.com>; ccda_s...@googlegroups.com
Cc: 'Structured Documents WG' <stru...@lists.hl7.org>; phar...@lists.hl7.org
Sent: Friday, July 12, 2013 2:41 AM
Subject: RE: MEDICATION SECTION EVIL_TS VOCABUALRY set url
Hello Kumara,
Copying Pharmacy WG, because there should be consensus across the Pharmacy domain about this value set.
You will find the HL7 code system TimingEvent (OID 2.16.840.1.113883.5.139), as part of the most recent HL7v3 ballot package, here:
This contains the following codes:
TimingEvent [2.16.840.1.113883.5.139] | |||
Lvl- Typ | Concept Code | Print Name | Definition, Properties, Relationships |
AC | AC | before meal (from lat. ante cibus) | |
ACD | ACD | before lunch (from lat. ante cibus diurnus) | |
ACM | ACM | before breakfast (from lat. ante cibus matutinus) | |
ACV | ACV | before dinner (from lat. ante cibus vespertinus) | |
C | C | meal (from lat. ante cibus) | |
. CD | CD | lunch (from lat. cibus diurnus) | |
. CM | CM | breakfast (from lat. cibus matutinus) | |
. CV | CV | dinner (from lat. cibus vespertinus) | |
HS | HS | Prior to beginning a regular period of extended sleep (this would exclude naps). Note that this might occur at different times of day depending on a person's regular sleep schedule. | |
IC | IC | between meals (from lat. inter cibus) | |
ICD | ICD | between lunch and dinner | |
ICM | ICM | between breakfast and lunch | |
ICV | ICV | between dinner and the hour of sleep | |
PC | PC | after meal (from lat. post cibus) | |
PCD | PCD | after lunch (from lat. post cibus diurnus) | |
PCM | PCM | after breakfast (from lat. post cibus matutinus) | |
PCV | PCV | after dinner (from lat. post cibus vespertinus) | |
WAKE | WAKE | Upon waking up from a regular period of sleep, in order to start regular activities (this would exclude waking up from a nap or temporarily waking up during a period of sleep) | |
I’ve done some clean-up in the descriptions to condense the size of the table. It is a known fact that this code system is far from perfect, since it mixes actual events of daily living with intervals relative to those events (e.g. ‘lunch’ and ‘between lunch and dinner’) This has been fixed in data types R2, but that’s out of scope for CDA R2 implementations. It would be good to agree about which value set to apply (full code system or a subset), to identify use cases that are not covered by the code system, and to specify implementation guidelines for how to use the codes (possibly combined with a quantified offset, e.g. ‘30 minutes before lunch’).
These guidelines are needed in a wider context than just MU2, since it’s known that many ongoing and planned implementations struggle with this
Hope this helps to start off the creation of these implementations guidelines,
Best,
Tom.
Van: owner-s...@lists.hl7.org [mailto:owner-s...@lists.hl7.org] Namens Kumara Prathipati
Verzonden: vrijdag 12 juli 2013 9:15
Aan: Benjamin Flessner; Boone, Keith W (GE Healthcare); Brian Weiss; ccda_s...@googlegroups.com
CC: Structured Documents WG
Onderwerp: MEDICATION SECTION EVIL_TS VOCABUALRY set url
Can some one point me to the full vocabulary value set for EVIL_TS data type in medication entry?
I need to use medication administration timings like
1 hr before meal
with meal
1 hr after meal
at bed time
etc..
I was able to locate FHIR value set (http://hl7.org/implement/standards/fhir/v3/TimingEvent/) for this purpose but not sure if it is official to use in CCDA for MU2?
Once again, it is useful if CCDA IG provides url to find required value sets used in each section/entry
Kumara
From: Benjamin Flessner <Benj...@epic.com>
To: "Boone, Keith W (GE Healthcare)" <keith...@ge.com>; Brian Weiss <brian...@cdapro.com>; "ccda_s...@googlegroups.com" <ccda_s...@googlegroups.com>
Cc: Structured Documents WG <stru...@lists.hl7.org>
Sent: Thursday, July 11, 2013 8:37 AM
Subject: {C-CDA Samples} RE: No Information Allergy Section
Is Allergy Severity a SHALL 1..1 (per CONF 16341 in Dec. Errata package) or a SHOULD 0..1 (per CONF 9961)?
(This is errata comment 296 which was resolved as ‘will fix’ on the 7/11 call, but it doesn’t state which direction it will be fixed in).
From: owner-s...@lists.hl7.org [mailto:owner-s...@lists.hl7.org] On Behalf Of Boone, Keith W (GE Healthcare)
Sent: Monday, July 01, 2013 8:29 AM
To: Brian Weiss; ccda_s...@googlegroups.com
Cc: Structured Documents WG
Subject: RE: No Information Allergy Section
I completely agree that "SHOULD" shall be ignored in these cases. ;-)
You have a good reason not too, which is exactly the kind of exception that SHOULD is designed to allow for.
Keith
From: owner-s...@lists.hl7.org [mailto:owner-s...@lists.hl7.org] On Behalf Of Brian Weiss
Sent: Monday, July 01, 2013 6:41 AM
To: ccda_s...@googlegroups.com
Cc: Structured Documents WG
Subject: Re: No Information Allergy Section
Even though it results in warning messages, I decided it was better to leave off the sub-elements of the Allergies section that are marked SHOULD (the Reaction Observation template, the Severity Observation template, and the participant element).
In general, SHOULD elements "should" be included (one HL7 leader has said something to the effect that a "SHOULD" is often a "SHALL" that was debated and as a compromise became a "SHOULD" - but ideally would have been a "SHALL"). And we all prefer "clean" validator results with no warning messages...
However, there are times where we have to take advantage of the "SHOULD" not being a "SHALL". I think the "No Information" case is one of those times. Adding more data fields about which we have no information (and then struggling to figure out the best way to say "i have no idea" in a C-CDA compliant way) doesn't really make sense. So, we'll accept the warning messages, and keep the sample smaller...
Attached is the latest version (of just the Allergy section). I think we are almost there, pending some possible comments about how to handle the effectiveTime elements (there are now only two of them). Now I will go back and refresh the whole "baseline sample" using the same principles here (leaving off the "SHOULD" elements, etc.).
However, I will also create a version that includes all the SHOULD elements in a way that passes the validator with no warnings. This is helpful purely for testing purposes as a "clean-validating baseline" that we can use to test other samples - even though I don't think it is the right way to actually say "no information" (the right way to do that, as noted above, is to leave off the SHOULD sub-elements and accept the warning messages).
Brian
On Friday, June 21, 2013 1:49:37 PM UTC+3, Brian Weiss wrote:
The detour related to handling Coded Concepts is still being travelled, but I wanted to return now to the original first sample-creation effort, which was a “Baseline C-CDA CCD” containing “no information”. While there is no clinical use case I am aware for creating an entire C-CDA CCD that says nothing, this exercise is believed useful for two reasons:
· The need to report individual section templates with “no information” is very prevalent. A sample scenario would be an EMR generating a CCD off of a patient record with only partial information.
· For testing purposes, it’s helpful to have a generic “baseline” CDA document that validates OK and into which test snippets can be “dropped in”.
Attached is one page containing one section (the Allergy Section) of the “Baseline No Information C-CDA CCD”. It validates on TTT with no errors and no warnings.
Happy if anyone wants to actually through it, but not expecting anyone to do that. Instead, I will call out the key questionable areas (called out with embedded comments in the document) that I think warrant feedback here:
1) Use of nullFlavor="NI" on an observation or act element. The C-CDA IG does this for observation in Figure 213 on page 450. It does it for substanceAdministration in Figure 12 on page 40. So, need some guidance on when that is and isn’t appropriate.
2) What to use in a “no information” effectiveTime range. There is an effectiveTime element in the Allergy Problem Act (the concern), the Allergy – Intolerance Observation, and the Reaction Observation. Seems pretty clear that the low sub-element of effectiveTime should get a nullFlavor="NI". My intuition was to do the same for the high sub-element (in general, I prefer to put NI anywhere possible, to just indicate “have nothing to say here – just filling out this element because I am forced to for conformance” – sort of like marking “NA” all over a form that I know nothing about from a content perspective… but Lisa made a good case for putting in the date of the CCD creation for high, as it is better to cap the date for the concern and the allergy observation, as otherwise there might be an incorrect inference about some concern or observation being “ongoing”. Challenging to deal with “semantics of no information”… I’d welcome any guidance.
3) With minimal narrative text to work with (“No Information About Allergies”) it was challenging to fill in all the SHAL/SHOULD text and originalText elements. Basically kept pointing to that same line.
4) As per my earlier mail, there are two places (one in the Severity Observation template – which itself appears twice, once in reference to the Allergy Observation and once in reference to the Reaction Observation – and one in the Reaction Observation) where the value element is used for a coded concept and the language of the conformance statements (7335 and 7356) seems to want to disallow nullFlavor by placing a SHALL on the @code attribute. As per my earlier mail, I nonetheless used nullFlavor="NI" with the (questionable) rationale that the “…where the @code SHALL…” language provides some wiggle room relative to “…SHALL have exactly one @code which SHALL…” language. Validator is happy with it and the alternative (picking something from the Value Set when we have “no information”) doesn’t seem good.
I think those four were the main challenges. Could be I needed to call out something else questionable here that I wasn’t sensitive enough to realize/understand was problematic…
Brian
***********************************************************************************
Manage your subscriptions | View the archives | Unsubscribe | Terms of use
***********************************************************************************
Manage your subscriptions | View the archives | Unsubscribe | Terms of use
--
You received this message because you are subscribed to the Google Groups "C-CDA Samples" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ccda_samples...@googlegroups.com.
To post to this group, send email to ccda_s...@googlegroups.com.
Visit this group at http://groups.google.com/group/ccda_samples.
For more options, visit https://groups.google.com/groups/opt_out.
***********************************************************************************
Manage your subscriptions | View the archives | Unsubscribe | Terms of use
Geen virus gevonden in dit bericht.
Gecontroleerd door AVG - www.avg.com
Versie: 2013.0.2904 / Virusdatabase: 3204/6480 - datum van uitgifte: 07/10/13
***********************************************************************************
Manage your subscriptions | View the archives | Unsubscribe | Terms of use
***********************************************************************************
Manage your subscriptions | View the archives | Unsubscribe | Terms of use
Geen virus gevonden in dit bericht.
Gecontroleerd door AVG - www.avg.com
Versie: 2013.0.2904 / Virusdatabase: 3204/6480 - datum van uitgifte: 07/10/13
Kumara Prathipati
Appreciate your quick help on "use as directed".
I am still hoping some one can help on maxdosequantity
<maxDoseQuantity nullFlavor="UNK">
<numerator nullFlavor="UNK" />
<denominator nullFlavor="UNK" />
</maxDoseQuantity>
I searched in Google to find some good example but none available.
All I see is the 'same" null flavor example in many web sites.
Morphine 25 mg tablet, take 1 tablet Q4hrs , not to exceed 3 tablets per day
max dose quantity = 3 per day
Need example
Prescriptions like this are written at least 1 million times per year.
It is surprising and unfortunate to find example is nor readily available (may be I am searching in the wrong place).
I will appreciate if any one can share an example for this scenario.
Kumara
Sent: Monday, July 15, 2013 8:30 AM
Subject: {C-CDA Samples} How to represent 'use as directed' [was RE: MEDICATION SECTION EVIL_TS VOCABUALRY set url]
Tom de Jong
Your renewed question crossed my reply to the original one. I propose to use the other thread for any follow-up questions regarding <maxDoseQuantity>. Let’s try to have one topic per thread, with a subject that stays fixed as long as the topic of debate stays fixed.